Beagle2 Mission Report
Mission Report
Editor M.R. Sims, Beagle 2 Mission Manager
Contributors O. Blake J. Bridges E. Chester J.F. Clemmet S. Hall M. Hannington S. Hurst G. Johnson S. Lewis M. Malin I. Morison D. Northey D. Pullan G. Rennie L. Richter D. Rothery B. Shaughnessy M.R. Sims A. Smith M. Townend L. Waugh
With inputs from C.T. Pillinger, ESA MEX and CoI Analysis Teams
Lander Operations Control Centre University of Leicester National Space Centre Leicester, UK
Mission Report
Editor M.R. Sims, Beagle 2 Mission Manager
Lander Operations Control Centre University of Leicester National Space Centre Leicester, UK
Copyright ©University of Leicester 2004 This document may be used for research, private study, criticism or review, but may not otherwise be reproduced, stored or transmitted in any form or any means without the prior permission in writing of the publishers. Enquiries concerning reproduction outside these terms should be sent to the publishers.
First Edition Publication Published by University of Leicester, Leicester LE1 7RH, UK ISBN: 1 898489 35 1
Front cover image: Flight configuration model of Beagle 2 Rear cover images top to bottom: Flight Model Gas Analysis Package Flight Model Position Adjustable Workbench installed in Flight Model lander Flight Model Position Adjustable Workbench
All images Courtesy Beagle 2. All rights reserved See www.beagle2.com
Beagle 2 was the UK's first mission to another planet. The project is a partnership between the Open University, the University of Leicester and EADS Astrium (UK). Other funding partners included the European Space Agency (ESA), the Office of Science and Technology of the Department of Trade and Industry, the Particle Physics and Astronomy Research Council (PPARC), the Wellcome Trust, the National Space Centre and the Millennium Commission.
The National Space Centre, supported by the Millennium Commission with National Lottery funding, is the UK’s largest attraction dedicated to the excitement of space. Co-founded by the University of Leicester and Leicester City Council; its other funding partners include, the East Midlands Development Agency and BT. Beagle-2 Mission Report
Contents
List of Tables ...... ix List of Figures...... xi Acknowledgement ...... xiii Disclaimer ...... xiii Abbreviations and Glossary ...... xiv
1 Executive Summary...... 1 2 Introduction ...... 3 3 Mission Achievements ...... 5 4 Mission Timeline...... 7 4.1 Overview...... 7 4.2 Details...... 8 4.2.1 Cruise Phase...... 8 4.2.2 Surface Operations Phase ...... 9 5 Cruise Phase...... 11 5.1 June 2003 ...... 11 5.2 July 2003...... 12 5.3 September 2003...... 13 5.4 Oct 2003 ...... 13 5.5 November 2003 ...... 14 5.6 December 2003...... 15 6 Surface Operations Phase ...... 17 6.1 Entry Descent and Landing ...... 17 6.2 Autonomous Surface Operations: Sol 1 to 3...... 19 6.3 Communications Search Modes ...... 20 6.4 Post-landing Search Activities ...... 21 6.5 Commanding Operations...... 22 6.6 Power Modelling...... 24 6.7 Surface Operations Summary ...... 24 7 Post-Operations Analysis ...... 25 7.1 Post Operations Analysis ...... 25 7.1.1 The Search for Beagle 2 on the Surface...... 25 7.1.2 Investigation of Failure Modes...... 29 7.1.3 Ejection...... 46 7.1.4 Design Failure Modes...... 47 7.2 Summary of Failure Analysis Findings...... 49 7.3 Beagle 2 Failure Modes: Identification and Analysis ...... 51 7.3.1 Launch Phase ...... 52 7.3.2 Cruise Phase...... 53 7.3.3 Ejection Characteristics ...... 53 7.3.4 Coast Phase...... 55 7.3.5 EDLS – General...... 56 7.3.6 EDLS – Ballistic Entry...... 57 7.3.7 EDLS Algorithm...... 60 7.3.8 Pilot Chute Phase ...... 61 7.3.9 Main Chute Phase...... 63 7.3.10 Airbag Initiation to Impact ...... 65 7.3.11 First Impact to Software Handover...... 68 7.3.12 PSW to LSW Handover...... 70 7.3.13 Default Timeline...... 70 7.3.14 Communications...... 72
iii Beagle-2 Mission Report
Annexes
A Beagle 2 Mission and Flight Operations Personnel ...... 77 B Checkout reports...... 81 B.1 Early Operations and Checkouts A, B, C ...... 82 B.2 Checkout A (185,186)...... 84 B.3 Checkout B (186,187)...... 84 B.4 Checkout C (193,194)...... 85 B.5 Checkout D: Memory Scrub (1 Sep 2003) ...... 87 B.6 Checkouts E and F: LSW Image Upload Test (7, 9 Oct 2003) ...... 87 B.7 Checkouts G and H: LSW Upload (21, 22 Nov 2003)...... 88 B.8 Checkouts I and J: Pre-Ejection Preparation (17 Dec 2003)...... 90 B.9 Ejection Report (19 Dec 2003)...... 91 C VMC Image Analysis...... 93 C.1 VMC Images ...... 93 C.2 Conclusions from VMC Image Analysis ...... 96 C.2.1 Introduction and Background ...... 96 C.2.2 VMC Position and Field of View...... 97 C.2.3 Ejection Velocity and Spin Rate...... 97 C.2.4 Sun Angle on Beagle 2...... 97 C.2.5 Beagle 2 “MLI” Feature...... 98 C.2.6 “Debris” Object ...... 98 C.2.7 “Background” Objects...... 99 C.2.8 Conclusions ...... 99 D Post-landing Search Strategy ...... 101 D.1 Introduction...... 101 D.1.1 Beagle onboard Comms Session types (Modes)...... 102 D.1.2 Jodrell Bank Comms Session type ...... 103 D.1.3 NASA Odyssey Comms Session types ...... 103 D.1.4 ESA Mars Express Comms Session types ...... 103 D.2 NASA Odyssey Communication modes...... 103 D.3 ESA Mars Express Communication modes...... 104 D.4 Command Files prepared for each Contact...... 105 D.4.1 File 1 (Sol 1 - Sol 3 (am))...... 105 D.4.2 File 2 (Sol 3 (pm) - Sol 6 (pm)) ...... 105 D.4.3 File 3 (Sol 7 (pm) - Sol 9 (pm)) ...... 105 D.4.4 File 4 (Sols 14 - 19; 31 – 32)...... 105 D.4.5 File 5 ‘Blind’ (30/31st Jan)...... 105 D.5 Criteria for Transitions between Comms Search Modes ...... 105 D.5.1 Operational Mode (1)...... 105 D.5.2 Safe Mode(2)...... 106 D.5.3 Comms Search Mode 1 (3)...... 106 D.5.4 Comms Search Mode 2 (4)...... 106 D.5.5 AutoTransmit Mode (5)...... 106 D.6 Factors which may affect the Comms Sessions...... 107 D.6.1 Missed Comms Sessions...... 107 D.6.2 Battery State-of-Charge Assessment...... 107 D.6.3 Battery Protection Logic (BPL) ...... 107 D.7 Summary of Events ...... 108 D.7.1 Course of Action...... 109 D.8 Summary of Relevant Investigations ...... 110 D.8.1 Software Protections...... 110 D.9 MLT Communicating modes ...... 111 D.9.2 Full Deployment Test...... 111
iv Beagle-2 Mission Report
D.10 Detailed Summary of all Communications Sessions...... 112 D.10.1 December 2003...... 113 D.10.2 January 2004...... 114 D.10.3 February 2004 ...... 118 E Landed Phase Operational Power Modelling...... 119 E.1 Introduction...... 119 E.2 Power Subsystem (PSS) Definition ...... 120 E.2.1 Solar Array ...... 120 E.2.2 Battery ...... 120 E.2.3 Battery Charge Control ...... 120 E.2.4 Power Loads ...... 120 E.3 Beagle 2 Power Model Description ...... 121 E.3.1 Module 1: Power Subsystem Hardware...... 121 E.3.2 Module 2: Lander Orientation...... 121 E.3.3 Module 3: Solar Flux Incident ...... 122 E.3.4 Module 4: Thermal Input ...... 123 E.3.5 Module 5: Subsystem Load ...... 123 E.3.6 Model Outputs ...... 123 E.4 Operational Phase Power Modelling ...... 124 E.4.1 Beagle 2 Landed Phase Operations ...... 124 E.4.2 Power Model Operational Load Profile ...... 124 E.4.3 Sol 1 to 3 Operations Power Modelling - Phase 1...... 125 E.4.4 Operations Power Modelling - Phase 2...... 133 E.4.5 Other Power Modelling...... 138 E.5 Conclusions ...... 139 E.6 Power Modelling Supporting Data ...... 139 F Electrical Cruise Behaviour Report...... 143 F.1 Battery Voltage and State of Charge ...... 143 F.1.1 Power Subsystem Telemetry ...... 143 F.1.2 Battery State of Charge Variation during Cruise...... 143 F.1.3 Load Current - Beagle 2 Un-powered ...... 145 F.1.4 Taper Charge and BSOC at Ejection ...... 146 F.1.5 Estimating Beagle 2 Load and Charge Currents ...... 147 F.1.6 Beagle 2 Data Stream Telemetry Analysis...... 148 F.1.7 Battery State of Charge Conclusions...... 151 F.2 Software Counters and Errors during Cruise...... 152 F.2.1 Power-On Reset Counter (MSC0043)...... 152 F.2.2 Memory EDAC Counters...... 152 F.2.3 Single-bit RAM Events (MSC0052)...... 153 F.2.4 Miscellaneous Counters...... 153 F.3 APS Voltage History...... 154 F.3.1 APS Power Line Voltage Specification ...... 154 F.3.2 APS Power Line Variation during Cruise Phase ...... 154 F.3.3 APS Power Line Trend across Cruise Phase...... 155 F.3.4 APS power line trend during individual checkouts A to J...... 157 F.3.5 APS Voltage History Conclusions...... 160 F.4 Timer and Latches...... 161 F.4.1 Timer Operation...... 161 F.4.2 Timer Load for Ejection...... 161 F.4.3 Timer and Latch Cruise Testing ...... 161 F.4.4 Timer and Latch Testing Conclusions ...... 162 F.5 Data Volumes of Cruise Phase Checkouts ...... 162 F.6 LSW Upload Procedure ...... 163 F.6.1 File Structure (Files 1 to 25) ...... 163 F.6.2 File Structure (File 26) ...... 164 F.6.3 Group Structure (Groups 1 to 6) ...... 164 F.6.4 Group Structure (Group 7) ...... 165 F.6.5 LSW Upload Deviations from Procedure...... 166 F.6.6 LSW Checksum Record ...... 167 v Beagle-2 Mission Report
F.7 PSW and LSW Checksum History ...... 168 F.7.1 PSW Checksum Summary ...... 168 F.7.2 LSW Checksum Summary...... 169 F.8 EEPROM Integrity Check ...... 170 F.8.1 Overview of Final Checkout ...... 170 F.8.2 Scope of EEPROM Integrity Check...... 170 F.8.3 EEPROM Check: Commands and Results ...... 171 F.8.4 EEPROM Contents ...... 172 F.8.5 EEPROM Memory Map ...... 175 F.8.6 Timer Telemetry Changes Correlated to Command History...... 176 G Landing Site Hazard Analysis ...... 183 G.1 Counting coverage of hazards using recent THEMIS IR daytime 100m/pixel,19m/pixel ...... 183 G.2 Counting coverage of hazards using recent THEMIS IR daytime ~19m/pixel...... 185 G.3 Counting coverage of hazards using MOC image R13-00632 resolution ~2-4 m/pixel ...... 186 G.4 Updated Analysis of Merged Images ...... 187 G.5 Summary of previous remote data on Isidis landing site: MOLA, radar, rock abundance from thermal inertia model ...... 188 G.6 Gusev MER Landing Site Coverage...... 188 H Atmosphere and Aerodynamic Implications on Entry, Descent and Landing of Beagle 2 . 189 H.1 Introduction...... 189 H.2 Summary...... 190 H.2.1 Targeting using the final Mars Express state and covariance matrix ...... 190 H.2.2 Atmosphere variability study to assess the effects of possible anomalous density profiles following localised dust storms some time prior to the Beagle2 entry...... 190 H.2.3 A sensitivity study on system drag coefficients to find the lower limits for system operations . 191 H.2.4 A worst case aerothermal study, inducing early boundary layer transition and excess heat flux margins...... 192 H.2.5 An analysis of a localised TPS removal on the conical rear cover coinciding with the observed position of the optical anomaly at release from Mars Express ...... 192 H.2.6 Summary Conclusions...... 193 H.3 Detailed Assessment...... 194 H.3.1 Trajectory, Targeting and Entry Parameters...... 194 H.3.2 Atmosphere Density Profile Variability Study ...... 195 H.3.3 Atmosphere Near-Ground Turbulence ...... 204 H.3.4 Fluctuations in density with altitude ...... 206 H.3.5 Sensitivity Analysis on System Drag Coefficients...... 208 H.3.6 Thermal Protection System Performance ...... 209 H.3.7 Rear Cover ‘Hole’...... 210 H.4 Notes ...... 211 H.5 Conclusions ...... 211 H.6 Appendix A to Annex H ...... 212 H.6.1 Correlation of Excel Model with TRAJ3 Fort 10 Outputs...... 212 H.6.2 Correlation of Nominal Entry Case...... 213 H.6.3 Correlation of Excel Spreadsheet with FGE Fort 10 – 20km Blend ...... 214 H.6.4 Correlation with 0km Blend ...... 215 H.6.5 Correlation with 15km Blend...... 217 I EDLS Events: Parachute Deployment, Heatshield Separation and Landing ...... 219 I.1 Lander/Heatshield Separation...... 219 I.2 Drag Coefficients...... 223 I.3 Main Parachute material porosity...... 223 I.4 Main Parachute/Airbag re-contact after bouncing ...... 223 I.5 Overall Summary...... 225 J Lovell Telescope Observations of the Beagle 2 Landing Site...... 227 J.1 Summary of Observations ...... 227 J.2 The Receiver...... 228 J.3 The Digital Data Acquisition System ...... 229
vi Beagle-2 Mission Report
J.4 Software Processing and Real-time Data Display ...... 229 J.4.1 Post Processing...... 230 J.5 Test Observations ...... 230 J.6 Telescope Pointing ...... 232 J.7 Observation Planning...... 232 J.8 Observations...... 232 J.9 The JBO Beagle 2 Team...... 233 K Assessment of the Cruise Phase Thermal Performance...... 235 K.1 Introduction...... 235 K.1.1 Scope...... 235 K.1.2 Acronyms ...... 235 K.2 Reference Documents...... 236 K.3 Cruise Phase Thermal Performance...... 237 K.4 Coast, Entry, and Descent Phases...... 243 K.5 Landed Phase ...... 244 K.6 Conclusions ...... 245 L Prediction of Size of a Beagle 2 Impact Crater ...... 247 L.1 Crater Size Equation: Schmidt – Holsapple Pi scaling ...... 247 L.2 Beagle 2 Specifics for Crater Size Calculation ...... 248 L.3 EDL Modelling...... 249 L.4 Beagle 2 Crater Size Calculations ...... 253 L.4.1 General Calculation for Beagle 2 Impactor ...... 253 L.4.2 Incorporating Impact Results from EDL Modelling ...... 255 L.5 Beagle 2 Impact Feature Size ...... 255 L.6 Conclusions ...... 256 L.7 References ...... 257 M Beagle 2 Landing Ellipse Evolution ...... 259
vii Beagle-2 Mission Report
This page left intentionally blank.
viii Beagle-2 Mission Report
List of Tables
Table 5.1 - Checkout Summary ...... 11 Table 7.1 - Atmosphere Density Profiles Used for EDL Modelling ...... 34 Table 7.2 - Results of Modelling EDL Varying Drag Coefficient...... 36 Table 7.3 - Overpressure Events...... 41 Table 7.4 - Pressures for ARM Cover Dis-bond Events ...... 41 Table 7.5 - Delta-V Outgassing Events ...... 43 Table 7.6 - Rates measured by gyroscopes at ejection and resulting off-pointing...... 46 Table B.1 - Checkout Summary ...... 81 Table B.2 - Beagle 2 Heater Data (4th June 2003) ...... 82 Table C.1 - VMC Image Times...... 96 Table C.2 - VMC Analysis Results...... 98 Table D.1 - December Overflights...... 113 Table D.2 - January Overflights ...... 117 Table D.3 - February Overflights...... 118 Table E.1 - Power Model Run General Parameters Phase 1 Modelling...... 125 Table E.2 - Power Model Run Thermal Parameters Phase 1 Modelling...... 125 Table E.3 - Array Orientation Parameters Phase 1 Modelling ...... 125 Table E.4 - Load Profile for Initial Sol 1 to 3 Power Model Run ...... 127 Table E.5 - Array Offset Angles Phase 2 Modelling ...... 133 Table E.6 - Orientation Angle Model Runs Phase 2 Modelling ...... 134 Table E.7 - Lander Subsystem Unit Power Requirements ...... 139 Table E.8 - Lander Subsystem Short Duration Loads ...... 139 Table E.9 - Payload Unit Current and Power Requirements ...... 140 Table E.10 - Heater Power and Resistance Specification ...... 141 Table E.11 - Thermal Model Energy Requirement Predictions...... 141 Table F.1 - Beagle 2 Power Subsystem Telemetry...... 143 Table F.2 - Estimated Un-powered Load Current...... 145 Table F.3 - Battery and Bus Currents ...... 147 Table F.4 - Reset Counter...... 152 Table F.5 - EDAC Counters...... 152 Table F.6 - Single-Bit RAM Events...... 153 Table F.7 - Miscellaneous Event Counters...... 153 Table F.8 - APS Specification...... 154 Table F.9 - Power Line Variation during Cruise...... 154 Table F.10 - Voltage Variation during Cruise...... 155 Table F.11 - Power Line Trend during Cruise ...... 155 Table F.12 - APS Line Power Trend during Checkouts...... 157 Table F.13 - APS Maximum Temperatures ...... 158 Table F.14 - Timer Data for Ejection ...... 161 Table F.15 - Timer Tests...... 161 Table F.16 - Checkout Data Volumes ...... 162 Table F.17 - Probe Software Checksum Verifications ...... 168 Table F.19 - Lander Software Checksum Verification ...... 169 Table G.1 - Area Coverage of Potential Hazards in post-ejection (ESOC) Isidis Landing Ellipse based on ~100m/pixel THEMIS image (Figure G.1) ...... 183 Table G.2 - Area Coverage of Potential Hazards in post-ejection (ESOC) Isidis Landing Ellipse ...... 185 Table G.3 - Area Coverage of Potential Hazards in Isidis Landing Ellipse based on ~2-4 m/pixel MOC...... 186 Table G.4 - Merged Image Potential Hazards...... 187 Table G.5 - Area Coverage of Potential Hazards in Gusev ...... 188 Table H.1 - Entry Parameters ...... 194 Table H.2 - SPICAM blended atmosphere, RAT timings...... 197 Table H.3 - Drag Reduction levels effect on RAT timing...... 208 Table H.4 - Parachute and RAT parameter sensitivity with atmosphere and EDLS elements Cd ...... 208 Table H.5 - Heatshield sizing Parameters ...... 209 Table H.6 - Summary of Correlation Results ...... 212
ix Beagle-2 Mission Report
Table H.7 - Key Events in 15km Blend Case...... 217 Table K.1 - Mars Express monitored thermal telemetry for days 183 and 349 ...... 236 Table K.2 - Maximum and minimum temperatures during checkouts...... 238 Table K.3 - Maximum APS temperatures recorded during Lander checkouts ...... 239 Table K.4 - Predictions for day 183...... 239 Table K.5 - Predictions with increased heat loss at ARM...... 240 Table K.6 - Predicted minimum external surface temperatures of MLI during Cruise...... 242 Table K.7 - Summary of flight predictions for Coast...... 242 Table K.8 - Coast phase temperature requirements and predicted margins ...... 242 Table K.9 - Coast predictions with increased heat loss at ARM (Model COA311)...... 243 Table K.10 - Cruise thermal performance summary...... 245 Table L.1 - Impactor Parameters for Crater Size Calculation...... 248 Table L.2 - Target Surface Densities ...... 249 Table L.3 - Target Parameters for Crater Size Calculation ...... 249 Table L.4 - Impact Velocity and Angle for Varying Atmospheres ...... 252 Table L.5 - Impact Velocity and Angle for Varying Entry Parameters...... 253 Table L.6 - Crater Size Predictions. Impact Velocities 1, 3 and 7 km/s...... 254 Table L.7 - Beagle 2 Impact Crater Size for Differing EDL Scenarios ...... 255
x Beagle-2 Mission Report
List of Figures
Figure 4.1 - Overview of Beagle 2 Mission ...... 7 Figure 5.1 - Beagle 2 team members at the LOCC during the first Cruise checkout...... 16 Figure 5.2 - Mission Manager Mark Sims on PISA Console in ESOC during Beagle 2 checkout ...... 16 Figure 6.1 - Beagle 2 EDL sequence ...... 18 Figure 6.2 - Beagle 2 Operations Sol 1, Default Mission Events Timeline ...... 19 Figure 7.1 - THEMIS VIS/IRS mosaic of Beagle 2 Landing Uncertainty Ellipse ...... 25 Figure 7.2 - Enlargement of planning mosaic showing location and coverage of high resolution (1.5m/pixel) MOC images ...... 26 Figure 7.3 - Location of “best” candidate feature observed in eastern ellipse...... 27 Figure 7.4 - “Best” candidate feature for Beagle 2 location ...... 27 Figure C.1 - VMC 02 ...... 93 Figure C.2 - VMC 03 ...... 93 Figure C.3 - VMC 04 ...... 94 Figure C.4 - VMC 05 ...... 94 Figure C.5 - VMC 06 ...... 95 Figure C.6 - VMC 07 ...... 95 Figure C.7 - VMC Field of View and Orientation (Courtesy ESA)...... 97 Figure E.1 - Orientation Angle Geometry ...... 122 Figure E.2 - Panel Offset Angle Geometry...... 122 Figure E.3 - Currents Flowing in System Sol 1 to 3 Power Model Run...... 128 Figure E.4 - Battery State of Charge Sol 1 to 3 Power Model Run...... 129 Rx On time set to 80 Minutes and ODY Sol 3 Am Added ...... 129 Figure E.5 - Currents Flowing in System Sol 1 to 3, CSM 1 and CSM 2 (revised) Power Model Run ...... 131 Figure E.6 - Battery State of Charge Sol 1 to 3, CSM 1 and CSM 2 (revised) Power Model Run...... 132 Figure E.7 - Currents Flowing in System. Panel Offsets 20/North Facing Lander Power Model Run...... 135 Figure E.8 - Battery State of Charge. Panel Offsets 20/North Facing Lander Power Model Run ...... 135 Figure E.9 - Battery State of Charge, Panel Offsets 20/North Facing Lander. Optical Depth 0.7 ...... 136 Figure E.10 - Battery State of Charge, Panels 2 to 4 Offset 20. Panel 1 flat. North Facing Lander...... 137 Figure E.11 - Battery State of Charge. Panel Offsets 20 / North Facing Lander. Initial BSOC 81% ...... 138 Figure F.1 - Battery State of Charge and Battery Terminal Voltage Variation during Cruise Phase...... 144 Figure F.2 - Battery State of Charge and Battery Terminal Voltage Variation Cruise Checkout I...... 144 Figure F.3 - Difference between MA0013 Battery Voltage and NBEA0110 BEAGLE-Battery V. Checkout C. 149 Figure F.4 - Battery State of Charge and Battery Terminal Voltage Variation Cruise Checkout I. B2 TM.... 150 Figure F.5 - BSOC and Battery Voltage Variation B2 TM Checkout I. Modified Battery Voltage Calculation 150 Figure F.6 - APS 13V Variation over Cruise Phase...... 156 Figure F.7 - APS 13 Volt variation correlated with APS temperature variation...... 157 Figure F.8 - APS 5V line during Checkout F...... 159 Figure F.9 - APS 5V line during Checkout G ...... 160 Figure G.1(a) - THEMIS IR daylight mosaic around landing site...... 184 Figure G.1(b) - THEMIS Mosaic Hazards...... 184 Figure G.2 - 19m/pixel THEMIS image ...... 185 Figure G.3 - MOC image R13-00632...... 186 Figure G.4 - Merged image provided by Malin Space Science Systems ...... 187 Figure H.1 - Comparison of SPICAM atmosphere density profile with various Isidis density profiles...... 190 Figure H.2 - Comparison of SPICAM atmosphere with EMCD density profiles at 17.2N 268E...... 191 Figure H.3 - Landing Dispersion Evolution ...... 194 Figure H.4 - Approximate location of the SPICAM measurements ...... 195 Figure H.5 - SPICAM density profile blended to low dust scenario at 15km compared to EMCD mean profiles at landing site...... 196 Figure H.6 - SPICAM density profile blended to low dust at 0km compared to EMCD mean profiles at landing site ...... 196 Figure H.7 - SPICAM density profile compared to EMCD mean profiles (at date and location of SPICAM measurements, Stephen Lewis, Oxford University)...... 197 Figure H.8 - MGS-TES atmsosphere at Beagle2 site on 25th December compared with EMCDv3.1 profles 198 Figure H.9 - SPICAM data blended in to default MGS atmosphere at 10km above datum ...... 199
xi Beagle-2 Mission Report
Figure H.10 - SPICAM data blended in to default MGS atmosphere at 5km above datum ...... 199 Figure H.10a - SPICAM atmosphere profile,...... 200 a comparison between the initial and the improved accuracy data sets...... 200 Figure H.10b - Comparison between improved SPICAM profile and predicted atmosphere at 17.2 N 268 E 201 Figure H.10c - Various atmosphere profiles normalised to the default MGS atmosphere ...... 201 (SPICAM type atmospheres have been artificially blended back to the default at 5km altitude) ...... 201 Figure H.10d Comparison of Density Profiles (5km blend point)...... 202 Figure H10e Comparison of Velocity Profiles (5km blend point) ...... 203 Figure H10f Comparison of Deceleration Profiles (5km blend point) ...... 203 Figure H.11 - Atmospheric temperature fluctuations above the MER Spirit landing site ...... 205 Figure H.12a - Hypothetical atmospheric density fluctuations...... 206 Figure H.12b - Effect of hypothetical atmospheric density fluctuations on velocity profile...... 207 Figure H.12c - Effect of hypothetical atmospheric density fluctuations on deceleration profile ...... 207 Figure H.13 - Static Pressure Field near Peak Heat Transfer during Entry ...... 210 Figure H.14 - Correlation of Nominal Entry Case ...... 213 Figure H.15 - Correlation of 20km Blend Case...... 214 Figure H.16 - Correlation of 0km Blend ...... 215 Figure H.17 - Improved Correlation ...... 216 Figure H.18 - Correlation in 15km Blend Case...... 217 Figure J.1 - The 401 MHz Receiver cryostat...... 228 Figure J.2 - A post-processed Waterfall Display...... 230 Figure J.3 - The "Beagle in a Box" test transmitter ...... 231 Figure J.4 - A waterfall display showing 10 minutes of data ...... 231 Figure J.5 - Two Mars graphics showing Beagle 2 landing site...... 232 Figure J.6 - The Lovell Observing Room December 25th 2003...... 233 Figure K.1 - Heater current data for days 183 and 349 ...... 237 Figure K.2 - Impact of backshell MLI surface emissivity on battery and arm heater temperatures...... 239 Figure K.3 - Residence time of molecules as a function of surface temperature [RD3]...... 241 Figure L.1 - Beagle 2 Deceleration Profile Using MGS Atmospheric Density Values ...... 250 Figure L.2 - Beagle 2 Velocity Profile Using MGS Atmospheric Density Values ...... 250 Figure L.3 - Beagle 2 Flight Angle Using MGS Atmospheric Density Values ...... 251 Figure L.4 - Beagle 2 Deceleration Profile Using SPICAM Atmospheric Density Values...... 252 Figure L.5 - Beagle 2 Impact Crater Size. Impact Velocity = 219.13 m/s...... 253 Figure L.6 - Beagle 2 Impact Crater Size. Impact Angle = 45 Degrees...... 254 Figure L.7 - Beagle 2 Crater Ejecta Field Thickness Variation...... 256 Figure M.1 - Beagle 2 Landing Ellipse Location Isidis ...... 259 Figure M.2 - Beagle 2 Landing Ellipse Evolution ...... 260
xii Beagle-2 Mission Report
Acknowledgement
There are countless individuals in teams of all size that helped to get Beagle-2 to Mars during the mission. It is not possible to name them all here, but we particularly wish to acknowledge the following teams for their incredible level of dedicated support:
• the Mars Express team at the European Space Operations Centre (ESOC) in Darmstadt, Germany; • the team at JPL in Pasadena, California who assisted with Beagle-2 command and telemetry links, UHF overflight coordination, operations of the NASA Mars Odyssey spacecraft, and ground segment data services; • the team at Jodrell Bank Radio Telescope, U.K. • the team at the Lander Operations Planning Centre at the Open University in Milton Keynes, UK.
In addition we wish to acknowledge the combined academic and industrial team that built Beagle 2 without whom the mission would have never occurred.
The Beagle 2 operations team would also like to thank NASA, Malin Space Science Systems, the Beagle 2 industrial contractors and ESA for the work conducted in the failure analysis and the UK Particle Physics and Astronomy Research Council who financially supported this investigation. All those who contributed, (e.g. DTI, ESA, PPARC, Wellcome Trust, Millennium Commission and the Beagle 2 science team) to the building of Beagle 2, its instrument packages and its operations phase are also gratefully acknowledged. Jean-Loup Bertaux, the PI of the SPICAM instrument on Mars Express, is thanked for providing early mission data prior to its publication in a peer reviewed paper. Any misinterpretation of this data is solely due to the contributors to this report.
Disclaimer
This report contains summary information, technical data, details of analyses performed, engineering principles, spacecraft command information and telemetry reports, all related to the Beagle 2 mission. It has been provided by members of the Beagle 2 team, and by external groups and individuals to provide an accurate record of the mission following launch and to look at why the mission may have ultimately failed. All information is believed to be accurate at the time of writing however analysis undertaken by other parties and further analysis in the future may arrive at different conclusions. Any inaccuracies or errors within this report due to the editing and compilation process are the responsibility of the contributors and editor and are unintentional.
xiii Beagle-2 Mission Report
Abbreviations and Glossary
AIT Assembly Integration and Test (=AIV) FPGA Field Programmable Gate Array AIV Assembly Integration and FTS File Transfer System Verification FW Filter Wheel AOCS Attitude and Orbital Control System GAP Gas Analysis Package APS Auxiliary Power Supply GS Ground System ARM Aeroshell Release Mechanism GTM Ground Test Model (full set of working avionics along with BCR Battery Charge Regulator critical mechanisms) BEE Back-end Electronics BNSC British National Space Centre HEPA High Efficiency Particulate Filter BS(O)C Battery State of Charge ITAR International Trade in Arms CEM Common Electronics Module Regulations (USA State Department) CEP Central Electronics Processor CoI (ESA) Commission of Inquiry JPL Jet Propulsion Laboratory (of COTS Commercial Of The Shelf NASA, Pasadena California) CSM Communications Search Mode LCL Latching Current Limiter DDOR Delta Differential One-way LMST Local Mean Solar Time Range LOBT Lander On-Board Time DHS Data Handling System LOCC Lander Operations Control DoY Day of Year Centre, National Space Centre, DTI Dept. Trade and Industry (of Leicester the U.K. government) LOPC Lander Operations Planning Centre, Open University, Milton Keynes E2 See EEPROM LSW Lander Software EDL(S) Entry, Descent and Landing (Sequence) LTST Local True Solar Time (time of day on Mars at the Isidis EEPROM Electrically Erasable Planitia landing site) Programmable Read-Only Memory MBS Mössbauer Spectrometer EGSE Electrical Ground Support MCS Mission Control System Equipment (for controlling the MER Mars Exploration Rover GTM) MET Mission Event Timeline ELM Electronics Module MEX Mars Express EMC Electro-magnetic Compatibility MIC Microscope ESA European Space Agency MLI Multi-Layer Insulation ESD Electrostatic Discharge MLT Mars Lander Transceiver ESOC European Space Operations Centre, Darmstadt, Germany MPS Mission Planning System ESS Environmental Sensor Suite MTL Mission Timeline
xiv Beagle-2 Mission Report
NASA National Aeronautics and Space Sol Martian Day Administration (USA) SPICAM Spectroscopy for the NEV Near-Earth Verification Investigation of the Characteristics of the Atmosphere of Mars (a NSC National Space Centre spectrometer onboard Mars OBC On-board Clock Express) ODY NASA Mars Odyssey SSMM Solid State Mass Memory (on Mars Express) PPARC Particle Physics and Astronomy SSTSP Standard Spacecraft Time Research Council Source Packet PPS Payload Power Supply SUEM Spin Up and Eject Mechanism PROX-1 CCSDS Communication Protocol for Landers TM/TC Telemetry, Telecommand PSW Probe Software TOA Top of Atmosphere TPS Thermal Protection System RADFET Radiation (sensitive) Field Effect Transistor UART Universal Asynchronous RCG Rock Corer Grinder Receiver Transmitter RF Radio-frequency UHF Ultra-high Frequency RFI Radio-frequency Interference UoL University of Leicester UTC Universal Time Coordinated SBU Switch and Backup Unit (Clock and Memory) WAM Wide-Angle Mirror SCOS (ESA) Spacecraft Operating XRS X-Ray Spectrometer System
SCS Stereo Camera System SGICD Space to Ground Interface
xv Beagle-2 Mission Report
This page left intentionally blank.
xvi Beagle-2 Mission Report Executive Summary
1 Executive Summary
Mars Express represented at its inception a unique opportunity for Europe to explore Mars. The Beagle 2 probe was conceived to search for evidence of past and present life on the surface of the planet.
The Beagle 2 project achieved the following, namely:
• An innovative integrated design • World class, high return, low mass integrated instrumentation • Advancement of planetary lander technology in Europe • Pioneering of industry/academia collaboration within the UK • Delivery of Beagle 2 to Mars Express • Flight to Mars • Release/ejection from Mars Express • Unprecedented levels of public interest and support for planetary science
The Beagle 2 team has conducted an internal review of the mission, seeking possible reasons for the lack of communication on 25th December 2003 (and since), following its successful ejection from Mars Express on the 19th December 2003. No definitive cause could be identified due to the paucity of data. However, a review of pre-launch assembly, integration and test data and post launch cruise telemetry data, and some post mission theoretical studies, allows the team to assign likelihood to a particular failure mode irrespective of whether there is any absolute evidence for or against it.
The Beagle 2 internal investigation has derived a number of findings. These findings are presented in the order in which they might have occurred, and not in their likelihood of happening:
1. Beagle 2 was nominal at spin up and ejection with all parameters within required limits and a very small landing ellipse was predicted. 2. Analysis of the ejection images show that Beagle 2 was ejected at the correct velocity and spin rate. 3. There is evidence of water ice accumulation of the +Z side of Mars Express (MEX) due to spacecraft and/or probe outgassing. 4. The “debris” object within the first image recorded by a camera on MEX may be ice from the top of the +Z surface of Mars Express, however no definitive statement as to its nature or origin can be made. 5. The illuminated patch seen on the multilayer insulation on the side of Beagle 2 in the shadowed area at ejection may be an optical reflection or possible damage but would not have been significant thermally during the coast phase. 6. Entry analysis shows that Beagle 2 was robust over a range of atmospheric conditions. Data from a Mars Express instrument (SPICAM) shows evidence of anomalously low atmospheric density at 20- 40km above the surface in early January. If such a condition existed over the Beagle 2 landing site, it may have prevented a successful landing. However the SPICAM data is at variance with all known models of the atmosphere. Both Mars exploration rovers (Spirit and Opportunity) report surprisingly low atmospheric resistance during their landings. Improved characterisation of the Martian atmosphere is, in the view of the Beagle 2 team, critical to the success of future missions. 7. Break-up of the extended thermal protection tiles or structural failure of the front heat shield and corrupt aerodynamics during entry cannot be ruled out. 8. There is no reason to doubt the operation of the main parachute; all possible tests, with the exception of one at high altitude, were performed. 9. Malfunction of the electronics system during the various high shock environments encountered during EDL, however unlikely, cannot be ruled out. Electronics were tested as working nominally throughout the cruise and for the last time on December 18th.
1 Beagle-2 Mission Report Executive Summary
10. Lander electronics malfunction due to a variety of reasons (ESD, random component failure etc.) cannot be ruled out. 11. Collision of the lander within its airbags and the parachute following release is unlikely, but the probability of occurrence is not zero. 12. Provision was made for the separation of the back cover, the lander package and heatshield. The theoretical models predict dispersion. 13. Malfunction, and failure/puncture on landing of the gasbag system cannot be ruled out, again appropriate tests, if less in number than desirable, were conducted. 14. The hazard levels within the landing site ellipse were in line with pre-launch predictions although it should be noted that two large craters are within the final predicted landing ellipse derived following ejection. 15. Failure to deploy due to damage during landing cannot be ruled out. All stages of the deployment were tested on multiple occasions. 16. Damage to the antenna during landing cannot be ruled out.
A large number of programmatic, design and technical lessons have been learnt from Beagle 2 which will need to be applied to future missions. The primary lesson is however that a lander cannot be treated as an “instrument” i.e. as an addition to an orbiter. Appropriate priority including funding, schedule and resources (mass etc.) must be given to a lander in any future mission.
High resolution mapping of the landing ellipse by Mars Global Surveyor has detected no parts of the Beagle 2 probe to date. No detailed images of the landing site from Mars Express have been received.
The Beagle 2 team would like to thank NASA, Malin Space Science Systems, the Beagle 2 industrial contractors and ESA for the work conducted in the failure analysis and the UK Particle Physics and Astronomy Research Council who financially supported this investigation. All those who contributed, (e.g. DTI, ESA, PPARC, Wellcome Trust, Millennium Commission and the Beagle 2 science team) to the building of Beagle 2, its instrument packages and its operations phase are gratefully acknowledged. Jean-Loup Bertaux, the PI of the SPICAM instrument on Mars Express, is thanked for providing data prior to its publication.
The recent results from the NASA Mars Exploration Rovers, Spirit and Opportunity and data from ground- based telescopes show that the Beagle 2 payload would have made significant contributions to understanding Mars, particularly in the areas of geochemistry, geochronology, mineralogy and atmospheric science. It would have addressed the question of whether life has ever existed had it landed at a site similar to NASA’s Opportunity rover. Consequently it is a desire of the consortium to re-fly the payload as soon as possible using an improved and evolved design of the low mass, innovative and integrated lander.
2 Beagle-2 Mission Report Introduction
2 Introduction
Beagle 2 was the UK-led low-mass innovative lander delivered by Mars Express to conduct in-situ analysis of its landing site in Isidis Planitia near the Martian equator, including looking for signs of past and present life on the planet. The probe was launched with Mars Express on the 2nd June 2003 and ejected from the orbiter on the 19th December 2003. The Beagle 2 team has been conducting an internal review of the mission including possible failure causes since the lack of communication from the lander following its scheduled landing on 25th December 2003. The following document is the final report from the Beagle 2 mission operations team and has been compiled by the team with the following purposes:
• to summarise the facts of the Beagle 2 mission thereby establishing the background to the loss of the lander, • to summarise the search strategy used to try and locate Beagle 2 on Mars, • to investigate any anomalies and explore potential failures which could be relevant to the loss of the lander.
The report describes in detail the operations, achievements and performance of the probe during the mission and then attempts to identify likely failure mechanisms. It has been said by NASA and others that the difference between success and failure on a Mars mission is very small and in the end chance plays apart in terms of the atmosphere, winds, landing site rocks etc. This report attempts to summarise all possible failure mechanisms. No definitive cause of the failure can be identified due to the lack of any data (radio, telemetry or visual). However, a review of pre-launch assembly, integration and test data and post launch cruise telemetry data has allowed the team to assign likelihood to a particular failure mode and whether there is any evidence for or against it. This report, where possible, presents analysis which may reduce the likelihood of any given failure mode being the cause for the lander failing to communicate. The report contents are constructed from various technical and project documents and analysis produced during the mission and the post-landing search and investigation. Detailed analysis and other data are contained within the report annexes, and summaries are presented in the main text. Some key documents where appropriate are also contained within the main text. Some text is consequently repeated within the report. This approach allows the reader who only wishes to learn the basic facts to avoid the need to read the details of each analysis.
It is hoped that this report will be of interest and use to academics, industrial teams, space mission professionals, space enthusiasts and interested members of the public. Some parts are very technical in nature, however we hope that this document will form a permanent record of the Beagle 2 mission and will be of use for future missions. We wish all future projects every success in exploring the fascinating planet Mars.
3 Beagle-2 Mission Report Introduction
This page left intentionally blank.
4 Beagle-2 Mission Report Mission Achievements
3 Mission Achievements
Set against the colossal achievement of designing, building and making Beagle 2 ready for launch with Mars Express on a Soyuz-Fregat rocket in June 2003, with less than 3 years from the final ESA acceptance and against an extremely tight mass budget of 60 kg and highly constrained interface requirements, the post- launch mission achievements seem relatively modest. Nevertheless important lessons have been learnt during the 6½ month cruise phase – from launch on 2nd June until ejection from Mars Express on 19th December.
These achievements are chronicled below with their background and relevance to the mission. These are not in any order of significance.
a) Beagle 2 was switched on; checked out and switched off a total of 10 times during the Cruise. b) Thermal telemetry came in very close to predictions with no anomalies. This confirms structural and MLI integrity and thermal design philosophy. c) Operation of the electronics and software (PSW) were nominal with a number of relatively minor exceptions, all of which were analysed, understood, corrected or avoided in following checkouts. d) On two occasions telemetry from Beagle 2 was lost (for different reasons) but after real-time analysis of the data, successful recovery was achieved. e) Problems with telecommand control of the Coast phase Timer were overcome by correcting the ESOC database in time for the pre-ejection set up on 17 and 18th December which was performed successfully. f) Battery and energy management worked nominally throughout. g) A new version of Lander software (LSW) was successfully uploaded into the program memory on 21st November after having been tested on ground using the Ground Test Model (GTM). This was deemed necessary as a new failure mode had been highlighted after launch, which meant that the Lid and Solar Panel deployment might not complete correctly possibly leading to loss of mission. This was considered sufficiently serious that it was agreed (by ESA, BNSC and the Beagle 2 Consortium) to go ahead with the risky process of uploading a completely new software image after verification of the upload process (performed during the 7th and 9th October checkouts). h) A faulty heater circuit (XRS BEE) allocation was identified and corrected. i) Beagle 2 was subjected to the Solar Flare Storm in October/November 2003. Note that Beagle 2 was switched off at the time and no adverse effects were observed at the following checkout. j) Beagle 2 was fully prepared during pre-ejection checkouts and there were no anomalies at the time of ejection. k) The process for commanding Beagle, although slow and laborious at the beginning did improve throughout the cruise. This was because the full Beagle TM/TC database was not imported into the ESOC control system. This is a ‘Lesson Learned’. l) In-flight monitoring of the accelerometers outputs at zero-g led to updating the offset values in the monitoring algorithm to improve the pilot chute deployment. m) A Ground Test Model of Beagle 2 was set up in the Lander Operations Control Centre (LOCC) area during the early part of the cruise in order to validate Mission procedures and databases; allow Instrument Engineers/Scientists to validate on-surface Science operations; perform additional interface testing between Instruments and the processor; and to validate new software patches both for the Cruise and for the Landed phases of the mission. In the absence of a Beagle 2 software simulator, the GTM was indispensable and the concept and completeness of such a resource should be improved for any future missions. n) The very small Mission team was complemented by individuals from Industry to produce an effective group with good team spirit that was able to handle the Cruise phase activities. It is conjectural as to whether this small team would have been able to cope equally as well with the Landed phase and science operations. o) Excellent support was provided by Jodrell Bank in the post-landing communications search activities and also for co-ordinating the generous offers of help from other Radio Telescopes around the world.
5 Beagle-2 Mission Report Mission Achievements
p) The support from NASA JPL in preparing for and executing the on-surface communications via NASA Odyssey was excellent and bodes well for any potential re-flight mission. q) The support from the Mars Express operations team at ESOC was also excellent throughout the cruise phase and during the post-landing communications search phase.
The post launch mission activities were well controlled and executed by the Mission teams at LOCC, ESOC and JPL with good support from the academic and industrial partners. Areas for improvement or enhancement have been identified in a ‘Lessons Learned’ document. Overall it can be said that a great deal was achieved in getting Beagle 2 to Mars and the experience gained by the teams involved would benefit a re-flight of Beagle 2 or another small lander, as well as assist in the ESA Aurora programme.
Finally it should be noted that the public support for Beagle 2 both in the UK and all around the world has been overwhelmingly positive, from very young children to pensioners, many of whom said they were surprised that the UK “did Space” but that it was much more important to have tried than to have succeeded and they would support another attempt.
A Rubicon has been crossed.
6 Beagle-2 Mission Report Mission Timeline
4 Mission Timeline 4.1 Overview
Figure 4.1 - Overview of Beagle 2 Mission
2nd June 2003 153:17:45 UTC Mars Express Launch
5th June 2003 155:08:13 to 08:45 Beagle 2 Frangibolt Release
4th – 6th July 2003 Days 185 to 187 Beagle 2 Checkout
12th-13th July 2003 Days 193 to 194 Beagle 2 Checkout
1st September 2003 Day 244 Beagle 2 Memory Scrub power on
7th and 9th October 2003 Days 280 and 281 Beagle 2 battery charging, checkout, EDLS parameter upload and LSW image load test
21st and 22nd November 2003 Days 325 and 326 LSW image upload and Timer test
17th and 18th December 2003 Days 351 and 352 Pre-ejection checkouts including final parameter uploads, load and start timer
19th December 2003 Day 353 Beagle 2 ejection from Mars Express and 6 day Coast
25th December 2003 359:02:51:22 Beagle 2 predicted at entry interface to atmosphere (120km)
7 Beagle-2 Mission Report Mission Timeline
4.2 Details
4.2.1 Cruise Phase
The following is a summary of the main events between the launch of Beagle 2 on Mars Express and the predicted entry into the Martian atmosphere.
02/06/2003 17:45 Mars Express Launch
05/06/2003 08:13 Beagle 2 Frangibolt Release
04/07/2003 20:04 Cruise Checkout A - Post Launch Checkout
05/07/2003 19:03 Cruise Checkout B - Heater/Timer Tests
12/07/2003 16:46 Cruise Checkout C - Heater/Timer Tests
01/09/2003 12:40 Cruise Checkout D - Memory Scrub
07/10/2003 11:03 Cruise Checkout E - LSW Upload Test
09/10/2003 11:48 Cruise Checkout F - LSW Upload Test (cont)
21/11/2003 08:00 Cruise Checkout G - LSW Upload
22/11/2003 10:15 Cruise Checkout H - LSW Upload (cont)
17/12/2003 06:34 Cruise Checkout I - Pre-Ejection Timer Load
18/12/2003 06:33 Cruise Checkout J - Pre Ejection Timer Check
19/12/2003 08:31 Beagle 2 Ejection from Mars Express
25/12/2003 02:51 Beagle 2 Predicted Atmospheric Entry
8 Beagle-2 Mission Report Mission Timeline
4.2.2 Surface Operations Phase
The following is a list of all contact opportunities when commanding of Beagle 2 was attempted or listening was carried out by Mars Express or ground based receivers. In addition, the Odyssey spacecraft listened for a signal from Beagle 2 on every geometric overflight until the last date listed below.
25/12/2003 05:25 Odyssey Hailing
25/12/2003 22:20 Jodrell Bank Carrier Listening
26/12/2003 18:06 Odyssey Hailing
26/12/2003 20:09 Odyssey Hailing
26/12/2003 23:00 Jodrell Bank Carrier Listening
27/12/2003 06:49 Odyssey Hailing
27/12/2003 22:56 Jodrell Bank and Stanford Carrier Listening
29/12/2003 08:13 Odyssey Hailing
30/12/2003 07:57 Odyssey Hailing
30/12/2003 20:54 Odyssey Hailing
31/12/2003 09:38 Odyssey Blind Commanding
01/01/2004 22:19 Odyssey Hailing
02/01/2004 11:02 Odyssey Hailing
07/01/2004 12:12 MEX Hailing
07/01/2004 13:33 Odyssey Hailing
08/01/2004 02:31 Odyssey Hailing
09/01/2004 13:27 MEX Hailing
10/01/2004 14:04 MEX Hailing
12/01/2004 02:02 MEX Hailing
22/01/2004 22:10 MEX Listening in Canister Mode
24/01/2004 23:19 MEX Hailing
25/01/2004 22:53 MEX Hailing
28/01/2004 19:30 Odyssey Hailing
30/01/2004 04:58 Odyssey Blind Commanding
31/01/2004 04:41 Odyssey Blind Commanding
03/02/2004 04:35 MEX Invalid/Valid Hailing
10/03/2004 18:16 Last Listening Overflight of Odyssey
9 Beagle-2 Mission Report Mission Timeline
This page left intentionally blank.
10 Beagle-2 Mission Report Cruise Phase
5 Cruise Phase
Cruise Phase of Mars Express/Beagle 2 began with launch on June 2nd, 2003. This section provides a short overview of the operations and activities related to Beagle 2 from launch until ejection from Mars Express on December 19th 2003. For more complete reports, please consult Annex B: Checkout Reports later in this document. Here we present the key events of Cruise Phase and summarise the in-flight testing of Beagle 2 that demonstrate that the lander was in the expected condition at the time of ejection. Annex F: Electrical Cruise Behaviour Report is also relevant and provides details of Beagle’s electrical systems’ status during cruise.
Beagle 2 was switched on and operated ten times during the Cruise Phase. Each such occasion is called a ‘Checkout’ and each had a specific purpose and set of goals, to check particular spacecraft systems or to uplink or downlink information. The cruise checkouts are summarised below, with power-on and power-off dates and times. Each checkout is referred to by a letter (A through J) and the summary provided here, Annexes B and F and other sections of this Mission Report use the same scheme for referring to checkouts. ‘DoY’ in the following table (and elsewhere in the report) refers to ‘Day of Year’ in 2003.
Power -On Power-Off Checkout DoY Date Time (UTC) DoY Date Time (UTC) A First Cruise Checkout 185 2003-07-04 20:04:04 186 2003-07-05 04:54:00 B NEV Checkout 186 2003-07-05 19:03:00 187 2003-07-06 01:41:00 C 2nd NEV Checkout 193 2003-07-12 16:46:00 194 2003-07-13 01:18:00 D Mini Checkout, E2 Scrub 244 2003-09-01 12:40:17 244 2003-09-01 14:00:30 E LSW Upload Test 280 2003-10-07 11:03:00 280 2003-10-07 16:45:00 F LSW Test Continued 282 2003-10-09 11:48:00 282 2003-10-09 17:30:00 G LSW Upload 325 2003-11-21 08:00:00 325 2003-11-21 19:00:00 H LSW Upload Part 2 326 2003-11-22 10:15:00 326 2003-11-22 16:38:00 I Pre-Ejection Checkout 351 2003-12-17 06:34:00 351 2003-12-17 14:51:00 J Pre-Ejection Checkout 2 352 2003-12-18 06:33:51 352 2003-12-18 10:40:00
Table 5.1 - Checkout Summary
5.1 June 2003
Following launch at 17:45 UTC on June 2nd 2003, Mars Express and the upper stage arrived at a parking orbit and Mars Express started its journey to Mars at 19:17 UTC. While Mars Express was checked and tested after launch, Beagle 2 remained off but its temperatures were monitored and recorded in the data sent back from Mars Express (MEX) – temperature data is not presented here but matched expectations of the thermal model (see also Annex K). Similarly, the battery status was nominal and the timer active.
An initial problem with a star-tracker on MEX delayed early operations. The heaters on-board Beagle used to maintain operating temperature started working 32.29 hours after launch (at 2003.155.03.14.35.311) after Beagle 2 had cooled down sufficiently.
On the 5th of June Mars Express performed its first major manoeuvre to correct its course to intercept Mars. At this early stage Beagle 2 was un-clamped from MEX. Frangibolt clamps were used to secure Beagle 2 throughout the launch and these needed to be released as soon as possible. The un-clamping procedure began 62h 24m into the mission. The three frangibolts, while not fired in the pre-planned order (due to either a command database or wiring error), were released successfully at the first attempt. On the 6th of June Mars Express passed the 1 million km mark and a manoeuvre was performed to enable one of its cameras to take a picture of the receding Earth.
11 Beagle-2 Mission Report Cruise Phase
5.2 July 2003
Checkout A (185,186)
The first Beagle 2 checkout took place on July 4th. Melacom commenced hailing 19h50, Beagle 2 switch on 19h52. (Note communications with Beagle 2 relies on an international digital protocol the so-called CCSDS Proximity 1 protocol, firstly the spacecraft hails the lander/probe and only once the hail has been acknowledged can command and telemetry data flow. This happens whether the probe is attached and a physical umbilical link exists or detached when a modulated RF signal is used.)
Initial hail failed as link rates were set to 2/2 rather than 8/8 (8/8 required for umbilical). Telemetry as expected was received at 20h05, with SSMM data dumps every 10 minutes. A safety parameter was updated that controls the earliest time to start the entry sequence. Connection test commands were successful. Timer test uplinked and performed at 22h10.
Melacom link was dropped at 22h58 due to a clock roll-over event in the Melacom unit. Beagle 2 still generating telemetry packets. 23h04 Beagle 2 stopped generating telemetry as the buffer was full. Melacom was power cycled at 01h33 but still no telemetry was seen. 04h47 PSW Reboot command was sent to Beagle 2, Melacom link aborted and rehailed. Telemetry was received at 04h49 indicating successful reboot and recovery from the anomalous link dropout.
Beagle 2 was powered off at 04h54, with much relief.
Checkout B (186,187)
On July 5th/6th the second Beagle 2 checkout was conducted. Melacom commenced hailing at 18h59, Beagle 2 switch on occurred at 19h03. Correct operation of the timer was confirmed using three tests after the latches were disabled. Heaters A4, B1 and B2 were then checked out successfully apart from Heater A4 (which maintained temperature of the payload electronics supporting the X-ray spectrometer). Heater testing started 22h45 and ended at 23h27. This issue was later investigated further and the heater found to function correctly.
The process and commands for writing values into Beagle 2 memory and reading them back again were verified. EEPROM dumps, SSMM dumps and PSW patching were used to test memory.
Timer stopped at 19h21 and restarted at 19h26. Timer test repeated unsuccessfully using Mars Express command database, and every combination of correct and incorrect Ack flags, and single/double encapsulation was tested. Double encapsulated commands were successful.
Battery state of charge was managed at approximately 50%. Beagle 2 was switched off at 01h40.
Checkout C (193,194)
The third checkout started with Melacom beginning hailing at 16h31, with Beagle 2 power on at 16h45. In order to maintain the 50% state of charge level, the main power lines were switched off at 17h56 – Beagle was running on its battery in space for the first time. The timer was tested again, starting and stopping in different ways. The heaters on both banks were exercised, starting at 20h28. The next step in the checkout procedure was to update more of the EDLS parameters (22h42), and some further timer investigation. All data from all tests were downlinked with regular SSMM drip-tap reads.
The final activity of checkout C was to test uploading blocks of memory, started at 00h41, completed at 01h12. Beagle 2 was powered off at 01h18.
12 Beagle-2 Mission Report Cruise Phase
5.3 September 2003
Checkout D (244)
Beagle 2 was powered on Sept 1st in order to allow the PSW software to ‘scrub’ Beagle’s memory. This was undertaken as a precautionary measure, as temporary errors may have arisen due to radiation during the several weeks in the off state since the checkouts in the first half of July. The scrubbing process happens automatically as a background task whilst Beagle 2 is powered. At the time of power-on, all Beagle 2 systems showed the expected nominal status. No active testing was required, and so telemetry was monitored for 1h 20m before Beagle 2 was powered down again.
5.4 Oct 2003
Checkout E (280)
During Oct 7th, a number of important operations took place on Beagle 2. Firstly, a simple test of heater control channels confirmed that the electronics heater for the X-ray Spectrometer was operational as heater B4, on the spare heater circuit. This test was conducted as there was concern that the heater might not be operable. The battery charge level was raised by using both power supply lines from Mars Express, beginning to charge the battery towards the 100% level required at ejection.
During the entry and descent through the Martian atmosphere Beagle’s landing systems rely on accurate timing and acceleration levels to determine when to trigger various events. The set of data that includes these and other parameters are called ‘EDLS parameters’ and these are regularly updated during Cruise Phase with progressively more accurate values because estimates and models improve as the spacecraft nears Mars. The EDLS parameters were updated on 7th Oct and a minor software update was provided to the probe software (PSW).
Checkout F (282)
Once these activities were completed, a ‘dry-run’ test was performed of the procedure for uploading a new version of lander software (LSW) that is used on the surface after landing. The software is transmitted as a series of files, and the test uplink was performed using two such files uploaded into a safe area of Beagle’s memory. The procedure was successfully validated.
On 9th October, the software updated on 7th October was verified, and a more complete and representative test of the software loading procedure was conducted. 26 software load files were used, corresponding to the full size of the lander software. The procedure was thus validated using a representative volume of data. The existing software was checked to ensure that not only had the LSW uplink load been successful, but that it had not interfered with any other part of memory.
13 Beagle-2 Mission Report Cruise Phase
5.5 November 2003
Checkouts G, H (325, 326)
During November 2003 the largest solar flare ever recorded occurred, and adverse effects were reported from numerous spacecraft. As Beagle-2 was off at the time, there was only a very small chance that the flare would cause any damage to Beagle. At power-on on November 21st it was confirmed that there was no corruption in any of the memory systems.
Unlike earlier checkouts, Mars Express and Beagle-2 were so far away from Earth that it was not possible to directly command the spacecraft in real-time (with signals taking over 6 minutes to travel to the spacecraft at the speed of light). All events were controlled using the MEX on-board mission timeline.
During the 21st and 22nd November the lander software image (LSW) was successfully completely replaced in a total operating time of 17h 47m. During this time the operations team also uplinked data to support the early operations of Beagle-2 for its first few days on Mars. Together these days included the following tasks:
• Upload of new version 3 .0 of lander software • LSW upload was confirmed via checksums • 55% of the LSW image was dumped to ground and successfully compared • All mission ancillary data was uploaded • Ancillary data was confirmed via checksums and appropriate memory dumps • Probe software (PSW) was verified with correct checksums to ensure no corruption during the uploads • Other critical areas of the Beagle-2 memory systems were verified • The operation of the timer for arrival at Mars was tested and verified • A contingency procedure recovering from umbilical link failure was verified
Several contingencies arose on November 21st. An error in the Mars Express mission timeline prevented software files 3-7 inclusive from being uploaded at the first attempt. The timeline was stopped, rebuilt, and all the files were uploaded. Following verification of the checksums, it was found that three files (12, 14 and 22) contained checksum errors. In addition, a command error in file 19 was identified, and so file 19 was rebuilt and sent to ESOC for processing and uplink to Beagle-2.
During the memory dump anomalous behaviour was observed. After one set of housekeeping packets had been transmitted, the communications link failed. The link recovery contingency procedure was then activated and the link restored. The result was that only 225 memory dump packets were received on the ground from an expected total of 404 for one image. This anomaly was fully investigated and is understood. Every memory dump packet received was correct, including those corresponding to files 12 and 14.
On November 22nd it was confirmed that the earlier checksum error for these files arose from a timing problem (calculating checksums takes a relatively long time compared to other commands). The new version of file 19 was uplinked successfully and verified. Supporting data followed. At the time of switch-off on November 22nd, the battery was charged to approximately 99% of its capacity (planning for ejection with 100%).
A timer test was performed in parallel with the ancillary data load and timer ‘trip’ was verified. The ‘trip’ is the signal responsible for activating Beagle-2 on arrival at Mars, some six days after ejection from Mars Express. The timer was zeroed and reset, and then loaded with a value equivalent to a 36 minute timeout (1FFO) at 14:12. The timer was started and first ticked at 14:23:56. The timer ‘trip’ signal was activated at 15:01 – within a single cycle as expected and required.
14 Beagle-2 Mission Report Cruise Phase
5.6 December 2003
Checkouts I, J (351, 352)
Beagle 2 was powered on both 17th and 18th December in preparation for ejection on the 19th. At each switch-on, telemetry showed nominal spacecraft status and this was re-confirmed repeatedly throughout both days. Note that Beagle-2 is off for ejection.
On the 17th December some final safety parameters were updated to the values for ejection. Software patches were uploaded and downlinked again for verification. A set of ancillary data was also uploaded to Beagle-2, including the following information:
• Schedule of early communications sessions • Communications sessions parameters (default durations etc.) • Mission Events Timeline
A new command file was created during the checkout to request a dump of the communications sessions in order to verify that they had been correctly loaded. No new activity sequences were uploaded. After verifying the integrity of the memory again, the timers were prepared for ejection. The value commanded for switch-on time 2.5 hours before entry was 0E71 and this was confirmed in telemetry by the expected value 1CE.
The timer was started at 13:50:22 and confirmed as started in telemetry at 13:51:07. This was final action before setting the long-period latch for ejection.
The following day, 18th December 2003, Beagle-2 was powered on in the normal way. The timer was still running as expected and contained the expected values. This was reported in telemetry as 202. Nine transitions of the timer had occurred since the previous switch-off, with a reported telemetry period of 19m 12s. The long-period time-out latch was enabled and the short-period latch confirmed as disabled. Beagle-2 was commanded to switch off for the last time in Cruise at 10:40.
Later in the day, the final GO/NO-GO for ejection decision was made (‘GO’) by the Beagle-2 team at 22:28 via voice loop conference.
Beagle-2 Ejection
On December 19th 2003 the operations team confirmed GO for Beagle-2 ejection at 06:51. Ejection occurred at 08:31 and was verified using the following four methods:
• ‘Glitch’ in S-band doppler at pyro firing (Reported at 07:39) • Spacecraft telemetry showing Beagle 2 disconnected (10:32). • Appropriate responses in spacecraft AOCS data (verified by 11:12) • VMC images showing Beagle 2 separated from Mars Express (with a delta-V of 0.30m/sec).
At the time of ejection, the following spacecraft status was noted:
• Battery charge level was verified at over 97% • Memory checks confirmed software status and critical data area integrity • The entry, descent and landing system parameters were as expected
15 Beagle-2 Mission Report Cruise Phase
Figure 5.1 - Beagle 2 team members at the LOCC during the first Cruise checkout
Figure 5.2 - Mission Manager Mark Sims on PISA Console in ESOC during Beagle 2 checkout
16 Beagle-2 Mission Report Surface Operations Phase
6 Surface Operations Phase
The Surface Operations Phase includes 3 categories of operation:
1. Entry Descent and Landing (EDL) 2. Autonomous Surface Operations a. Sol 1 to 3 Mission Event Timeline (MET) b. Communications Search Modes (CSM) 3. Commanded Operations
EDL is managed by Probe Software (PSW) and is completely pre-programmed and autonomous.
Timed operations post landing are implemented via the Mission Events Timeline (MET) and a catalogue of Activity Sequences (AS) running in Lander Software (LSW). The MET is populated in three ways, via events stored in EEPROM for execution post landing (the Sol 1 to 3 MET), events selected by operators and added by telecommand (Commanded Operations) or events added automatically by LSW (Communications Search Modes, CSM).
6.1 Entry Descent and Landing
At the end of coast phase the long period timer switches the Auxiliary Power Supply (APS) on and wakes up Beagle 2 two and a half hours before the probe reaches the predicted top of atmosphere. The processor switches on and PSW boots into EDL mode.
Atmospheric entry is at 5.5 km/s and an angle of 16.5°±1° below the horizontal. Two accelerometers, a science accelerometer and a system accelerometer, measure the deceleration profile. The deceleration values are monitored by the PSW EDL algorithm. Either accelerometer reaching the critical trigger value activates the pyrotechnics which deploy the pilot chute and subsequently the main chute. A radar altimeter initiates airbag inflation at a height of around 200m above the surface. The EDL sequence is depicted in Figure 6.1.
17 Beagle-2 Mission Report Surface Operations Phase
Elapsed time 2:40 – 3:00 hours 2 – 3 min13 Secs 2 – 4 min 4 – 6 min 8 – 9min
Acc. State STARTBELOW THB RISING ABOVE THB FALLING BELOW THA ABOVE THC (0) (1) (2) (3) (4) (5) (6)
Deceleration
THC Mission Phase Transition initiated by Timeout OR 3 consecutive RAT signals THB
14 secs RAT Active Last min before THE, high rate collection of 5 secs THA science data
THE RAT_ENABLE_PT_TIME LANDING_ENABLE_PT_TIME RAT_TIMEOUT_PT_TIME STATIONARY_PT_TIME Tseparated Edge of Tdescent0 Main Parachute T Handover Wake up detected atmosphere RAT Tlanding Tstationary detected Deployment detected to LSW detected detected
MISSION CRUISE COAST ENTRY DESCENT NEAR SURFACE LANDING SURFACE PHASE (1) (2) (3) (4) (5) (6) (7)
Algorithm Tseparated TInitEntry Tdescent TNearSurface Tlanding Run detection detection detection detection detection
Energise/De- Energise/De-energise Fire PRM Pyro Battery energise Pyro Timeline Battery Fire Aeroshell Release Commands Pilot chute deploy Fire Airbag Gasing Mechanism Clamp Band 1 Execution Switch OFF RAT release Start/Stop Aero-shell release Switch OFF PPS (Main chute deploy) Clamp Band 2 Switch OFF PAW & GAP Airbag outer lacing release release Start/Stop Select 15V PPS Airbag gassing release Switch ON PPS Airbag inner lacing release Update EEPROM Switch ON RAT Pyro wire release Reboot to start LSW
Figure 6.1 - Beagle 2 EDL sequence
18 Beagle-2 Mission Report Surface Operations Phase
6.2 Autonomous Surface Operations: Sol 1 to 3
All planned operations for Sols 1 to 3 were pre-programmed and would have executed automatically following a successful EDL. A set of ‘default’ MET entries and Activity Sequences for Sols 1 to 3 were prepared and loaded into EEPROM during cruise phase checkout I. The default MET executes from the point of PSW/LSW handover. The following text describes the pre-programmed activities as planned, and as if they had executed.
Figure 6.2 shows the pre-programmed operations for Sol 1.
SOL 1 (DoY 359) 02:54z 03:53z 04:39z 04:50z 04:53z LSP Deployment ESS On ODY Comms
WAM Image Heater Management
MET Entry 3: MET Entry 5: MET Entry 1: Environmental Lid and solar panel Transceiver switched Sensor Suite On. on to listen for Mars deployment. Calls Activated for 2 hours. Activity Sequence 1 Odyssey hail at MET Entry 2: Duration of approx. MET Entry 4. 05:25z. Stereo Camera 30 minutes. Survival heater System image. modes switched from Calls Activity Landing Safe Sequence 30000. to Sol 1 Night Wide Angle Mirror configuration. Calls allows information on Activity Sequence lander base and local 24000. surroundings to be returned in one image.
Figure 6.2 - Beagle 2 Operations Sol 1, Default Mission Events Timeline
LSW inherits Beagle2 from PSW with the processor and electronics drawing power from the battery. At this time the transceiver is off, the solar panels folded into the lid covering the UHF antenna and the lid is closed.
The first landed phase operation is Lid and Solar Panel (LSP) deployment. The lid main hinge is driven to 180° to open Beagle. The solar panels held in place by 5 frangibolts are released and the four individual panel hinges are driven to 160° to deploy the arrays.
20 minutes after deployment is complete the first imaging Activity Sequence is executed. The right camera of the Stereo Camera System is used with the Wide Angled Mirror (WAM) to achieve a single black and white image of a 360° field then compressed by a factor of 10 to ensure return within data volume budget of the first Mars Odyssey pass. Operations for the first night on the surface included an attempt to image the transit of Phobos across the field of view of the left camera – an operation that would provide valuable data for determining landing site location.
During early afternoon on the second Sol, when predicted margins on available solar array power and battery state of charge are healthy, payload management activities were to execute. The Gas Analysis Package is powered and engineering telemetry parameters sampled to ascertain instrument health. The ARM and PAW are released by activating hold down frangibolts. This is carried out at the earliest opportunity in order to eliminate a thermal path and reduce heat losses from the area of the battery and electronics rather than as a precursor to ARM manoeuvres.
19 Beagle-2 Mission Report Surface Operations Phase
Unstowing and moving the ARM out of the protective envelope of the base can not be planned safely before the operations team has the opportunity to assess the landing locality using the returned WAM images. The Planetary Underground Tool (PLUTO) (more commonly called the mole) launch lock pin is released shortly after, as early as possible into the mission to reduce the possibility that dust build up could clog the mechanism.
15 pre-programmed transceiver operations were chosen to cover overflight opportunities up to the 17th of January. Communications sessions were selected based on predicted lander visibility times provided by the Mars Odyssey Navigation team at JPL and the Mars Express Flight Dynamics teams at ESOC. Entries to coordinate transceiver sessions with 8 Mars Odyssey overflights during the first 7 Sols on the surface were programmed.
Contact durations were of the order of 20 minutes and the programmed times for the initial communications session entries were configured in order that the receiver remained on for a significant period before and after the predicted overflight times to account for any deviations from the predictions.
3 comms sessions were pre-programmed to coincide with times when the Beagle landing site was visible from Jodrell Bank radio telescope. These would allow an earth-based search to be carried out in support of a search from Mars orbit should communications not be established immediately post-landing.
6.3 Communications Search Modes
LSW handles the situation where communications opportunities are missed by modifying existing programmed communications sessions or by entering one of three autonomous dedicated comms search modes.
The communications search modes implemented are:
1. CSM 1 2. CSM 2 3. Auto Transmit Mode
Communications with Beagle 2 used CCSDS proximity protocol 1 where the relay orbiter hails the lander to establish a link, and communications (returned data and forwarded commands) only take place once the hail is complete.
Comms Search Mode 1 Automatically adds am and pm Odyssey passes (2 per Sol) onto the Mission Event Timeline (MET). Stored Comms sessions are also executed. The inserted communication sessions are inserted at 03:35 and 15:35 Local True Solar Time (LTST) for a duration of 80 mins (see also section 6.4).
Comms Search Mode 2 (CSM2 = LSW mode 4) Automatically inserts sessions onto the timeline depending on whether the LSW has determined the Local True Solar Time (LTST), from the Solar Panel currents at Dawn/Dusk, to be either Daytime (10:00 to 18:00) or Nightime (18:00 to 10:00).
• Day Mode The Transceiver is switched ON for 59 minutes of every hour. The 59 min ON period consists of: o 10 seconds of Expedited mode Telemetry o ‘Listen’ mode (receiver ON) for the remainder of the 59 mins. but with the transmitter Carrier (unmodulated) cycled ON for 1 minute out of every 10 mins., starting with 9 mins of Carrier OFF.
• Night Mode The Transceiver is cycled ON for 1 minute every 5 minutes, starting with 4 minutes OFF. During the 1 minute ON time the transceiver is in ‘Listen’ mode with the Carrier (unmodulated) being cycled as in Day mode. However, because the transceiver is switched OFF after 1 minute it never gets to the Carrier ON part of the cycle (after 9 mins) - in order to conserve energy during the night.
20 Beagle-2 Mission Report Surface Operations Phase
AutoTransmit Mode (ATM = LSW mode 5) ATM is the same as CSM2 above except for the duration of the Expedited mode Telemetry portion. This is the final Comms Search mode so therefore Beagle-2 will attempt to transmit data for the duration of a session providing there is sufficient battery power available. This is to handle the situation where an orbiter is listening but for some reason cannot Hail. As described above in CSM2 the Expedited mode is active for 10 seconds during which time telemetry from the Summary Lander Status (Packet Id = 128) is transmitted for approximately 0.25 seconds. In ATM mode the Expedited mode transmission consists of all of the Housekeeping and some Science packets, sufficient to fill the 10 seconds duration of transmission.
All communications modes are dependant on the Battery State of Charge (BSOC) which is used to determine if a session can be vetoed if there is insufficient battery power available. A lower state of BSOC is invoked for CSM2 compared to all other modes. In CSM1 data transmission will occur if a hail is received but the BSC is lower than the set value - however its duration is limited to 3 minutes.
6.4 Post-landing Search Activities
Searching for a signal from Beagle post landing was supported by 3 different organisations:
• NASA JPL via MARS Odyssey • Jodrell Bank (and other Radio Telescopes) • ESA via MARS Express
The searches were complementary in that they did not overlap in time. Jodrell Bank was monitoring for a UHF Carrier at or around the Beagle 2 Transceiver (MLT) transmit frequency of 401.585625 MHz, whereas the 2 orbiters were attempting to Hail and establish duplex communications with Beagle-2 (transmit and receive). However, as described in Annex A.4, it is possible to configure the MELACOM UHF transceiver on Mars Express into a ‘listening’ mode like Jodrell Bank and store all the Receiver output into the onboard memory unit (SSMM) for later downlinking to ground where it can be processed and any coherent signal recovered from the noise. This process is, however, time limited by the amount of available memory on Mars Express to store the sample data and monitoring must be synchronised to occur at the time that Beagle2 is expected/supposed to be transmitting.
This mode is called CANISTER mode and was used twice in the search for a signal from Beagle. In addition as a validation of the respective systems, a test was performed between MELACOM on Mars Express and MER-Spirit on the surface of Mars to be sure that MELACOM was able to receive a UHF signal at 401.5 MHz. This was successfully completed on 11th January.
The Odyssey and Mars Express orbits have different characteristics:
• Odyssey’s orbit is Sun-synchronous i.e. has visibility of the landing site at least twice a day at reasonably constant times of Sol (Day on Mars) at constant altitude though at differing elevations and for varying durations. • Mars Express is in an elliptical orbit with an inclination angle to the Mars equator which means it’s pass times over the landing site are variable in terms of both times of Sol and also duration. Passes over the landing site vary between being Apocentred and Pericentred so altitudes vary from <400 km to >12,000 km.
The Hailing procedure on Odyssey is slightly different to that used by Mars Express, though both use the Proximity-1 protocol. This is possible because certain aspects of the protocol are not defined and are open to design interpretation.
After discussions on 30th December with NASA JPL about ‘Blind’ commanding i.e. Sending telecommands on the forward link without waiting for a response from Beagle (in case Beagle’s return link was non operational), it was realised and confirmed that during all the previous Odyssey attempted contacts the Command files that LOCC had prepared for each agreed pass had been transmitted without the Odyssey Transceiver first establishing that a valid communications session is in progress - it does this in anticipation
21 Beagle-2 Mission Report Surface Operations Phase that a communications session is about to be established. Reception of commands would naturally have altered the lander status and this led to uncertainty as to the Lander on-board mode at that time.
Details of the contents of each Command file are described in Annex D.4.
There are several different types of Communication sessions, both from the onboard Beagle point of view and also from the Orbiter/Ground operations point of view which are described in Annex D.
6.5 Commanding Operations
During the period of active commanding, from 25th December 2003 until 3rd February 2004, the Beagle 2 operations team at the LOCC identified and investigated possible recoverable failure cases.
Hypotheses were proposed, recovery strategies developed and rehearsed on the Ground Test model and contingency commanding was sent to the spacecraft via Mars Express or Mars Odyssey. Commands were sent to Beagle 2 on 23 occasions. In this manner the following identified recoverable failure modes were exercised.
• Clock reset / Out of synch • Comms Search Modes (CSM) entered • Mission Event Timeline (MET) empty of comms sessions • Solar power marginal • Battery State of Charge Monitor threshold exceeded • Power Subsystem Monitoring threshold exceeded • Lander orientation unfavourable • Sequential Commanding not available • Software requires reboot
Clock reset / Out of synch A reset or jump in Lander On Board Time (LOBT) would have shifted the timings for operations on the Mission Events Timeline, leaving communications sessions out of synch with orbiter overflights or Jodrell Bank observations.
For each commanding overflight a value of LOBT corresponding to an offset from the time of maximum elevation of the pass was calculated. Commanding to reset LOBT to this predicted value was included in every commanding session from Sol 3 onwards.
Comms Search Modes The possible routes through the communications search mode (CSM) tree involve complex permutations. A significant part of the operations team search strategy involved categorizing these permutations and planning contingency operations accordingly.
Beagle 2 was hailed on overflights coinciding with the widest range of possible comms search mode opportunities.
No attempt to hail was made between 12th January and 22nd January. With software parameters controlling the duration of comms modes as set, this allowed adequate time for Beagle 2 to enter CSM 2 via any of the possible routes.
Software parameters control the length of time or the number of missed communications opportunities that must pass before entry to each comms search mode. Commanding to set these parameters to their minimum values was sent from Sols 14 onwards.
22 Beagle-2 Mission Report Surface Operations Phase
Mission Event Timeline Empty of Comms Sessions A number of failure mechanisms could result in the Mission Event Timeline (MET) being emptied of future communications events.
Commanding to add additional communications sessions to the MET was included with every commanding opportunity.
Solar Power Marginal If power margins were small, the transceiver may have been prevented from switching on / responding to a hail correctly. This could occur if deployment of the solar arrays had been incomplete. Opportunities for successful communications would then be limited to times of peak available power.
Hail attempts were made at a wide range of times of the day and hence a cross section of power regimes.
Commanding to adjust the angle of one solar panel was included in all hail attempts. The array was commanded from 20° above the plane of the lander base (nominal deployment), to lie flat in the plane – a configuration more robust to different orientations of the lander on the surface. This has been confirmed by power modelling.
Commanding to drive all four solar panels flat was included in all commanding sessions from Sol 14 onwards.
Commanding to double the current threshold at which the solar panel motors stop driving was also included from Sol 14 onwards.
Battery State of Charge Monitor An anomaly was identified in the performance of the Battery State of Charge (BSOC) monitoring algorithm that could have prevented the transceiver switching correctly in certain modes, even at healthy levels of BSOC.
The command to reset the software flag instructing LSW to use the BSOC monitoring scheme was sent in all comms sessions from Sol 8 onwards thereby preventing the on-board use of BSOC as an inhibitor.
Power Subsystem Monitoring It was conceivable that in a situation of low power margins that the power subsystem protection logic could prevent the transceiver from switching correctly.
Commands to deactivate the battery current, bus current and bus voltage protection logic were sent from Sol 14 onwards.
Lander Orientation Unfavourable Lander lid orientation at an angle to the surface could affect the performance of Beagle 2 antenna.
Mars Express and Mars Odyssey overflights occurred at a wide series of elevations and azimuth angles from the nominal landing site. Jodrell Bank listened for a maximum period of 11 hours in a single continuous observation. Combined, these search strategies ensured that a range of possible antenna orientations and resulting beam patterns were allowed for. As the solar arrays are attached to the lander lid, an orientation of the lid such that communications were severely impaired would almost certainly have degraded the power subsystem capability to such an extent as to be un-survivable.
Sequential Commanding not Available Problems with transceiver operation could have prevented the handshaking required between orbiter and lander for sequential commanding. Mars Express commands are only released on confirmation of a successful hail.
With Mars Odyssey commands are released immediately following the hail PDU, and before the communications session is confirmed as active by return hail acknowledgment from the Lander. Mars Odyssey provides an additional facility for blind commanding where commands can be resent continually for the duration of the overflight, without confirmation of a successful hail or acknowledgement of transfer frame receipt by the lander. Blind commanding with Mars Odyssey was carried out on Sols 7, 36 and 37.
23 Beagle-2 Mission Report Surface Operations Phase
Software Requires Reboot Commanding to reboot LSW was sent on Sols 36 and 37.
Conclusions With no response from Beagle 2 on the surface it is not possible to draw any concrete conclusions from the evidence of surface operations. The fact that the contingency commanding carried out was not successful in establishing contact does eliminate some failure modes. Naturally for any contingency commanding to have been successful a forward communications path (at least) must first have been established. As such, a failure on the surface which completely prevented communications can not be ruled out.
6.6 Power Modelling
The viability of autonomous landed operations; the MET for Sols 1 to 3 and comms search mode strategies, with respect to the power subsystem was established using the SEA-developed Beagle 2 Power Model. It was essential to ensure battery state of charge was maintained above a level where night time operations (essentially heater cycling to keep battery and electronics within operational temperatures limits) remained practicable.
A period of review of autonomous operations accompanied the power modelling campaign. The operational strategies developed and implemented in the default MET, and the choice of software parameters managing CSM, retained healthy power margins in the probable landing configurations and landing site conditions.
Power margins could be significantly reduced by adverse environmental conditions at the landing site and a northward pointing lander orientation (ie away from the sun) following deployment. The operational strategies selected ensured best possible robustness to “worst case survivable” landing conditions i.e. an un- favourable lander orientation at a cold landing site with atmospheric optical depths at the top end of the non-dust storm regime. Raised optical depths levels at the landing site in the days preceding landing reported by Mars Global Surveyor caused concern but levels had reduced by the 25th of December.
Resource (mass) constraints limited the designed capability of the Beagle 2 power subsystem and an unfavourable landed orientation is an understood and accepted risk. However, even if the landing configuration was such that the battery was completely flattened during the first night, sufficient battery charge would have been retained before sunset to allow the first communications session to take place.
6.7 Surface Operations Summary
In summary, no signal either modulated or unmodulated was ever received from Beagle by any of the listening orbiters or Radio Telescopes at any time from the day of landing (25th December 2003) until the search was called off in mid-February 2004, after exhausting all the onboard Communications modes, all the listening modes and all the Hailing/Commanding modes from Odyssey and Mars Express. Listening attempts with Odyssey continued to March 10th.
A table of all overflights is given in Annex D.10.
24 Beagle-2 Mission Report Post-Operations Analysis
7 Post-Operations Analysis 7.1 Post Operations Analysis
Activities associated with the Post Operations phase in fact commenced in late January, overlapping with the last weeks of Surface Operations. As Surface Operations wound up the Post Operations investigations became the focus of the Beagle 2 team at the LOCC. The Post Operations Phase encompasses two main thrusts; the search to locate Beagle 2 on the surface and the investigation of possible failure modes of the Lander.
The ESA Beagle 2 Commission of Inquiry began proceedings on the 5th of February 2004. The Post Operations Phase activities were carried out in parallel with and in support of the Commission’s investigations.
7.1.1 The Search for Beagle 2 on the Surface
7.1.1.1 Observations of the Beagle 2 Landing Site
Mars Global Surveyor (MGS) Mars Orbiter Camera (MOC)
Michael C. Malin Malin Space Science Systems (All Images Courtesy Malin Space Science Systems and NASA)
Shortly after the anticipated landing of Beagle 2 and the subsequent absence of communication with the vehicle, L. Richter (Beagle 2 science team member, DLR) inquired of M. Malin (MGS/MOC Principal Investigator, Malin Space Science Systems) whether images could be acquired of the presumed Beagle 2 landing location using the MOC. A campaign was established to image a portion of the landing uncertainty ellipse, guided by considerations provided by Richter and the Beagle 2 team. It was decided to concentrate the limited available imaging resources on the downrange (eastern) one-half of the ellipse, using as guidance the observation that both Mars Exploration Rovers (MER-A and MER-B) landed downrange of the centres of their respective uncertainty ellipses owing to elevated, global atmospheric temperatures at altitudes affecting entry dynamics.
Figure 7.1 shows the Beagle 2 uncertainty ellipse over an image base generated by MSSS using the only pre- landing data available, 18m/pixel Mars Odyssey THEMIS Visible Imaging Subsystem (VIS) data for the western one-half and 100m/pixel THEMIS Infrared Subsystem (IRS) data for the eastern one-half.
Figure 7.1 - THEMIS VIS/IRS mosaic of Beagle 2 Landing Uncertainty Ellipse
25 Beagle-2 Mission Report Post-Operations Analysis
This map was incorporated within the MGS/MOC software tool used to target off-nadir observations, and those observations were placed relative to the surface features seen within the ellipse. A total of 10 images were acquired, one (the first) west and nine (subsequent) east of the centre of the ellipse. Slightly more than 36% of the total ellipse, and 72% of the downrange one-half of the ellipse, was imaged. Due to constraints imposed by MGS support of the MER missions, only one image (the western image) was taken in January; two were taken in February, six were taken in March, and the last, a special super-resolution image, was taken in April. Figure 7.2 shows a reduced resolution (originally 1.5m/pixel) version of the mosaic of the eight images acquired of the eastern ellipse on top of the planning mosaic.
Figure 7.2 - Enlargement of planning mosaic showing location and coverage of high resolution (1.5m/pixel) MOC images of the eastern half of the uncertainty ellipse
Raw, cosmetically-cleaned, and map-projected versions of each image were provided to the Beagle 2 team for their evaluation. Scientists at MSSS also inspected the images for indications of the Beagle 2 vehicle or its components. Based on their experience in previous searches for landed vehicles (Viking Lander 1, Mars Pathfinder, Mars Polar Lander, MER-A ‘Spirit’ and MER-B ‘Opportunity’), and on their knowledge of the characteristics of the MOC imaging system, only one reasonably-likely candidate feature was identified in the MOC mosaic. Figures 7.3 and 7.4 (next page) show the location and subsequent imaging of this candidate. In the original, 1.5m/pixel image, the feature is relatively small (about 20m in diameter), dark, roughly circular, and appears to have some interior structure. It does not appear to be an impact crater. However, 20m is somewhat larger than one might expect from the impact of Beagle 2 hardware (see following section), certainly when scaled to the features seen associated with the impact of the MER-A and MER-B heatshields. The higher resolution, higher signal-to-noise ratio image (Figure 7.4, right) taken in April clearly shows the feature to be a small, degraded impact crater, with a small, dark sand dune within it.
Based on this imaging campaign and subsequent analyses, no incontrovertible evidence of the Beagle 2 lander was found within coverage of >72% of the downrange half of the landing uncertainty ellipse.
26 Beagle-2 Mission Report Post-Operations Analysis
Figure 7.3 - Location of “best” candidate feature observed in eastern ellipse
Figure 7.4 - “Best” candidate feature for Beagle 2 location Left: Original image identifying the candidate; Right: Higher resolution and signal-to-noise ratio image showing that the feature is natural
27 Beagle-2 Mission Report Post-Operations Analysis
7.1.1.2 Beagle 2 Impact Crater Size In support of the analysis of images of the landing ellipse an estimate was required of the size of crater that Beagle 2 would have made on impacting with the surface if all or part of the Entry Descent and Landing sequence failed. Crater size was calculated given information about the impactor and the impacted surface using a technique developed by Schmidt and Holsapple, scaling from empirical data of terrestrial cratering.
Impact velocity and flight angle were determined from EDL modelling incorporating a range of available atmospheric density profiles. The following results have been calculated from what are considered the most probable impact velocity and angle.
Crater rim diameter is calculated at approximately 2m.
An ejecta field radius 1.4m to 2m, measured from the crater rim is estimated.
Therefore the predicted total crater feature size is 5m to 6m diameter.
An absolute upper bound on feature size is given at 9m corresponding to an EDL based on an Unblended SPICAM atmospheric density profile.
28 Beagle-2 Mission Report Post-Operations Analysis
7.1.2 Investigation of Failure Modes The focus of investigations was necessarily limited and driven by the availability of evidence. Analysis summaries have been arranged by area of investigation.
Each section is described by the analysis carried out, supporting documents, findings, conclusion drawn and any open items. The following are in no particular order.
• Electrical Performance During Cruise Operations • Thermal Performance During Cruise Operations • EDLS and Mars Atmosphere • VMC Image Analysis • Aeroshell Release Mechanism Access Cover Dis-bond Solution Evidence • BAe Decoy Bond Wires • Out-gassing and SUEM Icing Evidence • Ejection Evidence • EDLS Events Parachute Deployment, Heatshield Separation and Landing
7.1.2.1 Electrical Performance during Cruise Operations The behaviour and performance of the Beagle 2 probe was monitored throughout cruise phase as required by Cruise Phase Operations. A dedicated additional review of cruise phase telemetry was carried out as part of the Post Operations investigation culminating in the Electrical Behaviour During Cruise report. This section summarises the analysis presented in the report and associated documents.
Reference Documents
• Report on Pre-ejection Checkouts • Report on LSW Image Upload 21st and 22nd November • Beagle 2 Observations Reports 1 to 11
7.1.2.1.1 Analysis
All critical telemetry collected from the Beagle 2 probe during cruise phase was reviewed. The analysis of telemetry was directed towards identifying and eliminating any circumstance which may have contributed to the loss of mission. The following were the main areas of investigation.
• Battery Voltage and State of Charge • Software Counters and Errors • Auxiliary Power Supply Voltage History • Performance of Timer Circuit and Latches • Upload of Lander Software Image • PSW / LSW Checksum History and EEPROM Integrity Check
29 Beagle-2 Mission Report Post-Operations Analysis
7.1.2.1.2 Findings
• The battery voltage and state of charge profiles were entirely consistent with the battery management strategies employed and the predicted requirements of the system. Beagle 2 was ejected from Mars Express with a battery state of charge of 97%.
• There were no unplanned software resets during cruise. No multi-bit errors occurred in EEPROM or RAM. A total of 10 single-bit errors occurred in RAM.
• APS voltages were healthy and well within test tolerances for the duration of cruise and no overall trend was observed. Voltage variations are correlated with APS temperature. Some limited evidence points towards increased variability in the 5V line at APS temperatures above specified operational limit. The unit qualification temperature is 8°C above the maximum seen in telemetry.
• All changes in Beagle 2 timer telemetry were correlated with Mars Express command history. No spacecraft anomalies were found in any timer test or related activity. Early problems with timer commanding were positively attributed to errors in command construction. Observations raised on unexpected timer and latch telemetry were the result of misunderstanding about the complex telecommand 238,3 (Perform Activity of a Function).
• Three spacecraft observations occurred during the upload of the LSW image. As a result additional commanding was required to complete the load. The successful load of the full compressed image to EEPROM was confirmed by the verification of 103 correct checksums for the appropriate area of memory.
• An “EEPROM Integrity Check” was carried out following operations to upload sections of memory in order to verify the continuing integrity of critical sections of EEPROM. Critical sections included the PSW image and EDLS software parameters. The final check was carried out on December 17th and all areas of critical memory were verified correct.
7.1.2.1.3 Conclusions
The behaviour of all probe subsystems during cruise was as expected with the exception of a number of understood and documented spacecraft anomalies.
None of these anomalies are considered to impact the survivability of Beagle 2 on the surface of Mars.
No evidence to support any failure hypothesis was discovered in probe telemetry returned during cruise operations.
30 Beagle-2 Mission Report Post-Operations Analysis
7.1.2.2 Thermal Performance during Cruise Operations The Rutherford Appleton Laboratory, responsible for the thermal design of Beagle 2, provided an assessment of the Cruise phase thermal performance of the probe.
7.1.2.2.1 Analysis
The assessment of cruise phase telemetry and post operations phase thermal modelling concentrated on the following areas.
• Auxiliary Power Supply (APS) Temperature Observation • Aeroshell Release Mechanism (ARM) Heater Cycling Observation • Possible Effect of Out-gassing Contaminants
The full analysis is provided in Annex K.
7.1.2.2.2 Findings
The Beagle 2 thermal subsystem performed very close to expectations for the duration of the Cruise phase.
APS Temperature
• During checkouts the Common Electronics warmed at a rate quicker than expected and thus achieved higher temperatures than predicted. With the exception of the APS, all temperatures remained within applicable limits. During a Lander operation on 21/11/2003 the APS was allowed to exceed its upper operational temperature by 2°C. The Common Electronics showed no indication of degraded performance throughout Cruise and there is no reason to suspect that the APS would be damaged by exceeding its upper operational temperature by only 2°C.
ARM Heater Cycling
• With the exception of brief periods during checkouts, the Aeroshell Release Mechanism (ARM) heaters remained at 100% duty cycle (i.e continually on) throughout Cruise. Temperatures of the heated components therefore remained about 5°C colder than would have been attained if the heaters had cycled as required.
• An increased duty-cycle implies either a greater heat loss than expected locally at the ARM, or a greater heat loss than expected from the Probe as a whole (i.e. via MLI). Together these predictions imply that the modelled thermal links from the ARM are not consistent with the flight build (i.e. the heat transfer from the ARM to the Lander is too great and that to the aeroshell is too small). A simple sensitivity analysis has been made to assess whether changing thermal links at the ARM could in principle achieve the observed results. It appears likely that the Cruise phase observations could be replicated by careful adjustment of thermal links at the ARM. As the ARM is actually a clamped joint it is not too surprising that there are some deviations from the pre-launch predictions, heat flow being dependant on contact area, force etc. The thermal data is consistent with no major heat leaks from the structure or major deviations from the expected thermal behaviour.
Contamination of MLI surface finishes
• The radiative properties of the external surfaces would be affected if out-gassed products were to condense on them. Analysis has shown that it is unlikely that water ice could have formed on the backshell MLI, however it could potentially have formed on other MLI surfaces. Organic contamination could potentially have formed on any of the MLI surfaces. Increased surface emissivity alone cannot replicate the observed performance. Therefore it is most likely that if there was contamination, levels were low enough to not significantly alter the radiative properties of the aluminised backshell MLI, or were confined to the colder surfaces with existing high emissivity values. 31 Beagle-2 Mission Report Post-Operations Analysis
Impact of Observations on Coast and Landed Phase Performance
• The ability of the probe thermal design to meet Coast phase requirements was verified though thermal balance testing on the AAM probe. A sensitivity study showed that there were substantial margins between raw predictions and the applicable temperature limits.
• Increased heat loss from the ARM would not cause the probe to fall below applicable temperature limits.
• With the backshell MLI external emissivity increased to simulate severe contamination by water ice its surface temperature is predicted to be -125°C. At this temperature, the residence time of water ice is about 100s. Under such conditions the ice would sublime and the backshell MLI would return towards its natural finish. The time constant of the sublimation would be much quicker than that of the probe’s thermal response and so there would be no noticeable impact on the temperatures within the probe. With the backshell MLI external properties changed to simulate organic contamination its surface temperature is predicted to be -15°C. Under such conditions the contaminants would sublime very readily and the backshell MLI would return towards it natural finish. There would be no noticeable impact on the temperatures within the probe.
• There is nothing in the Cruise thermal data that suggests concern for the Landing phase.
7.1.2.2.3 Conclusion
The thermal subsystem performed very close to expectations. There is no indication in thermal telemetry of problems that could have contributed to any Beagle 2 failure mechanism.
32 Beagle-2 Mission Report Post-Operations Analysis
7.1.2.3 EDLS and Mars Atmosphere The Beagle Entry Descent and Landing system was designed in coordination with EDL modelling performed by Fluid Gravity Engineering Ltd (FGE). EDL modelling was revisited in the post-operations phase in light of possible evidence that the atmospheric density profile above Isidis on 25th December may have differed from that used in the design phase EDL model.
NASA MER Rover Team Experience (Extract from Rob Manning Correspondence)
“From December 20 with a minor global* dust storm in progress David Kass et al. used GSFC's TES retrievals to create thermal and density profiles that were used to remodel the Rovers EDL After Spirit landing, the team came to believe that the updated models were better predictors than the less dusty models they had been using in the 1-2 years prior to landing. After both landings it was found that the a priori (3 to 5-days before landing) TES-based density model was close to the reconstructed atmosphere.
The team was surprised at how substantial the total timeline margin turned out to be (from chute deploy to RAD fire) especially given the surprising 3 sigma delay in the attainment of our target parachute deploy dynamic pressure (g-level) from the entry point. In retrospect it did match (to within 2σ or 15%) David Kass’ mean density model when the recommended dispersions were used. It turned out that the atmosphere densities were not widely off the (predicted) mark.”
Both MER landers reported a significant delay in parachute deployment. The atmospheres experienced by the MER lander were within NASA’s specification but NASA has not provided (at the time of writing) an explanation for the late parachute deployments. The MER landers utilise a more complex EDLS system capable of recovering from late initiation; the Beagle 2 system is unable to respond to changed circumstances as the sequence runs on time alone once the deceleration trigger threshold has been detected.
*the dust storm would be better described as regional
SPICAM Data
A preliminary profile of CO2 density and pressure, obtained by SPICAM during a stellar occultation of Zeta Puppis, was kindly provided to the Beagle 2 team by Dr. Jean-Loup Bertaux of the SPICAM team.
The SPICAM atmospheric observations were made on 13th of Jan, on MEX orbit 17. The line of sight (+Zsc axis) rose above 20km at 2004-01-13, 15:17:49.47 and rose above 40km at 2004-01-13, 15:17:56.76. The latitude of the sub-tangent points observed during this time frame range from 17.184069° and 17.185283° (areocentric). Longitudes range from 267.85125° to 267.87851°. Pressure data was provided in the range 33.4km to 145.0km.
The surface at the location of the SPICAM sample is 2km above the Martian datum. The surface of Mars at Isidis is 3.3km below the datum.
Over comparable mid-altitudes the SPICAM density profile shows significantly lower pressures than the profiles derived from models based on other data. No data is available at lower altitudes and the error band in the data is significant. Comparison with the European Mars Climate Database run at the SPICAM data latitude and longitude using the default MGS dust scenario shows densities at upper altitudes to be higher in the SPICAM data and densities at mid altitudes to be lower, a pattern repeated in the atmospheres reconstructed by the MER EDL teams.
33 Beagle-2 Mission Report Post-Operations Analysis
7.1.2.3.1 Analysis
The EDL was remodelled by Fluid Gravity Engineering (FGE) and additionally by Jim Clemmet (JFC) a number of times with differing factors varied. Effort was expended to ensure correlation between the FGE and JFC modelling. The reviewed EDL analysis addressed the following areas of investigation.
1. Effect of Various Atmospheric Density Profiles 2. Sensitivity Study on System Drag Coefficients 3. Reappraisal of Thermal Protection System (TPS) 4. Effect of Low Altitude Turbulence
1. Atmospheric Density Profiles
The EDL model was rerun with density profiles corresponding to different models and data. The range of atmospheres included in the modelling is shown in Table 7.1.
Case Atmosphere Blend Altitude / km 1 Low Dust NA 2 MGS Dust NA 3 Viking Dust Scenario NA 4 Reduced Density NA 5 SPICAM Data 15 6 SPICAM Data 10 7 SPICAM Data 5 8 MGS Factored 5
Table 7.1 - Atmosphere Density Profiles Used for EDL Modelling
In case 4 the nominal atmospheric MGS density profile was reduced in percentage increments.
In cases 5 to 8, as the SPICAM data covers a limited range of altitudes the profile was “blended” to data from other models at a range of altitudes in order to construct a full density profile. The altitude at which this blend is instigated significantly effects the EDL characteristics.
In case 8 the density profile has been revised by applying a ‘correction factor’ to the MGS prediction with a blend back to MGS at 5km. The ‘ correction factor’ applied is the ratio of the density as measured by SPICAM with that predicted for the MGS dust scenario at 17.2 N 268 E.
2. Drag Coefficient Estimates
Estimates made of coefficients of drag (Cd) were reviewed. Different values of Cd are relevant in four phases of EDL; Hypersonic, transonic, drogue chute and main chute. The EDL model was rerun with Cd values initially at the lowest estimates, then Cd values reduced by 10 % and 20 %. The revised estimates were modelled against various atmosphere cases.
3. Thermal Protection System (TPS)
The margins in the design of the Aeroshell thermal protection system were reviewed. An assessment was made of the effect of possible ingress of hot plasma into the internal volume resulting from the removal of a TPS tile from the ARM access cover.
34 Beagle-2 Mission Report Post-Operations Analysis
4. Low Altitude Turbulence
Data provided from the MER landing sites indicated a turbulent boundary layer in the lower atmosphere. A rough analysis was carried out to determine if local fluctuations in atmospheric density could cause premature triggering of pilot chute deployment. EDL was modelled with air pockets introduced into the nominal density profile.
7.1.2.3.2 Findings
1. Atmospheric Density Profiles
Both the MER Rover EDL experience and MEX SPICAM data provide evidence that the atmospheric density profile encountered during Beagle 2 EDL could have differed from the predicted model. Typically a less dense atmospheric profile results in delays to EDL event timings. A sufficient delay in triggering pilot parachute deployment device (PDD) would cause there to be insufficient time for the radar altimeter to trigger airbag inflation and result in EDL failure. All altitudes are referenced from the Martian datum (3.3km above Isidis surface).
Cases 1 to 3. EDL within margins and survivable for existing model profiles.
Case 4. Reduced Atmosphere: The onset of problems for the radar altimeter occurs at atmospheres reduced by about 15%.
SPICAM Blend 0 km: EDL not survivable. FGE model predicts the PDD to operate at 143.4 s after TOA at an altitude of - 1006.2km. JFC model predicts the PDD at less than 134s and below - 2000km.
SPICAM Blend 15km: Survivable. FGE Model predicts PDD at 138.6 seconds post TOA and altitude of 1793.6. JFC predicts 122.5 seconds and 1321 m. The FGE model shows that the system consumes 4640m between PDD and RAT ENABLE, i.e. surface detection software algorithm running (fixed time of 44s +29s). This compares with 4934m predicted for nominal entry analysis post separation.
SPICAM Blend 10km: PDD event is predicted at an altitude of 717m. It is unlikely that Beagle2 would survive if the SPICAM density profile blends in at 10km, even noting the pessimism of the JFC model.
SPICAM Blend at 5km: PDD event is predicted at an altitude of 183m. Not survivable
MGS SPICAM Revised Density Profile. Blend at 5km: PDD operation is predicted to occur at 1712m, 1.5km lower than for a nominal entry. This is just sufficient for RAT surface detection on the upper threshold.
35 Beagle-2 Mission Report Post-Operations Analysis
2. Drag Coefficients Estimates
Table 7.2 shows the impact of lower values of Cd on EDL survivability.
Case Drag Coefficient Cd Value Atmosphere Result Hypersonic Minimum Estimated Transonic Minimum Estimated Various No Problem A Drogue Chute Minimum Estimated Main Chute Minimum Estimated
Case Drag Coefficient Cd Value Atmosphere Result Hypersonic Minimum Estimated Transonic Minimum Estimated Reduced See Below B Drogue Chute Minimum Estimated Main Chute Minimum Estimated
Case Drag Coefficient Cd Value Atmosphere Result Hypersonic - 10 % Transonic - 20 % Viking No Problem C Drogue Chute - 10 % Main Chute - 20 %
Case Drag Coefficient Cd Value Atmosphere Result Hypersonic - 10 % Transonic - 20 % Viking D Not Survivable Drogue Chute - 20 % Main Chute - 20 %
Case Drag Coefficient Cd Value Atmosphere Result Hypersonic - 10 % Transonic - 20 % Nominal Just E Drogue Chute - 20 % Survivable Main Chute - 20 %
Table 7.2 - Results of Modelling EDL Varying Drag Coefficient
Case B. With Cd values set to the lowest estimate EDL failure can be induced with a lower atmospheric density. The lowered density required is outside model limits but within the SPICAM error bars.
The limit of survivability with existing model atmospheres occurs if three of the Cd values have been overestimated by 20% and one by 10%.
36 Beagle-2 Mission Report Post-Operations Analysis
3. Thermal Protection System
The re-appraisal of the TPS showed a better correlation between the FGE analysis and the experimental data provided by Martin Baker/EADS indicating that a slightly larger margin existed in the TPS design. Analysis showed that the TPS design was robust, even in the gap tile region where localized thinning of tiles was present.
Additional aero-thermal analysis was carried out to determine the effect of reduction in TPS integrity if a rear cover TPS tile section over the ARM access hole was separated before EDL (see section 7.2.4). It was shown that under these conditions the carbon fibre superstructure would exceed its design temperature by a significant margin.
4. Low Altitude Turbulence
At altitudes where g levels are approaching the PDD trigger point, a 1 second software filter in the EDL algorithm will cause local air pockets of less than around 330m in depth to be ignored. An air pocket of depth 600m introduced 1km above the nominal PDD altitude was shown to cause PDD to trigger prematurely. Whether the existence of such atmospheric features is suggested by data available has not been established.
7.1.2.3.3 Conclusions
EDL is robust to the minimum drag coefficient assumptions with the nominal atmosphere. EDL failure can occur if three of the Cd values have been overestimated by 20% and one by 10%. EDL failure can occur with a combination of the minimum Cd assumptions and a very low density atmosphere – within the SPICAM data error bars.
The thermal protection system if implemented as designed would provide adequate protection to survive EDL. Degradation of the TPS due to removal of a tile covering an ARM access cover would make a failure of structural or internal components very likely.
A rough analysis has shown that low altitude air pockets can result in premature parachute deployment. The nature and likelihood of such pockets has not been characterized.
A reduced atmospheric density profile impacts on EDL survivability. The onset of EDL problems occurs at densities reduced by 15% of the nominal atmosphere model.
Atmospheric profiles constructed from SPICAM data that result in failure scenarios have been demonstrated. With a SPICAM profile blend back into a nominal atmosphere profile at 10km above the datum airbag inflation could fail to occur in time.
It is clear that uncertainty in atmospheric characteristics leaves scope for a resulting loss of the Beagle 2 mission during EDL.
A conclusion that an atmosphere of reduced density led to the loss of Beagle 2 depends on the acceptance or otherwise of the veracity of atmospheres constructed for modelling in the post operations phase and the currently unexplained MER EDL experience.
37 Beagle-2 Mission Report Post-Operations Analysis
7.1.2.4 VMC Image Analysis Beagle 2 ejection was imaged by the Mars Express VMC camera. Eight images were delivered to the Beagle 2 operations team.
Reference Documents
Conclusions from VMC images, Issue 1, 22nd March 2004, Dr. M.R. Sims BEA2.ICD.00005.S.MMS, Issue 1 Rev0, MEX Lander Delivery Module Beagle 2-ESOC Interface
7.1.2.4.1 Analysis
Images 3 to 6 were analysed in detail by Virtual Analytics Ltd. with information from Beagle 2 design engineers placing the analysis in context. See Annex C for details of the analysis and images.
Analysis concentrated on determining the ejection velocity and solar aspect angle at ejection, identifying a bright image on the probe body in image 3 and a number of background “debris” objects visible in more than one image.
The image analysis made use of the following information.
• Image 3 was taken 67 seconds after ejection, with images 4, 5 and 6 at 50 second intervals. • Beagle 2 subtended 40, 23, 16, 12.5 pixels in images 3,4,5,6 respectively. Typical errors in the image analysis are ±0.5 pixels. • The Beagle 2 heat shield diameter was 934mm (allowing for 10mm of MLI) • Image 3 shows a bright feature within expected shadow on the side of Beagle 2 of image size 2.5 by 3.75 pixels. • Images 3,4,5 show an object apparently receding from Beagle 2. • 40 background objects are seen within the images 3 through 6.
7.1.2.4.2 Findings
Ejection velocity and solar aspect angle
• The distance of Beagle 2 was calculated as 20.54m ±0.26m in Image 3, 35.74m ±0.79m in Image 4, 51.37m ±1.65m in Image 5, and 65.75m ±2.74m in Image 6. • Ejection velocity was calculated at 0.3025 ±0.0083ms-1 (1σ) (excluding systematic errors). Ejection velocity of 0.31ms-1 is defined in the Beagle 2 ESOC ICD. • No data on spin rate could be derived from the VMC images. • A Solar Aspect Angle (SAA) of 133 ±5° with a worst case of ~±10° was estimated from image analysis. Data provided to RAL from Astrium Toulouse estimated SAA to be ~122.5° at time of ejection. ESOC have stated a figure of 124° is appropriate given changes resulting since the original estimate was given. Direct evidence in spacecraft telemetry verifying the pointing of MEX during ejection has not to date been provided to the Beagle 2 team.
Bright Feature in Image 3
• The bright feature appears to be a real object of maximum dimensions 5.8cm by 8.7cm (+~15% for halation). Physical height is estimated to be comparable to the objects linear dimensions. The feature is situated ~0.17m along the surface of rear cover as measured from the rear surface. • The feature position is consistent with the position of an MLI stud. This appearance of the feature is consistent with an object on the MLI, or loose or damaged MLI on the side of Beagle 2.
38 Beagle-2 Mission Report Post-Operations Analysis
Debris Object
• The “debris object” in images 4, 5 and 6 is not an image artefact and reduces in apparent size between images. The object has a consistent size and recedes from the VMC. It is apparently not physically shrinking e.g. sublimating/evaporating in the sunlight. • The distance of the object from the VMC is unknown but the minimum distance is estimated as 970mm ±130mm. • The size of the object is unknown. Size estimates were made for different distances between Mars Express and Beagle 2. These estimates place the size of the object in the range 20mm by 20mm to 60mm by 60mm. However it could be very close to the camera in which case it can be very small, ~2-3mm. • The object may be consistent in size with the feature on the side of Beagle 2; a MLI stud cover of ~20mm across; ice from the +Z surface of Mars Express; or very small debris from the SUEM mechanism/pyro/frangibolts. This latter option is thought to be unlikely as no debris was seen during testing.
Background objects
• The motion of the background objects seen within images 3 through 6 appears to be consistent with a direction of motion away from Mars Express along the Beagle 2 vector of ejection. Assuming these objects are real and not camera noise (cosmic ray hits) and are at a minimum distance (see section 5 above) and occupy a single pixel they have a physical size of >0.94mm and may be consistent with debris or ice from the spacecraft and/or SUEM.
ARM Cover Dis-bond Theory
• The position of Bright Feature in Image 03 is not consistent with being directly over an ARM cover position however it is potentially consistent with a “jet” of gas emerging from a gap at the lower end of an ARM cover position.
7.1.2.4.3 Conclusions
• Beagle 2 was ejected at, or very close to, it’s nominal velocity. • Sun angle is approximately correct but a large error/discrepancy exists with what was expected. This is not surprising given the accuracy expected from the VMC images. • A feature is seen within the shadow of the back cover which appears real and should not be present. This feature appears consistent with an object on the MLI, or loose or damaged MLI, cause unknown. • An object appears to recede from Mars Express apparently following Beagle 2. Its physical size is greater than 2.86mm but may be consistent in size with the feature on the side of Beagle 2 or a MLI stud cover of ~20mm across or even possibly very small debris from the SUEM mechanism/pyro/frangibolts etc. • Small “objects” are seen within the VMC images which may be real and would if real have a minimum physical size of >0.94mm and may be ice from the +Z side of the Mars Express spacecraft.
39 Beagle-2 Mission Report Post-Operations Analysis
7.1.2.5 Aeroshell Release Mechanism Access Cover Dis-bond Investigation The three Aeroshell Release Mechanism Access points on the probe are closed with bonded covers in the final phase of AIV. The possibility was considered that one or more of these covered became dis-bonded during flight. If this proved to be the case, the lack of complete isolation of the probe internal volume from the severe environment during EDLS could have led to a catastrophic failure.
Reference Documents
Arm Cover Theory 1.ppt, Richard Slade, EADS Astrium
7.1.2.5.1 Analysis
The ARM Cover Dis-bond Theory:
• HEPA blocked during launch due to parachute or other material shifting, ice, or damage in the pre- launch over-pressurisation incident. • Vent path through bio-seal blocked due to launch g forces. ARM cover dis-bonds due to pressure differential. • ARM cover dis-bonding damages MLI; MLI protrudes or is even ripped accounting for bright spot and debris in VMC images. • Local poor performance of MLI results in the observed ARM heater on through cruise rather than cycling.
Analysis carried out by EADS Astrium:
Evidence was collated from AIV records and design documentation. Pressure differential regime during launch was estimated. The strength of the ARM cover bond was modelled. Predictions were made on the possibility of ARM cover dis-bond for different scenarios.
ARM Cover Construction
ARM Cover is a four-ply carbon fibre reinforced plastic lozenge shaped: 75mm long x 33mm wide. The cover is surfaced with bonded Norcoat Liege thermal protection. The ARM Cover is bonded to the ARM access point with one-part silicone adhesive.
7.1.2.5.2 Findings
Evidence from AIV
• The original covers were removed following an over-pressurisation event during the FM vibration test in order to remove Front Shield.
• No record of the actual adhesive used was kept in the Working Log for either the first or repeat bonding. Time-stamped AAF video record shows correct primer and adhesive used, correct cure time and confirms CFRP plates actually bonded. The possibility of silicone contamination degrading bond strength (e.g. if epoxy was incorrectly used for 2nd bond) is therefore discounted.
40 Beagle-2 Mission Report Post-Operations Analysis
Aeroshell venting
• Internal pressures at which the aeroshell vents via different paths, if the HEPA filter is blocked, was investigated analytically. Table 7.3 shows the results.
Pressure differential / bar Event
Estimated aeroshell self vented via bio-seal during over- 0.19 to 0.28 pressurisation event
Maximum estimated pressure at which aeroshell self vents via 0.29 bio-seal during launch conditions.
Parachute Deployment Device (PDD) break-out patch predicted 0.54 to blow off
0.46 to 0.56 Front Shield structure predicted to fail
ARM Cover Design and HEPA Filter Design Case (aeroshell 0.055 venting normally via HEPA). However original ARM covers saw pressure differentials of 0.19 to 0.28 bar without debonding.
Table 7.3 - Overpressure Events
• Mechanism by which bioseal is “blocked” during launch is not evidenced.
Predicted ARM Cover Strength
Table 7.4 summarizes the results of analysis into pressures at which ARM Cover dis-bond mechanisms occur.
Pressure differential / bar Event
ARM Cover Dis-bonds 51 to 67 Simple calculation ignoring “peel” or “cleavage” failure ARM Cover Dis-bonds 28 Peel calculation including best case values for material properties (adhesive and cover structures). ARM Cover Dis-bonds 2.38 Peel calculation including worst Case values for material properties (adhesive and cover structures). ARM Cover Dis-bonds 0.42 Peel calculation including worst Case values for material properties and thermal protection layer completely dis-bonded.
Table 7.4 - Pressures for ARM Cover Dis-bond Events
• Even in the very conservative case the ARM covers remain bonded at pressures below those at which the aeroshell is expected to self vent via the bio-seal.
• The higher pressures required to dis-bond the covers could be caused by an ammonia leak from gas generator or premature airbag inflation. However at these pressure differentials the Front Shield structure is expected to fail long before the ARM Covers dis-bond
41 Beagle-2 Mission Report Post-Operations Analysis
7.1.2.5.3 Conclusions
The pressure differential required to dis-bond the ARM covers is higher than bio-seal self-venting pressure differential under launch acceleration. The ARM covers should not dis-bond during launch depressurisation. If the bio-seal cannot vent for some other reason, the Front Shield structure is expected to fail before the ARM covers dis-bond.
7.1.2.6 Electronics Bond Wires The Beagle 2 team was presented with evidence from BAe Systems unrelated to the mission, but detailing their problematic experience with testing a military decoy system. Failures to correctly deploy parachutes had been traced to shock induced software resets. Similarities between the system and the Beagle 2 EDLS led to the hypothesis that the Beagle 2 electronics may have exhibited the same problem.
The decoy has a mass of about 60kg propelled by a small rocket motor. A gas generator separates the motor from the main payload under software control. A drogue chute is deployed followed a few seconds later by a main chute. During test firings it was observed that the software occasionally failed to deploy the drogue and chute. The failures were attributable to a software reset co-incident with the firing of the gas generator. The cause of the reset was identified and reproduced on the ground. Further diagnostics revealed that a memory chip was the root cause. This chip (nominally specified to 1000G) malfunctioned for a few µs when subjected to a sudden shock in the region of 300-400G. Internally its memory address wires could touch causing corruption which re-set the running program with no permanent damage. After the software reset the software was unable to recover and so it did not complete the timing sequence of the drogue – parachute deployment.
7.1.2.6.1 Analysis
Beagle 2 shock testing of the Flight Model before launch focussed on survivability, requiring function pre- and post-shock to be demonstrated. The test campaign did not include shock testing while the processor was active and running software.
X-rays of the original tested memory chips as used on Beagle 2 were available for analysis.
During the post operations phase a shock test was carried out with the Beagle 2 Electronic Test Model at the LOCC. The ETM includes a flight representative processor and memory. The system was subjected to a rough approximation of the predicted shock regime while Probe Software was running in EDL mode.
7.1.2.6.2 Findings
Visible and X-ray inspection of the beagle 2 memory chips showed that the bond wires were very short and that touching during a shock was impossible without serious damage to the device.
No interruption to the operation of Probe Software could be induced during the ETM shock test.
7.1.2.6.3 Conclusion
The tests indicate that the Beagle 2 memory chips and associated electronics are robust in a high shock regime. A shock induced reset of PSW is not considered a likely failure mode.
42 Beagle-2 Mission Report Post-Operations Analysis
7.1.2.7 Out-gassing and SUEM Icing During cruise phase Mars Express experienced an unexpected delta-V. The cause was positively attributed to out-gassing from the +Z face.
Consideration was given to whether the out-gassing provided any evidence to support possible failure modes of Beagle 2, in particular the icing of the SUEM or a leak from the Airbag Gassing System (AGS).
7.1.2.7.1 Analysis
MEX Delta-V
On 5th June 2003 A delta-V of 2.5 mm/s was experienced by Mars Express following the slew for the earth moon picture during which the +Z face of the spacecraft was exposed to the sun. A delta-V was experienced during a number of subsequent periods where the +Z face was illuminated as summarised in Table 7.5. Data was kindly provided by ESA and ESOC.
Sun on Angle Delta-V Earth to Date Event Delta V +Z from +Z Spacecraft mm/s
04/06/03 STR Checkout No Yes 85
04/06/03 Test OCM No Yes 85
05/06/03 TCM-1 Yes Yes 70 2.5
06/06/03 VMC Picture Yes Yes 60
09/06/03 STR & SA Calibration No Yes 85
03/07/03 HRSC Yes Yes 65 3.5
05/09/03 1st Slew Yes Yes 0.15
2nd Slew No Yes 0
09/09/03 Slew Yes Yes 1.75
11/09/03 Slew Yes Yes 0.23
Melacomm 18/09/03 Yes Yes 1.3 /Stanford
2nd Slew No? Yes
Table 7.5 - Delta-V Outgassing Events
Total delta-V seen between June and December was 9.8mm/s.
Detectable out-gassing and a resultant delta V from the +Z face occurred on many but not all occasions it was illuminated by sunlight.
SUEM Icing
In the period preceding ejection, the possibility that the function of the SUEM could be impaired by icing was raised. Evidence from hypothesized out gassing were considered and the merits of warming the SUEM to remove ice were debated.
43 Beagle-2 Mission Report Post-Operations Analysis
7.1.2.7.2 Findings
Evidence from AIV indicated that the water content within Beagle 2 on arrival at the launch site was low however water would have been absorbed back into structure during pre-launch operations. MLI on the MEX +Z face and Beagle 2 would have been very cold as the face is in shadow throughout most of cruise therefore the +Z face will act as a getter for spacecraft out-gassing products. Considerable out-gassing may occur when the face is illuminated by the Sun and this provides the possible source of delta-V seen by MEX.
The estimation of percentage of out-gassing onto +Z face from a source in Beagle 2 was 50%. Calculation by ESA ESTEC showed that 115 grams of water sublimating from a surface would produce 9.8mm/s delta-V. Review of materials used in Beagle 2 shows that the water content at launch of at least 300 grams. Tiles are the main contributor but air bags and parachute are significant. 50% of the water is expected to be out- gassed in the first 30 days but this rate is slowed down with low temperatures.
Water is the strongest candidate for out-gassing from MEX and Beagle, there is no evidence supporting the possibility that ammonia was released.
Warming the SUEM structure was determined to be a prolonged and difficult task. On weight of evidence it was decided not to attempt an operation to de-ice the SUEM prior to ejection.
Debris seen in VMC images could be debris from ice on the SUEM or the +Z surface of Mars Express.
There was a small delta-V seen in December after the release of Beagle possibly indicating remaining ice in the area where Beagle was attached to MEX.
7.1.2.7.3 Conclusions
The delta V experienced by MEX is attributed to the out-gassing of water, and possible organics, from the +Z face. The investigation of out-gassing provides no evidence indicating a leak of ammonia from Beagle 2.
The SUEM may have experienced some icing, however as described in section 7.1.3 the Beagle 2 ejection was nominal and no impact to SUEM performance is indicated.
44 Beagle-2 Mission Report Post-Operations Analysis
7.1.2.8 EDLS Events Parachute Deployment, Heatshield Separation and Landing
References
Main Parachute Inflation Loads Software Detailed Definition Document - BEA2.SOF.00008.S.MMS Iss.0A Rev. D2
7.1.2.8.1 Analysis
The following elements of the EDL Modelling were revisited.
• Heatshield and Rear Cover Separation • Ballistic Coefficients of Parachute, Heatshield and Rear Cover • Main Parachute Material Porosity • Main Parachute / Airbag Re-contact After First Bounce
A full description of the analysis is given in Annex I
7.1.2.8.2 Findings
The ballistic drag coefficients applicable during the separation events were shown to be sufficiently different to ensure clean separations would occur with high probabilities. Ballistic coefficients calculated for the main parachute result in acceptable deployment strain.
The material comprising the canopy of the Main Parachute would be effectively non-porous due to the weave dimensions being smaller than the Martian upper atmosphere mean-free-path length.
Simple first bounce analysis demonstrated certain conditions whereby re-contact was possible, however the analysis makes the unlikely assumption that both airbag and parachute continue in same straight path.
7.1.2.8.3 Conclusions
Review of separation events provides no indication of a mechanism which could lead to EDL failure. Simple analysis of first bounce suggests an additional 10m on the strop length would have reduced the probability of re-contact.
45 Beagle-2 Mission Report Post-Operations Analysis
7.1.3 Ejection
References
Spacecraft Rates at Beagle 2 Ejection and Entry Trajectory Predictions, Input to the Beagle 2 Inquiry Commission, 12 February 2004
7.1.3.1.1 Analysis
The Mars Express operations team analysed the reaction of the MEX AOCS subsystem to the ejection of Beagle 2.
Fluid Gravity Engineering revisited the Monte Carlo determination of the landing ellipse based on the entry interface state and covariance provided by ESOC post ejection.
7.1.3.1.2 Findings
The reported angle between the sun and the MEX +Z axis (equivalent to Beagle 2 +Z axis) = 124°.
Maximum Rate Maximum Deflection Axis Cause Radians / Second Radians X +0.00332572 + 0.00885081 rotational component cross coupled from Z Y – 0.00004765 -0.00000544 and X axis ejection velocity Z +0.00886909 +0.11285782 component
Table 7.6 - Rates measured by gyroscopes at ejection and resulting off-pointing
FGE determined a small drift in landing site nominal from 11.6N 90.74E to 11.53N and 90.53E. The entry angle predicted was -16.6° with 3σ = 0.14. This compares to a requirement of -16.5° and 3σ of 1.0.
7.1.3.1.3 Conclusions
Ejection was nominal with slight over-performance of 1.29% with uncertainty of ±0.5%. The FGE analysis concludes that Beagle 2 was perfectly targeted by Mars Express with only a very small error in entry parameters and landing location.
46 Beagle-2 Mission Report Post-Operations Analysis
7.1.4 Design Failure Modes In order to track potential failure modes a spreadsheet format was used. An original version was produced by Dr. J. Clemmet (Beagle 2 Chief Engineer 1999 to 2003) to which other comments and work was added by all members of the Beagle 2 operations team. The final version was produced and edited by the Mission Manager Dr. M.R. Sims. (See section 7.3).
The spreadsheet attempts to list all potential failure modes for Beagle 2. For any given potential primary cause the various variants and consequences are given. Mitigation steps taken and any relevant comments are given were appropriate.
For each mode the Beagle 2 team assigned two parameters on the basis of analysis and engineering judgement. The first was look at any possible evidence for or against a particular mode and assign a weak or strong label if possible. For some modes no data is available for a variety of reasons so an unknown label is attached. For each mode likelihood or probability e.g. ‘Low’ of this being the cause is assigned based on knowledge of the design, test and mission. This of course is subjective and represents a collective view. Consequently others may reclassify the various modes in terms of different values or even perhaps via different parameters as they see fit. It is believed however that even this process is unlikely to substantially change the list of possible causes. For some modes additional important comments are given separately.
Only failure modes with some evidence for and those with a unknown label have been considered as likely/possible failures modes although any mode on the list could be the cause, given the complete lack of data since ejection.
The modes split into two areas: those that occur during entry descent and landing, and those that occur on the surface.
Following classification potential failure mechanisms fall into two classes: those with some evidence even though it may be weak, and those which have an unknown label. These are labelled using colours yellow and grey respectively in the spreadsheet. The summary of likely failure mechanisms in the Beagle 2 team’s view is given in the next section.
47 Beagle-2 Mission Report Post-Operations Analysis
This page left intentionally blank.
48 Beagle-2 Mission Report Post-Operations Analysis
7.2 Summary of Failure Analysis Findings
It is not possible given the circumstances to define a most likely failure mode. It is however very probable that failure occurred during entry, descent, and landing (EDL) or surface deployment. Given the complex nature of the former process this is perhaps the phase most likely to have caused the subsequent loss of communication and the mission.
The following potential causes have been identified; provisos where appropriate will be given. Some of the declared modes also cover and closely relate to modes where evidence for and against is unknown or not available to the Beagle 2 team these are integrated into the summary mode and are details for these are given in italics.
The following are not in priority or probability order.
Electronics too cold for start up after coast due to MLI damage during mission • Evidence for: bright spot on Beagle 2 in VMC Image. • Probability very low, thermal modelling shows unlikely to be significant.
Lander electronics malfunction and failure to operate one or more systems during EDL. • Evidence for: Electrostatic discharge (ESD) damage during final phases of assembly integration resorted in several components being replaced. • Probability low, no degradation of performance seen during cruise • Evidence for: Lack of comprehensive shock tests • Probability low, system robust with restart capability and limited shock testing in 2004 shows no problem • Other related modes: ESD during entry, Points weld on critical relay, a random component failure in electronics* probabilities thought to be very low. (note * not included on spreadsheet but effects whole sequence)
Excessive velocity during entry due to unusual atmospheric conditions resulting in a variety of possible failures (e.g. RAT not triggered, hard landing etc.) because terminal velocity on parachute was not reached. • Evidence for: SPICAM data shows low density at 20-40km above surface. • Probability low / medium, a large deviation from standard models of the atmosphere is required however this is not impossible. The SPICAM data is at variance with atmospheric data derived by temperature measurements but may be compatible with extremes that can be derived from the temperature data sets. • Conclusion: A better characterisation of atmosphere is required via a variety of methods for future missions.
Front heatshield breakup or aerodynamics corrupted leading to failure at hypersonic part of entry. • Evidence for: Heatshield damaged during over-pressurisation experienced in final vibration test. • Probability very low, all data shows shield was intact and correct shape prior to launch
Parachute envelopes airbags after first bounce leading to problems in airbag release etc. • Evidence for: rebound analysis not repeated following parachute redesign (latter required to limit landing velocity following airbag development problems). • Probability low/medium?, simple spreadsheet models show that it may be a concern however parachute motion and collapse following release requires extensive modelling to ascertain true probability of such a scenario and parachute velocity vector must closely match airbag vector for a collision to occur.
Airbags fail on impact or during subsequent bounces due to high velocity or are punctured by rocks • Evidence for: Limited testing of final design due to development problems • Probability low to medium dependant on terrain, winds and atmosphere. • Note this mode relates to the atmospheric density mode above but is separated due to the differing evidence and variants
49 Beagle-2 Mission Report Post-Operations Analysis
The following additional modes where evidence for and against is unknown or not available need also to be considered at the time of writing.
Beagle 2 was ejected in the wrong direction. • No definitive evidence available from spacecraft telemetry (i.e. matched startracker patterns) and no confirmed error budget allowing for thermo-elastic distortion has been provided to Beagle 2 by ESA. • Probability assigned extremely low • Not thought to be a credible failure scenario but cannot be definitively ruled out at this time.
Thermal protection tiles detachment from aeroshell during entry due to poor bonding. • Probability assigned very low.
Parachute(s) inflation problems. • Probability assigned very low due to testing carried out however no high altitude test conducted or was in fact deemed and is thought necessary. • Probability assigned low.
Airbag/ airbag gassing system leak at connection point resulting in incomplete inflation or failure following landing. • Impossible to test on assembled system. Probability assigned low.
Airbag jettison problems or problem resulting in subsequent damage to lander as part of release process. • Probabilities assigned low.
Damage to lid or clampband following impact of lander with ground causing consequent problems for release and deployment of lid and solar panels. • This is hard to assess and no lander drop test was conducted due to schedule, cost and equipment availability issues. • Probabilities assigned low/medium.
Antenna damaged on impact, see above. • Probability assigned low.
Return or forward link failure causing an unknown protocol problem, random component failure. • Probability very low, intensive testing and use during cruise tends to mitigate this.
In conclusion a large number of failure modes are possible. A limited subset have been identified by the Beagle 2 team as being probable causes. However it should be noted that a mode currently eliminated by the Beagle 2 team may have been the cause of the mission loss.
50 Beagle-2 Mission Report Failure Tree Summary
7.3 Beagle 2 Failure Modes: Identification and Analysis
Based upon Issue 9 of the Failure Modes spreadsheet, 7th May 2004.
Compiled by J.F. Clemmet and M.R. Sims
Notes
1. Failure modes are presented in a nominal chronological order, i.e. they are NOT in order of likelihood 2. Individual failure modes may have more than one possible primary cause 3. A primary cause may itself have variants 4. Failure modes are presented in a nominal mission chronological order 5. Failure Modes evidence for against and likelihood estimated 6. Failure modes classified as follows: a. Evidence For (strong, medium) RED b. Unknown or possible evidence GREY c. Evidence For (weak) YELLOW 7. In some failure cases mitigation notes are included. In other cases tests were conducted or Beagle 2 was designed and/or qualified to survive such conditions.
51 Beagle-2 Mission Report Failure Tree Summary
EDLS PHASE
7.3.1 Launch Phase Potential Likelihood Failure Mode Primary Cause Variant Consequence Evidence Based on For/Against Experience EDLS 0.0 Launch Phase 0.1 primary structure failure launch loads vibration Against Low mitigation, actions & comments Comments extensive ground testing and no abnormal reports from Mex post launch Noting: Pre-launch Over pressurisation. MeX AOCS shows no evidence depressurisati Against Low on extensive ground testing and no abnormal reports from Mex post launch Noting: Pre-launch Over pressurisation. MeX AOCS shows no evidence 0.2 secondary structure failure launch loads vibration Low
depressurisati see Item 5.5 Against Low on (weak) Depressurisation should occur through HEPA and bioseal. VMC image 3 could be interpreted to show evidence of ARM access cover blowout (illuminated zone within shadow suggesting malformed MLI blanket). Analysis shows that this failure is high unlikely given the construction and attachment of the access panels (see 9.2f and 5.5) and bioseal. Airbag is tightly packed against lander and retained by bag. Main chute material is sufficiently porous and should not cause significant flow resistance. Could flow through HEPA have been sufficiently degraded by accidental overpressurisation during final vibration test? Note: MRB minutes MIN/BEA2/0012/2003 do not address HEPA damage.
52 Beagle-2 Mission Report Failure Tree Summary
7.3.2 Cruise Phase 1.0 During Cruise Phase 1.1 Internal temperatures outside limits a flight build characteristics not as component failure Against Low during Cruise predicted (eg APS, ARMs, PDD, (strong) etc. With the exception of the APS, all temperatures monitored during Cruise checkouts appeared close to expectations; ARM heater circuit activity near 100% duty cycle but adjacent AD590 temperature in range -10 to -8C, i.e ARM temperature OK but more heat loss than modelled.(ARM limit -50C - good margin). ARM heater on-time higher than expected. Battery heater operation as predicted. Back Cover endplate heater initially less active. APS PCB temperatures during prolonged checkouts exceeded the set warning levels In qual temp testing operated at +65C and expected to be able function (within spec) to at least +75C. b outgassing products corrupting Against Low alpha/epsilon on aeroshell MLI (strong) MLI alpha/epsilon ratio is very sensitive to change over the contaminant thickness range 0.1 to 10 micron. Cruise Phase temperatures compare favourably with prediction. Contamination is therefore most unlikely.
7.3.3 Ejection Characteristics 2.0 Ejection Characteristics 2.1 incorrect delta-V from SUEM off target Against Low (strong) Initial assessment by ESOC suggests deltaV within 1.3% of nominal; from SUEM testing, including representative test pieces, finalising on a value was not easy, in particular settling on a coeff friction value in vacuum; finally values of 0.31m/s and 14.2rpm both +/- 5% were adopted. VMC image analysis indicates ejection velocity 0.3025+-0.0083m/s (1 sigma excluding systematic errors) i.e. within spec. 2.2 Off target a incorrect navigation trajectory error Against Low See 2.2d Noting: All MeX TCM's MOI's nominal b thermo-elastic distortion of Mex landing terrain Unknown Very Low bus characteristic 'unfriendly' Earlier comment from Beagle2 project to Mex querying thermoelastic analysis remains unanswered c thermo-elastic distortion of Mex off target Unknown Very Low star trackers Beagle2 assessment of star tracker mispointing 0.15° resulting in ~5km off target
53 Beagle-2 Mission Report Failure Tree Summary
d Mex attitude error Unknown Very Low Note although predicted star tracker patterns are available for ejection no actual patterns are available from telemetry Noting: VMC analysis indicates SAA of 133±5° c.f. spec 124° e entry angle outside spec too shallow high soak back Against Low (currently weak) ESOC report received: ESOC email states eject direction uncertainty 0.5° (3σ) too steep burn up Against Low (currently weak)
landing terrain airbag failure Against Low too severe (currently weak) Too steep may prevent reaching terminal velocity on parachute, RAT initialisation failure due to timing lander IPS failure on Against Low final fall (currently weak)
f trajectory perturbations from Against Phobos and Demos (strong) Not Included however nearest moon Phobos at 6869km during coast Effect from moons needs to be included in navigation for future missions 2.3 Angle of attack a thermo-elastic distortion of MEx unstable entry Unknown Very Low bus see 2.2 b thermo-elastic distortion of Mex Unknown Very Low star trackers earlier comment from Beagle2 project to Mex queryng thermoelastic analysis remains unanswered c Mex attitude error Unknown Very Low ESOC report received:ESOC email states eject direction uncertainty 0.5° (3σ) Noting: VMC analysis indicates SAA of 133±5 c.f. ESOC report of 124°. Pre- launch data indicated 122.5°.
54 Beagle-2 Mission Report Failure Tree Summary
d solar pressure Against Very Low (strong) VMC image analysis indicates ~133±5°. Noting: Solar pressure allowed for in navigation e breakup of extended TPS edge Against Low tiles (weak) Cross section unchanged from baseline; supported by extended backcover when under dynamic pressure; initial assessment (FGE) is that there is sufficient margin to accommodate local breakups w.r.t angle of attack but may present local heating problem on back cover f tip off due to Back Cover MLI/MLI Against Low collar frozen contact Possible debris noted in Mex VMC image sequence. Could this be a piece of MLI collar being torn off having been frozen to Noting: No evidence in BackCover MLI or ice from spacecraft or frangibolt debris. VMC images 2.4 low spin rate - unstable entry incorrect SUEM function unstable entry Against Low (strong) ESOC report received; from SUEM testing, including representative test pieces, finalising of a value was not easy, in Noting: VMC and ESOC particular settling on a coeff friction value in vacuum; finally values of 0.31m/s and 14.2rpm both +/- 5% were adopted delta v are good and spin rate related
7.3.4 Coast Phase 3.0 During Coast Phase 3.1 excessive cold internal temperature a MLI damage during ejection eg frozen to front shield start For (weak) Very Low during coast MLI collar temperature at TOA excessive Possible debris noted in Mex VMC image sequence. See 2.3f Noting: VMC shows a "bump" but MLI present. Local damage unlikely to be thermally significant b MLI not detaching from Back detaches after pilotchute malfunction Low Cover during entry - or delayed Pilot chute inflation
detaches no concern Low before PDD
55 Beagle-2 Mission Report Failure Tree Summary
does not possible corruption of Low detach airflow to pilot
c outgassing products corrupting Probe becomes too Against Low alpha/epsilon on aeroshell MLI cold resulting in frozen battery Degradation in MLI alpha/epsilon could lead to serious reduction in internal temperatures with frozen battery as worst case. As explained in 1.1b, Cruise Phase contamination is most unlikely and therefore also very unlikely for Coast. 3.2 wakeup timer malfunction Probe becomes too Against Low cold resulting in a failure of electronics to wake up or wakeup too late Timer setting and counter operation and rate checked during Cruise and correct countdown confirmed - but actual wakeup not tested due to ESOC safety concerns. FM tested during AIV & numerously tested on GTM
7.3.5 EDLS – General 4.0 EDLS – General 4.1 lander electronics malfunction a ESD degradation damage during For (weak) Low build All multiplexers on CEP board, except Return MUX, replaced during AIV at AAF after ESD, none replaced on CEM - potential Noting: AIV ESD damaged for slow degradation therefore remained on those not replaced, but no sign of any such degradation visible in Cruise TM data components replaced and no evidence seen during cruise b ESD damage during Cruise Against Low No sign of any such degradation visible in Cruise TM data Noting: No anomalies seen during cruise that can be associated with a ESD cause 4.2 wiring connection errors Untested H/W-S/W interfaces on non-specific operational failure Against Low FM lander/probe (weak) Note: accelerometers were tested during AIV Noting: AIV testing verified all testable EDLS electrical system aspects
56 Beagle-2 Mission Report Failure Tree Summary
4.3 excessive velocity throughout descent reduced atmospheric density due high velocity at key For (weak) Low/ to high temperature events Medium? Entry & descent analysis comparing low dust, MGS & Viking high dust scenarios have not shown any significant sensitivity - Note: Low density a 15% reduction in density is required to cause insufficient time for RAT operation. With Viking dust atmosphere (a hot case) atmosphere means Beagle require a forebody Cd at -125 & parachute Cd at -20% for no time for RAT operation. Assuming hypersonic Cd -10%, 2 would land down range transonic -20% & parachute -20% would give higher dynamic pressure (75Pa) & 17.9m/s descent rate - making likely to fail. as did both MERs. Atmospheric density profile data has been received from Mex (SPICAM) and ODYSSEY (THEMIS). SPICAM data indicates Assuming SPICAM type a possible low density at 20-40km compared to standard models however the measurement errors are very large at low atmosphere landing point altitudes. A number of atmospheric models are consistent with THEMIS data. All possible model variations consistent with would be shifted ~10km the THEMIS temperature data have not been provided by NASA. It is possible that SPICAM data may be consistent with one east. such model admittedly at a very high deviation to the mean. A low density continuing down to 10km above datum would be fatal to Beagle 2. There is some evidence from MER in terms of deviation of density away from models and currently unexplained late opening of the parachute. key events at lower For (weak) altitudes
4.4 aerodynamics malfunction a incorrect cone angle selected for Against Low Front Shield Beagle2 adopted 60deg half cone from Huygens - but JPL (MER, etc) used 70deg. Entry analysis shows 60 deg half cone Noting: All analysis shows it should work should be OK b aerodynamics modelling not fully Against accounting for all significant (weak) effects JPL report that, for both MER Spirit & Opportunity, the main (supersonic) parachute inflation was later than expected after Noting: Could Cd's be compensating for local atmosphere characteristics by s/w parameter updates (Spirit +1mile, Opportunity +9sec). Further wrong and allowance was reports indicate that it is within modelled atmospheric parameters/variation. only made for atmospheric gravity waves at relatively large scales (>km)
7.3.6 EDLS – Ballistic Entry 5.0 EDLS - Ballistic Entry 5.1 Front heatshield structural failure overpressure during final testing break up on entry For (weak) Very Low Predicted overpressure less than structural failure limit. See analysis re ARM covers 5.2 Front heatshield overheating a TPS undersized or damaged break up on entry Against Low FGE has investigated the effect of 2mm roughness with full turbulent flow, internal temperatures still appear acceptable Noting: Analysis shows margins
57 Beagle-2 Mission Report Failure Tree Summary
high internal Against Low temperatures Noting: Analysis shows margins b excessive dust erosion atmospheric Against Low conditions Noting: Analysis shows margins c corrupted aerodynamics front shield For (weak) Very Low misshapen from overpressure Front shield shape measured following incident and found to have recovered poor TPS tile geometry
5.3 Back Cover overheating TPS undersized break up on entry Against Low see 2.3e & 5.2a Noting: Analysis shows margins high internal Against Low temperatures Noting: Analysis shows margins 5.4 Plasma ESD induce damage a leakage at umbilicals (G&H and lost electronics Against Low Cinch connectors) Cinch connectors on lander side designed for deadfacing b charge build up inside aeroshell Unknown Low
5.5 Breach of Aeroshell a structural failures ingress of hot gases; Unknown Low excessive internal b ARM access cover leakage temperatures, material Against Low damage, loss of functions
58 Beagle-2 Mission Report Failure Tree Summary
Mex VMC image 3 shows illuminated area within shadow; this may correlate with an ARM access cover location or MLI studs; in addition apparently a possible substantial sized debris item nominally follows the probe trajectory on separation [these observations are based upon certain assumptions]; other possible particulate debris is also apparent in the image sequence. This other and the large orbit debris is thought to be ice from the +Z axis of Mars Express. After further analysis it is unlikely that an ARM access cover (cfrp skin &/or the small TPS tile) could have failed under internal pressure during launch depressurisation or via AGS NH3 leakage due to bond strength and internal pressures that may be generated. Structural failure of the front shield would probably occur first and consequently this mode is deemed highly unlikely. See 9.2f. c ARM access cover blowout launch ingress of hot gases; Against Low depressurisati excessive internal on temperatures, material See above note for (b) damage, loss of AGS functions Against Low leakage/failure See above note for (b) d TPS tile detachment poor bond excessive CFRP Unknown Very Low temperature(>500C) leading to structural failure & hot gas ingress
aerodynamic excessive CFRP Against Low erosion temperature(>500C) leading to structural failure & hot gas ingress Noting: Analysis shows margins
59 Beagle-2 Mission Report Failure Tree Summary
7.3.7 EDLS Algorithm 6.0 EDLS Algorithm - General 6.1 Pyrotechnic Initiator failure faulty batch no pyro activity Against Low Serial numbers recovered and redundant pyro initiators included on all pyro systems. 6.2 Pyro battery relay failure a wiring reversed Against Low (strong) Extensive GTM & FM testing mitigates b incorrect command Against Very Low (strong) Report from Logica indicates correct command sent. This is confirmed via H/W S/W IDD. c points welded during power surge Pyro battery & Unknown Very Low on contact operation avionics power supply not isolated from each other - pyro commanding could then affect CEP power supply
6.3 EDLS Algorithm failure a incorrect parameter values Against Very Low assigned (strong) Extensive testing on ground through simulation at Logica b processor reset due to high shock recovered PSW re-establishes Unknown Very Low sequence by reference to context data, could cause delay of up to 2sec in subsequent events Not considered to be a problem - PDD firing timing is the most critical but being the first cause event would not itself suffer. Noting: Recovery in Xtal Oscillator may be susceptible to very high shock environments. See permanent algorithm failure below. sufficient time not fatal. Recovery expected to be within time.
60 Beagle-2 Mission Report Failure Tree Summary
permanent algorithm failure Against Low (strong) BAe Systems have provided details relating to similar failure mode seen during decoy trails due to fault propagation in Noting: DPA's and shock memory chip; samples from Beagle2 flight batch have been subjected to X-ray & decapping for comparison with type used tests show not permanent by BAe. Bond wires within Beagle 2 chips found to be very short and failure of memory chips assumed to be unlikely under failures high shock environments. Crystal Oscillator may be susceptible to very high shock environments however initial shock tests of a running CEP board show no problems. 6.4 accelerometer failure Against Very Low (strong) Sensible near zero readings through Cruise checkouts; MODE2 where both accelerometers was used during entry data from both accelerometers is used and triggering can occur from either one i.e. redundancy used. MODE2 tested on ground in PSW S/W tests. 6.5 APS off power glitch, e.g. due to pyro relay await short or long Against Low operation timer to switch back (medium) on crash if occurs during entry Note: Pyro relay not tested during early cruise due to ESA safety concerns. Test proposed early December but cancelled due Noting: tested on ground to spacecraft operational constraints 6.6 APS malfunction damage/degraded during Cruise higher crash if occurs during Against Very Low temperatures entry (strong) than expected during extended checkouts Unlikely as APS always within operational limits, see 1.1a
7.3.8 Pilot Chute Phase 7.0 Pilot Chute Phase 7.1 PDD operation failure no command, mechanical, but electrically no pilot chute Unknown Very Low electrical…. redundant deployment battery current anomaly observed; however this earthing problem within Beagle 2 would not cause failure of PDD operation Noting: electrical and command tested pre-launch loss of Against Low connections at (medium) Cinch never seen during ground testing, including several qual vib tests
61 Beagle-2 Mission Report Failure Tree Summary
7.2 Pilot chute failure to inflate material cold welded during Cruise no deceleration Against Low (strong) Considered unlikely as materials similar to main chute 7.3 Pilot Chute prolonged inflation period characteristics not as expected in Unknown Very Low Mars atmosphere MER reported problems with disc gap band inflation which involved some failures to inflate which was resolved by changes in central hole diameter. Pilot chute on Beagle 2 was subjected to limited testing due to similarity to Huygens and is a disc gap band parachute. 7.4 Pilot chute failure to fully inflate atmospheric conditions inadequate Unknown Very Low deceleration See 7.3 7.5 Pilot chute structural failure a atmospheric pressure conditions no deceleration For (weak) Low see 4.3 b above M1.5 reduced For (weak) Low atmospheric density due to high temperature
c dust overloading delta mass Against Low
reduced Against Low porosity/higher drag Noting: Analysis assumed no porosity at Martian conditions
62 Beagle-2 Mission Report Failure Tree Summary
7.3.9 Main Chute Phase 8.0 Main Chute Phase 8.1 ARM operation failure a no command, mechanical, but electrically no aeroshell Unknown Very Low electrical…. redundant separation see 7.1 Noting: electrical and commands tested pre- launch b outside temperature limits during Against Low Cruise or Coast (strong) see 1.1 & 3.1 8.2 Main chute failure to inflate a no aeroshell separation no deceleration Unknown Very Low
b material cold welded during Cruise Against Very Low (strong) ESA have a report covering long term storage of nylon, Kevlar, (as used on Beagle2 but not Spectra) under cryogenic & vacuum conditions. The document reports on materials strengths and pull-out forces. Materials were packed to significantly higher densities than used on Beagle2. No unacceptable degradations were recorded, relevant to Beagle2. Additional tests performed on Spectra with no degradation problems seen. Design allowables allow for any degradation. 8.3 Main Chute prolonged inflation period characteristics not as expected in Against Low Mars atmosphere Ring sail used with good inflation characteristics 8.4 Main chute failure to fully inflate poor back cover separation/flight inadequate declaration Against Low path See 8.3 and 4.3 (latter re density fluctuations in atmosphere) 8.5 Main chute structural failure a atmospheric pressure conditions no deceleration Against Low see 4.3 b above M0.4(tbc) reduced Against Low atmospheric density due to high temperature ONGOING ANALYSIS c dust overloading delta mass Against Very Low
63 Beagle-2 Mission Report Failure Tree Summary
reduced Against Very Low porosity/higher drag Mean free path length of the Martian atmosphere comparable to canopy material porosity, results in an effectively non- Noting: Analysis assumed porous parachute. Possible that the presence of dust may result in apparent increase in density leading to Newtonian flow no porosity at Martian and higher dynamic pressure. This has not been analysed and would require the development of a new model. Parachute conditions aerodynamic tests were conducted using coated fabric to simulate non-porous performance. e long term degradation resulting Against Very Low from gamma irradiation & vacuum (strong) strip processes BEA2.RPT.00051 & 00077 provide data on material test samples and QM parachute samples after vacuum stripping and gamma irradiation and testing. No discernable degradation was noted from the QM samples. No mechanism is foreseen that could lead to a long term slow degradation. The materials used are uncoated and chemically stable. Gamma radiation attacks cross-linking & does not result in residual radio-activity. Nylon and Spectra are not highly cross-linked and are less susceptible to initial exposure degradation than highly cross-linked materials such as Kevlar, confined to reinforcements were there are high RFs. Fusion in fabrics is not an issue since no coating are employed and surface energy is low. Similar arguments apply to the vacuum stripping process. 8.6 heatshield fails to diverge on flight path a mechanical hang up Against Low see also 8.7 b characteristics not as expected in Against Low Mars atmosphere (weak) No mechanism is foreseen that could lead to a long term slow degradation. The materials used are uncoated and chemically Noting: Analysis shows stable. Gamma radiation attacks cross-linking but does not result in residual radio-activity. Nylon and Spectra are not highly divergence cross-linked and are therefore less susceptible to initial exposure degradation than highly cross-linked materials such as Kevlar. Parachute seam threads, risers and strop are all Spectra, Kevlar is confined to reinforcements were there are high RFs. Warp/weft fusion in fabrics is not an issue since no coating are employed and surface energy is low. Similar arguments apply to lack of susceptibility to the vacuum stripping process. Finally packing pressures in Beagle2 are low. 8.7 Back cover & Main Chute collide incorrect ballistic coefficient Snagging, poor Against Low differentials inflation, tear… (weak) Worst case modelling showed separation should occur; hardware feature (pull chord) included to perturb back cover flight Noting: Analysis shows path to further improve margins. Analysis was not repeated with the extended aeroshell geometry resulting from the airbag divergence packing problem. 8.8 Aerodynamic performance not as Against Low expected Aerodynamic drop tests were designed to generate the correct velocity at bag strip and the correct dynamic pressure for flight performance characterisation. It was not feasible to simulate the correct inflation load in this test series unless at very high altitude, hence the introduction of the ground based (tow truck) structural load tests.
64 Beagle-2 Mission Report Failure Tree Summary
7.3.10 Airbag Initiation to Impact 9.0 Airbag Initiation to Impact 9.1 premature airbag inflation premature airbag trigger Unknown Very Low high integrated leakage - low pressure on impact; unacceptable material degradation due to prolonged contact time with NH3 front shield Against Low proximity (weak)
9.2 airbag inflation failure a RAT failure to detect local no airbag protection Against Very Low reflectivity (Strong) different from expected Noting: Insensitivity to reflectivity b RAT surface detection too late local terrain Against Low slope (weak) excessive - crater Images show that only large craters probably have excessive slopes along with a few isolated cones Noting: Probability of hitting craters H and I low and surface characteristics similar to expected pendulum Against Very Low effect (strong) Period ~20secs based on simple pendulum. Frequency ~0.05Hz, amplitude unknown. RAT sampling at higher frequency so Noting: Wide FOV of RAT probably unlikely c RAT damaged by Solar Flare Against Very Low (strong) Solar flare only ~1-10Rad. All electronics would have been exposed and no effect likely given dose. Noting: Design rad hard d insufficient time due to excessive reduced For (weak) Low descent velocity atmospheric density due to high temperature see 4.3 e AGS operation failure no command, Unknown Very Low mechanical, electrical….
65 Beagle-2 Mission Report Failure Tree Summary
Noting: electrical and commands tested pre-launch f AGS leaked NH3 during Cruise – NH3 leakage see Item 5.5 Against Low out-gassing products (weak) leaked NH3 could in principle condense in HEPA filter where temperature will be close to 0degC ; this would prevent venting Noting: Out-gassing seen (liquid or ice would be sufficient - similar to PMD/bubble trap in propellant tanks) leading to build up of pressure which would throughout cruise (June in reality be released by bioseal or failure of front aeroshell before ARM cover. However blockage of HEPA requires rapid through September) leak explosive release and big loss of gas through HEPA in order to start process which would then probably result in probably due to water from insufficient gas to condense and block and allow pressure build up. S/C and Beagle 2. AGS test data shows no reason to leak g high leakage airbag leakage Against Low (medium) Airbags underwent leakage tests. Material used as per MER. airbag/AGS Unknown Medium connection failure no leak test performed during AIV due to being impossible to achieve with design h burst on inflation Against Low (strong)
I bag segments internally 'fused' insufficient Against Low under 6 month vacuum talcum powder (strong)
9.3 main parachute failure to release a No command, mechanical, prevents lander Unknown Low electrical…. deployment Noting: electrical and commands tested pre-launch b hang-up Unknown Low
9.4 main chute envelops lander/airbags lanyard too short prevents lander For (weak) Low/ deployment Medium? lanyard length 41.5m; with 34m. Recontact probability ~1.6% with old design of parachute. Full analysis not redone following chute design. probability unknown but thought to be low, surface wind and parachute collapse following release mitigates
66 Beagle-2 Mission Report Failure Tree Summary
9.5 airbag fails to protect lander a high airbag leakage early inflation Against Low
degradation Against Low during Cruise
b structural failure at AGS outlet Against Low pipe Test data shows that pipes and AGS were correctly supported and all units inspected and tested as required 9.6 airbag failure on impacts a velocity too high reduced high lander impact For (weak) Low atmospheric density due to high temperature See 4.3; nominal predicted velocity based upon main chute nominal Cd =0.98 derated from test mean of 1.04, i.e. extra margin. Note second bounce due to slopes etc. may be more severe. b high winds local weather Unknown Low/ conditions Medium? Wind data difficult to obtain , landing analysis relied on Global Circulation Models and landing site moved to reduce risks of Noting: Landing site moved wind velocities to reduce wind effects. However gusts and turbulence not taken into account c high dynamic internal pressure local terrain Against Medium characteristics (weak) -slopes, rocks, etc
d punctured local terrain Unknown Medium characteristics -slopes, rocks, etc See also 10.1b. Analysis shows terrain consistent with expected risks. See note above re second bounce. Limited testing on final design due to development problems. e not able to sustain multiple Against Medium bounces/impacts (weak)
67 Beagle-2 Mission Report Failure Tree Summary
f material degradation from NH3 Against Low exposure (medium) Exposure time limited via RAT and consistent with expected operation requirements and ground testing exposed material for longer periods than operation without problems. g bladder material quality problem Against Low Sufficient material was acquired , see 9.2g
7.3.11 First Impact to Software Handover 10.0 First Impact to Software Handover 10.1 airbag failure to jettison a No command, mechanical, lander not released Unknown Low electrical…. Noting: electrical and commands tested pre- launch b low internal pressure high leakage Unknown Medium
at least one Check valves are not Unknown Medium segment has true NRVs, they only gross leakage, close when the airbag note problem separates from the of second AGS; reverse flow will bounce see occur between airbag 9.4 above. segments.
c lacing hang-ups Against Low/ (weak) Medium?
d AGS failure to release no command, Unknown Low mechanical, electrical…. Noting: electrical and commands tested pre- launch rod hang-up Against Low
68 Beagle-2 Mission Report Failure Tree Summary
10.2 airbag jettison incorrect a lander released too early high dynamic high lander impact Unknown Low impact loads on lander
b lander imparted upward energy or high lander impact Unknown Low excessive downward energy
10.3 high lander impact on fall from airbags airbag jettison incorrect internal equipment Unknown Low damage
local terrain characteristics - Unknown Low/ slopes, rocks, etc medium?
10.4 lid frangibolt fails to operate damage on impact lander not deployed Unknown Low
10.5 Clampband release failure damage on impact lander not deployed Unknown Low
10.6 PSW fails to hand over to LSW Against Low (weak) No cause yet identified. Handover tested on GTM with Flight standard software, note without shocks
69 Beagle-2 Mission Report Failure Tree Summary
SURFACE PHASE
7.3.12 PSW to LSW Handover 1.0 PSW to LSW Handover 1.1 APS off power glitch, eg due to pyro relay await short or long Against Low operation timer to switch back on (medium) Note: pyro relay not tested during early cruise due to ESA safety concerns. Test proposed early December but cancelled due Noting: tested on ground to spacecraft operational constraints 1.2 APS malfunction damage/degraded during Cruise higher Against Very Low temperatures (strong) than expected during extended checkouts APS switch on and off seen without problems during cruise. APS temperatures within tested limits. 1.3 LSW fails to boot no surface operations Against Very Low (strong) No cause yet identified. Handover tested on GTM with Flight standard software note without shocks
7.3.13 Default Timeline 2.0 Default Timeline 2.1 LSW failure incorrect upload during cruise Against Very Low (strong) Checksums all good post LSW image upload. Identical software tested on GTM without problems including handover from PSW 2.2 Processor/memory fault a solar flare induced damage degraded Against Very Low through (4) (strong) power cycles during checkout No RAM, EEPROM errors seen during post flare checkout 21/22 Nov '03. Calculated flare dose shows low dose. No EEPROM errors seen during cruise.
70 Beagle-2 Mission Report Failure Tree Summary
b damage/degraded during Cruise higher Against Very Low temperatures (strong) than expected during extended ch/outs Only APS seen to be a high temperatures but within operational and tested limits. Processor and memory did not exceed +40C during cruise. 2.3 Deployment activity sequence a LSW upload error/corruption software Against Very Low malfunctions upload failure (strong) Tested twice on GTM post LSW upload without anomaly b local terrain characteristics, eg Unknown Low/ rocks Medium?
2.4 lid opening failure a damage on impact lander not deployed Unknown Low/ Medium?
b mechanical hang-up Against Low
2.5 solar panels release failure damage on impact antenna blockage Unknown Low
2.6 solar panel deployment failure a mechanical hang-up antenna blockage Unknown Medium
b Motor? Gearbox? Lubrication? Against Low Note early problems on GTM however tested twice without problem post LSW upload without anomaly 2.7 lid deployment causes damage to SAU stiction between adjacent solar loss of insulation& fall Against Very Low panel & SAU (esp. over battery of internal pop-up section) temperatures below limits (frozen battery) unlikely to cause problem before first night, i.e. should be OK for first planned comms session and probably OK for Jodrell Bank
71 Beagle-2 Mission Report Failure Tree Summary
7.3.14 Communications 3.0 Communications 3.1 antenna incorrect function damage on impact broken coax or Unknown Low/ termination Medium? see Qinetiq EIDP Annex K - degraded operation 3.2 antenna not operating a damage on impact frequency shift no comms Unknown Low
broken coax Unknown Low/ Medium?
b dust dust Against Low attenuation (weak) Dust tests show adequate performance even with large amounts 3.3 transceiver not operative a damage on impact no comms Against Low
b MET corrupted transceiver not Against(stron Very Low being g) commanded ON CSM1, CSM2 modes should have been activated however no signal seen c multiplexer ESD damage Against(stron Very Low g) see Qinetiq EIDP - ESD protected 3.4 battery SOC too low ref Cruise phase anomaly: B2 data v Mex no power Against Very Low volts data (strong) Battery seen at >95% of charge at ejection, verified by operation of taper charge and measured voltage 3.5 battery short circuit ref Cruise phase anomaly: B2 data v Mex no power Against Very Low volts data (strong) No battery problems seen during cruise however note pyro relay operation not tested, see 3.6 below and earthing error would not cause a short circuit 3.6 incorrect battery configuration pyro relay failure insufficient Against Low power available Note: pyro relay not tested during early cruise due to ESA safety concerns. Test proposed early December but cancelled due to spacecraft operational constraints
72 Beagle-2 Mission Report Failure Tree Summary
3.7 comms link not operative a LOBT corrupted/reset loss of Against Low synchronisation of comms sessions CSM1, CSM2 modes should have been activated however no signal seen Noting: Not seen in any ground testing b Odyssey link malfunction at B2 end prevents Against Low completion of (medium) handshaking Odyssey interface tested between an MLT and bench electronics at LMA Noting: Not seen in any ground testing c high attenuation in atmosphere (dust) inadequate prevents Against Low link budget detection by (weak) margin to Jodrell Bank, accommodat etc, of carrier e excessive dust attenuation Optical depth at landing not exceptional. d return link failure prevents Unknown Low/Medium completion of ? handshaking Cause: shock damage, protocol, random component failure e forward link failure prevents Unknown Low/Medium completion of ? handshaking Odyssey interface allows first packet of commands to be "received" independent of hail and allows a blind commanding Cause: shock damage, option which has been tried without apparent success protocol, random component failure f frequency shift in h/w Damage of prevents Against Low/Medium tuning detection by (strong) ? cavities on Jodrell Bank, impact etc, of carrier Shock tests conducted on MLT
73 Beagle-2 Mission Report Failure Tree Summary
g antenna mispointing forward / prevents Against Very Low return beams completion of (strong) not aligned handshaking Overflights by Odyssey and MeX have occurred at a number of elevations and azimuth angles and Jodrel bank has observed Noting: Wide range of for a duration of 11 hours eliminating tilt of antenna elevation covered in Odyssey and MeX passes h solar flare damage to transceiver RF Tx section no forward link Against Very Low section (strong) see Qinetiq EIDP Annex L. Design radiation hard i Rx section no return link Against Very Low (strong) see Qinetiq EIDP Annex L. Design radiation hard j solar flare damage to transceiver BBU BBU/RF RF section Against Very Low interface inoperative (strong) (untested during Cruise) see Qinetiq EIDP Annex L. Design radiation hard
74 Beagle-2 Mission Report Annexes
Annexes
A Beagle 2 Mission and Flight Operations Personnel...... 77 B Checkout reports ...... 81 C VMC Image Analysis ...... 93 D Post-landing Search Strategy...... 101 E Landed Phase Operational Power Modelling ...... 119 F Electrical Cruise Behaviour Report ...... 143 G Landing Site Hazard Analysis...... 183 H Atmosphere and Aerodynamic Implications on Entry, Descent and Landing of Beagle 2 ..... 189 I EDLS Events: Parachute Deployment, Heatshield Separation and Landing...... 219 J Lovell Telescope Observations of the Beagle 2 Landing Site ...... 227 K Assessment of the Cruise Phase Thermal Performance...... 235 L Prediction of Size of a Beagle 2 Impact Crater ...... 247 M Beagle 2 Landing Ellipse Evolution ...... 259
75 Beagle-2 Mission Report Annexes
Testing commands with the GTM
76 Beagle-2 Mission Report Annexes
A Beagle 2 Mission and Flight Operations Personnel
The following personnel have contributed to the mission operations, and this report, through direct or indirect input.
Mark Sims, University of Leicester, Mission Manager Colin Pillinger, Open University, Beagle 2 Lead Scientist
University of Leicester
Oliver Blake Flight Operations Team Gillian Butcher XRS Scientist Ed Chester Flight Operations Team Rosemary Danson Space Research Centre Secretary George Fraser Space Research Centre Director Mark Hannington Flight Operations Team John Holt AIV Technician Nick Nelms Electrical Engineer Guy Peters Project Assistant Derek Pullan Instrument Manager John Pye Space Research Centre Manager Jon Sykes PAW Engineer Dean Talboys XRS Ph.D. Student Daniel Thompson Ph.D. Student Alan Wells Former Director Space Research Centre
Open University
Simeon Barber GAP Manager Rebecca Baxter LOPC Secretary John Bland Mössbauer Scientist John Bridges Landing Site Characterisation Duncan Carter LOPC IT Support Jed Cawthorne DPA Manager / LOPC IT Support Joss Knight GAP Engineer Mark Leese GAP Manager Michele Lightfoot PA / Secretary to Colin Pillinger Taff Morgan GAP Scientist Andrew Morse GAP Engineer Manish Patel ESS Scientist Judith Pillinger Beagle 2 PR and Planetary Protection Scientist Becky Reynolds PSRI Secretary Tim Ringrose ESS Scientist Dave Rothery Landing Site Characterisation Andy Spry Planetary Protection Officer 77 Beagle-2 Mission Report Annexes
Martin Towner ESS Scientist Ian Wright GAP Leader John Zarnecki ESS Team Leader
EADS Astrium
Cliff Ashcroft Structural Engineer Stuart Hurst Beagle 2 Chief Engineer Nigel Philips Instrument Arm Engineer Richard Slade Structural Engineer Lester Waugh Communications and Software
ESA - ESOC
Michel Denis MEX Ops Team SOM Mike McKay MEX Flight Director Alan Moorhouse MEX Ops Team SOM Sybille Peschke MEX Mission Planning Team Erhard Rabeneau MEX Mission Planning Team Jonathan Schulster MEX Ops Team Michael Wittig Beagle 2 SUEM and Mechanical Engineer
ESA – ESTEC
Con McCarthy ESA Lander Manager Roland Trautner ESA Beagle 2 Operations Support Scientist Alistair Winton ESA RF Interface Engineer
SciSys
Les Baldwin Lander Software Engineer Glenn Johnson Lander Software Engineer Stewart Hall Ground Segment Engineer Mark Hann Lander Software Engineer Martin Townend Ground Segment Manager Roger Ward Lander Software Manager
SEA
Geoffrey Matt Power Modelling Engineer Alan Senior Processor Engineer Nigel Wright SEA Space Projects Manager
Qinetiq
Matt Cosby Transceiver Design Duncan Fortune Transceiver Project Manager Ted Gomm Antenna Design
78 Beagle-2 Mission Report Annexes
Steve Kynaston Transceiver Design
Engineering
Jim Clemmet Surrey Satellite Technology Ltd. Former Beagle 2 Chief Engineer Formerly EADS_Astrium Jonathan Gebbie Logica CMP Ltd. Probe Software Engineer Barrie Kirk Formerly Astrium Industrial Project Manager Stephen Lewis Oxford University Atmospheric Analysis David Northey Analyticon Parachute Engineer Brian Shaughnessy Rutherford Appleton Laboratory Thermal Modelling Engineer Arthur Smith Fluid Gravity Entry and Aerodynamic Modelling Martin Symonds Logica Probe Software Manager
Operations
Dave Barnes University of Wales Aberystwyth VR Modelling Mike Malin Malin Space Science Systems MGS MOC PI Ian Morrison Jodrell Bank Jodrell Bank Project Manager for Beagle 2 Operations Gerhard Paar Joanneum Research DEM Generation and Analysis Guy Rennie Virtual Analytics Image Analysis Eddie Taylor University of Wales Aberystwyth VR Modelling
Payload
Matt Balme UCL Stereo Camera Team Member Bodo Bernhardt University of Mainz Mössbauer Engineer Melody Carter MSSL Stereo Camera Team Member Andrew Coates MSSL Stereo Camera Leader Andrew Griffiths MSSL Stereo Camera Manager Beda Hofmann NMBE Microscope Team Stubbe Hviid MPAe Microscope Engineer Jean-Luc Josset Space-X Camera Systems Goestar University of Mainz Mössbauer Team Leader Klingelhoefer Ben Leuthi University of Bern Microscope Scientist TC Ng Hong Kong RCG Team Leader Lutz Richter DLR PLUTO ‘Mole’ Leader Peter Rueffer Braunschweig Image Compression Wolfram Sies DLR PLUTO ‘Mole’ Operations Nick Thomas University of Bern Microscope Team Leader KL Yung Hong Kong Polytechnic University RCG Engineer
... and many others who left the project prior to either launch or landing. We apologise for any name that we have accidentally left off the list or any misrepresentation of their role within the project.
79 Beagle-2 Mission Report Annexes
This page left intentionally blank.
80 Beagle-2 Mission Report Annexes
B Checkout reports
The Beagle 2 checkouts during cruise phase are summarised below, with power-on and power-off times. Each checkout is referred to by a letter (A through J).
Power -On Power-Off
Checkout DoY Date Time DoYDate Time
A First Cruise Checkout 185 2003-07-04 20:04:04 186 2003-07-05 04:54:00
B NEV Checkout 186 2003-07-05 19:03:00 187 2003-07-06 01:41:00
C 2nd NEV Checkout 193 2003-07-12 16:46:00 194 2003-07-13 01:18:00
D Mini Checkout, E2 Scrub 244 2003-09-01 12:40:17 244 2003-09-01 14:00:30
E LSW Upload Test 280 2003-10-07 11:03:00 280 2003-10-07 16:45:00
F LSW Test Continued 282 2003-10-09 11:48:00 282 2003-10-09 17:30:00
G LSW Upload 325 2003-11-21 08:00:00 325 2003-11-21 19:00:00
H LSW Upload Part 2 326 2003-11-22 10:15:00 326 2003-11-22 16:38:00
I Pre-Ejection Checkout 351 2003-12-17 06:34:00 351 2003-12-17 14:51:00
J Pre-Ejection Checkout 2 352 2003-12-18 06:33:51 352 2003-12-18 10:40:00
Table B.1 - Checkout Summary
81 Beagle-2 Mission Report Annexes
B.1 Early Operations and Checkouts A, B, C Report on Beagle 2 Early Mission Operations BGL2-LUX-RE-229 2nd June to 6th June 2003 Issue: 1 18th June 2003
Dr. M.R. Sims, Space Research Centre, University of Leicester, Leicester, LE1 7RH
Note: Beagle 2 Observation Reports ref OBS-1 etc. are not contained within this report for sake of clarity but can be made available for detailed technical enquiries for appropriate organisations.
2nd June 2003
Mex launch 17:45:24 UTC. Parking orbit burn 1755 UTC. TransMars Injection burn 1903 UTC. 1917 UTC spacecraft separation. Acquisition of S/C signal from New Norcia station in Australia. Solar panel deployment nominal.
3rd June 2003
0800 UTC spacecraft in Safe mode following AOCS problems due to startracker problem. Glint or glare detected in startracker. Spacecraft recovered from safe mode. Heater and trickle charge restored during recovery, heater only inhibit at one point.
Beagle 2 temperatures monitored and recorded, see separate spreadsheet.
Timer “ticking” and battery voltage as expected.
Startracker problems overcome by offset pointing whilst investigation underway.
4th June 2003
Spacecraft slewed 45° overnight startracker worked acquired stars moved into nominal mode (03:44 UTC)
Spacecraft slewed back and at 15° offset works nominally. Incremental slews underway. Glare problem would hold for 2 weeks until sun angle changes enough. However could not use HGA without slews. Software patch to startrackers investigated.
At 15° offset 6 stars seen by startrackers, at 20° offset 9 stars seen.
Beagle 2 heaters seen to start switching at 32.29 hours after launch 2003.15.03.14.35.311.
MEX Beagle 2 Heaters as follows:
Heater Name Resistance Switch On Switch Off Power Current at (Ω) (°C) (°C) (W) 28V (A) SUEM 216 -22.5 -17.5 3.629 0.129 Backplate 157 -8.3 -3.3 5 0.178 ARMs 156 -8.3 -3.3 5 0.178 Battery ? -8.3 TBC -3.3 TBC 8 0.286
Table B.2 - Beagle 2 Heater Data (4th June 2003)
14:12 UTC 0.18A drawn on heater circuit, probably ARMs as coldest point?
82 Beagle-2 Mission Report Annexes
5th June 2003
Startracker patching and testing underway. Test on large thrusters for TCM-1 burn successful.
Plot of Beagle 2 power shows two heaters on overnight, 0.18A continuous plus 0.28A intermittently, for ~10 minutes every 135 minutes. Assume battery plus ARMs.
Launch +62:24:30 start Beagle 2 unclamping.
LCL2 on 156.08.13.35.977 Prime Circuit armed 08.13.39.957
Frangibolt 1 sequence sent however Frangibolt 3 released! Wiring or database error??
Frangibolt Current On 08.20.07.959 3 Untied 08.21.23.958 Current off 08.21.55.968 Time to release 75.999 secs
Frangibolt Current On 08.30.07.970 2 Untied 08.31.31.962 Current off 08.32.07.959 Time to release 83.992 secs
Frangibolt Current On 08.40.07.960 1 Untied 08.41.23.958 Current off 08.41.55.962 Time to release 75.998 secs
LCL2 off 08.43.55.968 Prime Unarmed 08.44.51.958
Redundant fire sequence not used.
Current used ~3.82A with slight decrease during on-time presumably due to resistance changes as heaters gets warm.
6th June 2003
TCM-1 Burn successful. Startracker patch OK, written into RAM first then pm into EEPROM. One startracker has patch in EEPROM other in RAM. 106 km distance from Earth passed. Spacecraft slewed to take VMC picture of Earth (includes Beagle 2 in FOV). This includes 20 minutes on Beagle 2 front aeroshell. Successfully slewed back. S/C running on high gain antenna.
Beagle 2 timer ticking, battery heater cycling as yesterday, ARM heater on continuously.
83 Beagle-2 Mission Report Annexes
B.2 Checkout A (185,186) Report on First Beagle 2 Checkout, 4-5th July 2003 (Extracted from log book notes)
Melacom hailing 19h50, Beagle 2 on 19h52.
Initial hail failed as link rates were set to 2/2 rather than 8/8 (8/8 required for umbilical).
Telemetry received as expected 20h05, with SSMM data dumps every 10 minutes.
Launch configuration of safety sequence was moved, and connection test commands were successful.
Timer test uplinked and performed at 22h10.
Melacom link was dropped at 22h58 due to a clock roll-over event (see OBS-002). Beagle 2 still generating telemetry packets.
23h04 Beagle 2 stopped generating telemetry as the buffer was full. Melacom was power cycled at 01h33 but still no telemetry was seen.
04h47 PSW Reboot command was sent, Melacom link aborted and rehailed. Telemetry was received at 04h49 indicating successful reboot and recovery from the anomalous link dropout.
Beagle 2 was powered off at 04h54, with much relief.
B.3 Checkout B (186,187) Report on First Beagle 2 Checkout, 5-6thth July 2003 (Extracted from log book notes)
Melacom hailed at 18h59, Beagle 2 switch on at 19h03. Connection tests were successful. SSMM data dumps periodically throughout checkout.
Timer stopped at 19h21 and restarted at 19h26. Timer test repeated using Mars Express command database unsuccessfully, and every combination of correct and incorrect Ack flags, and single/double encapsulation was tested. Double encapsulated commands were successful.
Battery state of charge was managed at approximately 50%.
Heater testing started 22h45 and ended at 23h27. EEPROM dumps, SSMM dumps and PSW patching were used to test memory.
Beagle 2 was switched off at 01h40.
84 Beagle-2 Mission Report Annexes
B.4 Checkout C (193,194) Time Event Comments Day hr min sec 193 16 31 Melacomm On and Hailing 16 45 54 APS On 16 59 Connection Test Command 17 5 SSMM Data Dump 17 15 SSMM Data Dump 17 30 SSMM Data Dump 17 42 SSMM Data Dump 17 50 SSMM Data Dump 17 56 PFU power off To manage battery state of charge. Beagle 2 running on battery 18 0 Timer Test Start 18 9 11 commands sent 18 13 SSMM Data Dump 18 22 Stop timer tick, timer loading 18 33 SSMM Data Dump 18 38 42 Timer started 18 53 SSMM Data Dump 18 58 Stop timer tick, timer loading 19 5 SSMM Data Dump 19 10 44 Timer started 19 18 SSMM Data Dump 19 23 Stop timer tick, timer loading 19 33 SSMM Data Dump 19 38 Timer started 19 55 SSMM Data Dump 20 1 22 Timer tripped 20 5 SSMM Data Dump 20 12 26 PFU power on Beagle 2 running on Spacecraft power Timer stops 20 15 SSMM Data Dump 20 28 6 Heater Test started 20 32 0 XRS heater on 20 42 Timer start commanded 20 44 Timer confirmed as running 20 52 Heater commanded off 20 55 SSMM Data Dump 21 4 Heater Bank A off B on Heater B1 On 21 7 Heater B1 Off 21 15 Heater B2 On 21 17 Heater B2 Off 21 25 Heater Banks A and B off 21 29 SSMM Data Dump 21 40 17 Stop timer tick commanded MEX database command. Timer continues to run 21 43 32 Disable latches Restore safety interlocks (precaution) 21 47 SSMM Data Dump 21 59 SSMM Data Dump 22 36 SSMM Data Dump 22 42 5 EDLS Parameter Upload Start 22 42 19 EDLS Parameter Upload Stop
85 Beagle-2 Mission Report Annexes
Time Event Comments Day hr min sec 22 47 SSMM Data Dump 23 4 25 Stop timer commanded LOR 23 5 15 Timer confirmed as stopped 23 9 Start Timer 23 10 10 Timer confirmed as running 23 14 SSMM Data Dump 23 17 33 Stop timer tick commanded MEX database command. Timer continues to run 23 18 40 Timer confirmed as running Command fails to stop timer 23 20 Timer start commanded 23 23 SSMM Data Dump 23 27 Stop timer commanded LOR 23 30 Timer stops 23 31 Timer start commanded via LOR 23 32 Timer confirmed as running 23 34 SSMM Data Dump 23 38 Stop timer tick commanded MEX database command. Timer continues to run 23 40 Timer confirmed as running 23 41 Timer start commanded 23 44 SSMM Data Dump 23 51 Stop timer commanded LOR 23 52 Timer stops 23 55 Timer start commaned via LOR 23 56 37 Timer confirmed as running 23 57 SSMM Data Dump 194 0 18 44 EDLS parameters confirmed OK 0 41 Memory Load test part 1 0 45 23 SSMM Data Dump 0 51 2 EB85 checksum confirmed 0 57 32 Memory Load test block 2 1 2 3 Memory Load test block 3 1 5 57 Memory Load test block 4 1 ? Spacecraft problem with block 5 1 12 50 SSMM Data Dump 1 14 SSMM Data Dump 1 18 APS Off 1 50 APS confirmed off
86 Beagle-2 Mission Report Annexes
B.5 Checkout D: Memory Scrub (1 Sep 2003) Report on Beagle 2 Memory Scrub Power On 1st September 2003 Dr. M.R. Sims, Space Research Centre, University of Leicester, 13th January 2004
Beagle 2 was powered on Day 244 to allow the PSW software to scrub the memory to remove SEUs (cosmic ray hits in computer memory that may change the data contents) that may have accumulated in Beagle 2 since the checkouts in July. This was undertaken as a precautionary measure. This scrubbing process happens automatically as a background task whilst Beagle 2 is powered.
Beagle 2 telemetry showed nominal status following power up.
Light travel time of 1 min 29.28 secs at ~12:00 Day 244.
Beagle 2 was powered on at 12:40 and powered off at 14:00 on Day 244 giving a total on time for that day of 1 hour 20 minutes. SSMM drip tap reads were used to verify Beagle 2 status throughout the operations.
B.6 Checkouts E and F: LSW Image Upload Test (7, 9 Oct 2003) Report on LSW Image Upload Test 7th and 9th October 2003 Dr. M.R. Sims, Space Research Centre, University of Leicester, 13th January 2004
Beagle 2 was powered on Days 280 and 282 for the LSW image upload test and an investigation of XRS heater problem seen earlier in the mission.
Beagle 2 telemetry showed nominal status following each power up.
Light travel time of 3 min 2.56 secs at 10:08 Day 280 and 3 mins 10 secs at 12:59 Day 282.
Beagle 2 was powered on at 11:03 and powered off at 16:45 on Day 280 giving a total on time for that day of 5 hours 43 minutes.
Beagle 2 was powered on at 11:48 and powered off at 17:34 on Day 282 giving a total on time for that day of 5 hours 46 minutes.
The following tasks were achieved on day 280:
• Beagle 2 heater test to confirm XRS heater was connected to spare heater circuit • Boost charging of battery using second redundant power line • EDLS parameter upload • PSW patching • Image Upload test of two files
The following tasks were achieved on day 282:
• Boost charging of battery using second redundant power line • Verification of PSW patches by correct operation of PSW and checksums • Image Upload test of 26 files equivalent to full image load • PSW checksums verified to ensure no corruption of PSW following upload test
87 Beagle-2 Mission Report Annexes
B.7 Checkouts G and H: LSW Upload (21, 22 Nov 2003) Report on LSW Image Upload 21st and 22nd November Dr. M.R. Sims, Space Research Centre, University of Leicester, 25th November 2003
Beagle 2 was powered on Days 325 and 326 for the LSW image upload.
Beagle 2 telemetry showed nominal status following each power up indicating no EEPROM or RAM corruption from early November 2003 Solar Flares.
Light travel time of 6 min 2.56 secs (108,693,328km) at 0800 Day 325 and 6 min 7.45 secs (110,159,280km) at 09:45 Day 326 necessitated power up, software load and power down to be accomplished via the Mission (On-Board) Timeline.
Beagle 2 was powered on at 08:00 and powered off at 19:25 on Day 325 giving a total on time for that day of 11 hours 25 minutes.
Beagle 2 was powered on at 10:00 and powered off at 16:22 on Day 326 giving a total on time for that day of 6 hours 22 minutes.
The following tasks were achieved over the two days:
• Upload of LSW V3.0 • Load confirmed via checksums • 55% of the LSW image was dumped to ground and compared (see below) • All ancillary data was uploaded • Ancillary data was confirmed via checksums and appropriate memory dumps • PSW was verified with correct checksums following upload of the LSW • The EEPROM Allocation Table was verified • The Boot Context Table was verified • The Common software context was verified • PSW Parameters were also verified • The Timer trip operation was verified with a timer test • The Umbilical Link telemetry recovery procedure was verified
Several contingencies arose on Day 325.
An error in the MTL prevented LSW files 3,4,5,6,7 not being uploaded at the first attempt.
The MTL was stopped, rebuilt and all files were uploaded.
Three files 12, 14, and 22 (from 1-26) showed checksums errors on upload.
File 19 failed to load due to command errors in the LOR provided by the LOCC. This was rebuilt and resent to ESOC for processing.
During the memory dump anomalous behaviour was seen on the telemetry. Analysis of the Melacom data by QinetiQ shows that after generation of one set of housekeeping packets at S/C time of 30188-30192 the memory dump process effectively died along with the telemetry and command link (?) at 30209. The link recovery contingency procedure was then activated and the link restored. The result was that only 225 memory dump packets were received on the ground from an expected total of 404 for one image.
However image comparison showed that the recovered fraction ~55% was correct in all aspects at variance with the reported checksum errors. In particular files 12,14 memory areas were seen to be correct , unfortunately file 22 was not contained within the partial dump. It should be noted that checksum calculation is a so called slow command where other commands will take precedence. Consequently a checksum calculation can fail due to timing problems.
Following recovery of the umbilical link the EEPROM integrity check was performed.
88 Beagle-2 Mission Report Annexes
Due to the extended operations on day 325 the Beagle 2 battery reached its maximum limit of 35°C however was recharged to 94% of capacity.
On Day 326 Beagle 2 was re-powered and nominal telemetry received indicating strongly that PSW had not been corrupted during activities on Day 325.
Beagle 2 was commanded to return checksums for files 12,14 and 22 and checksums were found to be correct indicative of a timing problem with activities on day 325.
The corrected file 19 was then uploaded and its checksum was seen to be correct. The ancillary data was then uploaded and EEPROM integrity checks performed.
During Day 326 the battery was recharged to ~99% capacity and the taper charge limiter was seen to become active both in MEX and Beagle 2 telemetry.
A timer test was performed in parallel with the ancillary data load and timer trip was verified. The timer was zeroed and reset. The timer was loaded with 1FFO equivalent to 36min timeout at 14:12 and started at 14:23. The first timer tick transition was detected at 14:23:56. The timer trip was seen to activate at 15:01 within one 143 second cycle of the expected time.
89 Beagle-2 Mission Report Annexes
B.8 Checkouts I and J: Pre-Ejection Preparation (17 Dec 2003) Report on Pre-ejection Checkout and Beagle 2 Ejection 17-19th December 2003 Dr. M.R. Sims, Space Research Centre, University of Leicester, 12th January 2004
Beagle 2 was powered on Days 351 and 352 for its pre-ejection checkout and final parameter and software loads. Beagle 2 was ejected successfully from Mars Express on day 353 19th December 2003.
Beagle 2 telemetry showed nominal status following each power on and showed nominal and expected responses throughout all parts of the pre-ejection checkouts. Drip tap reads of the SSMM were scheduled at regular intervals throughout both days when Beagle 2 was powered to verify status and health.
Note all times given below are S/C executed time. Data was received on the ground following the one-way light travel time.
One way light travel time was 8 min 5.43 secs (145,529,376km) at 0453 Day 351 and 8 min 10.55 secs (147,061,728km) at 06:00 Day 352.
Beagle 2 was powered on at 06:34 and powered off at 14:51 on Day 351 giving a total on time for that day of 8 hours 17 minutes. Beagle 2 was powered on at 06:34 and powered off at 10:40 on Day 352 giving a total on time for that day of 4 hours 6 minutes. Beagle 2 was not powered on day 353 ejection day, ejection was conducted via Mars Express and verified using Mars Express telemetry.
Day 351 - 17th December 2003 Activities
The following activities were carried out on day 351 using Beagle 2 procedure BOP-022-DEC_17_CHECK:
1. Switch-on Beagle 2 verify health using ESOC procedure BE-FCP-001. Note timer was stopped at switch on and restarted. 2. PSW Parameter Upload to set safety parameters to eject values. Executed at 351:07:53, verified during EEPROM Integrity check completed by 351:10:58. Beagle 2 procedure BOP016- PSW_Params_EDLS was used. 3. Upload LSW patches 1 to 10 and dump data from patch area. Executed at 08:30 and verified at 08:55. Beagle 2 procedure BOP_019_LSW_FM_patches_issue_2 was used. 4. Upload early phase landed data namely updated communications sessions, updated timeline and dump checksums as verification. Executed 09:12 and verified by 10:58. Beagle 2 procedure BOP_020_Upload_Landed_Phase_Data was used. Note: new LOR was constructed, sent and uploaded as original did not contain dump command for communications sessions. Note: step 4 of procedure upload default activity sequences was not executed due to problems identified on 16/12/03 with revised activity sequences. The activity sequences placed on-board during LSW image upload 21st/22nd November being adequate for first three days of on-board mission timeline. 5. Conduct EEPROM integrity check. Executed at 10:20 and verified by 10:58. Beagle 2 procedure BOP_010_EEPROM_Check_Issue_4 was used. 6. Load Beagle 2 wakeup timer/counter prior to ejection following calculation and verification of timer load value (0E71 commanded, 1CE in returned telemetry). Executed by Mars Express using ESOC procedure EJ-FCP-162 and verified at 12:05 following an SSMM drip tap read. Timer loading executed nominally. 7. Start timer using ESOC procedure EJ-FCP-162 at 13:50:22, timer verified as started at 13:51:07 in S/C telemetry. 8. Charge Beagle 2 battery in parallel with above activities. Battery voltage at 25.13V at 351:13:28. 9. Switch off Beagle 2 using ESOC procedure BE-FCP-102 and verify Beagle 2 off. APS off at 14:51 and verified by 15:40.
90 Beagle-2 Mission Report Annexes
Day 352 - 18th December 2003 Activities
The following activities were carried out on day 352.
1. Switch on using ESOC procedure BE-FCP-101 and verify health. Note: Two RAM errors seen in drip tap read at 07:40, both errors had been washed out by PSW. 2. Check timer still running and at correct and expected values. By 352:09:40 nine transitions of timer state had been observed giving a 19min 12 sec period in terms of telemetry and expected values were seen at switch on (202) and switch off (20F). 3. Set long latch to enable, short latch to disable using ESOC procedure EJ-FCP-163. Sent at 09:00 verified following drip tap read of SSMM at 09:10 and verified at all subsequent reads prior to switch off. 4. Charge Beagle 2 battery in parallel with above activities. Battery Voltage at 25.13V at switch off. 5. Switch off Beagle 2 using ESOC procedure BE-FCP-102 and verify Beagle 2 off. APS off at 10:40 and verified by 11:19. 6. Go/No Go decision for ejection made at 22:28 via an operations team voice loop conference using ESOC procedure EJ-FCP-200.
B.9 Ejection Report (19 Dec 2003) Day 353 - 19th December 2003 Activities
Beagle 2 ejection Go was confirmed in an operations team voice loop conference at 06:51 using again ESOC procedure EJ-FCP-200. Beagle 2 ejection occurred at 353:08:31 using a ESOC Mars Express generated on-board timeline and was subsequently verified by 4 methods, namely:
“Glitch” in S-band Doppler at pyro firing (Reported at 07:39) Spacecraft telemetry showing Beagle 2 disconnected (10:32). Appropriate responses in spacecraft AOCS data (verified by 11:12) VMC images showing Beagle 2 separated from Mars Express with a delta-V.
Other Notes
Battery charging was verified by cycling of taper charge seen both within Beagle 2 internal telemetry and current drawn from LCL5A on Mars Express. 25.02-25.13V corresponds to battery charged to > 98%.
EEPROM integrity check verified following: • EDLS Parameters • PSW Image Checksum • Allocation Table • Boot Context • Common Software Context • PSW Parameters
LSW Image checksum was as expected.
Annexes/Reference Documents
Beagle 2 Procedures Mars Express FOP Issue 4.3 18th December 2003 Hard copy of S/C telemetry following ejection Beagle 2 Go/No Go for Ejection Timer Setting Calculation Mars Express Go/No Go for Ejection
91 Beagle-2 Mission Report Annexes
This page left intentionally blank.
92 Beagle-2 Mission Report Annexes
C VMC Image Analysis
Dr. M.R. Sims, Space Research Centre, University of Leicester G. Rennie Virtual Analytics Ltd. E. Chester NSC LOCC University of Leicester B.M. Shaughnessy RAL S. Hurst and M. Bonnar EADS-Astrium H. Eggel ESA ESTEC A. Moorhouse and M. Denis ESA ESOC J. Clemmet Surrey Satellite Technology Ltd. C.1 VMC Images Figure C.1 - VMC 02
IM2 (02) reconstructed.jpg
Time = 353.08.32.17
Figure C.2 - VMC 03
IM3 (03) reconstructed.jpg
Time = 353.08.33.07
All images courtesy ESA
93 Beagle-2 Mission Report Annexes
Figure C.3 - VMC 04
IM4 (04) reconstructed.jpg
Time = 353.08.33.57
Figure C.4 - VMC 05
IM5 (05) reconstructed.jpg
Time = 353.08.34.47
All images courtesy ESA
94 Beagle-2 Mission Report Annexes
Figure C.5 - VMC 06
IM6 (06) reconstructed.jpg
Time = 353.08.35.37
Figure C.6 - VMC 07
IM7 (07) stretched.jpg
Time = 353.08.36.39
All images courtesy ESA
95 Beagle-2 Mission Report Annexes
C.2 Conclusions from VMC Image Analysis
C.2.1 Introduction and Background Beagle 2 ejection 19th December 2003 was imaged by the VMC camera on Mars Express. The VMC camera (400-650nm) has a pixel size of 0.0014mm with a focal length of 12.32mm and a FOV of 40° by 30° offset from +z axis of the spacecraft giving an image of 640 pixels by 480 pixels. The diagrams enclosed as in the annex show the position and FOV of the VMC on Mars Express.
The time for the pyro firing which ejected B2 was scheduled in the Mars Express MTL at:
353.08.31.10 PFire_A_N (Pyro A / PDU A, RTU I/O Nominal)
For completeness, the other pyro/paths were fired as follows:
353.08.31.17 PFire_B_R 353.08.31.51 PFire_A_R 353.08.31.58 PFire_B_N
From the records of the commands uplinked to the MTL the VMC Imaging sequence was as follows:
Times of DMS TC ZVM00001 (DMS Read VMC Image)
DMS Read Time Reference Exposure /ms Image number 353.08.14.49 Read VMC Image Attached 1 4600 Image 2 353.08.32.17 Read VMC Image Eject 1 600 Image 3 353.08.33.07 Read VMC Image Eject 2 600 Image 4 353.08.33.57 Read VMC Image Eject 3 600 Image 5 353.08.34.47 Read VMC Image Eject 4 600 Image 6 353.08.35.37 Read VMC Image Eject 5 600 Image 7 353.08.36.39 Read VMC Image Eject 6 4600 Image 8
Table C.1 - VMC Image Times
Images 2-6 have been analysed in detail. Images 3 through 6 are crucial in that a shape of diameter > few pixels exists for Beagle 2.
Image 3 was taken 67 seconds after ejection, with images 4, 5, etc. at 50 second intervals.
r The formula D = f s is ps where D = real distance, f = camera focal length, is = image size in pixels, rs = real size, ps = pixel size; was used to derive distance of Beagle 2. Beagle 2 subtended 40, 23, 16 and 12.5 pixels in images 3, 4, 5 and 6 respectively.
Assuming a heat shield diameter of Beagle 2 of 924mm plus an allowance 10mm for MLI (giving a total diameter of 934mm), the distance of Beagle 2 was calculated as 20.54m in Image 3, 35.74m in Image 4, 51.37m in Image 5, and 65.75m in Image 6. Typical errors in the image analysis are ± 0.5 pixels. This equates to ± 0.26, 0.79, 1.65 and 2.74 m in errors in distance respectively for images 3 though 6.
96 Beagle-2 Mission Report Annexes
C.2.2 VMC Position and Field of View
Figure C.7 - VMC Field of View and Orientation (Courtesy ESA)
C.2.3 Ejection Velocity and Spin Rate From the distance (given above) and timing of the images it is possible to measure the ejection velocity. Based on camera characteristics and an assumed heat shield diameter of Beagle 2 of 924mm plus an allowance 10mm for MLI (giving a total diameter of 934mm) an ejection velocity of
0.3025 ± 0.0083 ms-1 (1σ) (excluding systematic errors) can be derived from the images compared with a nominal ejection velocity of 0.31 ms-1 as defined in the Beagle 2 ESOC ICD (Document BEA2.ICD.00005.S.MMS Issue 1 Rev0 MEX Lander Delivery Module Beagle 2- ESOC Interface) and 0.31 ± 1.295 ms-1 as reported by ESOC post separation.
No data on spin rate from the VMC images can be derived as no features are visible on Beagle 2 apart from in image 3, see section C.2.5 below.
C.2.4 Sun Angle on Beagle 2 By construction of a solid model of Beagle 2 and modelling the lighting it is possible to estimate the sun angle or solar aspect angle (SAA) on Beagle 2 by matching the resulting imaged shape to image 03. Analysis indicates a SAA of 133 ± 5° with a worst case of ~±10°. This assumes that the Beagle 2 axis is pointed in the correct direction and is the angle measured from the front tip of the front shield.
From data deduced from the Mechanical Thermal ICD (PID-A annex) provided to RAL from Astrium/Toulouse for thermal modelling purposes the SAA was estimated to be ~122.5° at time of ejection. The SAA at ejection has been stated at 124o by ESOC. Measurement from the VMC image is only just compatible with this (assuming maximum errors).
97 Beagle-2 Mission Report Annexes
C.2.5 Beagle 2 “MLI” Feature Image 03 shows a bright feature within expected shadow on the side of Beagle 2. This feature is estimated at a maximum size of 5.8 by 8.7cm (2.5 by 3.75 pixels) without allowance for halation (image blooming). An allowance of >~15% is thought to be appropriate for this effect. This feature appears to be a real object feature with a physical height comparable to its linear dimensions in order to be visible within the shadow, unless reflection of light from another source perhaps the MLI front shield overlap is illuminating this part of the rear cover having been pushed outwards by venting at launch.
The feature appears to be situated ~ 0.17m along the surface of rear cover Beagle 2 as measured from the rear (SUEM mounting) surface, the exact position around the axis cannot be ascertained without making assumptions on the rotation rate. The position of this feature is not consistent with being directly over an ARM cover position but does appear, assuming some assumptions are made on rotation rate i.e. nominal SUEM rate, consistent with an MLI stud position. This position is however still potentially consistent with a “jet” of gas emerging from a gap at the lower end of an ARM cover position. This feature appears to be consistent with an object situated on the MLI, or disturbed, loose or damaged MLI on the side of Beagle 2 from whatever cause or even illumination of an object or reflection of light from the multi-layer insulation on the front aeroshell.
C.2.6 “Debris” Object Images 3,4,5 show an object apparently receding from Mars Express. This object is not an image artefact and reduces in size between images. The distance of the object from the VMC is unknown.
Taking data from images 3 and 4 where the object has a size > 1 pixel the following sizes can be derived assuming the object is at the distance of 25%, 50% and 75% respectively of Beagle 2 itself and is consequently moving away at the same velocity.
Beagle 2 Assumed Object X size of Object Y size of Object Distance Distance /mm /mm 20.54 m 5.14m (25%) 20.54 24.86 (Image 3) 10.27m (50%) 41.08 49.72 15.41m (75%) 61.62 74.57 35.74m 8.94m (25%) 20.31 20.31 (Image 4) 17.87m (50%) 40.61 40.61 26.81m (75%) 60.92 60.92
Table C.2 - VMC Analysis Results
The analysis shows that the object has a consistent size and just recedes from the VMC without a significant change in size or orientation, i.e. it is apparently not physically shrinking e.g. sublimating/evaporating in the sunlight given the assumption of receding at a constant (Beagle 2) velocity.
Given the sun aspect angle it is possible to estimate the minimum distance from the camera when an object is sunlight. This minimum distance is estimated as 970 ± 130 mm. Assuming the object is just in sunlight i.e. at minimum distance in Image 03 then it would have physical dimensions of ~ 3.31 ± 0.45mm by 4.08 ± 0.55mm.
As the physical size of the object cannot be derived the only conclusion that can be drawn its size is > 2.86mm but may be consistent in size with the feature on the side of Beagle 2, see section 4, or a MLI stud cover of ~20mm across or even possibly very small debris from the SUEM mechanism/pyro/frangibolts etc. or debris/ice shed from the spacecraft. The former is thought to be unlikely as no debris was seen in ground testing.
98 Beagle-2 Mission Report Annexes
C.2.7 “Background” Objects A total of 40 background objects are seen within the images 3 through 6. These appear to be consistent with a direction of motion away from Mars Express along the Beagle 2 vector of ejection. Assuming these objects are real and not camera noise (cosmic ray hits) and are at a minimum distance, see section 5 above, and occupy a single pixel they have a physical size of > 0.94mm and may be consistent with debris or ice from the spacecraft and/or SUEM.
C.2.8 Conclusions The following conclusions can be reached from analysis of the VMC images:
1. Beagle 2 was ejected at or very close to its nominal velocity 2. Sun angle is approximately correct but a large error/discrepancy exists with what was expected. 3. A feature is seen with the shadow of the back cover which appears real. This feature appears consistent with an object on the MLI, or loose or damaged MLI, cause unknown. 4. An object appears to recede from Mars Express apparently following Beagle 2. Its physical size is greater than 2.86mm but may be consistent in size with the feature on the side of Beagle 2 or a MLI stud cover of ~20mm across or even possibly very small debris from the SUEM mechanism/pyro/frangibolts etc. 5. Small “objects” are seen within the VMC images which may be real and would if real have a minimum physical size of > 0.94mm. These may be consistent with ice or small debris from the spacecraft.
99 Beagle-2 Mission Report Annexes
This page left intentionally blank.
100 Beagle-2 Mission Report Annexes
D Post-landing Search Strategy
S. Hurst, EADS-Astrium O. Blake, NSC LOCC, University of Leicester
Post Landing Comms Search and relevant Investigations and Testing 09 February 2004
D.1 Introduction Searching for a signal from Beagle post landing was supported by 3 different organisations:
• NASA JPL via MARS Odyssey. • Jodrell Bank (and other Radio Telescopes) • ESA via MARS Express.
The searches were complementary in that they did not overlap in time. Also Jodrell Bank was only looking for a UHF Carrier at or around the Beagle 2 Transceiver (MLT) transmit frequency of 401.585625 MHz, whereas the 2 spacecraft were attempting to Hail i.e. Communicate with Beagle (Transmit and Receive). However, as can be seen later, it is possible to configure the unit (MELACOM) on Mars Express which communicates with Beagle into a ‘listening’ mode like Jodrell Bank and store all the Receiver output into the onboard memory unit (SSMM) for later downlinking to ground where it can be processed and any coherent signal recovered from the noise.
This mode is called CANISTER mode and was used twice in the search for a signal from Beagle. In addition as a validation of the respective systems, a test was performed between MELACOM on Mars Express and MER-Spirit on the surface of Mars to be sure that MELACOM was able to receive a UHF signal at 401.5 MHz. This was successfully completed on 11th January.
Apart from the orbits being different:
• Odyssey is Sun-synchronous i.e. It passes over the landing site twice a day at sensibly constant times of Sol (Day on Mars) and at sensibly constant altitudes. Due to orbit parameters the landing site is seen up to 4 times however the other passes are at very low elevation. • Mars Express is in an elliptical orbit with an inclination angle to the equator which means its pass times over the landing site are variable in terms of both times of Sol and also duration. Some passes are Apocentred while others are Pericentred and so altitudes vary from <400Km to >12,000Km. it was found that the Hailing procedure on Odyssey is different to Mars Express, although both are compliant to the Proximity-1 protocol.
After discussions on 30th December with NASA JPL about ‘Blind’ commanding i.e. Sending TC’s on the forward link without waiting for a response from Beagle (in case Beagle’s return link was non operational), it was found that all the previous Odyssey contacts had in fact sent the Command files that LOCC had prepared for each agreed pass.
Details of the contents of each Command file are described below in Section D.4.
There are several different types of Communication sessions, both from the onboard Beagle point of view and also from the Orbiter/Ground operation.
101 Beagle-2 Mission Report Annexes
D.1.1 Beagle onboard Comms Session types (Modes).
D.1.1.1 Operational mode (LSW mode 1)
Executes stored or uploaded Comms sessions.
D.1.1.2 Comms Search Mode 1 (CSM1 = LSW mode 3)
Automatically adds am and pm Odyssey passes (2 per Sol) onto the Mission Event Timeline (MET). Stored Comms sessions are also executed. The inserted communication sessions are inserted at 03:35 and 15:35 Local True Solar Time (LTST) for a duration of 80 mins. The transmit carrier is on for 60 secs and off for 540secs on a repetitive cycle with the receiver continuously on for the whole period. This as in other comms modes described below is dependant on the LSW having determined LTST from the Solar Panel currents at Dawn and Dusk.
D.1.1.3 Comms Search Mode 2 (CSM2 = LSW mode 4)
Automatically inserts sessions onto the timeline depending on whether the LSW has determined the Local True Solar Time (LTST), from the Solar Panel currents at Dawn/Dusk, to be either Daytime (10:00 to 18:00) or Nightime (18:00 to 10:00).
• Day Mode The Transceiver is cycled On for 59 minutes every hour. The 59 min. ON cycle consists of : o 10 seconds of Expedited mode Telemetry (ie transmitted regardless of the hail status of the communications session). o ‘Listen’ mode for the remainder of the 59 mins, but with the Carrier (unmodulated) cycled ON for 1 minute every 10 mins., starting with 9 mins of Carrier OFF.
• Night Mode The Transceiver is cycled ON for 1 minute every 5 minutes, starting with 4 minutes OFF. During the 1 minute ON time the MLT is in ‘Listen’ mode with the Carrier (unmodulated) being cycled as in Day mode, but because the MLT is switched OFF after 1 minute it never gets to the Carrier ON part (after 9 mins) in order to conserve energy during the night.
D.1.1.4 AutoTransmit Mode (ATM = LSW mode 5)
ATM is the same as CSM2 above except for the duration of the Expedited mode Telemetry portion. This is the final Comms Search mode so therefore it is assumed that it may as well transmit as much data as it can before the Battery goes flat, in case someone is listening but for some reason cannot Hail. As described above in CSM2 the Expedited mode is active for 10 seconds during which time telemetry from the Summary Lander Status (Packet Id=128) is transmitted for approximately 0.25 seconds. In ATM mode the Expedited mode transmission consists of all of the Housekeeping and some Science packets, sufficient to fill the 10seconds duration of transmission.
All communications modes are dependant on the Battery State of Charge (BSC) which can veto a session if it is low. A lower state of BSC is invoked for CSM2 compared to all other modes. In CSM1 a transmission will occur if a hail if received but the BSC is lower than the set value however it is limited to a duration of 3 minutes.
102 Beagle-2 Mission Report Annexes
D.1.2 Jodrell Bank Comms Session type Three Jodrell Bank Comms sessions were pre-programmed into the memory (EEPROM) prior to ejection from Mars Express with times when the expected Beagle landing site was visible from Earth, see table in Annex J.8.
They were programmed as normal ‘listening’ sessions with unmodulated Carrier being cycled ON for 10 seconds and OFF for 50 seconds. The LSW continues with the Carrier cycling throughout the ‘pass’ time when the MLT is ON until it receives a valid Hail, when it switches TM modulated Carrier ON for the remainder of the pass, assuming the MLT Carrier Lock stays present.
Since Jodrell Bank is not able to transmit it was expected that the unmodulated Carrier cycling would be seen at the appropriate times throughout the passes. Also since unmodulated Carrier has a better power level than with modulation, the link margin was expected to be greater than 10dB when both Antennas (Beagle and Jodrell Bank) were aligned.
D.1.3 NASA Odyssey Comms Session types All 8 pre-programmed Odyssey Comms sessions were configured with unmodulated Carrier ON/OFF cycles of 10 seconds ON ; 50 secs OFF in the listening mode for 40 minutes duration.
CSM1 auto-inserted Odyssey sessions are configured for starting at 03:35 (am) and 15:35 (pm) LTST and durations of 80 minutes with unmodulated Carrier cycles of 9 mins OFF and 1 minute ON.
Note * denotes that this is the nominal duration, but will be extended to 80 mins if the previous session(s) were ‘missed’ i.e. Unsuccessful.
D.1.4 ESA Mars Express Comms Session types All 4 pre-programmed Mars Express Comms sessions were configured with unmodulated Carrier ON/OFF cycles of 10 seconds ON ; 50 secs OFF in the listening mode.
Because of the nature of the Mars Express orbit and its consequent pass times over the Beagle landing ellipse, it is not possible to arrange for auto-insertion of MEx comms sessions in either CSM1 or 2.
Because of the MEX power anomaly and required adjustment to the MEX orbit, the first MEX communications session occurred on day 13 after landing rather than day 10 as planned, rendering some of the pre- programmed comms sessions on board Beagle 2 obsolete. D.2 NASA Odyssey Communication modes Odyssey has 3 different ways of communicating to Beagle :
1) Normal - Sequence controlled mode where Odyssey Hails and waits for a response from Beagle and then performs Proximity-1 protocol communication.
2) Unreliable Bit stream - Expedited mode where Odyssey Hails but does not wait for a Beagle response before sending the forward link data (telecommands) in other words ‘Blind’ Commanding. Beagle recognises the Expedited mode Hail and returns telemetry data without waiting for a Prox-1 Acknowledge from Odyssey i.e. ‘Blind’ Telemetry.
3) Listen - Odyssey receiver sweeps over the normal band and records Carrier and Bit Lock statuses throughout the pass.
103 Beagle-2 Mission Report Annexes
The Bit Rates specified in the Hail sequences of both the Normal and the Unreliable Bit Stream modes were always Fwd : 8Kbps and Rtn: 32Kbps except on 5 passes between 26th and 28th Dec when the Return link was set to 8Kbps.
These are noted in section D.10 which defines all the Contacts and the results obtained.
D.3 ESA Mars Express Communication modes Mars Express has 3 ways of communicating with Beagle via the MELACOM unit in UHF mode.
• Normal - Sequence controlled mode where MELACOM Hails and waits for a response from Beagle and then performs Proximity-1 protocol communication.
• Doppler Listen - MELACOM receiver sweeps over a larger frequency range than normal and records the Carrier Lock and Bit Lock statuses throughout the listening period. No attempt to Hail is performed so MELACOM is only looking for Carrier (modulated or not) from Beagle.
• Canister Listen - MELACOM receiver sweeps over the Doppler frequency range , but as well as storing the Carrier and Bit Locks it also stores the Receiver output directly into Memory, where it is later downlinked to Earth for post-processing in the same way as Jodrell Bank does i.e. It can recover a Carrier signal embedded in Noise. This gives a 20 – 40dB gain over the normal Receiver sensitivity, depending on how much post-processing (iterative) is performed.
The Bit Rates specified in the Hail sequences of the Normal Comms mode passes were always Fwd : 8Kbps and Rtn : 8Kbps except on 2 passes. On 7th Jan it was set to 2K/2K and on 10th Jan it was first set to 2K/8K and then 8K/2K during the pass.
These are noted on the Table in Annex D.10 which defines all the Contacts and the results obtained.
104 Beagle-2 Mission Report Annexes
D.4 Command Files prepared for each Contact
D.4.1 File 1 (Sol 1 - Sol 3 (am)) • 3x Connection Test TCs (17,1) • Group F data logging during Comms session. • Add Comms Sessions (Sols 8-9) to MET. • Adjust Solar Panel 3 from 160º to 180º • Request TM from Group F logging.
D.4.2 File 2 (Sol 3 (pm) - Sol 6 (pm)) • 5 x Connection Test TCs (17,1) • Force LOBT to expected Comms session start time. • Group F logging • Add Comms Sessions (Sols 5 -19am) to MET. • Adjust Solar Panel 3 from 160º to 180º
D.4.3 File 3 (Sol 7 (pm) - Sol 9 (pm)) • As File 2 plus: • Reset LOBT0-to-Midnight correlation. • Reset BSC_Usage_Enable_Flag to disable. • Add Comms sessions for Odyssey and MEx.
D.4.4 File 4 (Sols 14 - 19; 31 – 32) • As File 3 plus: • Max_Comms_Sessions_missed set to 1 • CSM1_Search_Period set to 1 • Comms_Silence_Duration set to 86400 secs (1 Day) • Battery Protection Logic changed : • Battery Current set to Disabled • Main Bus Current set to Disabled • Main Bus Voltage set to Disabled • LSP Motor Current threshold set to 200mA (from 100) • All 4 Solar Panels adjusted from 160º to 180º • Add Comms sessions (MEx Sols 23-32; ODy Sols 24-38)
D.4.5 File 5 ‘Blind’ (30/31st Jan) • Expedited mode Hail • 1 x Connection test (17,1) • LSW Reboot • Expedited Re-Hail • As File 4 above.
D.5 Criteria for Transitions between Comms Search Modes
D.5.1 Operational Mode (1) LSW initialises itself into mode 1, where all functions are enabled and stored Comms Sessions are put onto the MET. It will stay in mode 1 until one or more of the following criteria are met :
• 10 of the 15 stored Comms sessions are ‘missed’ - CSM1
105 Beagle-2 Mission Report Annexes
• The Comms Silence Watchdog (no Prox-1 Active signal) times out after 5 days - CSM1 • Processor Hardware Reset (with no SBU clock) - to CSM2 • Processor Software Reset - Safe mode (2)
D.5.2 Safe Mode(2) In Safe mode interfaces between the Processor and all experiments (GAP, PAW, ARM) are disabled and only Comms sessions are executed from the MET. Requires a TC command to return to Operational or CSM1 modes. If a soft Reset occurs during a Comms session then that session is restored to the MET after recovery, however a Re-Hail will be required to continue the TC/TM links.
D.5.3 Comms Search Mode 1 (3) In CSM1, Odyssey morning and afternoon Comms sessions are automatically inserted onto the MET and will be executed as well as the stored sessions. However if LSW has not yet been able to determine local Martian time (LTST) from Dawn/Dusk transitions of the Solar Panels it will jump to CSM2 directly. It will remain in CSM1 until one or more of the following criteria are met:
• A Comms session is successful (Prox-1 goes Active) - Safe mode • 10 ‘inserted’ Odyssey sessions are missed (5 Sols) - CSM2
D.5.4 Comms Search Mode 2 (4) In CSM2 Comms sessions are automatically inserted onto the MET depending on whether it is day or night time. If LSW has not been able to determine local time it will default to Night. The Comms sessions for day and night are as defined in D.1.1 above.
It will remain in CSM2 until one or more of the following criteria are met:
• A Comms session is successful (Prox-1 goes Active) - Safe mode • 10 days elapse - ATM mode
D.5.5 AutoTransmit Mode (5) This is the final Comms Search mode and so there is only one recovery:
• A Comms session is successful (Prox-1 goes Active) - Safe mode.
106 Beagle-2 Mission Report Annexes
D.6 Factors which may affect the Comms Sessions. Several factors can influence the characteristics of the Comms sessions that are executed on the MET. These factors are summarised below together with the particular characteristic(s) that is/are influenced (see Lander Operations Manual –SSSl-BGL2-LOM-0001 for details). However the characteristics fall into 3 groups :
• Timing of sessions - Start and End times can be adjusted. • Carrier-only transmission - Disabling totally or partially is possible. • Expedited mode Telemetry – Disabling totally is possible.
D.6.1 Missed Comms Sessions If a Comms session is missed in Operational mode (1) then the next session is set to ‘extended’ duration which means starting 20 minutes earlier and ending 20 minutes later than the stored values.
D.6.2 Battery State-of-Charge Assessment. The Battery state of charge (BSC) is continuously being calculated by LSW every minute from the terminal voltage (TM parameter A13), the charge/discharge current (TM parameter A17), the Battery temperature (TM parameter A?) and a look-up table (LUT), stored in EEPROM, defining the internal resistance vs. temperature of the Battery. The resulting value of the real-time calculation is then compared to another look-up table (stored in EEPROM) which defines the BSC for that particular time of day (LTST) which is necessary as a minimum to service the essential loads. This comparison yields a boolean result :
Good means there is excess Battery charge to service non-essential loads. Bad means that non-essential loads will be disabled.
If Carrier-only operation has been scheduled either:
• As part of the pre-stored (EEPROM) Comms session characteristics in mode 1. • In the CSM1 Comms session characteristics (1 minute every 10) • In the CSM2 or ATM Comms session characteristics, depending on Day or Night.
Then this will be disabled (suppressed) and so Odyssey or MEx or Jodrell Bank will not see any Carrier. Also any Expedited mode Telemetry in CSM2 or ATM will be suppressed. However, in addition in CSM1 normal transmission will be limited to 3 minutes.
D.6.3 Battery Protection Logic (BPL) In addition to the above defined BSC protection, there is another set of logic to protect the Battery if any of the following parameters fall below given limits (commandable)
• Battery voltage • Battery current • Main Bus current
The result is that all non-essential loads ,including the MLT are switched off. However the MLT is considered as a nearly essential load and is switched back on again, but of course the transmitter will be off, conserving power until it is re-Hailed. Any/all of the above parameters can be disabled by command.
107 Beagle-2 Mission Report Annexes
D.7 Summary of Events The full timetable of events together with the results obtained is shown in the table in this annex in section D.10. However, it is instructive to summarise the results and the logic for proceeding on the course of action (or inaction) actually taken.
1. None of the pre-stored 15 Comms sessions (Ody, JB and MEx) or other passes supported (>30) resulted in seeing either a clean (unmodulated) or modulated Carrier (Hail response or Expedited).
Possible Reasons
• MLT transmitter not working correctly but hearing Hails, so in mode 1 or 2 • MLT not being switched on at expected times, so missing sessions. • BSC and BPL maybe causing suppression of Carrier-only and also switching MLT off whenever Tx is switched on after receiving a Hail(Tx allow). This does not account for Hail failures. • Could have jumped to CSM1 (10 missed sessions) on 31st Dec • Could have jumped to CSM2 (h/w reset) on 25th Dec or later. • LSP deployment may have failed so Antenna cannot receive or radiate. • Beagle may have not landed in the expected ellipse.
2. None of the ‘Blind’ command files (Odyssey on 31st Dec and 30/31st Jan) has resulted in seeing either a clean (unmodulated) or modulated Carrier(Hail response or Expedited)..
Possible Reasons
• MLT transmitter not working correctly but hearing Hails, so in mode 1 or 2 • LSP deployment failed so Antenna cannot receive or radiate. • Beagle did not land in the expected ellipse.
3. None of the different versions of Hailing (Bit rates – valid and invalid; Sequential or Expedited modes) has resulted in seeing either a clean (unmodulated) or modulated Carrier(Hail response or Expedited)
Possible Reasons
• As for (2) above.
4. Keeping Radio Silence from 12th to 22nd Jan (10 days) to force CSM2 has not resulted in seeing either a clean (unmodulated) or modulated Carrier (Hail response or Expedited).
Possible Reasons
• As for (2) above.
108 Beagle-2 Mission Report Annexes
D.7.1 Course of Action From the above descriptions of the Comms Search mode logic and the factors affecting the execution of the sessions it is clear that it is impossible to be sure of what mode the Beagle Comms logic was in at any time post-landing.
For instance, it could be that Beagle went into CSM2 directly on completion of LSW initialisation and LSP Deployment on Sol 1 (25th Dec) due to a hard reset.
Or it could have remained in Operational mode (1) because it could hear the Hails but could not respond (broken MLT Transmitter or Antenna malfunction only affecting the Tx function).
Or it could have jumped into CSM1 on Sol 7 pm (31st Dec) having missed 10 sessions.
All the above options are also overlaid with uncertainty about the effects of BSC assessment and BPL logic on the Carrier-only mode making the situation more uncertain.
It was decided that after the first attempt by MEx to Hail on 12th Jan was unsuccessful, the only way to be sure of what state Beagle was in was to maintain radio-silence for 10 days in order to force Beagle into CSM2 then try to make contacts with both Odyssey and MEx at the appropriate times when the MLT transceiver would be ON.
On the first Mex contact after the enforced radio-silence (22nd Jan) it was agreed (B2 and ESA) that it was better to select a ‘Canister mode’ listen and try to detect a non-nominal (low level or off-tune frequency) carrier rather than a Hail which might put Beagle back into ‘Safe’ mode without necessarily sending any telemetry. MEx Hails were performed on the passes of 24th and 25th Jan.
Finally it was agreed that on the passes around end Jan/beginning Feb that we should try more ambitious ‘Blind’ commanding files from Odyssey including an LSW Reboot TC and an ‘Invalid’ Hail/Re-Hail attempt by MEx.
• 30th Jan Ody LSW Reboot ‘Blind’ command file. • 31st Jan Ody LSW Reboot ‘Blind’ command file. • 3rd Feb MEx peripass ‘Invalid’ Hail/Re-Hail attempt.
None of the attempts was successful, leaving 4 possible scenarios:
1) Catastrophic failure during EDLS. 2) Failure to complete Deployment before Battery went flat. 3) Failure of Antenna to Transmit and Receive. 4) Failure of MLT to Transmit.
109 Beagle-2 Mission Report Annexes
D.8 Summary of Relevant Investigations During the post landing period various investigations were conducted on the Ground Test Model (full set of Beagle 2 avionics plus working lander mechanisms) to clarify the possible reasons for not obtaining any response from Beagle. These fall into 3 categories:
• Software protections. • Transceiver (MLT) communicating modes. • Full Deployment test
D.8.1 Software Protections There are 2 categories of protection as defined above in paragraphs D.6.2 and D.6.3:
• BSC assessment. • BPL actions.
D.8.1.1 BSC assessment
On 7th Jan a test was conducted on the GTM to investigate the actual response when in CSM2 Night mode. This was induced by performing a h/w Reset on the processor and because the SBU clock was not working (as on FM) it forced a jump into CSM2 and defaulted to Night mode because no Dawn/Dusk transitions had been simulated.
No Carrier was seen even though MLT had been set to Expedited mode (status= 1446hex). Day mode was then provoked by simulating a Dusk transition on the Solar Panels, which resulted in seeing a clean Carrier on the Spectrum Analyser for 1 minute every 10 mins as expected, but still no Expedited Carrier for 10 seconds at the beginning of the 1 hour cycle.
On 8th Jan 2 patches were provided by SciSys (GJ) to (i) change the default times for cycling the Carrier in CSM2 Night (ii) set the BSC Usage Enable Flag to Disable in order to confirm the above results. Again clean Carrier was seen as expected but not Expedited Carrier.
D.8.1.2 BPL Actions
The original validation test for Battery protection had been performed on the ETM in September while investigating another MLT anomaly. The ETM had been set up with a ~15% BSC and LSW started. Because there were no temperature sensors for missing equipment (experiments, EDLS etc.) the LSW thermal control software read the temperature parameters as very low and switched on the appropriate heaters. These had been simulated with resistive loads on the ETM so drawing the correct currents from the Battery. When CSM2 night was induced it was seen that the switching on of the MLT caused sufficient extra current drain from the Battery as to cause the Battery protection logic to switch off all the Heaters and the MLT. The MLT came on again after ~2 seconds and continued working normally because by now the Battery had recovered from the overload and could support the reduced load satisfactorily. The MLT was hailed successfully and the History Log recovered in order to determine what had happened. The above explanation confirmed that it was the Battery Current that had tripped the BPL threshold. During the tests on 7 and 8th Jan for testing the BSC, TCs were sent to the GTM to disable 3 of the BPL parameters:
• Battery/Main Bus voltage • Battery current • Main Bus current
110 Beagle-2 Mission Report Annexes
D.9 MLT Communicating modes Two particular aspects of MLT UHF operation were tested post-landing on the GTM and ETM:
• Expedited and Carrier-only modes. • Blind Commanding configuration.
D.9.1.1 Expedited and Carrier-only mode testing
A series of tests had been performed in September by Astrium and Qinetiq on the ETM together with the MELACOM EQM with the MLT in RF mode while investigating the Cruise in-flight anomalies. At that time the flight version of LSW was V2.2-0. It was found that the LSW did not configure the MLT correctly when transferring from Expedited mode back to normal Sequence controlled mode, such that Hails would not be accepted and hence loss of control of Beagle would ensue.
A new version of LSW was produced to correct this problem (and others): V2.5-0 and this was tested also on the ETM and MELACOM in October. These tests were successful in that it was confirmed that Hailing after transitioning back from Expedited to Sequenced mode worked correctly. However it was not verified at this time that a carrier was actually produced in Expedited mode, only that the status of the MLT indicated that it was correctly in expedited mode for the expected duration (8-10 seconds) in CSM2.
During the BSC tests on 7/8th Jan using the GTM, which now had the EM MLT (from ETM) integrated, it was found that the Expedited mode part of CSM2 did not produce a modulated Carrier (see 8.1.1 above) so SciSys produced 2 new patches (13 = set Bypass type to EGSE: was MEx; 14 = set Tx Allow ON: was OFF). These were tested both separately and together to see if they corrected the anomaly but without success. However, unmodulated Carrier was always seen at the expected times.
Based on these results it is concluded that the AutoTransmit mode will not differ from the CSM2 mode of operation in that in both cases there will be no Expedited mode telemetry, only clean, unmodulated Carrier cycling.
However this does not account for the lack of any return signal, so there must be at least one other anomaly which has occurred.
D.9.1.2 Blind Commanding configuration testing
On 28th Dec the GTM was set up to support a ‘Blind’ commanding session. Qinetiq (MC) was present to configure a Transfer Frame file to include an expedited mode Hail followed by a series of TCs ( File 3 as defined in 4.3 above).
As this file included a command to adjust Solar Panel 3 from 160º to 180º it was possible to validate the successful transmission of this Transfer Frame through the MLT in ‘Blind’ mode i.e. without relying on a return signal by observing the movement of the panel. It had to be assumed that the other commands were executed successfully since there was no way of verifying them without a telemetry signal. On 27th Jan, after it had been decided to perform more ‘blind’ command attempts via Odyssey on the 30/31st Jan passes, the files were set up on the GTM EGSE.
This time the file was to include an LSW Reboot TC, considered to be a last ditch attempt to resolve any software induced problem onboard Beagle. Since the LSW Reboot TC would cycle power to the MLT it was necessary to include a re-Hail in the Transfer Frame in order to re-configure the MLT back into expedited mode for the remainder of the blind command file (File 4 as defined in 4.4 above).
D.9.2 Full Deployment Test On 27th January a full Deployment test was performed on the GTM. It was felt that although Deployment tests had been previously successfully completed on the GTM, they had been done with earlier versions of 111 Beagle-2 Mission Report Annexes
LSW or without all the current patches for PSW and LSW and so this test would validate the current configuration of the FM Beagle as closely as possible.
As before, the test started in PSW Cruise mode, through Separation into EDLS, through landing to Handover and through Deployment into landed operations including all the default tasks on the timeline (MET) which included the default Comms sessions.
The first Comms session was configured for ‘EGSE’ mode so that the MLT could be Hailed from the EGSE and the telemetry returned. As expected the EDLS data was returned together with other log files. However because the QM PAW was not available no images were taken or returned but the LSW correctly went into Safe-Op mode as expected.
Since there was a TC link available File 1 commands (as in 4.1 above) were sent as if it was the FM. Solar Panel 3 was seen to move as expected.
The second default Comms session (Jodrell Bank) was configured for UHF mode so that the Carrier on/off cycling could be observed on a Spectrum Analyser. This duly occurred at the expected time and for the expected duration.
One difference from the last test done (after the LSW V3.0-0 was issued) was that the Handover time (LOBT zero) provided by PSW to LSW was significantly different (~1000 seconds vs. 668 seconds). This was assumed to be due to the different timeouts and Algorithm parameters now in the EDLS due to updated information on Accelerometer performance (offset, drift etc. during Cruise) and not considered critical. On the FM, timing of events on the MET were calculated with respect to the estimate of LOBT zero to UT correlation provided by ESOC from the MEX trajectory and atmospheric descent predictions for Pilot Parachute deployment time.
D.10 Detailed Summary of all Communications Sessions The following tables detail the characteristics of each communication session.
112 Beagle-2 Mission Report ANNEXES
D.10.1 December 2003
1000-1800 LTST Bold= Order of Pass Pass Pass Compatible Compatible Elevation Compatible Cmds Default Sol Time Time Approx Time MeX Day/ Beagle 2 Measured with with Year Date Day Elevation Pass UTC UTC with Comms Comms No. Start Max Elev LTST End Range Night Response AGC CSM1 CSM1 Quality CSM2 Type Session UTC UTC UTC LTST 03:35 LTST 15:35 Day 2003 25-Dec 359 O 1 1 05:25 05:34 16:08 87.90 05:43 Day Good No Data No Carrier 17:19 Yes Yes O 2 16:27 16:34 02:51 11.50 16:41 Night Poor No Carrier CSM1:AM 2 17:19 18:39 Night T O 2 18:23 18:31 04:44 26.70 18:40 Night Marginal No Carrier Yes J 2 2 22:20 23:00 09:06 23:40 Day No Carrier 26-Dec 360 O 2 05:09 05:17 15:13 19.70 05:25 Day Poor No Carrier 17:58 05:38 Yes CSM1:PM 2 05:38 06:58 Day O 2 07:07 07:14 17:07 16.60 07:22 Day Poor No Carrier Yes CSM1:AM 3 17:58 19:18 Night O 3 3 18:06 18:15 03:49 63.40 18:23 Night Good No Data No Carrier Yes F O 3 20:09 20:12 05:44 1.70 20:15 Night Poor No Data No Carrier J 4 3 23:00 23:40 09:07 00:20 Day No Carrier 27-Dec 361 O 3 04:59 05:00 14:18 0.20 05:01 Day Poor No Carrier 18:38 06:18 Yes CSM1:PM 3 06:18 07:38 Day O 5 3 06:49 06:58 16:13 77.2 07:07 Day Good No Data No Carrier Yes Yes O 4 17:51 17:58 02:55 13.4 18:05 Night Poor No Carrier CSM1:AM 4 18:38 19:58 Night S O 4 19:48 19:55 04:49 23.3 20:03 Night Marginal No Carrier (Close) J and S 6 4 22:56 23:34 08:24 00:16 Day No Carrier 28-Jan 362 O 4 06:33 06:41 15:18:00? 22.6 06:49 Day Marginal No Carrier 19:18 07:37 Yes CSM1:PM 4 07:37 08:57 Day O 4 08:31 08:38 17:12:00? 14.4 08:46 Day Poor No Carrier Yes Yes CSM1:AM 5 19:18 20:38 Night Su O 7 5 19:30 19:39 03:55 73.2 19:48 Night Good No Data No Carrier Yes O 5 21:35 21:37 05:49 0.5 21:38 Night Poor No Carrier 29-Dec 363 O 5 06:21 06:25 14:23 1.5 06:27 Day Poor No Carrier 19:57 07:37 Yes CSM1:PM 5 07:37 08:57 Day O 8 5 08:13 08:22 16:17 67.1 08:31 Day Good No Data No Carrier Yes Yes M O 6 19:15 19:23 03:00 15.5 19:30 Night Poor No Carrier CSM1:AM 6 19:57 21:17 Night O 6 21:12 21:20 04:54 20.4 21:38 Night Marginal No Carrier (Close) 30-Dec 364 O 9 6 07:57 08:05 15:22 25.8 08:14 Day Marginal No Data No Carrier 20:37 (Close) 08:17 Yes CSM1:PM 6 08:17 09:37 Day T O 6 09:56 10:03 17:17 12.4 10:09 Day Poor No Carrier Yes CSM1:AM 7 20:37 21:57 Night O 10 7 20:54 21:03 03:59 83.8 21:12 Night Good No Data No Carrier Yes 31-Dec 365 O 7 07:45 07:45 14:27 2.8 07:52 Day Poor No Carrier 21:17 08:57 Yes CSM1:PM 7 08:57 10:17 Day O 11 7 09:38 09:47 16:22 58.2 09:55 Day Good No Data No Carrier Yes Yes W O 8 20:39 20:47 03:04 17.8 20:54 Night Poor No Carrier CSM1:AM 8 21:17 22:37 Night O 8 22:36 22:44 04:59 17.8 22:52 Night Poor No Data No Carrier (Close) Yes
Table D.1 - December Overflights
113 Beagle-2 Mission Report ANNEXES
D.10.2 January 2004
1000-1800 LTST Bold= Pass Pass Pass Compatible Compatible Order of Default Elevation Compatible Cmds Sol Time Time Approx Time MeX Day/ Beagle 2 Measured with with Year Date Day Comms Elevation Pass UTC UTC with Comms No. Start Max Elev LTST End Range Night Response AGC CSM1 CSM1 Session Quality CSM2 Type UTC UTC UTC LTST 03:35 LTST 15:35 Day 2004 01-Jan 1 O 8 09:21 09:30 15:28 29.5 09:38 Day Marginal No Carrier 21:56 Yes 09:36 Yes CSM1:PM 8 09:36 10:56 Day O 8 11:20 11:27 17:22 10.6 11:33 Day Poor No Carrier Yes CSM1:AM 9 21:56 23:16 Night Th MC Scrubbed 9 22:25 22:55 Night Yes O 9 22:19 22:27 04:04 85.20 22:36 Night Good No Data No Carrier Yes 02-Jan 2 O 9 09:08 09:13 14:33 4.1 09:17 Day Poor No Carrier 22:36 10:16 Yes CSM1:PM 9 10:56 12:16 Day O 9 11:02 11:11 16:27 50.5 11:19 Day Good No Data No Carrier Yes Yes F O 10 22:03 22:11 03:09 20.3 22:19 Night Marginal No Data No Carrier (Close) MC Scrubbed 10 22:25 22:55 Night Yes CSM1:AM 10 22:36 23:56 Night 03-Jan 3 O 10 00:01 00:08 05:04 15.4 00:16 Night Poor No Carrier 23:15 10:56 O 10 10:45 10:54 15:32 33.6 11:02 Day Good No Data No Carrier Yes Yes CSM1:PM 10 10:56 12:16 O 10 12:45 12:51 17:26 8.9 12:57 Day Poor No Carrier Yes S O 11 21:53 21:55 02:15 0.8 21:57 Night Poor No Carrier CSM1:AM 11 23:15 00:35 Night O 11 23:43 23:51 04:09 74.6 00:00 Night Good No Data No Carrier Yes 04-Jan 4 O 11 10:32 10:37 14:37 5.5 10:42 Day Poor No Carrier 23:55 11:35 Yes CSM1:PM 11 11:35 12:55 Su O 11 12:26 12:35 16:31 44 12:43 Day Good No Data No Carrier Yes Yes O No Hailing 12 23:27 23:34 03:14 23.2 23:43 Night Marginal No Carrier (Close) CSM1:AM 12 23:55 01:15 Night 05-Jan 5 O No Hailing 12 01:25 01:32 05:08 13.3 01:39 Night Poor No Carrier 12:15 O No Hailing 12 12:09 12:19 15:37 38.4 12:27 Day Good No Carrier Yes Yes CSM1:PM 12 12:15 13:35 Day O No Hailing 12 14:09 14:15 17:31 7.3 14:21 Day Poor Yes M O No Hailing 13 23:15 23:19 02:19 2 23:22 Night Poor No Carrier MC 13 23:00 23:30 Night 06-Jan 6 CSM1:AM 13 00:35 01:55 Night O No Hailing 13 01:07 01:15 04:13 64.8 01:24 Night Good No Carrier Yes 00:35 12:54 M 12 13 11:00 11:40 14:22 12:10 Day N/A Yes T O No Hailing 13 11:55 12:01 14:42 7 12:07 Day Poor No Carrier Yes CSM1:PM 13 12:54 14:14 Day O No Hailing 13 13:50 13:58 16:36 38.4 14:07 Day Good No Carrier Yes Yes 07-Jan 7 O No Hailing 14 00:51 00:59 03:18 26.4 01:07 Night Marginal No Carrier (Close) 01:14 13:34 CSM1:AM 14 01:14 02:34 Night O No Hailing 14 02:49 02:56 05:13 11.4 03:03 Night Poor No Carrier W M Hailing 14 12:12 12:15 14:14 90? 12:18 800 Day Good No Data No Carrier Yes O Hailing 14 13:33 13:42 15:42 44 13:50 Day Good No Data No Carrier Yes Yes CSM1:PM 14 13:34 14:54 Day O No Hailing 14 15:34 15:39 17:36 5.9 15:44 Day Poor No Carrier Yes 08-Jan 8 O No Hailing 15 00:38 00:42 02:24 3.3 00:47 Night Poor 01:54 14:14 CSM1:AM 15 01:54 03:14 Night O Hailing 15 02:31 02:39 04:18 56.3 02:48 Night Good No Data No Carrier Yes Th M Cannister 15 12:50 12:52 14:13 90? 12:55 750 Day Good No Carrier Yes O No Hailing 15 13:19 13:25 14:47 8.5 13:31 Day Poor No Carrier Yes CSM1:PM 15 14:14 15:34 Day O No Hailing 15 15:14 15:22 16:41 33.7 15:31 Day Good No Carrier Yes Yes 09-Jan 9 O No Hailing 16 02:15 02:23 03:23 30 02:31 Night Good No Carrier (Close) 02:33 14:53 CSM1:AM 16 02:33 03:53 Night O No Hailing 16 04:14 04:20 05:18 9.7 04:27 Night Poor No Carrier F M Hailing 16 13:27 13:29 14:10 90? 13:33 750 Day Good No Data No Carrier Yes CSM1:PM 16 14:53 16:13 Day O No Hailing 16 14:57 15:06 15:47 50.5 15:14 Day Good No Carrier Yes Yes O No Hailing 16 16:58 17:03 17:40 4.5 17:08 Day Poor No Carrier Yes 10-Jan 10 O No Hailing 17 02:02 02:06 02:29 4.6 02:12 Night Poor No Carrier 03:13 15:33 CSM1:AM 17 03:13 04:33 Night O No Hailing 17 03:55 04:03 04:23 48.9 04:12 Night Good No Carrier Yes S M Hailing 17 14:04 14:06 14:10 90? 14:09 750 Day Good No Data No Carrier Yes O No Hailing 17 14:42 14:49 14:51 10.2 14:56 Day Poor No Carrier Yes CSM1:PM 17 15:33 16:53 Day 114 Beagle-2 Mission Report ANNEXES
1000-1800 LTST Bold= Pass Pass Pass Compatible Compatible Order of Default Elevation Compatible Cmds Sol Time Time Approx Time MeX Day/ Beagle 2 Measured with with Year Date Day Comms Elevation Pass UTC UTC with Comms No. Start Max Elev LTST End Range Night Response AGC CSM1 CSM1 Session Quality CSM2 Type UTC UTC UTC LTST 03:35 LTST 15:35 Day O No Hailing 17 16:38 16:46 16:45 29.6 16:55 Day Marginal No Carrier Yes Yes 11-Jan 11 O No Hailing 18 03:38 03:46 03:28 34.2 03:55 Night Good No Carrier Yes 03:52 16:12 CSM1:AM 18 03:52 05:12 Night S O No Hailing 18 05:38 05:44 05:23 8.1 05:50 Night Poor No Carrier CSM1:PM 18 16:12 17:32 Day O No Hailing 18 16:21 16:30 15:51 58 16:38 Day Good No Carrier Yes Yes O No Hailing 18 18:22 18:27 17:45 3.2 18:31 Day Poor Yes 12-Jan 12 M Hailing 19 02:02 01:08 76 03:14 15000 Night Marginal No Data No Carrier 04:32 16:52 O No Hailing 19 03:25 03:30 02:34 6 03:36 Night Poor No Carrier M CSM1:AM 19 04:32 05:52 Night O No Hailing 19 05:19 05:27 04:28 42.6 05:36 Night Good No Carrier Yes M 13 19 03:10 15:58 14:03 04:20 Day N/A Yes O No Hailing 19 16:06 16:13 14:56 12 16:20 Day Poor No Carrier Yes CSM1:PM 19 16:52 18:12 Day O No Hailing 19 18:02 18:10 16:50 26.1 18:18 Day Marginal No Carrier Yes Yes 13-Jan 13 O No Hailing 20 05:02 05:10 03:32 39 05:19 Night Good Yes CSM1:AM 20 05:12 06:32 Night T O No Hailing 20 07:03 07:08 05:27 6.6 07:14 Night Poor M 14 20 15:38 15:58 14:03 16:18 Day N/A CSM1:PM 20 17:31 18:51 Day O No Hailing 20 17:45 17:54 15:56 66.7 18:03 Day Good Yes Yes O No Hailing 20 19:47 19:51 17:50 1.9 19:54 Day/Night Poor 14-Jan 14 M Not available 21 03:18 04:04 01:05 72 04:51 13857 Night Marginal 05:51 18:11 O No Hailing 21 04:48 04:54 02:38 7.5 05:00 Night Poor W CSM1:AM 21 05:51 07:11 Night O No Hailing 21 06:43 06:51 04:32 37.3 07:00 Night Good Yes O No Hailing 21 17:30 17:37 15:01 13.9 17:44 Day Poor Yes CSM1:PM 21 18:11 19:31 Day O No Hailing 21 19:26 19:34 16:55 22.9 19:42 Day Marginal Yes Yes 15-Jan 15 O No Hailing 22 06:26 06:34 03:37 44.7 06:43 Night Good Yes 06:31 18:50 CSM1:AM 22 06:31 07:51 Night T O No Hailing 22 08:27 08:32 05:32 5.1 08:37 Night Poor M Listen 22 16:29 16:31 13:28 16:33 825 Day CSM1:PM 22 18:50 20:10 Day O No Hailing 22 19:09 19:18 16:01 76.5 19:27 Day Good Yes Yes O No Hailing 22 21:13 21:15 17:55 0.7 21:17 Night Poor 16-Jan 16 M Not available 23 04:36 05:25 01:55 06:15 13112 Night 07:10 19:30 O No Hailing 23 06:12 06:18 02:43 9 06:25 Night Poor F CSM1:AM 23 07:10 08:30 Night O No Hailing 23 08:07 08:15 04:37 32.7 08:24 Night Good Yes O No Hailing 23 18:54 19:01 15:06 16 19:08 Day Poor Yes CSM1:PM 23 19:30 20:50 Day O No Hailing 23 20:51 20:59 17:01 20.1 21:06 Day Marginal (Close) Yes 17-Jan 17 CSM1:AM 24 07:50 09:10 Night 07:50 20:09 O No Hailing 24 07:50 07:59 03:43 51.4 08:07 Night Good Yes S O No Hailing 24 09:52 09:56 05:37 3.8 10:01 Night Poor M 15 24 18:10 18:30 16:31 18:50 Day N/A CSM1:PM 24 20:09 21:29 Day O No Hailing 24 20:33 20:42 16:06 87 20:51 Day Good Yes Yes 18-Jan 18 M Not available 25 05:31 06:29 01:38 07:27 13235 Night 08:29 20:49 O No Hailing 25 07:36 07:42 02:48 10.8 07:49 Night Poor S CSM1:AM 25 08:29 09:49 Night O No Hailing 25 09:31 09:39 04:42 28.7 09:48 Night Marginal Yes O No Hailing 25 20:17 20:25 15:11 18.4 20:33 Day Poor Yes CSM1:PM 25 20:49 22:09 Day O No Hailing 22:15 22:23 17:06 17.6 22:30 Day Poor (Close) Yes 19-Jan 19 CSM1:AM 26 09:09 10:29 Night 09:09 21:28 O No Hailing 26 09:14 09:23 03:48 59.3 09:32 Night Good Yes M O No Hailing 26 11:17 11:20 05:42 2.5 11:24 Night Poor CSM1:PM 26 21:28 22:48 Day O No Hailing 26 21:57 22:06 16:11 82.2 22:15 Day Good Yes Yes 20-Jan 20 O No Hailing 27 08:59 09:06 02:53 12.6 09:13 Night Poor No Carrier 09:48 22:08 CSM1:AM 27 09:48 11:08 Night T O No Hailing 27 10:55 11:03 04:47 25.2 11:12 Night Marginal No Carrier Yes O No Hailing 27 21:41 21:49 15:16 21 21:57 Day Marginal No Carrier Yes CSM1:PM 27 22:08 23:28 Day O No Hailing 27 23:39 23:47 17:10 15.3 23:54 Day Poor No Carrier Yes 21-Jan 21 M Not available 28 07:25 07:59 01:11 08:34 13130 Night 10:28 22:47 CSM1:AM 28 10:28 11:48 Night W O No Hailing 28 10:38 10:47 03:53 68.3 10:56 Night Good No Carrier Yes 115 Beagle-2 Mission Report ANNEXES
1000-1800 LTST Bold= Pass Pass Pass Compatible Compatible Order of Default Elevation Compatible Cmds Sol Time Time Approx Time MeX Day/ Beagle 2 Measured with with Year Date Day Comms Elevation Pass UTC UTC with Comms No. Start Max Elev LTST End Range Night Response AGC CSM1 CSM1 Session Quality CSM2 Type UTC UTC UTC LTST 03:35 LTST 15:35 Day O No Hailing 28 12:42 12:44 05:47 1.2 12:47 Night Poor No Carrier M Listen 28 20:30 20:33 13:29 20:36 874 Day Yes O No Hailing 28 21:30 21:32 14:21 0.7 21:34 Day Poor No Carrier Yes CSM1:PM 28 22:47 00:07 Day O No Hailing 28 23:21 23:30 16:15 71.9 23:39 Day Good No Carrier Yes Yes 22-Jan 22 M Not available 29 07:56 09:00 01:31 10:04 12636 Night 11:07 23:27 O No Hailing 29 10:23 10:30 02:58 14.6 10:38 Night Poor No Carrier T CSM1:AM 29 11:07 12:27 Night O No Hailing 29 12:20 12:27 04:52 22.1 12:36 Night Marginal No Carrier Yes M Hailing 29 22:10 22:12 14:26 22:14 845 Day No Data No Carrier Yes O No Hailing 29 23:05 23:13 15:20 23.9 23:21 Day Marginal No Carrier Yes CSM1:PM 29 23:27 00:47 Day 23-Jan 23 O No Hailing 29 01:04 01:11 17:15 13.2 01:18 Day Poor No Carrier 11:47 - Yes M Not available 30 08:42 09:48 01:39 10:55 11737 Night F CSM1:AM 30 11:47 13:07 Night O No Hailing 30 12:02 12:11 03:58 78.3 12:20 Night Good No Carrier Yes O No Hailing 30 14:08 14:09 05:53 0 14:09 Night Poor No Carrier O No Hailing 30 22:53 22:56 14:26 2 23:00 Day Poor No Carrier Yes 24-Jan 24 CSM1:PM 30 00:06 01:26 Day 12:26 00:06 O No Hailing 30 00:45 00:54 16:20 62.5 01:03 Day Good No Carrier Yes Yes S M Not available 31 09:26 10:31 01:42 11:36 11726 Night O No Hailing 31 11:47 11:54 03:03 16.8 12:02 Night Poor No Carrier CSM1:AM 31 12:26 13:46 Night O No Hailing 31 13:44 13:52 04:58 19.4 13:59 Night Poor No Carrier Yes M Hailing 31 23:19 23:21 14:14 23:24 874 Day No Data No Carrier Yes 25-Jan 25 O No Hailing 31 00:29 00:38 15:26 27.2 00:46 Day Marginal No Carrier 13:06 00:46 Yes CSM1:PM 31 00:46 02:06 Day S O No Hailing 31 02:28 02:35 17:20 11.4 02:41 Day Poor No Carrier Yes M Not available 32 09:39 10:51 01:23 12:03 11285 Night CSM1:AM 32 13:06 14:26 Night O No Hailing 32 13:26 13:35 04:03 88.9 13:44 Night Good No Carrier Yes M Hailing 32 22:53 22:56 13:10 23:00 890 Day No Data No Carrier 26-Jan 26 O No Hailing 32 00:16 00:21 14:31 3.2 00:25 Day Poor No Carrier 13:45 01:25 Yes CSM1:PM 32 01:25 02:45 Day M O No Hailing 32 02:09 02:18 16:25 54.2 02:27 Day Good No Carrier Yes Yes M Not available 33 10:06 11:04 00:57 12:03 11511 Night O No Hailing 33 13:11 13:18 03:08 19.2 13:26 Night Poor No Carrier CSM1:AM 33 13:45 15:05 Night O No Hailing 33 15:08 15:16 05:03 16.9 15:23 Night Poor No Carrier (Close) 27-Jan 27 O No Hailing 33 01:53 02:02 15:31 31 02:10 Day Good No Carrier 14:25 Yes 02:05 Yes CSM1:PM 33 02:05 03:25 Day O No Hailing 33 03:52 03:59 17:25 9.6 04:05 Day Poor No Carrier Yes O No Hailing 34 13:01 13:02 02:14 0.3 13:04 Night Poor No Carrier CSM1:AM 34 14:25 15:45 Night O No Hailing 34 14:50 14:59 04:08 80.5 15:08 Night Good No Carrier Yes 28-Jan 28 O No Hailing 34 01:40 01:45 14:36 4.6 01:49 Day Poor No Carrier 15:04 02:44 Yes CSM1:PM 34 02:44 04:04 Day O No Hailing 34 03:33 03:42 16:21 47.2 03:51 Day Good No Carrier Yes Yes O No Hailing 35 14:35 14:42 03:13 21.8 14:50 Night Marginal No Carrier CSM1:AM 35 15:04 16:24 Night O No Hailing 35 16:32 16:40 05:08 14.7 16:47 Night Marginal No Carrier (Close) 29-Jan 29 O No Hailing 35 03:17 03:26 15:36 35.3 03:34 Day Good No Carrier 15:43 Yes 03:24 Yes CSM1:PM 35 03:24 04:44 Day O No Hailing 35 05:17 05:23 17:30 8 05:28 Day Poor No Carrier Yes Not M 36 12:56 13:56 01:50 14:57 10800 Night available O No Hailing 36 14:23 14:26 02:19 1.5 14:29 Night Poor No Carrier CSM1:AM 36 15:43 17:03 Night O No Hailing 36 16:14 16:23 04:13 70.5 16:32 Night Good No Carrier Yes Not 30-Jan 30 M 36 02:03 02:06 13:40 02:09 900 Day 16:23 04:03 Yes available O No Hailing 36 03:03 03:09 14:41 5.9 03:14 Day Poor No Carrier Yes CSM1:PM 36 04:03 05:23 Day O Hailing 36 04:58 05:06 16:35 41.2 05:15 Day Good No Data No Carrier Yes Yes Not M 37 12:34 13:42 00:57 14:50 11000 Night available O No Hailing 37 15:58 16:06 03:18 24.8 16:14 Night Marginal No Carrier CSM1:AM 37 16:23 17:43 Night O No Hailing 37 17:57 18:04 05:12 12.8 18:11 Night Poor No Carrier 31-Jan 31 O Hailing 37 04:41 04:50 15:41 40.3 04:58 Day Good No Data No Carrier 17:02 Yes 04:43 Yes
116 Beagle-2 Mission Report ANNEXES
1000-1800 LTST Bold= Pass Pass Pass Compatible Compatible Order of Default Elevation Compatible Cmds Sol Time Time Approx Time MeX Day/ Beagle 2 Measured with with Year Date Day Comms Elevation Pass UTC UTC with Comms No. Start Max Elev LTST End Range Night Response AGC CSM1 CSM1 Session Quality CSM2 Type UTC UTC UTC LTST 03:35 LTST 15:35 Day CSM1:PM 37 04:43 06:03 Day O No Hailing 37 06:41 06:46 17:34 6.5 06:52 Day Poor No Carrier Yes O No Hailing 38 15:46 15:50 02:24 2.7 15:54 Night Poor No Carrier CSM1:AM 38 17:02 18:22 Night O No Hailing 38 17:38 17:47 04:18 61.4 17:56 Night Good No Carrier Yes
Table D.2 - January Overflights
117 Beagle-2 Mission Report ANNEXES
D.10.3 February 2004
1000-1800 Bold= Order of Pass Pass Pass Compatible Compatible LTST Sol Elevation Ye Da Cmds Default Time Time Approx Time MeX Day/ Beagle 2 Measured with with Compatible Date No Elevation Pass UTC UTC ar y Comms Comms Start Max Elev LTST End Range Night Response AGC CSM1 CSM1 with . Quality Type Session UTC UTC UTC LTST 03:35 LTST 15:35 CSM2 Day 01-Feb 32 O No Hailing 38 04:27 04:33 14:46 7.4 04:38 Day Poor No Carrier 17:42 05:22 Yes CSM1:PM 38 05:22 06:42 Day O No Hailing 38 06:22 06:30 16:40 36.1 06:38 Day Good No Carrier Yes Yes O No Hailing 39 17:22 17:30 03:23 28.1 17:39 Night Marginal No Carrier (Close) CSM1:AM 39 17:42 19:02 Night O No Hailing 39 19:21 19:28 05:18 10.9 19:34 Night Poor No Carrier 02-Feb 33 CSM1:PM 39 06:01 07:21 Day 18:21 06:01 O No Hailing 39 06:05 06:14 15:46 46.1 06:22 Day Good No Carrier Yes Yes O No Hailing 39 08:05 08:10 17:39 5.1 08:15 Day Poor No Carrier Yes Not M 40 15:35 16:31 01:47 17:27 11000 Night available O No Hailing 40 17:09 17:14 02:29 3.9 17:18 Night Poor No Carrier Yes CSM1:AM No Hailing 40 18:21 19:41 Night No Carrier O No Hailing 40 19:02 19:11 04:23 53.5 19:20 Night Good No Carrier Yes 03-Feb 34 M Hailing 40 04:35 04:38 13:34 04:42 1000 Day No Data No Carrier 19:01 06:41 Yes O No Hailing 40 05:50 05:57 14:51 9 06:03 Day Poor No Carrier Yes CSM1:PM 40 06:41 08:01 Day O No Hailing 40 07:46 07:54 16:45 31.7 08:02 Day Good No Carrier Yes Yes Not M 41 15:07 16:17 00:55 17:27 10800 Night available O No Hailing 41 18:46 18:54 03:28 31.8 19:03 Night Good No Carrier Yes CSM1:AM 41 19:01 20:21 Night O No Hailing 41 20:45 20:52 05:23 9.2 20:58 Night Poor No Carrier 04-Feb 35 CSM1:PM 41 07:20 08:40 Day 19:40 07:20 O No Hailing 41 07:29 07:37 15:50 52.8 07:46 Day Good No Carrier Yes Yes O No Hailing 41 09:30 09:34 17:44 3.8 09:39 Day Poor No Carrier Yes O No Hailing 42 18:33 18:38 02:34 5.2 18:43 Night Poor No Carrier CSM1:AM 42 19:40 21:00 Night O No Hailing 42 20:26 20:35 04:28 46.7 20:44 Night Good No Carrier Yes 05-Feb 36 O No Hailing 42 07:14 07:21 14:56 10.6 07:27 Day Poor No Carrier 20:19 08:00 Yes CSM1:PM 42 08:00 09:20 Day O No Hailing 42 09:10 09:18 16:50 27.9 09:26 Day Marginal No Carrier Yes Yes O No Hailing 43 20:10 20:18 03:33 36.1 20:27 Night Good No Carrier Yes CSM1:AM 43 20:19 21:39 Night O No Hailing 43 22:10 22:16 05:28 7.7 22:22 Night Poor No Carrier
Table D.3 - February Overflights
118 Beagle-2 Mission Report Annexes
E Landed Phase Operational Power Modelling
M. Hannington, NSC LOCC, University of Leicester
E.1 Introduction This document describes the Beagle 2 power subsystem modelling carried out as part of operations planning activities of the Beagle 2 Flight Operations Team in late 2003 and early 2004. The Beagle 2 Power Model developed for the Beagle 2 team by SEA Ltd. was utilised and modelling focussed entirely on landed phase operations.
Modelling work was undertaken in the first instance over the period 08/09/2003 to 16/09/2003 as part of the development and validation of pre-programmed operations and software parameters for the landed phase which were uploaded during cruise phase checkouts. This work is referred to as Phase 1 Modelling in the report and was carried out using the penultimate version of the model “FullModelAug8_2003”. An additional period of modelling was carried out between 07/12/2003 and 12/12/2003 to verify the pre-programmed operations against the final version of the model “FullModelNov6_2003” and a fuller and more representative set of landing conditions. This is referred to as Phase 2 Modelling.
The period of power modelling carried out during the design phase of Beagle 2 is not covered in this report.
Reference Documents
The documents listed below are referred to in this annex.
Reference Title Document ID / Version Author / Date Geoffrey Matt A Beagle 2 Power Modelling Manual 99 / TM / 2514 / Issue 3 03 / 08 / 2000 http://www.giss.nasa.gov/tools/m B NASA Mars Time Calculator ars24 Preliminary Power Requirements Darren Chaney C for Beagle 2 19 / 10 / 2000 K Arnold D Power Model Runs BGL2-MMS-TN-00029/0AD1 2003 SSSL-BGL2-LOM-0001 E Lander Operations Manual 11 / 08 / 03 Draft B Amendment to Beagle 2 Power Model Geoffrey Matt F SEA/03/TN/4241 / Issue 2 User Manual – Second Delivery Sept 2000 Modelling of Battery Effective G LION-AEA-BD-RP-0012 / Issue 1 26 / 06 / 01 Resistance
119 Beagle-2 Mission Report Annexes
E.2 Power Subsystem (PSS) Definition The Beagle 2 power subsystem consists of 4 orientation adjustable solar arrays providing power to lander and payload units and to charge a Lithium Ion battery. The arrays are connected to the bus and battery via a Battery Charge Regulator (BCR). Three bus sets are supplied; the unregulated Main Bus Supply and two groups of regulated supplies derived from the Main Bus by the Auxiliary Power Supply (APS) and Payload Power Supply (PPS) converters.
E.2.1 Solar Array The solar cells are Hughes Spectrolab triple junction Gallium Arsenide units of dimensions 6.92 cm by 3.13 cm. Each array comprises strings of 13 cells, with 3 strings per section and 2 sections per panel. The four array panels are connected to the lander lid by individual motorised hinge mechanisms which allow the arrays to be stowed for the launch, cruise and coast phases and deployed once landing is complete. The hinge motor can be commanded to adjust the orientation of the array from stowed in the lid: zero degrees, through deployed flat in the plane of the lid: 180° to around 200° maximum deployment. The 4 array hinges and the main lid-base hinge are equi-spaced around the lid with an angle of 72° between each.
E.2.2 Battery The Beagle 2 battery is a 54 cell Lithium Ion battery built by AEA Technology Ltd. Cell voltage at full charge is 4.2V and arrangement of 6 cells per string provides a battery voltage at full charge of 25.2V. Individual cell capacity is 1.5 Ahrs and 9 parallel strings provide a battery nameplate capacity of 13.5 Ahrs.
E.2.3 Battery Charge Control No dedicated main bus voltage regulation is implemented and the bus voltage follows the battery voltage. During daylight operation the BCR allows all available solar array current above load requirements to charge the battery. While the battery state of charge is low the battery is charged at a rate limited by the array supply available and the load demand. As the battery approaches full charge the battery terminal voltage reaches a reference value (preset by telecommand) initiating taper charging.
The BCR circuit connects the solar arrays to the battery and load via an FET switch. During taper charging the FET is switched by a pulse width modulation method, cycling charge current connectivity and maintaining the terminal voltage at the reference value. Full charge is thus approached in a controlled manner to avoid overcharging.
E.2.4 Power Loads The power requirements of lander and payload subsystem units are tabulated in Appendix 1.
120 Beagle-2 Mission Report Annexes
E.3 Beagle 2 Power Model Description The core function of the Beagle 2 Power model can be described in 5 modules. More detailed description of the power model function can be found at Reference A
E.3.1 Module 1: Power Subsystem Hardware The Beagle 2 power sub-system hardware as detailed in Section 3 is represented in the model.
The solar array is described by the number, configuration and dimensions of solar cells, the cell IV curve and the variation of cell performance with temperature.
The battery is represented by the number of series and parallel cells, the capacity charge per cell, the relationship between battery state of charge and internal voltage during charging and discharging and the variation of battery internal resistance with internal voltage and temperature.
The functional state of the subsystem at each time division within the model run is determined. The system stable point voltage is established and the current produced by the array in response to varying illumination level is calculated. The current flow into and out of the battery corresponds to changes to battery state of charge. The voltage regulation during the taper charge regime is also modelled.
The user has the facility to modify any of the system parameters implemented. During the Beagle 2 design phase this allowed trade off on the performance of different designs. For the operations planning phase a set of flight representative parameters was determined and utilised. Changes to power subsystem parameters would be made only in response to predicted / observed degradation over life.
E.3.2 Module 2: Lander Orientation. Critical to the determination of the solar power available to the system is the orientation of the solar panels with respect to the sun. The physical relationship of the solar arrays to the lander and the lander to the surface of Mars are represented by two sets of operator controlled variables.
E.3.2.1 Orientation Angle
This represents the angle between the direction to Mars North in the model and a lander datum line taken to be a line drawn between the lander lid and centre of solar array panel 1. Rotation of Orientation Angle determines the ‘facing’ of the lander as a unit with respect to the position of the sun in the sky at the landing site at any given time. Orientation angle is depicted in Figure E.1.
E.3.2.2 Offset Angle
For each solar panel an Offset Angle is defined. This angle is the angle between the plane of the lander (lid and base are taken to be in the same plane) and the plane of the array. Offset Angle for each array may be adjusted separately correlating to the angle that the array hinge is driven to in the spacecraft. In the model full deployment is represented as zero degrees and angles above the lander / base plane as negative. The convention in Beagle 2 Lander Software (LSW) operation however defines hinge angles of zero degrees as arrays fully stowed and 180 degrees as fully deployed.
121 Beagle-2 Mission Report Annexes
Figure E.1 - Orientation Angle Geometry
Figure E.2 - Panel Offset Angle Geometry
E.3.3 Module 3: Solar Flux Incident It is required to define the relative geometry of Mars to the Sun and the position of the Sun in the sky above the lander at any time. To achieve this a simple model of Mars orbit and rotation is implemented. At a given UTC time the position of Mars in its orbit, the distance to the sun and the seasonal tilt of the planet’s axis with respect to the Sun is calculated. With the location of the landing site in latitude and longitude the geometry between the sun and the lander can be determined. Combining with the Orientation and Offset Angles of the lander and panels, the angle between each panel and the direction of Sun is evaluated. As the Sun rises and sets the angle of incidence of solar flux varies for each panel with consequent variation in the total array supply current available.
Two other environmental factors influence the flux incident. The Atmospheric Optical Depth and Dust Transmission Factor will both be dependent on the dust content of the atmosphere at the landing site. The flux incident at the top of the atmosphere varies with the calculated distance between Mars and the Sun. Setting the Optical Depth fixes the extent to which the solar flux at the top of the atmospheric corridor above the landing site is attenuated before it is incident on the solar arrays. Changing the Dust Transmission Factor allows the user to introduce the effect of dust build up on the solar arrays over a number of days.
122 Beagle-2 Mission Report Annexes
E.3.4 Module 4: Thermal Input Both the performance of the solar array cells and the variation of battery internal resistance with state of charge are temperature dependent. The effect of temperature on these values is determined internally to the model. A temperature profile for the solar arrays and battery over the period modelled is required as a user input.
E.3.5 Module 5: Subsystem Load The model requires information on the expected instantaneous load being seen at any time by the system. The model recognises loads defined as constant power or loads defined as resistive. The user input is a UTC timeline of load levels representative of unit switching on board the spacecraft in response to automatic and commanded operation.
E.3.6 Model Outputs The model provides output showing the variation of operational parameters calculated during the model run, such as power available from the soar arrays and battery terminal voltage, by which the performance of the system can be determined. One main operational driver determines the feasibility of an implemented power profile; a level of battery state of charge must be maintained such that night time loads remain supportable.
123 Beagle-2 Mission Report Annexes
E.4 Operational Phase Power Modelling
E.4.1 Beagle 2 Landed Phase Operations The landed phase in this sense refers to all operations following handover of control of the lander from Probe Software (PSW) to Lander Software (LSW) and completion of the deployment of the solar arrays. The landed phase inherits its start boundary condition from the power demands of the EDLS and deployment phase and the predictions of the thermal environment during cruise and descent. The profile begins with a battery temperature higher than the normal landed daily cycle and a predicted battery state of charge of 70% - depleted from near 100% at top of atmosphere (TOA) by the current required for pyro and frangibolt firing and driving the lid and solar array hinges. The start epoch for all the landed phase power modelling was taken at 2003.359.03.24.00z allowing 30 minutes from the time of predicted first bounce at 2003.359.02.54.00z and corresponding to a mid afternoon Local True Solar Time (LTST) of 14:02.
The design phase power modelling is described in References C and D. Necessity of using estimates for unit power demands and a fabricated operations plan to construct a load profile limited the applicability of the design phase modelling with respect to the details of the operations phase. However the results indicated the following factors impacting on survivability in early landed phase.
• Initial battery state of charge on landing must be above 50% • Landing in unfavourable weather, a dust storm, dramatically increases the probability of the battery state of charge falling below acceptable limits. • The subsystem performance is sensitive to tilt of the lander on the surface. Angles between the lander base and the surface of Mars above 20 degrees engender orientation configurations which are not survivable past the first night.
Resource limitation required that the power subsystem was not designed with margin to survive in a ‘worst case’ landing. Worst case might be considered to be landing with a heavily discharged battery, with the lander tilted in a northerly direction in a duststorm.
E.4.2 Power Model Operational Load Profile The load profile is interfaced to power model as a timeline of power levels with start time and duration. The power subsystem loads modelled can be divided into three categories; continuous loads, varying autonomous loads and Mission Event Timeline (MET) loads.
The continuous loads on Beagle 2 are the Auxiliary Power Supply (APS), the Common Electronics Module (CEM) board and Common Electronics Processor (CEP) board. The APS, CEM and CEP are powered for the duration of nominal operations.
Loads may be switched as a background task by the Lander Software. The only autonomously switched load of consideration in the timeframe modelled is the duty cycling of the battery heaters. In this case thermal modelling carried out by Rutherford Appleton Laboratories predicted that in a cold landing site the duty cycle to maintain the battery at -27.5 degrees throughout the night would require 26 Whrs of energy. It wasn’t practicable to exactly map the varying profile of the duty cycle in the power model and a continuous load equivalent was adopted.
Beagle 2 programmed operations are implemented through the addition of timed entries on to the Mission Events Timeline in LSW. Power loads controlled via the MET include the operations of the transceiver, frangibolt actuation, and the operation of the ARM and all payload instruments. The MET constitutes the operator primary interface with Beagle 2. All operations controlled by MET entries will have a tagged start time and duration. These times are used to locate the load within the load profile in the power model. An associated power requirement for the MET entry provides the magnitude of the load.
124 Beagle-2 Mission Report Annexes
E.4.3 Sol 1 to 3 Operations Power Modelling - Phase 1 The scope of landed phase power modelling covered pre-programmed operations for Sols 1 to 3 and subsequent autonomous communications search modes.
During cruise phase a pre-programmed ‘default MET’ was loaded into memory on Beagle 2 for execution following landing. The default MET comprised a timeline of all operations for execution on the first three Sols on the surface. The first segment of operations phase power modelling focussed on the verification of this Sol 1 to 3 timeline before it was uploaded.
E.4.3.1 Initial Sol 1 to 3 Model
Tables E.1 to E.3 show the parameters for the initial power runs of Sols 1 to 3
Power Model Parameters: General
Parameter Value
Start of Run 2003.359.03.24.00z
Initial state of charge 70%
Optical Depth 1
Dust Accumulation Losses 1
Table E.1 - Power Model Run General Parameters Phase 1 Modelling
Power Model Parameters: Thermal
Parameter Value Note
Battery Temperature Profile 245 to 253 K Cold Case Landing Site Solar Array Temperature Profile 191 to 266 K Cold Case Landing Site
Table E.2 - Power Model Run Thermal Parameters Phase 1 Modelling
Power Model Parameters: Orientation
Parameter Value
Orientation Angle Not Relevant
Offset Angle: Solar Array 1 0
Offset Angle: Solar Array 2 0
Offset Angle: Solar Array 3 0
Offset Angle: Solar Array 4 0
Table E.3 - Array Orientation Parameters Phase 1 Modelling
125 Beagle-2 Mission Report Annexes
The start of the run is timed from predictions of landing time and the duration of the deployment.
Initial state of charge follows prediction from Astrium design taking into account very small battery currents during cruise phase when only the timer is powered, and significant battery loads seen on pyro firings during EDLS and frangibolt and motor driving during deployment.
Optical depth data as reported at the Viking 1 and 2 landing sites over 900 Sols was available at the LOCC. Inspection showed that during extended periods of dust storms optical depth values ranged from 1 to 3.93. During quiescent times optical depths were seen to average around 0.6 with occasional excursions above 1. Optical depth of 1 was chosen for the modelling as being the top end of ‘non dust storm’ weather conditions.
The modelling carried out in Phase 1 was to exercise the effect of position and duration of Mission Event Timeline entries on the load profile and relative impact on the power subsystem. To limit the number of permutations the solar array offset angles were set to 0 (flat arrays). This removed the variability of the model runs with spacecraft facing – the model does not change with varying Orientation Angle. Phase 2 analysis reintroduced non zero array offset angles.
Table E.4 shows the load profile input to the model for the initial Sol 1 to 3 MET period. The continuous load of the CEP / CEM / APS is overlain with the loads introduced by the MET. Night time battery duty cycling is represented by continual load present from sunset to sunrise which draws an equivalent energy to the duty cycle.
The Environmental Sensor Suit (ESS) represents the longest duration MET load, frangibolts firing represent peak short term loads. The segment of the comms session where the transmitter is powered represents the most significant daily power load to be managed.
126 Beagle-2 Mission Report Annexes
Default MET Events and Power Model Load Equivalents
D / N DOY Time UTC EVENT LOAD MODE 359 02:54 LANDING COMPLETE 03:24 DEPLOY COMPLETE CEP APS CEM 1 03:53 WAM SOL 1.1 CEP APS CEM SCS 2 04:00 CEP APS CEM 1 04:39 ESS ON CEP APS CEM ESS 3 05:10 WAM SOL 1.2 CEP APS CEM ESS SCS 4 05:15 CEP APS CEM ESS 3 05:25 B2 RECEIVER ON CEP APS CEM ESS COMMS RX 5 05:36 AOS ODY CEP APS CEM ESS COMMS TX 6 05:46 LOS ODY CEP APS CEM ESS COMMS RX 5 05:56 B2 RECEIVER OFF CEP APS CEM ESS 3 07:06 CEP APS CEM 1 07:08 SSET BAT HEATER CEP APS CEM BAT 7 07:26 WAM NIGHT 1 CEP APS CEM BAT SCS 8 07:31 CEP APS CEM BAT 7 19:55 WAM SOL 2.1 CEP APS CEM BAT SCS 8 20:00 SUNRISE ESS ON CEP APS CEM ESS 3 20:05 ESS OFF CEP APS CEM 1 22:53 JB TX ON CEP APS CEM ESS COMMS TX only 9 23:03 JB TX OFF CEP APS CEM 1 23:54 WAM SOL 2.2 CEP APS CEM SCS 2 23:59 CEP APS CEM 1 360 00:56 ARM FRANG PLUTO PIN CEP APS CEM FRANGI 10 00:59 PLUTO LOCK PIN CEP APS CEM PUT PIN 14 01:04 CEP APS CEM 1 01:58 GAP READING CEP APS CEM GAP 11 02:07 ESS ON CEP APS CEM ESS 3 04:32 WAM SOL 2.3 CEP APS CEM ESS SCS 4 04:37 CEP APS CEM ESS 3 06:04 B2 RECEIVER ON CEP APS CEM ESS COMMS RX 5 06:14 AOS ODY CEP APS CEM ESS COMMS TX 6 06:25 LOS ODY CEP APS CEM ESS COMMS RX 5 06:35 B2 RECEIVER OFF CEP APS CEM ESS 3 07:48 CEP APS CEM ESS BAT 12 07:53 WAM SOL 3.2 CEP APS CEM ESS BAT SCS 13 07:58 BATTERY HEATER CEP APS CEM ESS BAT 12 09:51 ESS OFF CEP APS CEM BAT 7 18:37 ESS ON CEP APS CEM ESS BAT 12 20:41 SUN RISE CEP APS CEM ESS 3 22:44 ESS OFF CEP APS CEM 1 23:32 JB TX ON CEP APS CEM ESS COMMS TX only 9 23:42 JB TX OFF CEP APS CEM 1 361 06:34 ESS ON CEP APS CEM ESS 3 06:44 B2 RECEIVER ON CEP APS CEM ESS COMMS RX 5 06:54 AOS ODY CEP APS CEM ESS COMMS TX 6 07:04 LOS ODY CEP APS CEM ESS COMMS RX 5 07:14 B2 RECEIVER OFF CEP APS CEM ESS 3 08:37 SUN SET BAT HEATER CEP APS CEM ESS BAT 12
Table E.4 - Load Profile for Initial Sol 1 to 3 Power Model Run
127 Beagle-2 Mission Report Annexes
Currents Flowing in System
4
3
2
1 I Solar [A] I Load [A] I Battery [A]
Current [A] 0 359.14375 359.64375 360.14375 360.64375 361.14375 361.64375 362.14375
-1
-2
-3 Time [UTC days]
Figure E.3 - Currents Flowing in System Sol 1 to 3 Power Model Run
Figures E.3 and E.4 show example power model outputs for the initial Sol 1 to 3 run (including receiver on times set to 80 minutes, frangibolt firings relocated and the Sol 3 AM Odyssey comms session added; see section E.4.3.2 for discussion). Figure E.3 shows the solar array supply current, the load current and the battery charge and discharge currents. Figure E.4 shows the battery state of charge profile.
For a power load profile to be accepted the battery state of charge cycle must not fall below a critical value at its minimum, the trend in maximum state of charge should be upwards towards 100 %, trend in minimum state of charge should be upwards and battery current and bus current must stay within onboard software limits. An acceptable limit on battery minimum state of charge was selected at around 30% to ensure battery voltages retain margin above load shedding limit of 18 V and well above critical level of 15 V at which the APS will power off.
128 Beagle-2 Mission Report Annexes
Battery Charge in Percentage Format
120
100
80
60 Charge [%] Charge
40
20
0 359.14375 359.64375 360.14375 360.64375 361.14375 361.64375 362.14375 Time [UTC days]
Figure E.4 - Battery State of Charge Sol 1 to 3 Power Model Run. Rx On time set to 80 Minutes and ODY Sol 3 Am Added
E.4.3.2 Iterations to the Sol 1 to 3 Model
The initial Sol 1 to 3 run showed a power profile with very healthy margins. Battery state of charge reached a minimum of 39% on the first night with slightly higher values for subsequent Sols. Maximum state of charge remained close to 100% and the overall trend in maximum and minimum state of charge was upwards. Taper charging was reached by midday each day. As the landing local solar time is in early afternoon, the sunlight time available for charging the battery is less on Sol 1 than on subsequent Sols. Battery state of charge is consequently lowest on the night of the first Sol on the surface.
Three significant changes were made to the default timeline during development and preparation for upload. The receiver On times were widened, the location of frangibolt firing was adjusted and an additional Mars Odyssey communications session was inserted on the morning of Sol 3.
E.4.3.2.1 Iteration 1. Widened Receiver On Times and Adjusted Frangibolt Firings Communication sessions inserted on the MET were originally constructed to have a receiver on time duration of 20 minutes corresponding to the maximum time horizon to horizon of an Odyssey overflight. Within this 20 ‘listen’ duration a transmitter on time of 10 minutes was estimated. The initial Sol 1 to 3 power run used these timings.
The receiver on time was revised to 80 minutes to increase the probability of successfully establishing comms should the relative timing of MET entry to orbiter overflight be unsynchronised as the result of an offset from predictions in on board time or landing location.
It was noted from the current flowing profile that frangibolt firing, a peak power load, had not been ideally located at the time of maximum available power and battery state of charge. Receiver load duration was increased to 80 minutes, the timings of frangibolt loads adjusted and the model re-run. Battery minimum state of charge reduced to 36 percent on night of Sol 1, but recovered to 46 percent on night Sol 2. Maximum state of charge approached 100% and taper charge is reached before midday.
129 Beagle-2 Mission Report Annexes
E.4.3.2.2 Iteration 2. Odyssey Session Sol 3 AM The Beagle 2 baseline communications concept was to limit the choice of communications sessions to those that occurred during daylight hours. This restriction was made with respect to perceived power limitations but led to the selection of poor communications opportunities in daylight in place of some high elevation overflights occurring at night.
Due to the poor quality of the pass at 16:00 LTST on Sol 2, the comms opportunity at the 04:00 LTST pass on the morning of Sol 3 was considered as a replacement. The Sol 3 morning comms session was added to the load profile and the model rerun to determine feasibility. Battery state of charge recovered to 43% on night Sol2, down only 3% from the profile without the night time session.
The Sol 1 to 3 default MET was frozen following this trade-off between power margins and increasing the capability to secure first comms. The profiles in figures E.3 and E.4 represent this frozen configuration.
E.4.3.3 Communications Search Modes
Three Communications Search Modes are defined in Beagle 2 Lander Software; Communications Search Mode 1 (CSM 1), Communications Search Mode 2 (CSM 2) and Auto-transmit Mode. Each mode’s framework strategy is fixed but software parameters modifying the strategy are operator selectable. A full description of each comms search mode is available at Reference E.
E.4.3.3.1 Comms Search Modes Function and Software Parameters Comms Search Mode 1 inserts additional communications sessions at the predicted times of Odyssey overflights (around 04:00 and 16:00 LTST). A software parameter determines the duration of the inserted comms session listen (receiver on) period.
Comms Search Mode 2 insets communications sessions continually throughout the day and night. The frequency with which sessions are inserted, and the duration of the sessions are set by software parameters. The day and night cycles can be set separately.
For any communication session inserted onto the MET a Carrier Only mode is available. In Carrier Only mode the transmitter is switched on and off at regular intervals and an unmodulated carrier transmitted until Beagle 2 is hailed or the communication session times out. Carrier only mode aids location of Beagle from Mars orbit or earth based antennas in the scenario where initial contact is not made. Software parameters set the transmitter on/off cycle for carrier only mode. A separate cycle can be chosen for planned MET comms sessions, and CSM 1 / CSM 2 sessions. In all cycles the carrier off segment precedes the carrier on.
The choice of software parameters controlling CSM 1 and 2 and the Carrier Only segment has significant impact on the power profile. Survivability and effectiveness in establishing comms required trade off.
In order to determine viability of the search modes and exercise the necessary selections of default values for search mode software parameters the power model was extended to cover periods of Comms Search Mode 1 and Comms Search Mode 2. From a power viewpoint Comms Search Mode 2 and Auto-transmit Mode are identical.
E.4.3.3.2 Power Modelling CSM 1 Sols 4 and 5 were added to the timeline with a CSM 1 inserted Odyssey AM and PM communications session each Sol. The carrier on / off duty cycle for CSM 1 and 2 had been originally selected as 6 minutes transmitter off 6 minutes transmitter on. An average power level representative of the 6 minute carrier only duty cycle was chosen for the model load.
Battery state of charge fell to 40% on night Sol 3 and further to 33% on night of Sol 4. A strategy was identified that could ensure that the carrier only mode segment of CSM 1 comms sessions would only occur in the PM comms slot, and be prevented in the AM session (or whenever Battery State of Charge fell below a certain level). This significantly reduced the load in the AM when array power is not available. The model run with this strategy showed minimum battery state of charge recovered to 45% and 40% on night Sol 3 and 4 respectively. 130 Beagle-2 Mission Report Annexes
E.4.3.3.3 Power Modelling CSM 2 Sols 6 and 7 were added to the timeline to model CSM 2. An average power level load representative of the CSM 2 night and CSM 2 day cycles was calculated.
Power model runs of the 7 Sols showed battery state of charge reducing to 39% and 34% on night Sol 5 and 6 respectively. In light of these reduced battery state of charge levels the comms search mode strategies were revised as follows.
• A strategy for CSM 2 Night preventing the carrier only segment of the communications session occurring was implemented by selecting the transceiver on cycle time to be shorter than the carrier off cycle time.
• The period of CSM 2 Night was selected to last from sunset to 10 am LTST to delay CSM 2 day cycle until increased solar panel power margin was available.
• CSM 2 Day strategy cycles the transceiver on for 59 minutes and off for one, with carrier only duty cycle originally set to 6 minutes off / 6 minutes on – effectively an extended CSM 1 session lasting throughout the day. The carrier only duty cycle for CSM 1 and CSM 2 was revised from 6 minutes off / 6 minutes on to 9 minutes off / 1 minute on. This significantly reduces the power demand while retaining the probability that the transmitter will be on at some time during a 20 minute Odyssey pass.
Modelling with these revisions showed the minimum battery state of charge levels recovered to 46 %, 44%, 42% and 41% on the nights of Sols 3, 4, 5 and 6 respectively.
The model timeline was increased to add two more Sols of CSM 2, minimum battery state of charge stabilising at just over 40%.
Figures E.6 and E.7 show the power model output for the finalised timeline and Comms Search Strategies.
Currents Flowing in System
4
3
2
1 I Solar [A] I Load [A] I Battery [A]
Current [A] 0 359.1458333 360.1458333 361.1458333 362.1458333 363.1458333 364.1458333 365.1458333 366.1458333 367.1458333
-1
-2
-3 Time [UTC days]
Figure E.5 - Currents Flowing in System Sol 1 to 3, CSM 1 and CSM 2 (revised) Power Model Run
131 Beagle-2 Mission Report Annexes
Battery Charge in Percentage Format
120
100
80
60 Charge [%] Charge
40
20
0 359.1458333 360.1458333 361.1458333 362.1458333 363.1458333 364.1458333 365.1458333 366.1458333 367.1458333 Time [UTC days]
Figure E.6 - Battery State of Charge Sol 1 to 3, CSM 1 and CSM 2 (revised) Power Model Run
The most likely communications search timeline follows a pattern of 5 Sols of programmed communications sessions, followed by 5 Sols of CSM 1 before CSM 2 is entered. The model included only 3 Sols of programmed sessions followed by 2 Sols of CSM 1 and 4 Sols of CSM 2. While this does not match exactly the CSM timeline the model demonstrates that the Communications Search Mode strategy is supportable over an extended period.
132 Beagle-2 Mission Report Annexes
E.4.4 Operations Power Modelling - Phase 2
E.4.4.1 Final Version Power Model
The final version of the power model was delivered by SEA to the Beagle 2 Flight Operations Team on 06/11/03. A full description can be found at Reference F. The majority of the updates were to facilitate the models use in an operational mode, allowing quicker iteration of timelines and importing and exporting files to and from the Operations Planning System and Thermal Model.
The time format used was changed to standard UTC. The algorithm controlling sunrise/set times was checked against the algorithm used by the NASA Mars Time applet (Reference B) and correlation over the Sol 1 to 9 time period was better than 3 minutes.
A number of updates to system specifications which had arisen since the previous release, such as the solar cell IV curve, were made permanent. The model used in Phase 1 had been updated on an ongoing basis by the Flight Ops team with most of the changes included in the final release. Two remaining deltas from the Phase 1 version were expected to impact results. The battery internal resistance algorithm was revised to be accurate to -30 DgC and to include the effect of chemical diffusion. The system specification driving these changes is described at Reference G.
The full Sol 1 to 9 timeline was rerun with the new version of the model and compared with the results from Phase 1. Minimum battery state of charge was lower than Phase 1 by an average of around 2.5% with minimums on nights Sols 1 to 8 of 34%, 40%, 43%, 42%, 39%, 38%, 37% and 37%.
E.4.4.2 Solar Array Offset Angles at 20 Degrees
The Lid and Solar Panel Deployment Activity Sequence run by LSW deploys the solar panels to angles of 160 degrees. With the condition that the lander has come to rest flat on Mars this gives solar array angles of 20 degrees above the surface plane. Phase 1 modelling had set the solar panel angles to zero (fully deployed) in order to reduce required permutations of the model. It was therefore necessary as part of Phase 2 modelling to validate the final MET and CSM strategies against the real landing configuration with solar array angles as they would be post deployment. Tables E.5 and E.6 show the configurations exercised.
Power Model Parameters: Array Offsets
Parameter Value
Offset Angle: Solar Array 1 20
Offset Angle: Solar Array 2 20
Offset Angle: Solar Array 3 20
Offset Angle: Solar Array 4 20
Table E.5 - Array Offset Angles Phase 2 Modelling
133 Beagle-2 Mission Report Annexes
Power Model Parameters: Orientation Angle Model Run Parameter Value Facing 1 Orientation 0 South Angle 2 Orientation 90 West Angle 3 Orientation 180 North Angle 4 Orientation 270 East Angle
Table E.6 - Orientation Angle Model Runs Phase 2 Modelling
With the solar panels offset the profile becomes highly sensitive to the orientation angle of the lander. In an orientation in which the panel offset tilts them towards the south, and therefore the angle of incidence of solar flux (wrt panel normal) is reduced, the profile can be expected to be more favourable than the flat panel case. The converse holds with orientations where the panels are tilted towards the north. The orientation angles 0, 90, 180 and 270 are roughly correlated with south, west, north and east.
South Facing Case
Orientation angle of 0 degrees is closest to south facing and an expected significant improvement in power profile was observed when the model was run, with battery state of charge in a band 45 to 50%.
West Facing Case
With orientation of 90 degrees the profile resembles the flat panel case, with an improved battery state of charge on the first night with the westward facing benefiting from an afternoon landing.
East Facing Case
With an orientation of 270 degrees, east facing, the profile on Sol 1 is markedly degraded. Landing in mid afternoon the sun is already in the west and the effective available battery charge time is reduced. Battery state of charge falls to 20% on the night of Sol 1 in this configuration, with battery terminal voltages just short of falling to 18 V. Recovery on subsequent Sols is complete with minimum battery state of charge varying within a few degrees below 40%.
North Facing Case
Orientation of 180, with the spacecraft north facing and all panels tilted away from the sun is understandably the least favourable. The battery state of charge profile is severely degraded. State of charge falls to 20% on night Sol 1, but does not recover on subsequent Sols, trending down to a 7% minimum on night Sol 2 (battery terminal voltage around 15 V) and is flat by night Sol 3. This configuration is clearly not survivable within the system parameters modelled.
134 Beagle-2 Mission Report Annexes
Currents Flowing in System
3
2
1
I Solar [A] 0 I Load [A] 00:00 25/12/03 00:00 26/12/03 00:00 27/12/03 00:00 28/12/03 00:00 29/12/03 00:00 30/12/03 00:00 31/12/03 00:00 01/01/04 I Battery [A] Current [A]
-1
-2
-3 Date and Time [UTC]
Figure E.7 - Currents Flowing in System. Panel Offsets 20/North Facing Lander Power Model Run
Battery Charge in Percentage Format SA = - 20 0A = 180
80
70
60
50
40 Charge [%] Charge
30
20
10
0 00:00 25/12/03 00:00 26/12/03 00:00 27/12/03 00:00 28/12/03 00:00 29/12/03 00:00 30/12/03 00:00 31/12/03 00:00 01/01/04 Date and Time [UTC]
Figure E.8 - Battery State of Charge. Panel Offsets 20/North Facing Lander Power Model Run
135 Beagle-2 Mission Report Annexes
E.4.4.3 Recovering the North Facing Case
A rethink of the proposed power load profile was undertaken to investigate the possibility that the default MET was too ambitious. A run of the model with receiver on times reduced from 80 to 40 minutes and the Odyssey communications session for Sol 3 AM removed produced a very slightly improved profile with the battery flat on night Sol 4. Further reduction of the strategy to establish communications was considered risky.
It was concluded that the power profile of the north facing case may remain marginal for the conditions modelled. This was not an unexpected result given the findings of the design phase modelling. The conditions may be thought of as approaching the extreme end of the survivable landing scenario, that is an unfavourable orientation on a cold landing site with an optical depth at the top end of the non-dust storm outlook.
It had already been demonstrated that at optical depths of 2, the low end of the spread of dust storm optical depths, only the most favourable orientation is survivable. A repeat of the north facing run was carried out with an optical depth of 0.7, a level representative of the average non-dust storm optical depth from the Viking data. Figure E.10 shows the much more healthy battery state of charge profile for these conditions.
Battery Charge in Percentage Format SA = - 20 0A = 180 Optical Depth = 0.7
100
90
80
70
60
50 Charge [%] Charge 40
30
20
10
0 00:00 25/12/03 00:00 26/12/03 00:00 27/12/03 00:00 28/12/03 00:00 29/12/03 00:00 30/12/03 00:00 31/12/03 00:00 01/01/04 00:00 02/01/04 00:00 03/01/04 Date and Time [UTC]
Figure E.9 - Battery State of Charge, Panel Offsets 20/North Facing Lander. Optical Depth 0.7
The LSP deployment sequence as programmed thus proves robust to optical depths up to 1 in favourable landing orientation and to average expected optical depths at unfavourable orientation. A further run modelled the sensitivity of all 4 landing orientations to one panel being set flat to 180 degrees while the other panels remained at 160 degrees. The run showed battery state of charges healthy for all four orientations at optical depth of 1. A modification to the panel deployment strategy would clearly provide increased margin on survivability. Figure E.11 shows the north facing case with panel 1 set flat and an optical depth of 1.
136 Beagle-2 Mission Report Annexes
Battery Charge in Percentage Format SA 1 = 0 / SA 2 to 4 = - 20 / 0A = 180
100
90
80
70
60
50 Charge [%] Charge 40
30
20
10
0 00:00 25/12/03 00:00 26/12/03 00:00 27/12/03 00:00 28/12/03 00:00 29/12/03 00:00 30/12/03 00:00 31/12/03 00:00 01/01/04 00:00 02/01/04 00:00 03/01/04 Date and Time [UTC]
Figure E.10 - Battery State of Charge, Panels 2 to 4 Offset 20. Panel 1 flat. North Facing Lander.
E.4.4.4 Operational Implications of the North Facing Power Model Results
With an unmodified deployment activity sequence the north facing case had been seen to be robust only to the most probable landing site optical depths. A review of the strategy of landing with panels at 160 degrees was undertaken during the few days before the last Beagle 2 checkout before ejection.
Reports of atmospheric dust levels at Isidis provided by Malin Space Science Systems began to be received at the LOCC in the 2 weeks before ejection. While there was some concern at the initial reports, with optical depths up to 1.2, the daily levels and the forecast for day of landing levelled out at around the favourable 0.7.
The Lid and Solar Panel Deployment activity sequence, with the final solar array angles of 160 degrees inherent, underwent two extended and successful Ground Test Model testing periods, once before launch and once as part of the end to end landing / deployment / default MET testing. The integrity of the correct upload of the deployment activity sequence was established a number of times during cruise checkouts. The development and testing of an updated activity sequence and preparation of that for upload on the penultimate or final cruise checkout opportunities was not considered to be without risk.
The operational decision of the LOCC Team with respect to the results of the north facing power modelling was as follows. The tested activity sequence for Lid and Solar Panel Deployment would remain unmodified. The first command load sent to Beagle 2 on the surface, to be received before sunset on Sol 1 should initial comms be successful, was modified to include commanding to immediately drive solar array hinge motor 1 from 160 degrees to 180 degrees.
137 Beagle-2 Mission Report Annexes
E.4.4.5 Late Developments in Phase 2 Power Modelling
There were 2 late developments in the expected landing power regime.
A review of understanding relating to battery heater operation on the surface revealed that both prime and redundant battery heaters could be expected to duty cycle during the first nights on the surface. A revised battery heater energy requirement of an additional 20% was provided by thermal modelling at RAL. The impact to the power profile overall, a drop of a few percent in minimum battery states of charge, was not of a level that necessitated a review of battery heater management.
Astrium design support provided an updated expected post deployment BSOC of 81%; a significant improvement on the 70% initial BSOC that all Phase 1 and 2 power modelling had incorporated. Figure E.12 shows the dramatic effect that the extra 11% initial state of charge makes on survivability in the north facing case with all panels at 160 degrees and optical depth 1.
Battery Charge in Percentage Format SA = - 20 0A = 180
100
90
80
70
60
50 Charge [%] 40
30
20
10
0 00:00 24/12/03 00:00 25/12/03 00:00 26/12/03 00:00 27/12/03 00:00 28/12/03 00:00 29/12/03 00:00 30/12/03 00:00 31/12/03 00:00 01/01/04 Date and Time [UTC]
Figure E.11 - Battery State of Charge. Panel Offsets 20 / North Facing Lander. Initial BSOC 81%
E.4.5 Other Power Modelling The above sections describe the main phases of power modelling carried out as part of the operations planning by the Flight Operations Team at the LOCC. Power modelling on less intensive scale was also carried out in order to determine appropriate values for certain lander software parameters relating to the power subsystem and as part of investigations relating to contingency operations of the Gas Analysis Package. Description of the modelling associated with these areas is outside the remit of this report.
138 Beagle-2 Mission Report Annexes
E.5 Conclusions As part of the operations planning phase, power modelling exercised and validated operations from the standpoint of management of the Beagle 2 power subsystem.
The results of power modelling influenced the development of preloaded Mission Events Timeline, lid and solar panel deployment and selection of power and communications subsystem software parameters. The analysis has been necessarily reactive. There has not been an effort to optimise performance but rather to validate scenarios driven by other system requirements and to identify survivable strategies.
E.6 Power Modelling Supporting Data Tables E.7 to E.11 show Beagle 2 Power Requirements used in Phase 1 and 2 power modelling.
Lander Subsystem Unit Power Requirements.
Supply Main Pyro PPS APS Total Voltage Bus Bus +5V +6V 12V 15V +12V - 12V 5 V (W) (W) (W) (W) (W) (W) (W) (W) (W) (W) APS 1.57 1.57 CEP 5.15 5.15 CEM 2.44 2.44 Frangibolt 30 30 Circuits Motors (1) 7.5 Motors (2) 1.5 1.5 1.5 Motors (3) 1.2 Motors (4) 0.36 Receiver* 3.36 3.36 Transmiter* 23.54 23.54 Tx + Rx 26.9 26.9 *Including Baseband
Table E.7 - Lander Subsystem Unit Power Requirements
Lander Subsystem Short Duration Loads
ARM Frangibolt Circuits 3A for 3 min
Pluto Lock Pin 16 W for 300s
Table E.8 - Lander Subsystem Short Duration Loads
139 Beagle-2 Mission Report Annexes
Payload Unit Current and Power Requirements
Supply PPS APS Power Voltage +5 V +6 V 12 V 15 V +12V - 12V 5 V Total (mA) (mA) (mA) (mA) (mA) (mA) (mA) (W) PAW 76.4 2 9 7 140 1.304 Torch 75.6 124 9 7 140 3.13 SCS LH 230.4 60 9 7 140 2.944 SCS RH 201.7 59 9 7 140 2.7855 FW LH 75.7 95 9 7 140 2.6955 FW RH 74 95 9 7 140 2.687 MIC Image 210.7 60 9 7 140 2.8455 MIC Focus 75 77 9 7 140 2.422 MIC All LEDs 75.3 143 9 7 140 3.4135 MIC Red LEDS 76 41 9 7 140 1.887 MIC Green LEDS 76 40 9 7 140 1.872 MIC Blue LEDS 76 40 9 7 140 1.872 MIC UV LEDS 76.1 27 9 7 140 1.6775 RCG Drill 76 9 7 140 1.272 RCG Hammer 76 90 9 7 140 2.352 MBS Shutter 76.2 33 9 7 140 1.669 Mole Winch 75 59 9 7 140 1.975 Mole Hammer 75 55 9 7 140 1.927 XRS Spec 450 2.7 XRS + Heaters 540 3.24 MBS Switch On 115 0.69 MBS Spec 300 1.8 ESS 190 0.95 GAP TBD
Table E.9 - Payload Unit Current and Power Requirements
140 Beagle-2 Mission Report Annexes
Supply Main Bus (W) (Ohms) A1 Battery 13.5 40.5 A2 Transmitter 5.5 80 A3 GAP SE 5 88.1 A4 (B4) XRS MBS 2.1 215 B1 Battery 13.5 40.5 B2 Receiver 5.5 80 B3 GAP Spec 5.6 78.4
Table E.10 - Heater Power and Resistance Specification
Thermal Model Energy Requirement Predictions
Battery heater energy BOL 26 Whrs / SOL Prime Battery Heater Battery heater energy BOL 31.2 Whrs / SOL Prime and Redundant Battery Heaters
Battery heater energy EOL 48 Whrs / SOL
Table E.11 - Thermal Model Energy Requirement Predictions
141 Beagle-2 Mission Report Annexes
This page left intentionally blank.
142 Beagle-2 Mission Report Annexes
F Electrical Cruise Behaviour Report
E. Chester, M. Hannington and O.Blake, NSC LOCC, University of Leicester
This Annex is included from the report “Beagle 2 Cruise Report – Electrical Behaviour”, Issue 1, 3 March 2004. For a summary of checkouts, please see the start of Annex B.
F.1 Battery Voltage and State of Charge
F.1.1 Power Subsystem Telemetry To asses the health and function of the Beagle 2 power subsystem six telemetry parameters available in the Mars Express data stream and three telemetry parameters available in the Beagle 2 data stream were considered.
MEX Data Stream B2 Power Subsystem Telemetry Mnemonic Title Description NBEA0110 BEAGLE-Battery V Battery Terminal Voltage NDAD0526 BEAS_PWR_STAT_P PFU Main Charge Circuit Status NDAD0614 BEAS_CHARG_STAT PFU Trickle Charge Status NDAD0627 BEA_APS_STAT Beagle APS status NPWD2541 LCL 5A Curr Beagle Representative of Prime PFU Main Charge current NPWD2861 LCL 5B Curr Beagle Representative of Redundant PFU Main Charge current
Beagle 2 Data Stream Power Subsystem Telemetry Mnemonic Title Description MA0017 Main Bus Current Battery Terminal Voltage MA0019 Battery Current Battery Charge / Discharge Current MA0013 Battery Voltage Battery Terminal Voltage
Table F.1 - Beagle 2 Power Subsystem Telemetry
The full compliment of nine TM parameters was available for each of the 10 occasions on which Beagle 2 was powered. Outside the checkout periods the MEX suite of parameters were obtained each day in two samples, of half hour duration, at 08:30 and 20:30.
F.1.2 Battery State of Charge Variation during Cruise Battery state of charge (BSOC) was calculated using the following:
• NBEA0110 BEAGLE-Battery V • battery temperature, • Estimates of battery charge / discharge currents based on NPWD2541 LCL 5A Curr Beagle and NPWD2861LCL 5B Curr Beagle • Estimates of load current based on Beagle 2 power load definitions.
Variation of the battery terminal voltage in MEX TM and the calculated BSOC across cruise phase are shown in Figure F.1. Figure F.2 shows the variation during Checkout I on the 17th December.
143 Beagle-2 Mission Report Annexes
B2 Battery State of Charge and Battery Terminal Voltage (MEX TM). Cruise Phase
120 25.5
25 100
24.5 80
BSOC 24 terminal volts
60
BSOC / % 23.5
40
23 Battery Terminal Voltage / Volts
20 22.5
0 22 08/06/2003 28/06/2003 18/07/2003 07/08/2003 27/08/2003 16/09/2003 06/10/2003 26/10/2003 15/11/2003 05/12/2003 25/12/2003 14/01/2004 00:00 00:00 00:00 00:00 00:00 00:00 00:00 00:00 00:00 00:00 00:00 00:00 Date / Time
Figure F.1 - Battery State of Charge and Battery Terminal Voltage Variation during Cruise Phase
B2 Battery State of Charge and Battery Terminal Voltage (MEX TM) Checkout I
120 25.2
25.1
100 25
24.9 80
24.8
BSOC 60 24.7 terminal volts BSOC / % 24.6
40 24.5
24.4 20
24.3
0 24.2 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 04:48 06:00 07:12 08:24 09:36 10:48 12:00 13:12 14:24 15:36 16:48 Date / Time
Figure F.2 - Battery State of Charge and Battery Terminal Voltage Variation Cruise Checkout I
144 Beagle-2 Mission Report Annexes
The trend in battery state of charge across cruise follows a profile consistent with the battery management strategies implemented. During the first half of cruise phase control of main charge during checkouts and effect of trickle charge in the periods between checkouts maintained BSOC at around 50%. From the 7th of October the battery was charged towards full charge during each checkout. Trickle charge is ineffective by design at battery states of charge above 50% and BSOC falls accordingly between checkouts E, F, G, H, I and J. The rate of charge and discharge during each checkout is consistent with the application of one, two or no PFU Main Charge supplies. Note, load imposed by spacecraft telemetry sensors contributes to battery charge decrease between checkouts.
F.1.3 Load Current - Beagle 2 Un-powered From the trend in BSOC during the periods between checkouts an estimation of the load current drawn when Beagle 2 was un-powered was made.
Checkout B to C. Load Drawn When Beagle 2 Un-powered Time BSOC / % Start 06/07/2003 02:08 52 End 12/07/2003 16:48 52 Calculated Load Current = 0 A
Checkout C to D. Load Drawn When Beagle 2 Un-powered Time BSOC / % Start 13/07/2003 11:20 51 End 01/09/2003 12:48 50 Calculated Load Current = 0.00011 A
Checkout F to G. Load Drawn When Beagle 2 Un-powered Time BSOC / % Start 09/10/2003 17:36 97 End 21/11/2003 07:52 75 Calculated Load Current = 0.0029 A
Checkout H to I. Load Drawn When Beagle 2 Un-powered Time BSOC / % Start 22/11/2003 16:33 97 End 17/12/2003 06:24 84 Calculated Load Current = 0.0030 A
Table F.2 - Estimated Un-powered Load Current
For the periods between checkouts B and C and checkouts C and D over all current drawn was zero evidencing effective trickle charging.
For the periods between checkouts F and G and checkouts H and I, with trickle charge inactive, current drawn was 2.9 mA and 3.0 mA respectively. These values are to the lower end of the estimated current drawn by the timer circuit and circuits supporting MEX B2 telemetry acquisition.
145 Beagle-2 Mission Report Annexes
F.1.4 Taper Charge and BSOC at Ejection In order to avoid battery overcharging a taper charge scheme is implemented. Once battery terminal voltage exceeds a value, settable by telecommand, charge connectivity is cycled and full charge is approached in a controlled manner. Taper charge is observed in telemetry as on/off cycling of PFU main charge status, cycling of the LCL current between 0 and approximately 550 mA and cycling of the reported battery terminal voltage. PFU Main Charge remained enabled throughout checkouts F to J with intention to fully charge the battery. Taper charge was achieved on 4 occasions during checkouts F, H, I and J.
Checkout F
• Prime PFU MC (A) circuit enabled at 09/10/2003 11:52. • Redundant PFU MC (B) line enabled at 09/10/2003 12:20 with both LCL 5 A and B currents showing around 560 mA. • Taper charge commenced at 09/10/2003 14:50. • Redundant PFU MC (B) was switched off at 09/10/2003 17:00.
MC duty cycle of both PFU power lines reduced from around 74% to 33% from taper charge commencing to PFU MC (B) switch off. PFU MC (A) duty cycle remained at 66%, supplying the same average current as the duty cycle with both power lines, until switch off at 09/10/2003 17:30. BSOC at commencement of taper charge was 86%. A 10% increase in BSOC was estimated from taper charge duty cycle characteristics over 2 hours and 40 minutes. BSOC calculated from MEX TM following switch off was consistent with this calculation at 97%.
Checkout H
• Prime PFU MC (A) circuit enabled at 22/11/2003 10:25 with LCL 5 A current showing around 560 mA. • Taper charge commenced at 22/11/2003 13:14. • Beagle 2 switch off at 22/11/2003 16:38
PFU MC (A) duty cycle was around 60%. BSOC at taper charge commencement was 95%. A 2.5% increase in BSOC was estimated from taper charge duty cycle characteristics over 3 hours and 8 minutes. BSOC calculated from MEX TM following switch off was consistent with this calculation at 97%.
Checkout I
• Prime PFU MC (A) circuit enabled at 17/12/2003 06:35 with LCL 5 A current showing around 560 mA. • Taper charge commenced at 17/12/2003 14:04. • Beagle 2 switch off at 17/12/2003 14:49
PFU MC (A) duty cycle was around 55%. BSOC at taper charge commencement was 96%. A 0.5% increase in BSOC was estimated from taper charge duty cycle characteristics over 45 minutes. BSOC calculated from MEX TM following switch off was consistent with this calculation at 97%.
Checkout J
• Prime PFU MC (A) circuit enabled at 18/12/2003 06:58 with LCL 5 A current showing around 560 mA. • Taper charge commenced at 18/12/2003 07:29. • Beagle 2 switch off at 18/12/2003 10:40
PFU MC (A) duty cycle reduced from 68% to 58%. BSOC at taper charge commencement was 94%. A 2.5% increase in BSOC was estimated from taper charge duty cycle characteristics over 3 hours and 11 minutes. BSOC calculated from MEX TM following switch off was consistent with this calculation at 97%. Taper charge analysis supports MEX TM indication that Beagle 2 battery state of charge at ejection from Mars Express stood at 97%.
146 Beagle-2 Mission Report Annexes
F.1.5 Estimating Beagle 2 Load and Charge Currents Battery charge/discharge current is not available directly in MEX TM. Considering the change in battery state of charge over a period of continuous charging or discharging it was possible to make estimates of the battery charge current during checkouts. The difference between this value and the LCL 5A current TM then provides an estimate of the Beagle 2 bus current.
Battery and Bus Current Estimates
Charge/ Battery Current Bus Current / A Bus Current / A Checkout Discharge / A PFU MC in TM PFU MC = 0.7 A A Charge 0.2748 0.2721 0.4252 B Discharge 0 0.4203 0.4203 C Charge 0.2977 0.2519 0.4022 D Charge 0.3038 0.2609 0.3963 E Charge 0.8526 0.0190 0.5474 (1) (1) (1) (1) F G Charge 0.2003 0.3574 0.4997 H Charge 0.1994 0.3030 0.5006 (1) (1) (1) (1) I (1) (1) (1) (1) J
(1) Taper charge: calculation unreliable.
Table F.3 - Battery and Bus Currents
The estimates must be considered approximate but the following observations are noted.
Battery current estimates show reasonable correlation with the expected values, 0.2 to 0.23 A. An extended period of battery discharge occurred in checkout B and the estimated discharge current approaches the expected range, 0.48 to 0.5 A.
Bus current estimates during each of the charge periods were made using PFU MC telemetry, showing an average supply current of around 550 mA. These estimates were lower than the expected range by of the order of 200 mA.
Bus current estimates were recalculated with a PFU MC of 0.7 mA, the current limit for the supply. Revised estimates were more consistent with the expected range.
NPWD251 LCL 5A Curr Beagle showed values across cruise averaging at 550 mA. The PFU main charge supply is current limited at 700 mA. Analysis of BSOC provides some indication that the PFU is in fact providing 700 mA. A mechanism which would result in the inconsistency has not been identified.
147 Beagle-2 Mission Report Annexes
F.1.6 Beagle 2 Data Stream Telemetry Analysis During final electrical testing an anomaly was discovered in Beagle 2 current telemetry. Both MA0017 Main Bus Current and MA0019 Battery Current report anomalous values.
The anomaly restricts the evaluation of power subsystem health using the Beagle 2 data stream.
F.1.6.1 Beagle 2 Bus and Battery Current
Reference R2 hypothesised a mechanism for the anomaly and corresponding formulae for calculating main bus current and battery currents from MA0017 and MA0019.
1. Main Bus Current = 1.4 * MA0019 2. Battery Charge Current = 3.4 * MA0017 + 2.1 * MA0019
The formulae were used to calculate Bus and Battery current from Beagle 2 TM across cruise. Correlations of calculated and expected battery and bus currents were reasonable, thus validating the formulae and supporting the hypothesised anomaly mechanism. The multiplying factor in formula 1 may be a little small.
Battery and Bus Current Formulae Evaluation Average Bus Current 0.400 A Average Battery Current 0.2419 A (Charging with one PFU MC) Average difference battery / bus 0.0420 A current (battery supplying bus)
F.1.6.2 Beagle 2 Battery Volts
Inconsistency between battery terminal voltage reported in Beagle 2 and MEX TM, MA0013 Battery Voltage and NBEA0110 BEAGLE-Battery V respectively, was evident throughout cruise. The difference between the two telemetry parameters in checkout C is shown in Figure F.3.
148 Beagle-2 Mission Report Annexes
Difference Battery Volts in B2 / MEX TM. Checkout C
23.4
23.2
23
22.8
B2 Battery Volts 22.6 MEX Battery Volts
Battery Volts / V Volts Battery 22.4
22.2
22
21.8 12/07/2003 12/07/2003 12/07/2003 12/07/2003 12/07/2003 12/07/2003 13/07/2003 13/07/2003 13/07/2003 16:48 18:00 19:12 20:24 21:36 22:48 00:00 01:12 02:24 Date / Time
Figure F.3 - Difference between MA0013 Battery Voltage and NBEA0110 BEAGLE-Battery V. Checkout C
The offset remained constant throughout cruise with NBEA0110 higher than MA0013 by an average of 0.5289 volts.
MA0013 reached a maximum of 24.6383, a value lower than the threshold for initiation of taper charge. MEX telemetry, along with the variation of battery current in Beagle 2 telemetry, demonstrates the successful operation of taper charging. Taper charging initiates at the expected voltage as reported in MEX TM. MEX battery voltage telemetry is therefore considered representative of the voltage as measured by the taper charge monitor circuit, and by extension the actual terminal voltage of the battery.
The mechanism behind the observation has not been investigated.
F.1.6.3 BSOC from Beagle 2 Telemetry
Battery state of charge was calculated from battery voltage and temperature in TM and the calculated battery and bus currents as described in section 2.6.1. The BSOC profile during checkouts follows that calculated using MEX TM. Figure F.4 shows the change in battery state of charge and battery voltage during checkout I as demonstrated by B2 TM. The difference between BSOC calculated using MEX TM and B2 TM are attributable to the difference in telemetered battery voltages. Figure F.5 shows the BSOC profile during checkout I calculated using a battery voltage modified to remove the 0.5289 volt offset.
149 Beagle-2 Mission Report Annexes
Battery Voltage and Battery State of Charge. Beagle 2 TM Checkout I
100 24.7
90 24.6
80 24.5 70
24.4 60
BSOC 50 24.3 terminal volts BSOC / % 40 24.2 Terminal Voltage / Volts 30 24.1 20
24 10
0 23.9 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 06:00 07:12 08:24 09:36 10:48 12:00 13:12 14:24 15:36 Date / Time
Figure F.4 - Battery State of Charge and Battery Terminal Voltage Variation Cruise Checkout I. B2 TM
Battery Voltage and Battery State of Charge. Adjusted B2 Battery Volts Checkout I
120 25.2
25.1 100
25
80 24.9
BSOC 60 24.8 terminal volts BSOC / %
24.7
40 volts / voltage terminal
24.6
20 24.5
0 24.4 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 17/12/2003 06:00 07:12 08:24 09:36 10:48 12:00 13:12 14:24 15:36 Date / Time
Figure F.5 - BSOC and Battery Voltage Variation B2 TM Checkout I. Modified Battery Voltage Calculation
150 Beagle-2 Mission Report Annexes
F.1.7 Battery State of Charge Conclusions • Battery voltage and state of charge trend across cruise followed a profile consistent with the battery management strategies employed.
• Beagle 2 un-powered load currents were consistent with expected values for timer and telemetry circuits.
• Taper charging was initiated and operated correctly during 4 of the checkouts.
• Beagle 2 was ejected from Mars Express with a battery state of charge of 97%
• The effects of an anomaly on the telemetry returned for battery and bus current in the Beagle 2 telemetry stream were characterised. An anomaly in the values of battery voltage returned in Beagle 2 telemetry was identified and characterised. Inconsistency between expected and observed LCL 5 current telemetry was noted.
151 Beagle-2 Mission Report Annexes
F.2 Software Counters and Errors during Cruise The following sections present summary information about the values of software counters maintained by PSW. These are all reported in telemetry packet 8 of type 3,25.
F.2.1 Power-On Reset Counter (MSC0043) Parameter MSC0043 was examined during cruise and correctly increments at every planned power-on and reset of the system. The final value was as expected, confirming that there were no unplanned / unexplained resets during cruise phase operations.
Checkout Time Expected Count Value Actual Count Value A First Cruise Checkout 2003-07-04 20:04:04 95,96 95,96 B Next Checkout 2003-07-05 19:03:00 97,98 97,98 C 2nd NEV Checkout 2003-07-12 16:46:00 99 99 D Mini Checkout, E2 Scrub 2003-09-01 12:40:17 100 100 E LSW Upload Test 2003-10-07 11:03:00 101 101 F 2003-10-09 11:48:00 102 102 G LSW Upload 2003-11-21 08:00:00 103 103 H LSW Upload Part 2 (Fix) 2003-11-22 10:00:00 104 104 I PECO 2003-12-17 06:34:00 105 105 J PECO2 2003-12-18 06:33:51 106 106
Table F.4 - Reset Counter
F.2.2 Memory EDAC Counters The following parameters are reported by PSW:
Parameter Description Recorded Value(s) during Cruise MSC0051 EEPROM Multi-bit Error Counter 0 throughout cruise MSC0053 RAM Multi-bit Error Counter 0 throughout cruise MSC0052 RAM Single-bit Error Counter See RAM EVENTS section below
Table F.5 - EDAC Counters
152 Beagle-2 Mission Report Annexes
F.2.3 Single-bit RAM Events (MSC0052) The following table summaries the single-bit events in RAM for each checkout. Where no value is reported for a checkout, no RAM errors were observed. All times given are to within ±60s unless otherwise stated. If a larger uncertainty is given, it is due only to the time interval between consecutive telemetry packets reporting parameter MSC0052. This is typically seen when many packets of other types are being generated. It is assumed that a count of 2 corresponds to a single SEU.
Checkout Date/Time MSC0052 Count
A 04/07/2003 21:33 +2 05/07/2003 04:58 0 (following reset) C 13/07/2003 00:19 +2 D 01/09/2003 13:43 ±120s +2 E 07/10/2003 15:02 +2 G 21/11/2003 14:59 +4 H 22/11/2003 10:22 ±300s +2 22/11/2003 13:59 +4 J 18/12/2003 06:50 +2 SEU Total 10
Table F.6 - Single-Bit RAM Events
F.2.4 Miscellaneous Counters The following parameters are reported by PSW:
Parameter Description Recorded Value(s) during Cruise MSC0072 Seg Violation Counter 0 throughout cruise MSC0044 Watchdog Timeout 0 throughout cruise MSC0045 Constraint Error Counter 0 throughout cruise MSC0046 Storage Error Counter 0 throughout cruise MSC0073 Program Error Counter 0 throughout cruise MSC0074 Task Error Counter 0 throughout cruise MSC0075 Illegal Instruction Error Counter 0 throughout cruise MSC0048 Bus Error Counter 0 throughout cruise MSC0049 FPU Error Counter 0 throughout cruise MSC0076 Unknown Error Counter 0 throughout cruise
Table F.7 - Miscellaneous Event Counters
153 Beagle-2 Mission Report Annexes
F.3 APS Voltage History The Auxiliary Power Supply (APS) provides seven regulated power lines for which four voltage telemetries are available.
• MD0021 APS Analog V • MD0020 APS 13 V Monitor • MD0019 APS 5V Monitor • MD0018 APS 3.3 V Monitor
MD0018 APS 3.3 V Monitor, MD0019 APS 5V Monitor and MD0020 APS 13 V Monitor report the voltage of the respective individual power lines. The APS +12, -12, +5 and -5 volt analogue outputs are combined in a single telemetry channel reported in MD0021 APS Analog V.
Telemetry for the duration of the cruise phase was available and three areas of analysis were carried out.
• APS power line voltage variation across cruise phase, June 4th to December 17th • APS power line voltage trend across cruise phase June 4th to December 17th • APS power line trend during individual checkouts A to J
F.3.1 APS Power Line Voltage Specification
APS Specification Voltage Line Spec Voltage / V Spec Tolerance / % Tolerance Testing / %* APS 3.3 V (L) 3.3 3 2.87 APS 5 V (L) 5.12 5 2.5 APS 13.5 V 13.5 10 5.32
Table F.8 - APS Specification
Shown are the voltage and tolerance specifications and the revised tolerance estimation from unit level testing before launch. Estimated tolerance includes worst case initial tolerance, regulation, age drift, temperature drift, radiation degradation and ripple voltage.
F.3.2 APS Power Line Variation during Cruise Phase Variation across the duration of cruise phase was considered and the following summary obtained.
Voltage Variation Across Cruise MD0021 MD0020 MD0019 MD0018 APS analog V APS 13V monitor APS 5 V monitor APS 3.3V monitor Max Voltage / V 1.7901 13.5406 5.1353 3.3005 Minimum Voltage / V 1.7866 13.4562 5.0821 3.2902 Mean Voltage / V 1.7874 13.5245 5.1204 3.2982 Spread Max - Min / V 0.00343 0.0844 0.0531 0.01031 Telemetry bit size 0.00031251 0.0009705 0.00062504 0.00031252 Spread Max - Min / 11 87 84 32 RAW Bits
Table F.9 - Power Line Variation during Cruise
154 Beagle-2 Mission Report Annexes
Comparison of variation during cruise with the Unit Level Testing Specification was made. For each power line the delta from the voltage specification to maximum and minimum exhibited voltages were calculated and represented as percentages of the specified tolerance.
Comparison: Voltage Variation Across Cruise / Specification MD0021 MD0020 MD0019 MD0018 APS analog V APS 13V monitor APS 5 V monitor APS 3.3V monitor Maximum Voltage / V 1.7901 13.5406 5.1353 3.3005 Delta from Specification Not Available 0.0406 0.0153 0.0005 to Maximum Voltage / V Delta represented as % of Not Available 5.70 12.25 0.54 tolerance
Minimum Voltage / V 1.7866 13.4562 5.0821 3.2902 Delta from Specification TBD 0.044 0.0378 0.0097 to Maximum Voltage / V Delta represented as % of TBD 0.80 0.31 1.8 tolerance
Table F.10 - Voltage Variation during Cruise
All power line voltages were observed to be well within tolerance. The APS 5V line shows greatest variation at 12.25% of tolerance. The maximums of 5.135 seen for the 5.5V line correspond to only two telemetry samples (out of 1749). A maximum of 5.128 may be more representative giving a variation of 6% of tolerance.
F.3.3 APS Power Line Trend across Cruise Phase Application of least squares fit linear trend line to telemetry provided following trend summary.
MD0021 MD0020 MD0019 MD0018 APS analog V APS 13V monitor APS 5 V monitor APS 3.3V monitor
Trend Line Gradient 4.00E-07 5.00E-06 1.00E-06 3.00E-08 Number of Samples 1749 1749 1749 1749 Voltage Rise over Cruise / V 0.0006996 0.00875 0.001751 0.00005256
Table F.11 - Power Line Trend during Cruise
Very slight upward trend for all four voltage lines was observed over cruise.
Figure F.6 shows the trend of APS 13 V line across cruise phase. To allow the pattern to be discerned checkouts A to J have been plotted contiguously. There were 1749 telemetry samples of APS voltage over the period June 4th to December 17th.
155 Beagle-2 Mission Report Annexes
MD0020 APS 13 V Line Trend Across Cruise Phase
13.55 y = 5E-06x + 13.521 13.54
13.53
13.52
13.51
MD0020 13.5 Linear (MD0020)
Voltage / Volts 13.49
13.48
13.47
13.46
13.45 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Sample
Figure F.6 - APS 13V Variation over Cruise Phase.
Following analysis of voltage variation during each checkout (see section 4.4) it is considered that the trend over cruise is likely to be a feature of the voltage response to APS temperature rather than a real drift in regulated voltage.
156 Beagle-2 Mission Report Annexes
F.3.4 APS power line trend during individual checkouts A to J Voltage variation during each of checkouts A to J was analysed. All four voltages follow an exponential increase from switch on to switch off. The pattern is most marked on the 13V and 5V lines with the 13V line rising by over 80 mV on the longer checkouts.
MD0021 MD0020 MD0019 MD0018 Checkout APS analog V APS 13V monitor APS 5 V monitor APS 3.3V monitor Rise / V Rise / V Rise / V Rise / V A First Cruise Checkout 0.0065 0.0747 0.0381 0.0075 B Next Checkout 0.0053 0.0747 0.0343 0.0093 C 2nd NEV Checkout 0.0065 0.0844 0.0300 0.0090 D Mini Checkout, E2 0.0031 0.0669 0.0262 0.0096 Scrub E LSW Upload Test 0.0059 0.0747 0.0343 0.0096 F 0.0053 0.0805 0.0331 0.0100 G LSW Upload 0.0071 0.0815 0.0393 0.0090 H LSW Upload Part 2 (Fix) 0.0068 0.0708 0.0531 0.0081 I PECO 0.0056 0.0815 0.0368 0.0090 J PECO2 0.0065 0.0747 0.0318 0.0084 AVERAGE 0.0059 0.0764 0.0357 0.0090
Table F.12 - APS Line Power Trend during Checkouts
Figure F.7 shows the exponential rise of the 13 V line during checkout G correlating with the rise in APS temperature, parameter MT0011 APS Temp. Discernable correlation between APS temperature and voltage is evident with all 4 voltage telemetries.
MD0020 APS 13 V and MT0011 APS Temp. Checkout G. 21 / 11 / 2003
70 13.55
60 13.54
13.53 50
13.52 40
13.51 30 MT0011 13.5 MD0020
Temp DgC 20 VoltageV / 13.49
10 13.48
0 13.47 07:12 09:36 12:00 14:24 16:48 19:12 21:36
-10 13.46
-20 13.45 Date / Time
Figure F.7 - APS 13 Volt variation correlated with APS temperature variation
157 Beagle-2 Mission Report Annexes
Checkout Epoch Max Value of MT0011 A 05/07/2003 05:02:19 48.7083995 B 06/07/2003 01:39:37 47.2217556 C 13/07/2003 01:10:51 49.058198 D 01/09/2003 13:58:58 30.3439751 E 07/10/2003 16:45:00 47.6590038 F 09/10/2003 17:33:51 49.0144732 G 21/11/2003 19:24:00 57.322189 H 22/11/2003 16:38:09 54.1302771 I 17/12/2003 14:47:42 53.1246063 J 18/12/2003 10:39:47 47.2654804
Table F.13 - APS Maximum Temperatures
APS maximum temperature showed upward trend over cruise, correlated to on time duration. This pattern is likely to account for the trend across cruise in voltages reported in section F.3.4.
Cross-checking commands and telemetry by hand
158 Beagle-2 Mission Report Annexes
F.3.4.1 Observation OBS 3 APS Temperature Rise Rate
Following checkout A an observation was raised on APS temperature rise rate. Deviation from expected rise rates and maximum temperatures provided by thermal modelling were observed. The temperature profile was seen to be asymptotic to a value above the upper operational limit of 55 °C on each checkout. APS temperature approached operational limit during checkouts H and I and exceeded the limit by 2°C during checkout G. The qualification limit is 65°C and the unit design is good to greater than 75°C.
More detail about OBS 3 is provided in the observation report.
Observation Reference Description Status OBS-003 APS Temperature Rise Rate CLOSED
In light of the observation, the stability of the APS voltage lines at increased temperature was given consideration. No unusual pattern was seen in the 3.3 V, 13 V or analogue V telemetry at the higher temperatures.
The 5V line showed some increased variability while the APS temperature was above 55°C during checkout G. Figures F.8 and F.9 compare the 5V profile for checkouts F and G.
MD0019 APS 5V Checkout F. 09 / 10 / 2003
5.135
5.13
5.125
5.12
5.115 MD0019 5.11 VoltageV /
5.105
5.1
5.095
5.09 10:48 12:00 13:12 14:24 15:36 16:48 18:00 Date / Time
Figure F.8 - APS 5V line during Checkout F
159 Beagle-2 Mission Report Annexes
MD0019 APS 5 V Checkout G. 21 / 11 / 03
5.14
5.135
5.13
5.125
5.12
5.115 MD0019 VoltageV / 5.11
5.105
5.1
5.095
5.09 07:12 09:36 12:00 14:24 16:48 19:12 21:36 Date / Time
Figure F.9 - APS 5V line during Checkout G
The segment of increased variation correlates with the APS temperature rising above 55°C however the time also coincides with the highest level of processor and TT and C subsystem activity seen during cruise (upload and return of LSW image) and the occurrence of observation OBS 007 Loss of Memory Dump Packets During LSW Upload (see OBS 007 report).
Observation Reference Description Status OBS-007 Loss of Memory Dump Packets During LSW Upload CLOSED
The gap in telemetry from around 16:00 to 18:00 is due to unexpected transceiver switching as part of OBS 007.
No increased variation is seen during the 5V profile during checkout H, when the final APS temperature was also high at 54°C.
F.3.5 APS Voltage History Conclusions APS voltages were healthy and well within test tolerances for the duration of cruise and no overall trend was observed. Voltage variations are correlated with APS temperature. Some limited evidence points towards increased variability in the 5V line at APS temperatures above operational limit.
160 Beagle-2 Mission Report Annexes
F.4 Timer and Latches
F.4.1 Timer Operation The timer circuit is responsible for switching on the Beagle processor prior to entry into the top of the atmosphere.
F.4.2 Timer Load for Ejection The following table summarises the key information used by the Beagle team to load and verify the timer prior to ejection of Beagle 2 from Mars Express.
Estimated top-of-atmosphere time 359-02:50:22 Provided by ESOC- FD Assumed timer start time 351-13:50:22 Provided by ESOC- FCT APS On-time Prior to entry 02:30:00 Provided by Beagle2/Astrium Maximum timer value 325h 15m 0s Provided by Beagle2/Astrium Timer tick period 142.930s Provided by Beagle2/Astrium
Table F.14 - Timer Data for Ejection
F.4.3 Timer and Latch Cruise Testing The following table summarises the timer tests conducted post-launch.
Checkout DoY Date Latch Tick Start/Stop Preload / Reset Trip Notes A 185 2003-07-04 X X B 186 2003-07-05 X X X Anomalies with the Mars Express database were found (single encapsulation) C 193 2003-07-12 X X X X Timer was preloaded twice, and tripped at the correct time. D 244 2003-09-01 No timer activity other than start-up sequence E 280 2003-10-07 X X X F 282 2003-10-09 X X X Confirmed timer preload enables the long latch G 325 2003-11-21 X X X H 326 2003-11-22 X X X X I 351 2003-12-17 X X X X Timer load for ejection (0E71) J 352 2003-12-18 X Final latch enable for ejection
Table F.15 - Timer Tests
The end of this Annex contains a summary of all changes in Beagle 2 timer telemetry and a correlation with Mars Express command history, and other observations and notes.
161 Beagle-2 Mission Report Annexes
F.4.4 Timer and Latch Testing Conclusions No spacecraft anomalies were found in any timer test or related activity. The early problems with the Mars Express database for Beagle 2 commanding were overcome by assembling Beagle 2 commands as Lander Operations Requests (LORs) at the LOCC. The only other problems related to timer operation resulted from incorrectly constructed commands (single encapsulation with invalid source field), and a misunderstanding about the complex telecommand 238,3 (Perform Activity of a Function) explained the change in long latch status following a timer preload (Function ID 5) – this was investigated and confirmed in checkout F.
F.5 Data Volumes of Cruise Phase Checkouts The following table shows a summary of data volumes during the cruise phase checkouts.
For each checkout, the number of Beagle2 telecommands forwarded to the lander by Mars Express is shown, along with the volume in bytes of those commands. The volume in bytes of all returned Beagle2 telemetry is shown in the final column.
The most significant data in the table are checkouts F and G, where a large volume of forward commands are seen for the LSW Image Upload test and the LSW Image Upload itself.
Checkout Date No. of Cmds Fwd (TC) Vol (bytes) Rtn (TM) Vol (bytes)
A 2003-07-04 36 500 117416
B 2003-07-05 36 552 249836
C 2003-07-12 209 22648 332142
D 2003-09-01 3 36 47968
E 2003-10-07 198 26452 257794
F 2003-10-09 1237 213008 359264
G 2003-11-21 3811 660108 665404
H 2003-11-22 240 25920 326426
I 2003-12-17 82 1862 334794
J 2003-12-18 11 160 161278
Totals 5863 951246 2852322
Table F.16 - Checkout Data Volumes
162 Beagle-2 Mission Report Annexes
F.6 LSW Upload Procedure The procedure for uploading LSW is found in BOP-012, and is summarised here.
The entire LSW image was provided as a single stack of 2154 telecommands. Due to limitations of buffer size and checksum memory length, the stack was divided into 26 files each with the structure given in section7.1. The timing and structure of these files was verified in the LSW Image Upload Test checkout of 7th Oct 2003 (see BOP-009).
F.6.1 File Structure (Files 1 to 25)
Write memory telecommand 1 … Write memory telecommand 20 Check memory using absolute address 6,9 telecommand Perform connection test telecommand 17,1 BLOCK 1 BLOCK Perform connection test telecommand 17,1 Perform connection test telecommand 17,1 Write memory telecommand 1 … Write memory telecommand 20 Check memory using absolute address 6,9 telecommand Perform connection test telecommand 17,1 BLOCK 2 BLOCK Perform connection test telecommand 17,1 Perform connection test telecommand 17,1
FILE Write memory telecommand 1 … Write memory telecommand 20 Check memory using absolute address 6,9 telecommand Perform connection test telecommand 17,1 BLOCK 3 BLOCK Perform connection test telecommand 17,1 Perform connection test telecommand 17,1 Write memory telecommand 1 … Write memory telecommand 20 Check memory using absolute address 6,9 telecommand Perform connection test telecommand 17,1 BLOCK 4 BLOCK Perform connection test telecommand 17,1 Perform connection test telecommand 17,1
The block size above is determined by the maximum length of memory than can be checked by the checksum telecommand. The connection test commands are inserted to allow the checksum command sufficient time to complete. All files except the final file have the above structure. The final file is illustrated in the following subsection.
163 Beagle-2 Mission Report Annexes
F.6.2 File Structure (File 26)
Write memory telecommand 1 … Write memory telecommand 20 Check memory using absolute address 6,9 telecommand Perform connection test telecommand 17,1 BLOCK 1 BLOCK Perform connection test telecommand 17,1 Perform connection test telecommand 17,1 Write memory telecommand 1 … Write memory telecommand 20 Check memory using absolute address 6,9 telecommand Perform connection test telecommand 17,1 FILE 26 BLOCK 2 BLOCK Perform connection test telecommand 17,1 Perform connection test telecommand 17,1 Write memory telecommand 1 … Write memory telecommand 10 Check memory using absolute address 6,9 telecommand Perform connection test telecommand 17,1 BLOCK 3 BLOCK Perform connection test telecommand 17,1 Perform connection test telecommand 17,1
F.6.3 Group Structure (Groups 1 to 6) Groups are composed of four files (with the exception of group 7). Each file is released from MEX SSMM and then followed by abort link and rehails with appropriate timing. 2 minutes is allowed for the release of telecommands from the files (96 telecommands, 1 per second); 30 seconds is allowed for the link to be aborted; 1 minute allowed for the link to be established before continuing with the next file. The files are divided into groups as follows:
Block Files 1 1,2,3,4 2 5,6,7,8 3 9,10,11,12 4 13,14,15,16 5 17,18,19,20 6 21,22,23,24 7 25,26
164 Beagle-2 Mission Report Annexes
Despatch FILE 1 telecommand file from SSMM using MTL telecommand Wait 2 minutes Abort the Melacom/Beagle2 link using link abort command Wait 30 seconds FILE 1 Rehail Beagle2 at 8/8/Seq Wait 1 minute Despatch FILE 2 telecommand file from SSMM using MTL telecommand Wait 2 minutes Abort the Melacom/Beagle2 link using link abort command Wait 30 seconds FILE 2 Rehail Beagle2 at 8/8/Seq Wait 1 minute Despatch FILE 3 telecommand file from SSMM using MTL telecommand GROUP Wait 2 minutes Abort the Melacom/Beagle2 link using link abort command Wait 30 seconds FILE 3 Rehail Beagle2 at 8/8/Seq Wait 1 minute Despatch FILE 4 telecommand file from SSMM using MTL telecommand Wait 2 minutes Abort the Melacom/Beagle2 link using link abort command Wait 30 seconds FILE 4 Rehail Beagle2 at 8/8/Seq Wait 1 minute SSMM SSMM Drip-tap read DUMP
F.6.4 Group Structure (Group 7) The final group has a different structure as there are only 2 files remaining for upload.
Despatch FILE 25 telecommand file from SSMM using MTL telecommand Wait 2 minutes Abort the Melacom/Beagle2 link using link abort command Wait 30 seconds
FILE 25 Rehail Beagle2 at 8/8/Seq Wait 1 minute Despatch FILE 26 telecommand file from SSMM using MTL telecommand
GROUP 7 Wait 2 minutes Abort the Melacom/Beagle2 link using link abort command Wait 30 seconds
FILE 26 Rehail Beagle2 at 8/8/Seq Wait 1 minute SSMM SSMM Drip-tap read DUMP
165 Beagle-2 Mission Report Annexes
F.6.5 LSW Upload Deviations from Procedure During the execution of the LSW Upload Procedure (BOP-012) the following anomalies were encountered, and observation reports generated. The relevant Observation Reports should be consulted for detailed description and actions taken, but in the context of LSW software upload, each anomaly is summarised here along with the corrective actions.
Observation Reference Description Status OBS-007 Loss of Memory Dump (6,6) Packets during LSW Upload CLOSED OBS-008 Command Failures During LSW Upload CLOSED OBS-009 Checksum Errors During LSW Upload CLOSED
F.6.5.1 OBS-007 - Loss of Dump Packets
During the return of the LSW image from EEPROM a proportion of the expected 6, 6 Memory Dump using Absolute Address Packets were not received. The loss of packets was due to the MLT misinterpreting memory dump data as command instruction to switch off. An LSW patch has been implemented to avoid reoccurrence of the problem.
Approximately 52% of the LSW image was successfully dumped and verified as correct, and all checksums for entire software image were returned as correct and taken as sufficient evidence for a successful upload.
F.6.5.2 OBS-008 - Command Failures
During the LSW upload 17 memory load telecommands failed. These corresponded to the contents of Block 4 of File 19 of the LSW software. The source sequence count of the commands was in violation of the Beagle 2 SGICD due to an error in the MCS software. The SCOS2000 parameter CMD_MAX_PUS_SEQ_CNT was updated to a new lower value of 2047 (it was set to 16384). The failed commands were regenerated using the MCS and successfully executed by Beagle 2.
One of the failed commands was the checksum command for File 19 Block 4, and so one checksum was missing at the end of the first LSW upload checkout. Block 4 of File 19 was reloaded successfully and the correct checksum (E085) returned.
F.6.5.3 OBS-009 - Checksum Errors
During the upload of LSW three areas of memory failed checksum. The image dump packets however verified correct memory contents. The checksum commands were repeated and correct values were returned.
In BOP 12 LSW Image Upload commands to write each block of memory were immediately followed by the checksum commands for that area. These commands were executed from a TC file on board Mars Express with a spacing of 1 second. The most likely mechanism for the observation is that the checksum commands executed before the previous memory write operation had completed and thus returned an erroneous value. The 3 second delay for completion of the checksum command was marginal.
166 Beagle-2 Mission Report Annexes
F.6.6 LSW Checksum Record The expected and reported checksums for each file are given below. At the end of the image upload process at 16:30:00 on 22nd Nov 2003 (following the retransmission of block 19) all checksums were verified as correct.
File Block Checksum File Block Checksum File 1 1 a7e7 File 14 5 4a1a 2 63c8 6 1c2d 3 b8ea 7 8970 4 22b4 8 dec0 File 2 5 dad2 File 15 9 edc0 6 5c42 10 40df 7 9966 11 caad 8 59da 12 7fa9 File 3 9 729d File 16 13 9155 10 6d01 14 8bd2 11 f8fa 15 2f0a 12 11c9 16 2d8c File 4 13 94a1 File 17 1 d3a8 14 c587 2 818b 15 464f 3 176d 16 3b50 4 7f0d File 5 1 2ab1 File 18 5 dfc1 2 c6b1 6 5850 3 39b0 7 ffd4 4 2e6f 8 387f File 6 5 e446 File 19 9 d659 6 62e0 10 d881 7 9d7f 11 e3f7 8 3ce0 12 e085 File 7 9 3803 File 20 13 d136 10 e3bc 14 e7bd 11 6f1f 15 7294 12 fd0d 16 8c65 File 8 13 be59 File 21 1 7644 14 a1c8 2 1faa 15 cdc8 3 fc6e 16 c8e8 4 2add File 9 1 e1b8 File 22 5 fc6b 2 a589 6 7d5f 3 7f85 7 9fe8 4 279b 8 3f7e File 10 5 7f3d File 23 9 e34b 6 e6c0 10 75a1 7 8549 11 0b74 8 e7d5 12 bd1f File 11 9 4981 File 24 13 5a3a 10 5922 14 d209 11 4803 15 b292 12 02b5 16 e9a3 File 12 13 42a3 File 25 1 831e 14 b3bb 2 48c0 15 dfa7 3 d79a 16 b96f 4 26b6 File 13 1 2617 File 26 5 3e5a 2 adf2 6 d192 3 8696 7 424d 4 2fd6
167 Beagle-2 Mission Report Annexes
F.7 PSW and LSW Checksum History The following tables contain a summary of the Probe Software and Lander Software checksum checks.
F.7.1 PSW Checksum Summary The first table show the PSW checksums and when they were checked. As can be seen, PSW was checked at every checkout in which a large amount of data was written to the EEPROM. All checksums were correct at each time of checking.
Checkout PSW Checksum 2003- 2003- 2003- 2003- 2003- 2003- 2003- 2003- 2003- 2003- Block 07-04 07-05 07-12 09-01 10-07 10-09 11-21 11-22 12-17 12-18 1 ec7a OK OK OK OK 2 90a3 OK OK OK OK 3 ad11 OK OK OK OK 4 c002 OK OK OK OK 5 e025 OK OK OK OK 6 bacb OK OK OK OK 7 8371 OK OK OK OK 8 edaa OK OK OK OK 9 9fc5 OK OK OK OK 10 d934 OK OK OK OK 11 9038 OK OK OK OK 12 56ee OK OK OK OK 13 907c OK OK OK OK 14 c21e OK OK OK OK 15 056f OK OK OK OK 16 6799 OK OK OK OK 17 38df OK OK OK OK 18 5218 OK OK OK OK 19 1262 OK OK OK OK 20 8744 OK OK OK OK
Table F.17 - Probe Software Checksum Verifications
The table on the following page shows a summary of the LSW checksum checks during LSW 3.0 upload.
As is discussed in more detail elsewhere in this document, there were four incorrectly reported checksums during the first LSW Image Upload Checkout, but these were re-checked the at the following checkout and were seen to be correct.
168 Beagle-2 Mission Report Annexes
F.7.2 LSW Checksum Summary
Checkout File Block Expected 2003- 2003- File Block Expected Checkout Checksum 11-21 11-22 2003- 2003- 1 1 a7e7 OK 11-21 11-22 2 63c8 OK 14 5 4a1a OK 3 b8ea OK 6 1c2d OK 4 22b4 OK 7 8970 X - e876 OK 2 5 dad2 OK 8 36861 OK 6 5c42 OK 15 9 edc0 OK 7 9966 OK 10 40df OK 8 59da OK 11 caad OK 3 9 729d OK 12 7fa9 OK 10 6d01 OK 16 13 9155 OK 11 f8fa OK 14 8bd2 OK 12 11c9 OK 15 2f0a OK 4 13 94a1 OK 16 2d8c OK 14 c587 OK 17 1 d3a8 OK 15 464f OK 2 818b OK 16 3b50 OK 3 176d OK 5 1 2ab1 OK 4 7f0d OK 2 c6b1 OK 18 5 dfc1 OK 3 39b0 OK 6 5850 OK 4 2e6f OK 7 ffd4 OK 6 5 e446 OK 8 387f OK 6 62e0 OK 19 9 d659 OK 7 9d7f OK 10 d881 OK 8 3ce0 OK 11 e3f7 OK 7 9 3803 OK 12 e085 X OK 10 e3bc OK 20 13 d136 OK 11 6f1f OK 14 e7bd OK 12 fd0d OK 15 7294 OK 8 13 be59 OK 16 8c65 OK 14 a1c8 OK 21 1 7644 OK 15 cdc8 OK 2 1faa OK 16 c8e8 OK 3 fc6e OK 9 1 e1b8 OK 4 2add OK 2 a589 OK 22 5 fc6b OK 3 7f85 OK 6 7d5f X - 2129 OK 4 279b OK 7 9fe8 OK 10 5 7f3d OK 8 3f7e OK 6 e6c0 OK 23 9 e34b OK 7 8549 OK 10 75a1 OK 8 e7d5 OK 11 0b74 OK 11 9 4981 OK 12 bd1f OK 10 5922 OK 24 13 5a3a OK 11 4803 OK 14 d209 OK 12 02b5 OK 15 b292 OK 12 13 42a3 OK 16 e9a3 OK 14 b3bb OK 25 1 831e OK 15 dfa7 OK 2 48c0 OK 16 b96f X - 8f12 OK 3 d79a OK 13 1 2617 OK 4 26b6 OK 2 adf2 OK 26 5 3e5a OK 3 8696 OK 6 d192 OK 4 2fd6 OK 7 424d OK
Table F.19 - Lander Software Checksum Verification
169 Beagle-2 Mission Report Annexes
F.8 EEPROM Integrity Check
F.8.1 Overview of Final Checkout The final Beagle 2 checkout was conducted using Beagle2 operating procedure: BOP-022: Beagle 2 Dec 17th Pre-ejection Checkout
The checkout included the following activities:
• LSW Patches Upload • LSW Early Data Upload (MET, Communications sessions etc.) (N.B. Activity sequences were not uploaded) • EDLS Upload • EEPROM Integrity Check
The checkout was conducted on December 17th 2003 (day 351), with Beagle 2 switched on at 06:34:00. After successful completion of the procedure Beagle 2 was powered down at 14:51:00. This report details the checks and results of the integrity check of the EEPROM only. The EEPROM check was conducted using the procedure: BOP-010 Issue 4: Beagle2 EEPROM Integrity Check. This check was conducted on the basis that all other procedures in the checkout were completely successfully.
F.8.2 Scope of EEPROM Integrity Check The EEPROM integrity check procedure BOP-010 was developed to verify correct EEPROM contents for correct Beagle operation. It specifically checks critical areas of memory that are used during boot-up and initialisation of the spacecraft. Each critical area is replicated in memory, in a ‘mirror copy’. The software verifies and selects which mirror copy to use automatically. These copies are referred to as ‘mirror copy 1’ (MC/1) and ‘mirror copy 2’ (MC/2). Unless otherwise indicated, all checks are conducted on both MC/1 and MC/2.
To validate EEPROM contents, two kinds of check are performed. Areas of memory can be checked by the probe software and a checksum returned in telemetry (checksum), or the entire memory contents can be dumped in telemetry packets and the exact memory contents reconstituted on the ground (dump).
The areas checked and the methods used are as follows:
PSW Image Area Checksum Allocation Table Area Dump Boot Context Area Dump Common Software Context Area Dump PSW Parameters Area Dump PSW Work Parameters Area Dump
The communications sessions and MET data uploaded in the checkout during procedure BOP-022 Step 4 are not verified as part of the integrity check.
The PSW image area contents were not downloaded for verification (i.e. BOP-010 Issue 4 Step 4 was omitted). This was in order to avoid a reported anomaly with the MLT (packet data interpreted as command words). All checksums were generated and validated correctly.
170 Beagle-2 Mission Report Annexes
F.8.3 EEPROM Check: Commands and Results Return PSW Image Area Checksums Checksum commands (TC 6,9: Check Memory Using Absolute LOR 9,1,1 Address) were issued as follows, to check all 20 blocks of the PSW. Command Block Checksum FM matches GTM ? 1e5cc090000d1906090005010008000003c0b48d 1 EC7A Y 1e5cc091000d19060900050100080f0003c0ce9f 2 90A3 Y 1e5cc092000d19060900050100081e0003c040a9 3 AD11 Y 1e5cc093000d19060900050100082d0003c05960 4 C002 Y 1e5cc094000d19060900050100083c0003c04ce4 5 E025 Y 1e5cc095000d19060900050100084b0003c0f140 6 BACB Y 1e5cc096000d19060900050100085a0003c07f76 7 8371 Y 1e5cc097000d1906090005010008690003c066bf 8 EDAA Y 1e5cc098000d1906090005010008780003c0547e 9 9FC5 Y 1e5cc099000d1906090005010008870003c0b121 10 D934 Y 1e5cc09a000d1906090005010008960003c03f17 11 9038 Y 1e5cc09b000d1906090005010008a50003c026de 12 56EE Y 1e5cc09c000d1906090005010008b40003c0335a 13 907C Y 1e5cc09d000d1906090005010008c30003c08efe 14 C21E Y 1e5cc09e000d1906090005010008d20003c000c8 15 056F Y 1e5cc09f000d1906090005010008e10003c01901 16 6799 Y 1e5cc0a0000d1906090005010008f00003c0f85e 17 38DF Y 1e5cc0a1000d1906090005010008ff0003c0824c 18 5218 Y 1e5cc0a2000d19060900050100090e0003c022c1 19 1262 Y 1e5cc0a3000d19060900050100091d000227b2be 20 8744 Y
The following areas of memory were dumped using ‘Dump Memory Using Absolute Address (6, 5)’ telecommands despatched from LOR 11,10,10. FM matches GTM ? Allocation Table MC/1 and MC/2 See section 4.1 1e5cc0e2000d190605000501000000000038ded4 Y 1e5cc0e3000d190605000501000d8800003809a9 Y
Boot Context Area MC/1 and MC/2 See section 4.2 1e5cc0e4000d190605000501000001800002d248 Y 1e5cc0e5000d190605000501000d898000020535 Y
Common Software Context Area MC/1 and MC/2 See section 4.3 1e5cc0e6000d1906050005010000020000089e5d Y 1e5cc0e7000d190605000501000d8a0000084920 Y
PSW Work Parameters MC/1 and MC/2 See section 4.4 1e5cc0ea000d19060500050100000680001d9195 Y 1e5cc0eb000d190605000501000d8e80001d46e8 Y
PSW Parameters Area MC/1 and MC/2 See section 4.5 1e5cc0e8000d1906050005010000028000a4204f Y 1e5cc0e9000d190605000501000d8a8000a4f732 Y
All five areas above were compared with the corresponding dumps from the GTM and confirmed as identical. The values dumped follow in Section 4.
171 Beagle-2 Mission Report Annexes
F.8.4 EEPROM Contents The following tables detail the memory contents dumped as per the commands in section 3 above. These data are taken from the GTM, and were compared with the corresponding memory dumps from the FM.
F.8.4.1 EEPROM Allocation Table
Address Data 04000000 04000800 04000004 0001805E 04000094 00000800 04000008 FFAA5500 04000098 0055AAFF 0400000C 00000000 0400009C 00000000 04000010 00000000 040000A0 040A3800 04000014 00000000 040000A4 00001000 04000018 0055AAFF 040000A8 0055AAFF 0400001C 00000000 040000AC 00000000 04000020 00000000 040000B0 040A7800 04000024 00000000 040000B4 0000A000 04000028 0055AAFF 040000B8 0055AAFF 0400002C 00000000 040000BC 00000000 04000030 00000000 040000C0 040CF800 04000034 00000000 040000C4 00000200 04000038 0055AAFF 040000C8 0055AAFF 0400003C 00000000 040000CC 00000000 04000040 04069800 040000D0 040D0000 04000044 00002A00 040000D4 00001A00 04000048 0055AAFF 040000D8 0055AAFF 0400004C 00000000 040000DC 00000000 04000050 04074000
04000054 00003000
04000058 0055AAFF Start Address of MC/2: 0400005C 00000000 04000060 04080000 040D8800 04000064 00004967 04000068 FFAA5500 0400006C 00000000 04000070 0409B000 04000074 00000200 04000078 0055AAFF 0400007C 00000000 04000080 0409B800 04000084 00002000 04000088 0055AAFF 0400008C 00000000 04000090 040D6800
172 Beagle-2 Mission Report Annexes
F.8.4.2 Boot Context Area
Address Data 04000180 0055AAFF 04000184 FFAA5500
Start Address of MC/2: 040D8980
F.8.4.3 Common Software Context Area
Address Data 04000200 FFAA5500 04000204 00000000 04000208 00000000 0400020C 00000000 04000210 00020000 04000214 00000000 04000218 00026800 0400021C 00000000
Start Address of MC/2: 040D8A00
F.8.4.4 PSW Work Parameters
Address Data 04000680 00000000 MAY VARY 04000684 00000000 MAY VARY 04000688 00000000 MAY VARY 0400068C 00000000 MAY VARY 04000690 00000000 04000694 00000000 04000698 00000000 0400069C 00000000 040006A0 00000000 040006A4 00000000 040006A8 00000000 040006AC 00000000 040006B0 00000000 040006B4 00000000 040006B8 00000000 040006BC 00000000 040006C0 00000000 040006C4 00000000 040006C8 00000000 040006CC 00000000 040006D0 00000000 040006D4 00000000 040006D8 00000000 040006DC 00000000 040006E0 00000000 040006E4 00000000 040006E8 00000000 040006EC 00000000 040006F0 00000000
Start Address of MC/2: 040D8E80
173 Beagle-2 Mission Report Annexes
F.8.4.5 PSW Parameters Table MC/1
Address Data 04000368 00000000 04000454 3FFBD09E 04000280 C3860000 0400036C 00000000 04000458 BF77B376 04000284 430BAAC1 04000370 00000000 0400045C 3891ED24 04000288 00000002 04000374 3BA2497D 04000460 3911ED17 0400028C 00000002 04000378 3C22497E 04000464 3891ED24 04000290 00000C50 0400037C 3BA2497D 04000468 3FFC7019 04000294 00001F08 04000380 3FDC913A 0400046C BF78F272 04000298 00000054 04000384 BF3E34BE 04000470 389298AC 0400029C 000016CA 04000388 3BA5867C 04000474 3912989E 040002A0 40F851EC 0400038C 3C25867C 04000478 389298AC 040002A4 42200000 04000390 3BA5867C 0400047C 3FFD98D9 040002A8 42200000 04000394 3FE0F7EC 04000480 BF7B4408 040002AC 40A00000 04000398 BF471C0D 04000484 38937B2E 040002B0 BF19999A 0400039C 3BABDCA3 04000488 39137B2E 040002B4 3F19999A 040003A0 3C2BDCA4 0400048C 38937B2E 040002B8 BE816F00 040003A4 3BABDCA3 04000490 3FFF20B8 040002BC 424C44D0 040003A8 3FE9949F 04000494 BF7E53D7 040002C0 00000000 040003AC BF588827 04000498 3F800000 040002C4 00000000 040003B0 3BB4E918 0400049C 00000000 040002C8 00000000 040003B4 3C34E917 040004A0 00000000 040002CC 00000000 040003B8 3BB4E918 040004A4 00000000 040002D0 3F800000 040003BC 3FF5E0F8 040004A8 00000000 040002D4 39EAE18B 040003C0 BF716932 040004AC 3BA2497D 040002D8 00000000 040003C4 3F800000 040004B0 3C22497E 040002DC 00000000 040003C8 00000000 040004B4 3BA2497D 040002E0 18DAEA40 040003CC 00000000 040004B8 3FDC913A 040002E4 000124F8 040003D0 00000000 040004BC BF3E34BE 040002E8 000001A1 040003D4 00000000 040004C0 3BA5867C 040002EC 000001A1 040003D8 40F851EC 040004C4 3C25867C 040002F0 18DAEA40 040003DC 42200000 040004C8 3BA5867C 040002F4 0000104A 040003E0 42200000 040004CC 3FE0F7EC 040002F8 000005B2 040003E4 40A00000 040004D0 BF471C0D 040002FC 0000002A 040003E8 BFE66666 040004D4 3BABDCA3 04000300 00000EA6 040003EC 3FE66666 040004D8 3C2BDCA4 04000304 000000D0 040003F0 BF810AB3 040004DC 3BABDCA3 04000308 000061A8 040003F4 43237C83 040004E0 3FE9949F 0400030C 00000000 040003F8 00000000 040004E4 BF588827 04000310 389190F6 040003FC 00000000 040004E8 3BB4E918 04000314 39119104 04000400 00000000 040004EC 3C34E917 04000318 389190F6 04000404 00000000 040004F0 3BB4E918 0400031C 3FFBD09E 04000408 3F800000 040004F4 3FF5E0F8 04000320 BF77B376 0400040C 3AB955F8 040004F8 BF716932 04000324 3891ED24 04000410 00000000 040004FC 3F800000 04000328 3911ED17 04000414 00000000 04000500 00000000 0400032C 3891ED24 04000418 18DAEA40 04000504 00000000 04000330 3FFC7019 0400041C 000124F8 04000508 00000000 04000334 BF78F272 04000420 000001A1 0400050C 00000000 04000338 389298AC 04000424 000001A1 0400033C 3912989E 04000428 18DAEA40 Start Address of MC/2: 04000340 389298AC 0400042C 0000104A 040D8A80 04000344 3FFD98D9 04000430 000005B2 04000348 BF7B4408 04000434 0000002A 0400034C 38937B2E 04000438 00000EA6 04000350 39137B2E 0400043C 000000D0 04000354 38937B2E 04000440 000061A8 04000358 3FFF20B8 04000444 00000000 0400035C BF7E53D7 04000448 389190F6 04000360 3F800000 0400044C 39119104 04000364 00000000 04000450 389190F6
174 Beagle-2 Mission Report Annexes
F.8.5 EEPROM Memory Map
bytes kbytes words kwords start address end address EEPROM area 1572864 1536 393216 384 0400 0000 0417 FFFF EEPROM blocks Block 1 524288 512 131072 128 0400 0000 0407 FFFF Block 2 524288 512 131072 128 0408 0000 040F FFFF Block 3 524288 512 131072 128 0410 0000 0417 FFFF
Mirrored EEPROM Mirrored copy 1 686080 670 171520 167.5 0400 0000 040A 77FF Critical parameters area 2048 2 512 0.5 0400 0000 0400 07FF EEPROM allocation table 384 0.375 96 0.09375 0400 0000 0400 017F Boot context area 128 0.125 32 0.03125 0400 0180 0400 01FF Common software context area 128 0.125 32 0.03125 0400 0200 0400 027F PSW parameter table 768 0.75 192 0.1875 0400 0280 0400 057F LSW parameter table 256 0.25 64 0.0625 0400 0580 0400 067F PSW working parameters 256 0.25 64 0.0625 0400 0680 0400 077F Spare 128 0.125 32 0.03125 0400 0780 0400 07FF LSW image 430080 420 107520 105 0400 0800 0406 97FF LSW patches 43008 42 10752 10.5 0406 9800 0407 3FFF PSW patches 49152 48 12288 12 0407 4000 0407 FFFF PSW image 110592 108 27648 27 0408 0000 0409 AFFF LSW work area 2048 2 512 0.5 0409 B000 0409 B7FF LSW default timeline data 32768 32 8192 8 0409 B800 040A 37FF Spare 0 0 0 0 040A 37FF PSW work area 16384 16 4096 4 040A 3800 040A 77FF Context data 16000 15.625 4000 3.90625 040A 3800 040A 767F Spare 384 0.375 96 0.09375 040A 7680 040A 77FF Spare 0 0 0 0 040A 77FF
Single-copy EEPROM Single-copy data 200704 196 50176 49 040A 7800 040D 87FF HHD 163840 160 40960 40 040A 7800 040C F7FF Science data 131072 128 32768 32 040A 7800 040C 77FF HK data 26624 26 6656 6.5 040C 7800 040C DFFF Event data 6144 6 1536 1.5 040C E000 040C F7FF Non-critical boot context data 2048 2 512 0.5 040C F800 040C FFFF LSW history log backup 26624 26 6656 6.5 040D 0000 040D 67FF LSW XRS LUT 8192 8 2048 2 040D 6800 040D 87FF Spare 0 0 0 0 040D 87FF
Mirrored EEPROM Mirrored copy 2 686080 670 171520 167.5 040D 8800 0417 FFFF Critical parameters area 2048 2 512 0.5 040D 8800 040D 8FFF EEPROM allocation table 384 0.375 96 0.09375 040D 8800 040D 897F Boot context area 128 0.125 32 0.03125 040D 8980 040D 89FF Common software context area 128 0.125 32 0.03125 040D 8A00 040D 8A7F PSW parameter table 768 0.75 192 0.1875 040D 8A80 040D 8D7F LSW parameter table 256 0.25 64 0.0625 040D 8D80 040D 8E7F PSW working parameters 256 0.25 64 0.0625 040D 8E80 040D 8F7F Spare 128 0.125 32 0.03125 040D 8F80 040D 8FFF LSW image 430080 420 107520 105 040D 9000 0414 1FFF LSW patches 43008 42 10752 10.5 0414 2000 0414 C7FF PSW patches 49152 48 12288 12 0414 C800 0415 87FF PSW image 110592 108 27648 27 0415 8800 0417 37FF LSW work area 2048 2 512 0.5 0417 3800 0417 3FFF LSW default timeline data 32768 32 8192 8 0417 4000 0417 BFFF Spare 0 0 0 0 0417 BFFF PSW work area 16384 16 4096 4 0417 C000 0417 FFFF Context data 16000 15.625 4000 3.90625 0417 C000 0417 FE7F Spare 384 0.375 96 0.09375 0417 FE80 0417 FFFF Spare 0 0 0 0 0417 FFFF
175 Beagle-2 Mission Report ANNEXES
F.8.6 Timer Telemetry Changes Correlated to Command History The following table present changes in key timer and latch-related telemetry. This table was produced by extracting the key telemetry parameters for the entire cruise phase period. The following post-processing steps were applied:
• Convert SCOS timestamps to normal date formats • Include samples where a change in any parameter occurred. This change is usually the clock tick (high-low-high transitions) • The Mars Express command history was analysed using custom software to report Beagle 2 commands and timestamps • The commands were correlated with events seen in the telemetry sample list. • Remaining entries where the only parameter changing is ME0006 (timer tick) were removed. • Several such ‘tick-only’ samples were retained just to report status. • All samples were checked for changes in any of ME0007,8,9 (i.e. unexplained or uncorrelated events) and none were found.
The final list is presented here. Changes in ME0006 (TICK) are ignored unless the tick starts or stops. The conclusion of this analysis is that every item in Beagle 2 timer telemetry is understood, and correlates with either a telecommand or the normal behaviour of the timer circuit. No anomalies with the spacecraft were identified.
Beagle Telemetry Parameters Long Latch Short Latch Trip Tick Count Command Time TIME ME0009 ME0008 ME0007 ME0006 ME0005 Commanded Activity / Observations (from History) 04/07/2003 20:04:02 DIS DIS NOT TRIP LOW 0 ON 04/07/2003 20:15:33 DIS DIS NOT TRIP HIGH 27C Timer Set 04/07/2003 20:15:33 DIS DIS NOT TRIP LOW 27C 05/07/2003 04:58:28 DIS DIS NOT TRIP LOW 0 04:47:00 PSW Reboot 05/07/2003 05:00:23 DIS DIS NOT TRIP LOW 297 OFF 05/07/2003 19:02:59 DIS DIS NOT TRIP LOW 0 ON 05/07/2003 19:04:54 DIS DIS NOT TRIP HIGH 2C4 05/07/2003 19:22:11 DIS DIS NOT TRIP HIGH 2C5 05/07/2003 19:26:02 EN DIS NOT TRIP HIGH 2C5 19:25:00 Perform Activity of a Function - Start Timer Tick 05/07/2003 19:27:57 EN DIS NOT TRIP LOW 2C5 05/07/2003 19:27:57 EN DIS NOT TRIP HIGH 2C5 19:31:00 Direct Register Load - Reset Timer Clock Register (Failed because ESOC database contained single-encapsulated commands) 05/07/2003 19:31:47 EN DIS NOT TRIP LOW 2C5 05/07/2003 19:31:47 EN DIS NOT TRIP LOW 2C5 19:36:00 Perform Activity of a Function - Start Timer Tick (Failed because ESOC database contained single-encapsulated commands) 05/07/2003 19:31:47 EN DIS NOT TRIP HIGH 2C5 05/07/2003 19:47:09 EN DIS NOT TRIP HIGH 2C5 19:47:59 Direct Register Load - Reset Timer Clock Register (Failed because ESOC database contained single-encapsulated commands) 05/07/2003 19:47:09 EN DIS NOT TRIP LOW 2C5 05/07/2003 19:49:04 EN DIS NOT TRIP HIGH 2C5 05/07/2003 19:52:54 EN DIS NOT TRIP LOW 2C5 19:52:00 Perform Activity of a Function - Start Timer Tick 176
Beagle-2 Mission Report ANNEXES
Beagle Telemetry Parameters Long Latch Short Latch Trip Tick Count Command Time TIME ME0009 ME0008 ME0007 ME0006 ME0005 Commanded Activity / Observations (from History) (Failed because ESOC database contained single-encapsulated commands) 05/07/2003 19:54:50 EN DIS NOT TRIP HIGH 2C5 05/07/2003 19:54:50 EN DIS NOT TRIP LOW 2C5 05/07/2003 20:15:57 EN DIS NOT TRIP LOW 2C7 20:16:00 Direct Register Load - Reset Timer Clock Register (Failed due to error in ESOC database - command ACK fields incorrectly set) 05/07/2003 20:17:52 EN DIS NOT TRIP HIGH 2C8 05/07/2003 20:21:43 EN DIS NOT TRIP LOW 2C8 20:20:00 Perform Activity of a Function - Start Timer Tick (Failed due to error in ESOC database - command ACK fields incorrectly set) 05/07/2003 20:25:33 EN DIS NOT TRIP HIGH 2C8 05/07/2003 20:37:04 EN DIS NOT TRIP HIGH 2C8 20:37:00 Direct Register Load - Reset Timer Clock Register (Failed due to error in ESOC database - command ACK fields incorrectly set) 05/07/2003 20:42:50 EN DIS NOT TRIP LOW 2C8 20:42:00 Perform Activity of a Function Start Timer Tick (Failed due to error in ESOC database - command ACK fields incorrectly set) 05/07/2003 20:46:40 EN DIS NOT TRIP HIGH 2C8 05/07/2003 20:48:35 EN DIS NOT TRIP LOW 2C8 05/07/2003 20:48:35 DIS DIS NOT TRIP LOW 2C8 20:48:20 Reset long period timer timeout enable latch 05/07/2003 20:52:26 DIS DIS NOT TRIP HIGH 2C8 20:48:21 Reset long period timer timeout enable latch 05/07/2003 20:52:26 DIS DIS NOT TRIP LOW 2C8 06/07/2003 01:18:29 DIS DIS NOT TRIP LOW 0 01:18:00 PSW REBOOT 06/07/2003 01:20:24 DIS DIS NOT TRIP LOW 2D7 OFF 12/07/2003 16:45:51 DIS DIS NOT TRIP LOW 0 16:45:00 ON 12/07/2003 16:47:46 DIS DIS NOT TRIP LOW CD 12/07/2003 17:58:49 DIS DIS NOT TRIP HIGH D0 18:00:01 Reset short period timer timeout enable latch value 1 12/07/2003 18:00:44 DIS DIS NOT TRIP LOW D0 18:00:02 Reset short period timer timeout enable latch value 0 12/07/2003 18:00:44 DIS DIS NOT TRIP HIGH D0 18:00:02 Reset long period timer timeout enable latch value 1 12/07/2003 18:02:39 DIS DIS NOT TRIP LOW D0 18:00:03 Reset long period timer timeout enable latch value 0 12/07/2003 18:04:35 DIS DIS NOT TRIP HIGH D0 12/07/2003 18:04:35 DIS DIS NOT TRIP LOW D0 12/07/2003 18:04:35 DIS DIS NOT TRIP LOW 0 Reset Timer 12/07/2003 18:06:30 DIS DIS NOT TRIP HIGH 0 12/07/2003 18:08:25 DIS DIS NOT TRIP HIGH 0 18:08:01 Reset long period timer timeout enable latch value 1 12/07/2003 18:10:20 DIS DIS NOT TRIP LOW 0 18:08:02 Reset long period timer timeout enable latch value 0 12/07/2003 18:10:20 DIS DIS NOT TRIP HIGH 0 12/07/2003 18:23:47 DIS DIS NOT TRIP LOW 1 12/07/2003 18:23:47 DIS DIS NOT TRIP LOW 4 18:23:03 Perform Activity of a Function 1,5 - Preset Timer, value AAA 12/07/2003 18:23:47 DIS DIS NOT TRIP LOW 6 12/07/2003 18:39:08 EN DIS NOT TRIP HIGH 154 18:38:00 Perform Activity of a Function 1,6 - Start Timer Tick 12/07/2003 18:39:08 EN DIS NOT TRIP LOW 154 12/07/2003 18:58:20 EN DIS NOT TRIP HIGH 156 12/07/2003 18:58:20 EN DIS NOT TRIP LOW 156 18:58:04 Perform Activity of a Function 1,5 - Preset Timer, value 1555 12/07/2003 18:58:20 EN DIS NOT TRIP LOW 0 Timer set 12/07/2003 18:58:20 EN DIS NOT TRIP LOW 3 177
Beagle-2 Mission Report ANNEXES
Beagle Telemetry Parameters Long Latch Short Latch Trip Tick Count Command Time TIME ME0009 ME0008 ME0007 ME0006 ME0005 Commanded Activity / Observations (from History) 12/07/2003 19:04:06 EN DIS NOT TRIP HIGH 2AA Correct 12/07/2003 19:11:47 EN DIS NOT TRIP LOW 2AA 19:10:00 Perform Activity of a Function 1,6 - Start Timer Tick 12/07/2003 19:21:23 EN DIS NOT TRIP LOW 2AA 12/07/2003 19:23:18 EN DIS NOT TRIP HIGH 2AA 12/07/2003 19:23:18 EN DIS NOT TRIP LOW 2AA 12/07/2003 19:23:18 EN DIS NOT TRIP LOW 1 19:23:04 Perform Activity of a Function 1,5 - Preset Timer, value 1FF6 12/07/2003 19:23:18 EN DIS NOT TRIP LOW 4 12/07/2003 19:30:59 EN DIS NOT TRIP LOW 3FE Correct 12/07/2003 19:30:59 EN DIS NOT TRIP HIGH 3FE 12/07/2003 19:38:39 EN DIS NOT TRIP LOW 3FE 19:38:00 Perform Activity of a Function 1,6 - Start Timer Tick 12/07/2003 19:40:35 EN DIS NOT TRIP HIGH 3FE 12/07/2003 19:46:20 EN DIS NOT TRIP LOW 3FF 19:47:45 Set long period timer timeout enable latch value 1 12/07/2003 19:48:15 EN DIS NOT TRIP HIGH 3FF 19:47:47 Set long period timer timeout enable latch value 0 12/07/2003 19:48:15 EN DIS NOT TRIP LOW 3FF 12/07/2003 19:50:11 EN DIS NOT TRIP HIGH 3FF 12/07/2003 19:52:06 EN DIS NOT TRIP LOW 3FF 19:51:47 Reset long period timer timeout enable latch value 1 12/07/2003 19:52:06 EN DIS NOT TRIP HIGH 3FF 19:51:47 Reset long period timer timeout enable latch value 0 12/07/2003 20:01:42 DIS DIS NOT TRIP LOW 3FF 12/07/2003 20:01:42 DIS DIS NOT TRIP HIGH 0 Trip – Correct 12/07/2003 20:03:37 DIS DIS NOT TRIP LOW 0 12/07/2003 20:26:40 DIS DIS NOT TRIP HIGH 1 20:26:00 Perform Activity of a Function 1,5 - Preset Timer, value 14 (to switch BCR) (Failed because ESOC database contained single-encapsulated commands) 12/07/2003 20:30:30 DIS DIS NOT TRIP HIGH 1 12/07/2003 20:43:56 EN DIS NOT TRIP HIGH 1 20:43:06 Perform Activity of a Function 1,6 - Start Timer Tick 12/07/2003 20:43:56 EN DIS NOT TRIP LOW 1 12/07/2003 21:41:32 EN DIS NOT TRIP HIGH 4 21:40:17 Direct Register Load - Reset Timer Clock Value 1 (Failed due to checksum error in ESOC database – command with SSC=C0AF contained incorrect Beagle 2 command checksum (2DAF)– should have been D2AF) 12/07/2003 21:41:33 EN DIS NOT TRIP LOW 4 12/07/2003 21:45:24 DIS DIS NOT TRIP LOW 4 21:44:36 Reset long period timer timeout enable latch value 1 12/07/2003 21:45:24 DIS DIS NOT TRIP HIGH 4 21:44:36 Reset long period timer timeout enable latch value 0 12/07/2003 21:47:19 DIS DIS NOT TRIP LOW 4 12/07/2003 23:04:06 DIS DIS NOT TRIP HIGH 9 12/07/2003 23:09:52 EN DIS NOT TRIP HIGH 9 23:09:24 Perform Activity of a Function 1,6 - Start Timer Tick 12/07/2003 23:15:37 EN DIS NOT TRIP LOW A 12/07/2003 23:17:33 EN DIS NOT TRIP HIGH A 23:17:32 Direct Register Load - Reset Timer Clock Value 1 (Failed due to checksum error in ESOC database – command with SSC=C0AF contained incorrect Beagle 2 command checksum (2DAF)– should have been D2AF ) 12/07/2003 23:19:28 EN DIS NOT TRIP HIGH A 12/07/2003 23:21:23 EN DIS NOT TRIP LOW A 23:20:52 Perform Activity of a Function Start Timer Tick Double Encaps Mex DB 12/07/2003 23:27:09 EN DIS NOT TRIP HIGH A 12/07/2003 23:27:09 EN DIS NOT TRIP LOW A 178
Beagle-2 Mission Report ANNEXES
Beagle Telemetry Parameters Long Latch Short Latch Trip Tick Count Command Time TIME ME0009 ME0008 ME0007 ME0006 ME0005 Commanded Activity / Observations (from History) 12/07/2003 23:29:04 EN DIS NOT TRIP HIGH A 12/07/2003 23:32:54 EN DIS NOT TRIP LOW A 23:32:13 Perform Activity of a Function Start Timer Tick 12/07/2003 23:36:45 EN DIS NOT TRIP HIGH A 12/07/2003 23:38:40 EN DIS NOT TRIP LOW A 23:39:10 Direct Register Load - Reset Timer Clock Value 1 (Failed due to checksum error in ESOC database – command with SSC=C0AF contained incorrect Beagle 2 command checksum (2DAF)– should have been D2AF ) 12/07/2003 23:40:35 EN DIS NOT TRIP HIGH A 12/07/2003 23:42:30 EN DIS NOT TRIP HIGH A 23:42:56 Perform Activity of a Function Start Timer Tick 12/07/2003 23:42:30 EN DIS NOT TRIP LOW A 12/07/2003 23:52:06 EN DIS NOT TRIP HIGH A 12/07/2003 23:54:01 EN DIS NOT TRIP HIGH C 12/07/2003 23:57:52 EN DIS NOT TRIP LOW C 23:55:55 Perform Activity of a Function 1,6 – Start Timer Tick 12/07/2003 23:59:47 EN DIS NOT TRIP LOW C 13/07/2003 00:01:42 EN DIS NOT TRIP HIGH C 00:00:14 Reset long period timer timeout enable latch value 1 13/07/2003 00:01:42 DIS DIS NOT TRIP HIGH C 00:00:16 Reset long period timer timeout enable latch value 0 13/07/2003 01:05:04 DIS DIS NOT TRIP HIGH F 13/07/2003 01:07:00 DIS DIS NOT TRIP LOW F 01:17:40 Reset short period timer timeout enable latch value 1 13/07/2003 01:08:55 DIS DIS NOT TRIP HIGH F Reset short period timer timeout enable latch value 0 13/07/2003 01:08:55 DIS DIS NOT TRIP LOW F 01:17:45 Reset long period timer timeout enable latch value 1 13/07/2003 01:10:50 DIS DIS NOT TRIP HIGH F Reset long period timer timeout enable latch value 0 13/07/2003 01:16:36 DIS DIS NOT TRIP LOW 10 OFF 01/09/2003 12:40:14 DIS DIS NOT TRIP LOW 2F5 ON 01/09/2003 12:40:14 DIS DIS NOT TRIP LOW 0 Perform Activity of a Function 1,5 - Preset Timer, value 14 (to switch BCR) (Failed because ESOC database contained single-encapsulated commands) 01/09/2003 12:42:09 DIS DIS NOT TRIP LOW 2F5 01/09/2003 12:42:09 DIS DIS NOT TRIP HIGH 2F5 12:40:36 Perform Activity of a Function 1,5 - Preset Timer, value 14 (to switch BCR) (Failed because ESOC database contained single-encapsulated commands) 01/09/2003 13:58:58 DIS DIS NOT TRIP LOW 2FA OFF 07/10/2003 11:03:13 DIS DIS NOT TRIP LOW 190 ON 07/10/2003 11:03:13 DIS DIS NOT TRIP LOW 0 07/10/2003 11:05:08 DIS DIS NOT TRIP LOW 190 11:03:00 CHARGE ON 07/10/2003 11:05:08 DIS DIS NOT TRIP HIGH 190 07/10/2003 11:28:11 DIS DIS NOT TRIP LOW 192 07/10/2003 11:28:11 DIS DIS NOT TRIP LOW 1 07/10/2003 11:28:11 DIS DIS NOT TRIP HIGH 1 11:27:09 Perform Activity of a Function 1,5 - Preset Timer, value 14 (to switch BCR) 07/10/2003 11:30:06 DIS DIS NOT TRIP HIGH 1 Tick stopped 07/10/2003 11:41:37 DIS DIS NOT TRIP HIGH 1 07/10/2003 13:09:57 EN DIS NOT TRIP HIGH 1 13:09:24 Perform Activity of a Function 1,6 – Start Timer Tick 07/10/2003 13:11:52 EN DIS NOT TRIP LOW 1 Note - latch enabled 07/10/2003 16:43:04 EN DIS NOT TRIP LOW D 16:45:03 Reset short period timer timeout enable latch (toggled) 07/10/2003 16:44:59 EN DIS NOT TRIP HIGH D 16:45:05 Reset long period timer timeout enable latch (toggled) / OFF 09/10/2003 11:48:13 DIS DIS NOT TRIP HIGH 94 179
Beagle-2 Mission Report ANNEXES
Beagle Telemetry Parameters Long Latch Short Latch Trip Tick Count Command Time TIME ME0009 ME0008 ME0007 ME0006 ME0005 Commanded Activity / Observations (from History) 09/10/2003 11:48:13 DIS DIS NOT TRIP LOW 0 11:48:00 ON 09/10/2003 11:50:08 DIS DIS NOT TRIP HIGH 94 09/10/2003 11:50:08 DIS DIS NOT TRIP LOW 94 11:03:00 CHARGE ON 09/10/2003 11:53:59 DIS DIS NOT TRIP HIGH 94 09/10/2003 11:53:59 DIS DIS NOT TRIP HIGH 0 11:52:53 Perform Activity of a Function 1,5 - Preset Timer, value 14 (to switch BCR) 09/10/2003 11:53:59 DIS DIS NOT TRIP LOW 1 09/10/2003 11:53:59 DIS DIS NOT TRIP HIGH 1 09/10/2003 11:57:49 EN DIS NOT TRIP HIGH 1 11:57:45 Perform Activity of a Function 1,6 – Start Timer Tick 09/10/2003 11:59:44 EN DIS NOT TRIP LOW 1 09/10/2003 11:59:44 DIS DIS NOT TRIP LOW 1 11:59:15 Reset long period timer timeout enable latch (toggled) 09/10/2003 11:59:44 DIS DIS NOT TRIP HIGH 1 09/10/2003 17:31:55 DIS DIS NOT TRIP HIGH 13 17:30:03 Reset short period timer timeout enable latch (toggled) 09/10/2003 17:33:50 DIS DIS NOT TRIP LOW 13 17:30:05 Reset long period timer timeout enable latch (toggled) / OFF 21/11/2003 08:00:12 DIS DIS NOT TRIP LOW A6 21/11/2003 08:00:12 DIS DIS NOT TRIP LOW 0 08:00:01 ON 21/11/2003 08:02:07 DIS DIS NOT TRIP LOW A6 08:01:06 Perform Activity of a Function 1,5 - Preset Timer, value 14 (to switch BCR) (Failed because ESOC database contained single-encapsulated commands) 21/11/2003 08:02:07 DIS DIS NOT TRIP HIGH A6 08:01:06 Perform Activity of a Function 1,6 – Start Timer Tick 21/11/2003 08:02:07 EN DIS TRIP HIGH A6 21/11/2003 08:02:07 DIS DIS NOT TRIP HIGH A6 08:01:08 Reset long period timer timeout enable latch (toggled) 21/11/2003 08:04:03 DIS DIS NOT TRIP LOW A6 21/11/2003 09:32:22 DIS DIS NOT TRIP HIGH AB 09:32:08 Perform Activity of a Function 1,5 - Preset Timer, value 14 (to switch BCR) (Failed because ESOC database contained single-encapsulated commands) 21/11/2003 09:34:17 DIS DIS NOT TRIP LOW AB 21/11/2003 09:34:17 EN DIS TRIP LOW AB 09:33:07 Perform Activity of a Function 1,6 – Start Timer Tick 21/11/2003 09:34:17 EN DIS TRIP HIGH AC 21/11/2003 09:36:12 EN DIS TRIP LOW AC 21/11/2003 18:14:48 EN DIS TRIP LOW C7 18:14:55 Reset long period timer timeout enable latch (toggled) 21/11/2003 18:16:42 DIS DIS NOT TRIP LOW C7 21/11/2003 19:23:59 DIS DIS NOT TRIP LOW CA 19:25:04 OFF 22/11/2003 10:17:58 DIS DIS NOT TRIP HIGH F7 10:00:01 ON 22/11/2003 10:17:58 DIS DIS NOT TRIP LOW 0 22/11/2003 10:37:10 DIS DIS NOT TRIP LOW F9 Perform Activity of a Function 1,5 - Preset Timer, value 14 (to switch BCR) (Failed because ESOC database contained single-encapsulated commands) 22/11/2003 10:37:10 DIS DIS TRIP LOW F9 22/11/2003 10:37:10 EN DIS TRIP LOW F9 Perform Activity of a Function 1,6 – Start Timer Tick 22/11/2003 10:37:10 EN DIS TRIP HIGH F9 22/11/2003 11:00:13 EN DIS TRIP LOW FA 22/11/2003 11:00:13 EN DIS TRIP HIGH FA 22/11/2003 11:02:08 EN DIS TRIP LOW FA 11:03:00 Reset long period timer timeout enable latch (toggled) 22/11/2003 11:04:03 EN DIS TRIP HIGH FA 22/11/2003 11:17:29 EN DIS TRIP HIGH FA 180
Beagle-2 Mission Report ANNEXES
Beagle Telemetry Parameters Long Latch Short Latch Trip Tick Count Command Time TIME ME0009 ME0008 ME0007 ME0006 ME0005 Commanded Activity / Observations (from History) 22/11/2003 11:19:25 EN DIS NOT TRIP HIGH 0 Timer set for test 22/11/2003 11:19:25 EN DIS NOT TRIP HIGH 0 Reset and not trip as expected 22/11/2003 11:21:20 EN DIS NOT TRIP LOW 0 22/11/2003 11:29:01 DIS DIS NOT TRIP LOW 0 11:20:00 Perform Activity of a Function 1,6 – Start Timer Tick 22/11/2003 11:29:01 DIS DIS NOT TRIP HIGH 0 22/11/2003 11:30:56 DIS DIS NOT TRIP LOW 0 22/11/2003 11:38:37 DIS DIS NOT TRIP LOW 1 22/11/2003 11:38:37 EN DIS NOT TRIP LOW 1 22/11/2003 11:38:37 EN DIS NOT TRIP HIGH 1 22/11/2003 12:03:34 EN DIS NOT TRIP HIGH 1 22/11/2003 12:03:34 EN DIS NOT TRIP LOW 1 22/11/2003 12:05:30 EN DIS NOT TRIP HIGH 1 22/11/2003 12:20:51 EN DIS NOT TRIP LOW 3 22/11/2003 12:20:51 DIS DIS NOT TRIP LOW 3 12:03:01 Reset long period timer timeout enable latch (toggled) (time difference due to hail delay) 22/11/2003 12:22:46 DIS DIS NOT TRIP HIGH 3 22/11/2003 13:58:47 DIS DIS NOT TRIP HIGH 7 13:59:59 Set long period timer timeout enable latch (toggled) 22/11/2003 13:58:47 DIS DIS NOT TRIP LOW 7 22/11/2003 14:02:37 DIS DIS NOT TRIP HIGH 7 22/11/2003 14:04:32 DIS DIS NOT TRIP LOW 7 14:03:00 Reset long period timer timeout enable latch (toggled) 22/11/2003 14:04:33 DIS DIS NOT TRIP HIGH 7 22/11/2003 14:06:27 DIS DIS NOT TRIP LOW 7 22/11/2003 14:17:59 DIS DIS NOT TRIP HIGH 9 22/11/2003 14:18:00 DIS DIS NOT TRIP LOW 9 22/11/2003 14:18:00 EN DIS NOT TRIP LOW 9 22/11/2003 14:21:50 DIS DIS NOT TRIP LOW 9 22/11/2003 14:21:50 DIS DIS NOT TRIP HIGH 9 22/11/2003 14:29:30 DIS DIS NOT TRIP LOW 1 14:11:09 Perform Activity of a Function 1,5 - Preset Timer, value 1FF0 22/11/2003 14:29:30 DIS DIS NOT TRIP LOW 3 22/11/2003 14:41:01 EN DIS NOT TRIP HIGH 3FD 14:22:29 Perform Activity of a Function 1,6 – Start Timer Tick 22/11/2003 14:42:56 EN DIS NOT TRIP LOW 3FD 22/11/2003 15:17:30 EN DIS NOT TRIP LOW 3FF Timer trip test 22/11/2003 15:19:25 EN DIS TRIP HIGH 0 Roll-over and trip as expected 22/11/2003 15:25:11 DIS DIS NOT TRIP HIGH 0 15:05:59 Reset long period timer timeout enable latch (toggled) 22/11/2003 15:25:11 DIS DIS NOT TRIP LOW 0 22/11/2003 16:36:13 DIS DIS NOT TRIP HIGH 4 16:20:01 Reset short period timer timeout enable latch (toggled) 22/11/2003 16:38:08 DIS DIS NOT TRIP LOW 4 16:20:02 Reset long period timer timeout enable latch (toggled) 22/11/2003 16:38:08 DIS DIS NOT TRIP HIGH 4 16:22:05 OFF 17/12/2003 06:34:14 DIS DIS NOT TRIP LOW 346 06:34:00 ON 17/12/2003 06:34:14 DIS DIS NOT TRIP LOW 0 17/12/2003 06:36:09 DIS DIS NOT TRIP LOW 346 17/12/2003 06:36:09 DIS DIS NOT TRIP LOW 0 06:35:06 Perform Activity of a Function 1,5 - Preset Timer, value 0E71 (final value) 17/12/2003 06:36:09 DIS DIS NOT TRIP HIGH 1 06:35:06 Perform Activity of a Function 1,6 – Start Timer Tick 181
Beagle-2 Mission Report ANNEXES
Beagle Telemetry Parameters Long Latch Short Latch Trip Tick Count Command Time TIME ME0009 ME0008 ME0007 ME0006 ME0005 Commanded Activity / Observations (from History) 17/12/2003 07:08:48 EN DIS NOT TRIP HIGH 1 06:35:07 Reset long period timer timeout enable latch (toggled) 17/12/2003 07:08:48 DIS DIS NOT TRIP HIGH 1 07:07:24 Perform Activity of a Function 1,6 – Start Timer Tick 17/12/2003 07:08:48 DIS DIS NOT TRIP LOW 1 07:07:26 Reset long period timer timeout enable latch (toggled) 17/12/2003 11:29:55 DIS DIS NOT TRIP HIGH 10 17/12/2003 11:31:51 DIS DIS NOT TRIP LOW 10 11:31:59 Reset short period timer timeout enable latch (toggled) 17/12/2003 11:33:46 DIS DIS NOT TRIP HIGH 10 11:32:10 Reset long period timer timeout enable latch (toggled) 17/12/2003 11:33:46 DIS DIS NOT TRIP HIGH 0 17/12/2003 11:33:46 DIS DIS NOT TRIP LOW 0 17/12/2003 11:35:41 DIS DIS NOT TRIP HIGH 0 17/12/2003 11:35:41 DIS DIS NOT TRIP LOW 0 11:36:10 Perform Activity of a Function 1,5 - Preset Timer, value 0E71 (final value) 17/12/2003 13:52:00 EN DIS NOT TRIP HIGH 1CD 13:50:21 Perform Activity of a Function 1,6 – Start Timer Tick 17/12/2003 13:52:00 DIS DIS NOT TRIP HIGH 1CD 13:50:23 Reset long period timer timeout enable latch (toggled) 17/12/2003 13:52:00 DIS DIS NOT TRIP LOW 1CD 17/12/2003 14:47:41 DIS DIS NOT TRIP HIGH 1D1 14:50:01 Reset short period timer timeout enable latch (toggled) 17/12/2003 14:49:37 DIS DIS NOT TRIP LOW 1D1 14:50:12 OFF 18/12/2003 06:34:00 DIS DIS NOT TRIP LOW 202 06:33:47 ON 18/12/2003 06:34:00 DIS DIS NOT TRIP LOW 0 06:35:01 Reset short period timer timeout enable latch (toggled) 18/12/2003 06:35:55 DIS DIS NOT TRIP LOW 202 18/12/2003 08:59:56 DIS DIS NOT TRIP LOW 20A 09:00:00 Reset short period timer timeout enable latch (toggled) 18/12/2003 09:01:51 EN DIS NOT TRIP LOW 20A 09:00:10 Set long period timer timeout enable latch (toggled) 18/12/2003 09:01:51 EN DIS NOT TRIP HIGH 20A 18/12/2003 10:37:51 EN DIS NOT TRIP LOW 20F 10:40:00 Reset short period timer timeout enable latch (toggled) 18/12/2003 10:39:46 EN DIS NOT TRIP HIGH 20F 10:40:12 OFF
182
Beagle-2 Mission Report Annexes
G Landing Site Hazard Analysis
Potential landing hazards in recent MOC narrow angle image and THEMIS images
January 29th 2003
John Bridges, Dave Rothery PSSRI & Dept. of Earth Sciences, Open University
This report summarises our investigations into the currently available THEMIS and MOC imagery of the Beagle 2 landing site. We use a MOC narrow angle image of ~2-4 m/pixel and THEMIS daytime IR images of ~19m/pixel and ~100m/pixel. The 19m/pixel image covers all of the post-ejection ellipse, the 100m/pixel THEMIS image the central part of it. In addition, for comparison, we have conducted some hazard counts using the NASA Spirit descent image over Gusev Crater. The figures show the location of the images and the post-ejection landing ellipse (centre 11.53oN, 90.50oE), calculated by ESOC. The hazards are identified on the images.
We have also summarised the expected slope calculations etc. in section 4, prior to launch.
G.1 Counting coverage of hazards using recent THEMIS IR daytime 100m/pixel,19m/pixel These two images have been provided over the assumed post-ejection landing ellipse, with daytime imaging at resolutions ~100m/pixel and ~19m/pixel. Table G.1 contains counts for positive topographic features, recognizable craters and ejecta around the largest crater (1.6 km diameter) in the area. The MOC narrow angle image (total approximate width 3 km, length 9 km including area outside the landing ellipse) overlaps the northern and southern sides of the landing ellipse.
Area % Recognizable Craters (green) 2.1 Positive topographic features (red) 6.0 Ejecta around Largest Crater (blue) 2.6
Table G.1 - Area Coverage of Potential Hazards in post-ejection (ESOC) Isidis Landing Ellipse based on ~100m/pixel THEMIS image (Figure G.1)
183
Beagle-2 Mission Report Annexes
Figure G.1(a) - THEMIS IR daylight mosaic around landing site 0.5 squares = 30 km
Figure G.1(b) - THEMIS Mosaic Hazards Craters (green) and topographic features (red), ejecta blanket. (blue). White rectangle is area of 19m/pixel image. Centre of post-ejection landing ellipse is 11.53oN, 90.50oN, 7.6 km x 57 km
184
Beagle-2 Mission Report Annexes
G.2 Counting coverage of hazards using recent THEMIS IR daytime ~19m/pixel The ~19m/pixel THEMIS IR visible image includes the full N-S extent of Beagle 2’s landing ellipse but not the complete E-W extent.
Area % Positive topographic features (red) 7.4 Deepest Craters (yellow) 1.3 Craters (green) 1.5 Shallow craters (blue) 2.2
Table G.2 - Area Coverage of Potential Hazards in post-ejection (ESOC) Isidis Landing Ellipse based on ~19m/pixel THEMIS image (Figure G.2)
Figure G.2 - 19m/pixel THEMIS image. Red topographic features, yellow deepest craters, green other craters, blue shallow/degraded craters. THEMIS image 20031224a.
The approximate lower size limit of craters counted is 10 pixels diameter, i.e. 190m diameter. The largest crater is 38 pixels diameter = 700 m. In addition there are numerous smaller craters mostly of types identified too small to capture individually and also small scale positive topographic features contributing to a general roughness across an estimated 80% of the remainder of the terrain. This means that only about 10% of this total rectangle can be regarded as ‘smooth’ at this resolution although whether this constitutes a hazard is not clear.
185
Beagle-2 Mission Report Annexes
G.3 Counting coverage of hazards using MOC image R13-00632 resolution ~2-4 m/pixel This MOC image has been provided over the assumed landing ellipse (centre 11.53oN, 90.50oE), calculated post-ejection from Mars Express by ESOC.
Area % Deep Craters 0 Positive topographic features (red) 2.5 Other recognizable craters (green) 4.2
Table G.3 - Area Coverage of Potential Hazards in Isidis Landing Ellipse based on ~2-4 m/pixel MOC
Figure G.3 - MOC image R13-00632 Red topographic features, Green recognizable craters
186
Beagle-2 Mission Report Annexes
G.4 Updated Analysis of Merged Images Following landing, hazard analysis was performed upon new images provided by Malin Space Science Systems. The revised analysis was performed for a merged image of 3 strip images taken within the landing ellipse.
Area % Deep Craters 3.34 Positive topographic features (red) 4.85 Other recognizable craters (green) 2.77
Table G.4 - Merged Image Potential Hazards
Figure G.4 - Merged image provided by Malin Space Science Systems
187
Beagle-2 Mission Report Annexes
G.5 Summary of previous remote data on Isidis landing site: MOLA, radar, rock abundance from thermal inertia model • Mean rock abundance 11% (= area coverage of ground). Localised concentrations around or within fresh (=deep?) impact craters may be higher e.g. 17% whereas other areas will be lower e.g. 2%. 15% rock abundance is equivalent to approx. 5% of the ground surface being covered by rocks >0.2m height. • Radar data (10s cm wavelength and scale) rms slope 2.4 ± 1o. • MOLA slopes (approx 100m E-W scale, less N-S) rms =0.57o, median = 0.3o, Bridges et al. (2003) J. of Geophysical Res., 108, E1-16.
G.6 Gusev MER Landing Site Coverage (Descent image ‘MGSnew-location-tight-arrow-38’)
Field of view ~2km?
Table G.5 - Area Coverage of Potential Hazards in Gusev
Area % Deep Craters 0.0 Positive topographic features 3.0 Other recognizable craters 2.1
188
Beagle-2 Mission Report Annexes
H Atmosphere and Aerodynamic Implications on Entry, Descent and Landing of Beagle 2
Jim Clemmet, (formerly Beagle2 Chief Engineer) Arthur Smith, (Fluid Gravity Engineering Ltd.)
21 May 2004
ADAPTED FROM ABOVE REPORT, ISSUE 1 DRAFT 5
H.1 Introduction This report investigates the role the Martian atmosphere and the performance of the entry system aerodynamics may have had in contributing to the failure of Beagle2.
After the arrival of Mars Express in orbit around Mars in January 2004, the SPICAM instrument onboard Mars Express, took a direct measurement of the density of the carbon dioxide content in the atmosphere as a function of altitude. Figures H.1 and H.2 show how this data compares with nominal atmospheres at Isidis, the Beagle2 target landing site, and more importantly at the latitude and time of the SPICAM measurements.
Unfortunately the SPICAM data does not extend below an altitude of 30km above the Martian datum. Even at this altitude the error band in the data is quite significant. This report investigates the potential implications of this new data from the Mars Express instrument. To aid this, a simple excel spreadsheet model has been developed allowing a quick appreciation of different scenarios. The spreadsheet has been correlated with the fort 10 output of the FGE TRAJ3 code. The results of this correlation exercise are presented in Appendix A (presented here in section H.6).
The SPICAM data is taken at latitude 17.18 North, longitude 267.86 East. The surface at this location is nominally 2km above the datum. This compares with the Beagle2 target landing site in Isidis Planitia at 11.6N 90.7E , the opposite side of the planet, and 3.3km below the datum.
As a part of this correlation exercise, different projections from the SPICAM data down to the surface are investigated, without correction for the different latitudes. A nominal latitude correction is then introduced into the model and the implications assessed.
The margins in the design of the Aeroshell thermal protection system (TPS) are also reviewed and the implications of lower drag coefficients are investigated. Lastly, an assessment is made of the effect of possible ingress of hot plasma into the internal volume, due to a suggested local failure of the Aeroshell coincident with the illumination anomaly seen in the image following separation from Mars Express.
189 Beagle-2 Mission Report Annexes
H.2 Summary
H.2.1 Targeting using the final Mars Express state and covariance matrix We found during the approach to Mars and after final release that the landing ellipse continually reduced from the pre-launch estimates. Based upon the data provided by ESOC, we conclude that Beagle2 was perfectly targeted by Mars Express with only very small nominal error in entry parameters and landing location. Any future European Mars mission can now take advantage of this knowledge of the accuracy achievable during nominal mission design. This is particularly important for landing site selection. However the margins built in to Beagle2 because of the initial uncertainties proved to be useful as we shall see in the following sections.
H.2.2 Atmosphere variability study to assess the effects of possible anomalous density profiles following localised dust storms some time prior to the Beagle2 entry During Beagle 2 design phase a Monte Carlo analysis of entry, vehicle and atmosphere parameters utilised an array of 100 atmospheres, selected by Oxford University Department of Atmospheric Physics to represent the small and large scale variability on the landing day. There has been some speculation re- enforced by Mars Express SPICAM data that dust in the atmosphere caused particularly low density and thus a fatally low deployment sequence. Modelling has shown that Beagle2 is robust to all reasonable variations in the atmosphere due to dust using the available models. There have been various comments from NASA staff on entry and descent performance of the MER landers, Spirit and Opportunity. The US reconstructed atmosphere was within their pre-flight expectations but the late deployment of the parachute in both cases remains unexplained.
Figure H.1 compares the SPICAM data with a selection of atmospheres used for Beagle2 entry above Isidis.
0.001
Default Dust (0100) 0.0009 Default Dust (1400) Low Dust (0100) 0.0008 Low Dust (1400) Viking Dust (0100) Viking Dust (1400) 0.0007 SPICAM
0.0006
0.0005
Density (kg/m3) 0.0004
0.0003
0.0002
0.0001
0 30000 40000 50000 60000 70000 80000 90000 Altitude (m)
Figure H.1 - Comparison of SPICAM atmosphere density profile with various Isidis density profiles
The European Mars Climate Database (EMCD) has been run at the SPICAM data latitude/longitude for the default MGS dust scenario (1400). Figure H.2 compares this with the SPICAM data. The SPICAM data has large error bars, increasing as the altitude reduces, but taking the apparent trend this implies that the atmosphere at upper latitudes may be denser than predicted by the model but at mid-altitudes the density may be lower.
190 Beagle-2 Mission Report Annexes
This phenomenon has been used in predicting what may have happened to Beagle2 if applied to Isidis. It is possible that delays in triggering the pilot parachute deployment could lead to insufficient time for airbag inflation or to more severe impact velocity, either resulting in loss of mission. Perhaps too speculative, but this may also give an explanation for the late parachute deployment of the MER landers.
1.0E-03
1.0E-04 spicam
1.0E-05 MGS DUST MCD
1.0E-06
density (kg/m3) density 1.0E-07
1.0E-08
1.0E-09 0 20000 40000 60000 80000 100000 120000 140000 altitude (m)
Derived from EMCD data courtesy Stephen Lewis, Oxford University.
Figure H.2 - Comparison of SPICAM atmosphere with EMCD density profiles at 17.2N 268E
H.2.3 A sensitivity study on system drag coefficients to find the lower limits for system operations Here we chose to put all system drag coefficients to the lowest feasible values to determine at what level a system failure would occur. Failure could be in the parachute deployment loads or insufficient time for RAT operations. No errors were placed on the accelerometer measurement systems. Beagle2 was found to be robust to the minimum drag assumptions. Only when combined with a lower atmospheric density could a failure be induced. The lowered density required is outside the model atmosphere limits, but within the SPICAM error bars. We conclude that it may be possible that a combination of very low drag coefficients particularly for the parachute system, and very low density atmosphere could account for a failure.
191 Beagle-2 Mission Report Annexes
H.2.4 A worst case aerothermal study, inducing early boundary layer transition and excess heat flux margins A re-appraisal of the materials data supplied to FGE by MBA/ EADS of arc heater test results has been performed. Since FGE was not responsible for TPS system sizing or margins, the initial TPS analysis objective was to provide a quick check of the sizing to enable rapid assessment of the effects of trajectory or mass change and to allow an estimate of dust erosion. In the re-appraisal we checked our analysis and produced a slightly better fit to the experimental data, in fact this led to an assessment of slightly larger TPS margin than had been used in trajectory and mass change assessments during design. For the aerothermal analysis the final nominal trajectory was designed to allow a good probability of laminar flow with the increased mass of the probe over the initial design goal. Subsequently an increase in aeroshell diameter partially offset the mass increase, but the revised nominal entry angle was retained to give a large margin for descent and landing. Given the revised TPS parameters, a highly pessimistic early turbulent transition plus an additional safety margin on fluxes produced pyrolysis depth and bondline temperatures which were acceptable. We therefore conclude that the TPS design, if implemented correctly, was not the causes of the failure, even in the additional ‘gap’ tile region where some localised thinning of tiles was present.
H.2.5 An analysis of a localised TPS removal on the conical rear cover coinciding with the observed position of the optical anomaly at release from Mars Express It has been speculated that a rear cover TPS tile section over the ARM access ‘hole’ was separated by a gassing event of some kind during ascent or cruise. An analysis using entry surface heat fluxes in that region shows that the carbon fibre substructure would exceed its design temperature by a considerable margin. Hot plasma ingress and flow to or from the HEPA filter could not be ruled out since pressure gradients from the side to rear cover are present. We conclude that if the anomaly was indeed a tile separation, then a failure of structure or internal components was likely.
192 Beagle-2 Mission Report Annexes
H.2.6 Summary Conclusions
Overall it is concluded
• that Beagle2 was well targeted
• that the entry system is robust to reasonable variations in atmosphere but an anomalous atmosphere as may be hypothesized from SPICAM data could lead to mission loss
• that reduced aerodynamic drag coefficients are by themselves unlikely to lead to failure
• that gas-dynamic uncertainties in the absence of damage will not cause loss of mission.
The effects of accelerometer errors or dust erosion have not been reassessed, the former being part of the software system analysis and the second since no further data is available from EADS on erosion performance since our original analysis assumptions.
193 Beagle-2 Mission Report Annexes
H.3 Detailed Assessment
H.3.1 Trajectory, Targeting and Entry Parameters The target landing site co-ordinates are 11.6 N and 90.74 E. In the four figures below the landing ellipses are shown computed from 1000 Monte Carlo simulations from the entry interface state and covariance supplied by ESOC (Mars Express) assuming normal statistics. (The February 03 estimate shows the original landing site co-ordinates as 90.5E prior to a co-ordinates system change). Just before separation the nominal landing site is still on the target. Post separation after separation dynamics were taken into account by ESOC there is a small drift in the landing site nominal to 11.53N and 90.53E. Overall we can see that the downrange errors have halved with about one third reduction coming from the separation knowledge, whilst the cross range halved during the cruise and halved again with separation knowledge.
Later we undertook a Monte Carlo with 85000 entry simulations to allow an analytic fit to the landing ellipse without assuming a normal distribution to aid in the search for Beagle2. The estimates of input distribution shape and variances were a mix of subjective, worst case (rectangular), best guess and detailed estimates. Mars Express covariance interpretation assumes normality in our MC code. As such extrapolation to the tails of the final distribution is not to be taken too seriously unless the input data are also revised. As such we never considered to go beyond ~1000 solutions in our Monte Carlo code. The entry parameter requirements agreed in the interface document between Beagle2 and Mars Express and those achieved are shown in Table H.1. We can conclude that Mars Express and the spin eject mechanism performed an excellent delivery of Beagle2.
Requirement Achieved nominal 3 sigma +/- nominal 3 sigma +/- Entry angle -16.5 1.0 (nom) -16.6 0.14 Angle of attack at 100.5km 0.0 3.0 ~0.05 ~0.6 Table H.1 - Entry Parameters
landing estimate February 03 landing estimate 18/12/03 12.5 12.5
12 12
11.5 11.5 lattitude lattitude
11 11
10.5 10.5 89.5 90 90.5 91 91.5 92 89.5 90 90.5 91 91.5 92 longitude longitude
landing estimate 29/11/03 landing estimate 24/12/03 12.5 12.5
12 12
11.5 11.5 lattitude lattitude
11 11
10.5 10.5 89.5 90 90.5 91 91.5 92 89.5 90 90.5 91 91.5 92 longitude longitude
Figure H.3 - Landing Dispersion Evolution
194 Beagle-2 Mission Report Annexes
H.3.2 Atmosphere Density Profile Variability Study
H.3.2.1 Initial Assessment using FGE TRAJ3 Code
One of the uncertainties for the Martian entry is the atmosphere model. For the nominal Beagle2 ‘best guess’ atmosphere Oxford University Atmospheric Physics selected the dust scenario 1 ‘MGS dust’ from the EMCD V3.1. Unfortunately some weeks prior to the arrival of Beagle2 a series of localised dust storms were observed on the planets surface. This activity had ceased well before Christmas Eve, but some residual effects may have been present in the atmosphere which would affect performance. The starting point of this analysis is our confidence in the entry phase performance since the nominal entry angle was achieved giving considerable margins in the nominal scenario. We therefore examined the effects of differing atmosphere assumptions on the system performance.
The first step was to define the failure thresholds due to atmosphere density reduction and determine whether these are realistic. By simply increasing the temperature and reducing the density we maintain the pressure profile. This is very crude, but gives a rough 1st indication. We find that the onset of problems for the Radar Altimeter trigger at 280m (RAT) timing are at around –15% of density throughout the trajectory. This was compared to reports of +2.5% temperature at Gusev from the US. Also reported was that JPL use 21% error bars on their atmosphere profiles which is outside our design criteria.
During the study, measurements from Mars Express SPICAM indicated a very low density at the lower limit of measurement capability which has large error bars. To undertake a worst case scenario, the SPICAM mean data was blended back to the nominal low dust atmosphere at several altitudes. SPICAM data was taken at 15:17 UTC on 13th January 2004 at 17.18 North, longitude 267.86E (aerocentric). Local True Solar Time is estimated at ~01h at measurement site. (Zterrain = +2437m, Zaeroid=+1697m, Zterrain for Beagle2 landing site is –3367m, see Beagle2/ESOC ICD for definition). Figure H.4 shows an approximate location for this measurement in Tharsis (MOLA scaled altitudes). The measurement is close to the mountains. We have not considered whether there are large scale disturbance effects.
Figure H.4 - Approximate location of the SPICAM measurements
195 Beagle-2 Mission Report Annexes
The SPICAM data first was used at the Beagle 2 landing site with no adjustments for altitude or local time of day since these were not known precisely at the time of the analysis, but the density was factored for the mole fraction of CO2 to give a total density estimate. (since SPICAM only measures CO2 density).
Variation of Density w ith Altitude for 3 dust scenarios
1.0E+00 MGS dust Scenario 1.0E-01 Viking dust scenario 1.0E-02 Low Dust Scenario SPICAM 1.0E-03 15km blended 1.0E-04
1.0E-05 Density (kg/m3) 1.0E-06
1.0E-07
1.0E-08 0 20000 40000 60000 80000 100000 120000 Height above Surface (m)
Figure H.5 - SPICAM density profile blended to low dust scenario at 15km compared to EMCD mean profiles at landing site
For the 20km and 15km blended atmosphere parachute deployment sequences are lower in altitude but seem to be fine in terms of available time and parachute loads. Repeating the 20km blend with a 5% drag reduction on all elements also produces an acceptable sequence. The hypersonic deceleration occurs deeper into the atmosphere with a peak of just over 20g, and fluxes are also high with over 90W/cm2. These values are above the maximum given to MBA/EADS for design purposes, but not alarmingly, so we would expect the TPS to survive.
Variation of Density w ith Altitude for several dust scenarios
1.0E-01 MGS dust Scenario
1.0E-02 Viking dust scenario Low Dust Scenario 1.0E-03 SPICAM blended at 0 km 1.0E-04
1.0E-05
Density (kg/m3) 1.0E-06
1.0E-07
1.0E-08 0 20000 40000 60000 80000 100000 120000 Height above Surface (m)
Figure H.6 - SPICAM density profile blended to low dust at 0km compared to EMCD mean profiles at landing site
196 Beagle-2 Mission Report Annexes
Moving the blend point produces a failure somewhere between 0 and 15km (Table H.2), this is not surprising as we know from our initial analysis that an overall reduction of ~15% in density is a problem.
Blend point (km) 20 15 0 Main sequence initiation time (s) 184 183 Main fails to open RAT nominal operations time (s) 241 226 Main fails to open (280m above terrain)
Table H.2 - SPICAM blended atmosphere, RAT timings
As a reminder, the FGE modelling was primarily set up for entry simulation and so atmosphere definition is in 1km steps. For the terminal descent phase, the atmosphere definition should ideally be in finer steps, and so above results are considered approximate.
Later analysis carried out by Oxford shows that when compared to the EMCD at the correct latitude, longitude (i.e. altitude) and date, that the differences from SPICAM to EMCD are much smaller than we assumed initially above (Figure H.7).
Default Dust (0100) 1.0E-03 Default Dust (1400) Low Dust (0100) Low Dust (1400) Viking Dust (0100) 1.0E-04 Viking Dust (1400) Spicam+Error
Density (kg/m3) 1.0E-05
1.0E-06 30000 40000 50000 60000 70000 80000 90000 Altitude (m)
Figure H.7 - SPICAM density profile compared to EMCD mean profiles (at date and location of SPICAM measurements, Stephen Lewis, Oxford University)
197 Beagle-2 Mission Report Annexes
Finally, data from JPL generated from MGS-TES received after our analysis agrees very well with the Viking profile (Figure H.8), and as such the Viking profile was used in later analysis.
Variation of Density with Altitude for 4 dust scenarios
1.0E-01 MGS dust Scenario 1.0E-02 Viking dust scenario Low Dust Scenario 1.0E-03 MGS-TES
1.0E-04
1.0E-05 Density (kg/m3) 1.0E-06
1.0E-07
1.0E-08 0 20000 40000 60000 80000 100000 120000 Height above Surface (m)
Figure H.8 - MGS-TES atmsosphere at Beagle2 site on 25th December compared with EMCDv3.1 profles
198 Beagle-2 Mission Report Annexes
H.3.2.2 Excel model predictions with SPICAM data applied directly at Isidis
H.3.2.2.1 Spreadsheet Prediction with 10km Blend
acceleration comparison 5
0 -20000 0 20000 40000 60000 80000 100000 120000 140000
-5
FGE 20kmblend -10 "SPICAM"
-15 acceleration (m/s2) acceleration
-20
-25 altitude (m)
Figure H.9 - SPICAM data blended in to default MGS atmosphere at 10km above datum
The PDD event is predicted at an altitude of 717m above the Martian datum. With the Isidis landing site is approximately -3300m, it is unlikely that Beagle2 would survive if the SPICAM density profile blends in at 10km, even noting the pessimism of the spreadsheet model. The TRAJ3 predicted deceleration curve, with blend at 20km, is included to illustrate how the profile changes with blend point.
H.3.2.2.2 Spreadsheet Prediction with 5km Blend
acceleration comparison
5
0 -20000 0 20000 40000 60000 80000 100000 120000 140000
-5 FGE 20kmblend
"SPICAM 5km blend" -10
-15 acceleration (m/s2)
-20
-25 altitude (m)
Figure H.10 - SPICAM data blended in to default MGS atmosphere at 5km above datum
199 Beagle-2 Mission Report Annexes
PDD operation predicted by spreadsheet at just above the datum at 183m. Again the TRAJ3 predicted 20km blend deceleration curve is included to illustrate how the profile changes with blend point.
H.3.2.3 Excel model prediction with MGS/SPICAM revised density profile
Subsequent to the preceding analyses, the SPICAM instrument team have reprocessed the original data to an improved accuracy. Figure H.10a provides a visual comparison between the original and the reprocessed data sets. It can be seen that the oscillatory variation has been much reduced, resulting in a smoother curve. Importantly, the density in the middle atmosphere retains much the same magnitude.
All SPICAM, courtesy of Jean-Loup Bertaux, SPICAM Principal Investigator, has been received by the Beagle2 project with great appreciation.
SPICAM DATA 1.0E-03
1.0E-04 SPICAM - original SPICAM - revisited
1.0E-05
1.0E-06
1.0E-07 Density
1.0E-08
1.0E-09
1.0E-10
1.0E-11 0 20000 40000 60000 80000 100000 120000 140000 160000 altitude (m) Figure H.10a - SPICAM atmosphere profile, a comparison between the initial and the improved accuracy data sets
The density profile has then been revised by applying a ‘correction factor’ to the ISIDIS EMCD default MGS prediction with a blend back to MGS at 5km. The ‘ correction factor’ applied is the ratio of the density as presented by the updated SPICAM data set with that predicted by EMCD for the default MGS dust scenario at 17.2 N 268 E (refer to Figure H.10b, and compare with Figure H.2 using the initial SPICAM data set).
This SPICAM type atmosphere, together with a low dust scenario at ISIDIS and a Viking lander atmosphere is then normalised to the default MGS dust scenario. The results of this process are presented in Figure H.10c. The original SPICAM data set, ‘corrected’ to ISIDIS is also included. It is interesting to note that the SPICAM type atmospheres compares very favourably, with respect to profile, with that of the reconstructed atmospheric from the MER SPIRIT lander deceleration data. (The Beagle2 project has been given sight of this data but is not free to include the information in this report at this time). Clearly a SPICAM type atmosphere is viable and in many respects not so dissimilar to the Viking atmosphere. The MER team have also reported that the peak of the dust storm occurred on the 23rd December 2003, just two days before Beagle2’ entry and descent through the Martian atmosphere Spirit’s entry is some 10 days later and the SPICAM data is from Mars Express Orbit 17 in mid January 2004. It might be speculated that the atmosphere was near its worst from an entry viewpoint at around the time of the Beagle2 approach to Mars.
200 Beagle-2 Mission Report Annexes
1.0E-01 spicam factored MGS-revisited 1.0E-02 MGS DUST MCD
1.0E-03
compare with fig 2 in report 1.0E-04 3)
1.0E-05
density (kg/m 1.0E-06
1.0E-07
1.0E-08
1.0E-09 0 20000 40000 60000 80000 100000 120000 140000 altitude (m)
Figure H.10b - Comparison between improved SPICAM profile and predicted atmosphere at 17.2 N 268 E
140000 SPICAM factored MGS-revisited @ ISIDIS Viking
120000 Low Dust @ ISIDIS spicam original
100000
80000
Altitude 60000
40000
20000
0 0.0% 50.0% 100.0% 150.0% 200.0% density % normalised to default NGS profile
Figure H.10c - Various atmosphere profiles normalised to the default MGS atmosphere (SPICAM type atmospheres have been artificially blended back to the default at 5km altitude)
THE SPICAM derived atmosphere, based upon the improved accuracy data set has then been imported into the Excel model, and using a 5km blend point, entry and descent to the point of PDD initiation has been predicted (Figure H.10d).
201 Beagle-2 Mission Report Annexes
Peak g of 13.3 occurs at 19km, about 7km further into the atmosphere than for a nominal entry. PDD operation is predicted to occur 1778m, (about 1.5km lower than for a nominal entry) i.e. noting the pessimism in the excel model of about 500m, maybe just sufficient for RAT surface detection on the upper threshold. Uncertainty in landing site altitude itself is of the order of 200 to 300 metres.
These delays do not seem inconsistent with the MER experiences of late parachute deployment on both Spirit and Opportunity.
From an initial study of the velocity and deceleration profiles it would be tempting to state that the entry is not too different from the nominal prediction but the deployment of the drogue parachute in this scenario is significantly delayed and the system has little to no margin for success.
Sensitivity to blending a revised density profile, based upon SPICAM data, back into a nominal profile has been demonstrated in Appendix A. It would not be difficult to imagine an alternative blending scheme that either improves the margin or causes mission failure, with the RAT having no opportunity to trigger the airbag inflation.
density profiles 2.00E-02
1.80E-02
1.60E-02
1.40E-02
1.20E-02 SPICAM corrected MGS 1.00E-02 MGS
density spicam data
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00 0.0 20000.0 40000.0 60000.0 80000.0 100000.0 120000.0 140000.0 160000.0 altitude (m)
Figure H.10d Comparison of Density Profiles (5km blend point)
202 Beagle-2 Mission Report Annexes
Velocity comparison
6000
5000
4000 FGE 20km blend
SPICAM corrected MGS 3000 mgs nominal
velocity (m/s) velocity 2000
1000
0 -20000 0 20000 40000 60000 80000 100000 120000 140000 altitude (m)
Figure H10e Comparison of Velocity Profiles (5km blend point)
acceleration comparison
5
0 -20000 0 20000 40000 60000 80000 100000 120000 140000
-5 FGE 20kmblend(no correction) SPICAM corrected MGS default MGS -10
-15 acceleration (m/s2)
-20
-25 altitude (m)
Figure H10f Comparison of Deceleration Profiles (5km blend point)
203 Beagle-2 Mission Report Annexes
H.3.2.4 Summary
We can not absolutely rule out a low density at ground level, and we have to consider the error bars in the low altitude SPICAM data are very large. The analysis above, whilst not conclusive, suggests that the atmosphere alone could be the cause of failure in the entry and descent system with the RAT trigger having insufficient time to function. Airbag performance parameters have not considered, but it’s clear that reduction in parachute drag coefficient or atmosphere density or near surface winds increase impact speed.
H.3.3 Atmosphere Near-Ground Turbulence The data in Figure H.11 give measurements of near surface atmospheric temperature fluctuations seen at the Spirit landing site (Gusev).
The data appear to indicate a turbulent boundary layer. Whether it is driven thermally or by local roughness (fetch) needs to be investigated. The near surface structures in the colour map look quite large and long lived. Since Mars atmosphere density is lower than Earth, and wind speed lower (gradient wind speed on the UK is ~50m/s), flow Reynolds numbers will be much lower (~300 times lower). 'Open country' terrain roughness needs to be determined for the measurement sites. All of these influence the boundary layer profile. Without undertaking further analysis it’s difficult to judge, but with a lower Reynolds number the terrain looks pretty smooth. So it is probably 'natural' turbulence and since the boundary layer will be thicker than Earth and one would then expect mixing of the near surface thermal layer (solar warming or cooling) with the mean atmosphere to quite a height with large scale eddies present. It should be possible to correlate the mean gradient wind speed (i.e. at the boundary layer edge height) and the surface roughness with the turbulence intensity, turbulence length scale and gust speed and duration and frequency of occurrence . There is much measurement data on Earth used for building and pylon loads etc. (see BS8100 and CP3 + ESDU publications). To scale to Mars, it should be possible to use a turbulent boundary layer simulation, try and correlate with Earth data somewhere 'similar' then try and correlate the Mars data. Of course we have no measurements on Mars but the data above could be a place to start.
204 Beagle-2 Mission Report Annexes
Figure H.11 - Atmospheric temperature fluctuations above the MER Spirit landing site
(Courtesy NASA)
205 Beagle-2 Mission Report Annexes
H.3.4 Fluctuations in density with altitude It has been suggested that local fluctuations in atmospheric density, analogous with air pockets experienced when flying, could cause premature triggering of the pilot chute deployment. The Entry and Descent algorithm uses two software filters. To check for the PDD trigger threshold, the algorithm has first to be in the ‘falling’ mode, i.e. that the deceleration is over the peak value and is reducing. The slow filter is used for this check. Once confirmed that the sequence is in this phase, the fast filter is used to detect the trigger point. In both cases three consecutive reading over an elapsed period of 24ms have to meet the necessary criteria. The slow filter has a frequency equivalent to 0.7rads/s and the fast filter has 6.3rads/s (1Hz). Events occurring faster than 9s will not influence ‘falling’ confirmation and events faster than 1s will not cause false detection of the PDD trigger point. At the time of triggering the probe should be travelling with a velocity of about 330m/s and if ‘air pockets’ were to occur earlier the probe would be travelling even faster. ‘Air pockets’ of 330m depth or less will consequently be ignored. The group delay of the slow filter is quite large at 7.848s so falling confirmation relevant to trigger detection comes from events somewhat earlier, i.e. when conditions could be different.
The excel model has been used to predict what may occur if substantial air pockets were encountered (Figure H.12a-c).
Three ‘air pockets’ were incorporated into a single run. The upper altitude pockets do not cause premature triggering by themselves but do cause a further delay in the trigger, nominally a loss of a few hundred meters. The lower air pocket, inserted just 1km above the previously predicted trigger point and about 600m deep causes premature triggering of the PDD.
density profiles 2.0E-02
1.8E-02
1.6E-02
1.4E-02
1.2E-02 SPICAM corrected MGS
1.0E-02 airpocket density
8.0E-03
6.0E-03
4.0E-03
2.0E-03
1.0E-05 0.0 5000.0 10000.0 15000.0 20000.0 altitude (m)
Figure H.12a - Hypothetical atmospheric density fluctuations
206 Beagle-2 Mission Report Annexes
Velocity comparison 6000.0
5000.0
4000.0
SPICAM corrected MGS
3000.0 air pocket velocity (m/s) velocity 2000.0
1000.0
0.0 0.0 20000.0 40000.0 60000.0 80000.0 100000.0 120000.0 140000.0 altitude (m)
Figure H.12b - Effect of hypothetical atmospheric density fluctuations on velocity profile
acceleration comparison
2.000
0.000 0.0 20000.0 40000.0 60000.0 80000.0 100000.0 120000.0 140000.0
-2.000
-4.000 SPICAM corrected MGS
air pocket -6.000
-8.000
acceleration (m/s2) acceleration -10.000
-12.000
-14.000
-16.000 altitude (m)
Figure H.12c - Effect of hypothetical atmospheric density fluctuations on deceleration profile
Clearly such low density atmosphere pockets can be detrimental to the lower altitude entry performance. The consequences to pilot chute loading have not been assessed if premature deployment into a low density atmosphere were to occur, but it is worth noting that the effect on velocity is minimal. Dynamic pressure may therefore not be catastrophic. More precise analysis using the FGE TRAJ3 code is recommended but the true nature of such pockets needs first to be investigated and characterised.
207 Beagle-2 Mission Report Annexes
H.3.5 Sensitivity Analysis on System Drag Coefficients In the Monte Carlo analyses, drag errors of each element of the system (entry, drogue and main) are considered to be uncorrelated, thus to get a minimum in every phase should be a very rare event. In this analysis we assume a correlation coefficient of 1, i.e. we set all Drag coefficients to their minimum conceivable values. In fact these are below the 3 sigma limits used in the Monte Carlo simulations and close to implausible for the hypersonic entry. Table H.3 shows the results of several combinations of minimum Drag coefficient and atmosphere model. Columns in red (Cases 2 and 4) represent a RAT failure due to insufficient activation time being available after main deployment.
Monte Carlo Case number (-% of Cd) 3igma, 1 2 3 4 rectangular 100 variations by Nominal, Viking, Viking & Atmosphere Nominal Viking Oxford low dust nominal Hypersonic Entry 5 10 12 10 10 Supersonic / 5 10 12 20 20 Transonic 5 at M<=1, 10 at Drogue 10 20 10 20 M1.5 Main 4 10 20 20 20
Simulate late Ok for Comment trigger nominal
Table H.3 - Drag Reduction levels effect on RAT timing
Parachute inflation times were left at nominal values and g switch errors were assumed to be zero, parachute and RAT timing parameters are shown in Table H.4 for some combinations of atmosphere and error assumption.
Atmosphere Nominal Viking Nominal Nominal Viking Viking Viking Hypersonic Entry (-%Cd) 0 0 10 10 10 10 10 Supersonic / Transonic (-%Cd) 0 0 10 20 10 20 20 Drogue (-%Cd) 0 0 10 20 10 10 20 Main (-%Cd) 0 0 20 20 20 20 20 Main deployment initiation time (s) 189 191 192.5 191.5 194.7 193.8 193.9 Drogue dynamic pressure (Pa) 561 616 678 635 686 687 Main dynamic pressure (Pa) 67.5 67.8 75 83 75 75 83 RAT trigger Time (s) 314 299 239 231 225.6 228.3 216.3 Velocity at RAT trigger (m/s) 15.8 16.0 17.9 17.9 17.9 17.9 No trig
Table H.4 - Parachute and RAT parameter sensitivity with atmosphere and EDLS elements Cd
The above drag coefficient changes amount to the same change in atmospheric density since Drag = 0.5 density v2 Area Cd. We can expect therefore that a combination of density (say ~ –8%) and drag (say ~ - 8%) might lead to our initial criteria for overall failure i.e. –15% overall on density. Nevertheless we can conclude that such an overestimation in drag coefficient is unlikely from our 3 sigma evaluations on EDLS component drag, and that such a RAT failure scenario through lack of time appears unlikely.
208 Beagle-2 Mission Report Annexes
H.3.6 Thermal Protection System Performance The Beagle2 nominal trajectory was designed to allow a laminar flow boundary layer during peak heat transfer assuming a reasonable value of surface roughness for the granular cork Norcoat Liege Thermal Protection System (TPS) tiles. The TPS was not modified following mass increases from the initial specification of 60kg, which after the diameter increase gives a ballistic coefficient of 8% over the specification used for TPS design. To alleviate this, the nominal entry angle was reduced but the corridor remained within the original design envelope, thus some of the TPS margin allocated by MB/EADS was already used (estimated about 1%).
Methods for predicting the transition to turbulent flow in our trajectory code along the final nominal trajectory do not indicate transition. We noted in earlier reported analyses that a particle impact or char failure causing 1mm surface steps may cause transition, so we have examined the worst case turbulent transition, assuming a roughness of 2mm and pessimistic transition criteria of 80% of the nominal value. Further we have added an additional design margin of 20% to the fluxes. Margin policy was the responsibility of the TPS designer EADS and FGE supplied nominal fluxes (i.e. no added margin) for design purposes. Bond line temperatures remain well below the design criteria even assuming a warm start temperature of the structure.
In order to assess the reaction (pyrolysis onset) depth in the ablator a little better, we have re-examined the Norcoat Liege test data made available to us and refitted the material properties. It appears that the material performance is slightly better than our initial assumptions. For this worst case analysis the onset of pyrolysis penetrates deeper than would be desired ideally on the forward TPS (i.e. ~62% of TPS thickness c.f. 50% for a ‘nominal’ design), but still is acceptable. Cold wall flux levels at the expansion corner for the fully turbulent case with additional margin are about 130W/cm2, higher than the qualification level, but within the material capability (~200W/cm2). Beagle2 also had a much thicker expansion corner TPS tile due to the diameter increase, thus any increase in surface recession should not be a problem.
Table H.5 shows a comparison of forebody TPS sizing parameters. Design cases were 60kg with 0.9m diameter and all are evaluated on the nominal entry condition post flight excepting entry angle. Bracketed figures are the figures made at the time of the original analysis. As expected the flight nominal was within the original design parameters, whilst the worst case turbulent expansion corner is outside the qualified flux level by a small amount as discussed above. Using the Viking dust scenario which appears close to the MGS-TES brings the heat soak closer to the no margin design limit case.
Normalised ablator Entry angle Heat load Max flux thickness requirement degrees J/cm2 W/cm2 parameter 2508 65 Design (shallow) -15 1.00 (2513) (63) 2030 81 Design (steep) -21 0.84 (2041) (79) 2446 72 0.96 February 03 Nominal -16.5 (2455) (73) Post flight Nominal -16.6 2437 73 0.95 Post flight Viking -16.6 2646 72 0.98 atmosphere Post flight SPICAM blend 20km -16.6 2408 92 0.93 15km, 2436 85 0.93 0km 2518 81 0.94 Worst case roughness/ transition (expansion -16.6 3556 107 1.03 corner), nom atmosphere
Table H.5 - Heatshield sizing Parameters
209 Beagle-2 Mission Report Annexes
For the reduced density SPICAM atmospheres we notice a significant increase in flux, but still within the qualification levels if the boundary layer remains laminar, but since the heating pulse is shorter, there is no impact on the TPS thickness requirement.
For the rear cover there appears to be no problem even in the reduced local thickness regions over the extension tiles.
We conclude that the Beagle2 TPS is sufficiently robust as designed, even when applying much more pessimistic heating levels to the near nominal trajectory flown, or in the event of a reduced density atmosphere.
H.3.7 Rear Cover ‘Hole’ From the post separation photographs there appears to be an anomalous geometric feature on the MLI on the conic section. If this was a hole in the TPS around 1/3 of the way aft from the maximum diameter due to a TPS panel being lifted over the ARM access, we would see >500C on the CFRP structure during entry. This is based on the same aerothermal analysis as used for the rear cover and assumed the fluxes are applied directly to an exposed thin CFRP structure. This would cause a failure. Hot gas penetration and multiple systems demise could be assumed since there are pressure gradients (Figure H.13) between the conical rear cover and outside the base ring where the filter is located. It’s difficult to say which way the flow would go since angle of attack and spin rate cause base flow fluctuations and these tend not to be directly in phase with the motion. Of course the aerothermal analysis assumes a worst case level of flux, but even with this margin removed, gas plasma temperatures of over 2000 to 2500K are present and structure safe temperatures are still easily exceeded.
Figure H.13 - Static Pressure Field near Peak Heat Transfer during Entry
210 Beagle-2 Mission Report Annexes
H.4 Notes
1. Both MER landers reported a significant delay in parachute deployment. 2. NASA continues to claim that, from backward engineering, the atmospheres experienced by the MER lander were within their specification. Yet NASA has not published an explanation for the late parachute deployments. 3. The MER landers utilize a more complex EDLS system capable of recovering from late initiation; unfortunately the Beagle2 system is unable to respond to changed circumstances, the sequence running on time alone once the acceleration trigger threshold has been detected. 4. Analysis of the SPICAM data and comparing it to output from the EMCD at the same latitude suggests that the SPICAM data is viable although at the low bounds of model variations.
H.5 Conclusions
1. Entry trajectory is not seen as a likely cause of mission loss. 2. Taking the SPICAM density data as correct and applying it directly to Isidis, depending on how the profile is projected forward towards the surface, failure scenarios can be predicted Even if the SPICAM profile blends back into a nominal atmosphere profile as early as 10km above the datum (i.e. 13km above the landing site target) airbag inflation and could fail to occur in time, before impact. 3. Factoring the nominal MGS profile at Isidis by the ratio of ‘SPICAM to Nominal MGS’ at the SPICAM measurement latitude, then survival may be marginal if the profile recovers to a nominal atmosphere by an altitude of 5km. Blending in at below 5km, or applying a different blending scheme, would result in predicted mission loss. 4. Temperature and density fluctuation if of sufficient magnitude and at lower altitudes could lead to mission failure 5. The Thermal Protection System is sufficiently robust in its design to accommodate wide variations in entry 6. An over-estimation of drag coefficient of sufficient magnitude alone to cause mission loss is unlikely. But combined with lower than expected atmosphere density, low Cd could result in failure through late or no triggering of the airbag inflation or high impact velocity. 7. The uncertainty in the atmosphere characteristics could be real and if so could have caused the loss of the Beagle 2 mission.
211 Beagle-2 Mission Report Annexes
H.6 Appendix A to Annex H
H.6.1 Correlation of Excel Model with TRAJ3 Fort 10 Outputs
Correlations have been conducted between the Excel spreadsheet models for the following cases:
• Nominal entry into an ‘MGS atmosphere’ • Entry into a SPICAM atmosphere with the following blend points (i.e. the SPICAM data is bended back into the nominal MGS atmosphere at a defined altitude, below that altitude the atmosphere is defined by the MGS case):
20km blend point 15km blend point 0km blend point
Additional runs of the Excel model have been made with 10km and 5km blend points. No TRAJ3 fort 10 data is available for comparison for these cases.
The following table summarises the results:
Peak deceleration Altitude at PDD Case Altitude at peak g g 0.77g trigger TRAJ3 Excel TRAJ3 Excel TRAJ3 Excel Nominal 12.3 12.9 26127 25000 3154 2950 20km blend (1) 19.9 20.9 20371 20000 2100 1933 15km blend (2) 20.2 21.6 18662 17000 1794 1321 10km blend (2) - 19.6 - 17000 - 717 5km blend (2) - 18.6 - 15000 - 183 0km blend (2) - 17.6 - 15000 - -1102 0km blend (1) 16.3 17.6 17541 18000 -1006 -316
(1) blend profile common between TAJ3 and Excel model (2) uses a linear interpolation on a log density v altitude scale
Table H.6 - Summary of Correlation Results
As can be seem the excel model is generally 300 to 700m pessimistic with regard to the PDD trigger altitude. This is probably due to some inaccuracy in the ballistic coefficients used and in the coarse 1km step sizes used.
212 Beagle-2 Mission Report Annexes
H.6.2 Correlation of Nominal Entry Case The following 3 plots compare the output from an excel spreadsheet model [produced by Jim Clemmet] with the output from the formal aerodynamic model [produced by FGE]. This initial correlation case is with the MGS density profile. It can be seen that good correlation is achieved for both acceleration and velocity. The density profile is also illustrated.
acceleration comparison comparison Velocity comparison 6000.0 2.0
0.0 5000.0 0.00E+00 2.00E+04 4.00E+04 6.00E+04 8.00E+04 1.00E+05 1.20E+05 1.40E+05
-2.0
) 4000.0 -4.0
3000.0 -6.0 nominal entry nominal entry mgs nominal -8.0 2000.0 (m/s) velocity
acceleration (m/s2 acceleration mgs nominal -10.0
1000.0 -12.0
-14.0 0.0 altitude (m) 0.00E+00 2.00E+04 4.00E+04 6.00E+04altitude 8.00E+04(m) 1.00E+05 1.20E+05 1.40E+05
density profiles 2.00E-02
1.80E-02
1.60E-02
1.40E-02
1.20E-02 noinal entry MGS 1.00E-02 spicam data density
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00 -2.00E+0 0.00E+002.00E+044.00E+046.00E+048.00E+041.00E+051.20E+051.40E+051.60E+05 4 altitude (m)
Figure H.14 - Correlation of Nominal Entry Case
213 Beagle-2 Mission Report Annexes
H.6.3 Correlation of Excel Spreadsheet with FGE Fort 10 – 20km Blend This correlation case is with the SPICAM density data blended into the MGS density profile at 20km altitude above the datum. The spreadsheet uses the same density profile as the Fort 10 output.
acceleration comparison comparison Velocity comparison 6000 5
5000 0 -20000 0 20000 40000 60000 80000 100000 120000 140000
) 4000 -5
3000 -10 FGE 20kmblend FGE 20km blend "SPICAM"
velocity (m/s) 2000 -15 "SPICAM" acceleration (m/s2
1000 -20
-25 0 -20000 0 20000 40000 60000 80000 100000 120000 140000 altitude (m) altitude (m)
density profiles 2.00E-02
1.80E-02
1.60E-02
1.40E-02
1.20E-02 SPICAM+projection20 MGS 1.00E-02 spicam data density
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00 0.0 20000.0 40000.0 60000.0 80000.0 100000.0 120000.0 140000.0 160000.0 altitude (m)
Figure H.15 - Correlation of 20km Blend Case
214 Beagle-2 Mission Report Annexes
H.6.4 Correlation with 0km Blend With this much later blend at the datum of the SPICAM density profile into the MGS profile, the correlation at first sight seems not to be quite so good. But in fact it is because the blend profiles are not the same. Which is the correct method of blending is debatable but this does show the sensitivity. Whilst for both cases, Beagle 2 would not survive, the FGE model predicts the PDD to operate at 143.4 s after TOA at an altitude of -1006.2km. The spreadsheet model predicts the PDD at less than 134s and more importantly below -2000km. The spreadsheet is therefore a little more pessimistic with regard to the fate of Beagle2 than the FGE analysis.
acceleration comparison comparison Velocity comparison
6000 2
0 -20000 0 20000 40000 60000 80000 100000 120000 140000 5000 -2
-4 4000 )
-6
-8 FGE 0kmblend 3000 FGE 0km blend -10 "SPICAM"
-12 velocity (m/s) 2000 "SPICAM" acceleration (m/s2 -14
1000 -16
-18
0 -20 -20000 0 20000 40000 60000 80000 100000 120000 140000 altitude (m) altitude (m)
density profiles 2.00E-02
1.80E-02
1.60E-02
1.40E-02
1.20E-02 "SPICAM+projection0" MGS spicam data 1.00E-02 FE blend data density
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00 0.0 20000.0 40000.0 60000.0 80000.0 100000.0 120000.0 140000.0 160000.0 altitude (m)
Figure H.16 - Correlation of 0km Blend
A comparison of the two density profiles is quite subtle and is predominantly at the knee of the curve. Utilising the FGE 0km blend profile in the spreadsheet model gives much improved correlation.
215 Beagle-2 Mission Report Annexes
density profiles 2.00E-02
1.80E-02
1.60E-02
1.40E-02
"SPICAM+projection0" 1.20E-02 MGS spicam data 1.00E-02 FE blend data density
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00 -20000.0 0.0 20000.0 40000.0 60000.0 80000.0 100000.0 120000.0 140000.0 160000.0 altitude (m)
acceleration comparison comparison
2
0 -20000 0 20000 40000 60000 80000 100000 120000 140000 -2
-4 )
-6 FGE 0kmblend -8 "SPICAM" -10
-12 acceleration (m/s2 -14
-16
-18
-20 altitude (m)
Velocity comparison
6000
5000
4000
3000 FGE 0km blend
"SPICAM" 2000 velocity (m/s) velocity
1000
0 -20000 0 20000 40000 60000 80000 100000 120000 140000 altitude (m)
Figure H.17 - Improved Correlation
216 Beagle-2 Mission Report Annexes
H.6.5 Correlation with 15km Blend Correlation is again good. The following table presents data for key events (the spreadsheet does not include post PDD modelling):
FGE TRAJ3 EXCEL SPREADSHEET Time Altitude Velocity Time Altitude Velocity s m m/s s m m/s PDD 138.6 1793.6 312 122.51321 309 ARM 182.7 -2085.484 RAT ENABLED 211.7 -2847.6 16.5
Table H.7 - Key Events in 15km Blend Case
Again it can be seen that the spreadsheet is more pessimistic. The FGE model shows that, for this case, the system consumes 4640m operation between PDD and RAT ENABLE, i.e. surface detection software algorithm running (fixed time of 44s drogue shoot flight time +29s front shield separation /RAT ON delay and RAT warm up period). This compares with 4934m predicted for nominal entry analysis post separation (nominal entry, nominal MGS atmosphere). For this ‘blend case’, the descent still has just sufficient time for airbag inflation and achievement of terminal descent rate (the landing site is ~3300m below the datum).
acceleration comparison comparison Velocity comparison 6000
5
5000
0 -20000 0 20000 40000 60000 80000 100000 120000 140000 4000 ) -5
FGE 15kmblend 3000 -10 "SPICAM" velocity (m/s) 2000 FGE 15km blend -15 acceleration (m/s2 "SPICAM" 1000
-20
0 -20000 0 20000 40000 60000 80000 100000 120000 140000 -25 altitude (m) altitude (m)
density profiles 2.00E-02
1.80E-02
1.60E-02
1.40E-02
1.20E-02 SPICAM+projection15 MGS 1.00E-02 spicam data density
8.00E-03
6.00E-03
4.00E-03
2.00E-03
0.00E+00 0.0 20000.0 40000.0 60000.0 80000.0 100000.0 120000.0 140000.0 160000.0 altitude (m)
Figure H.18 - Correlation in 15km Blend Case
217 Beagle-2 Mission Report Annexes
This page left intentionally blank.
218 Beagle-2 Mission Report Annexes
I EDLS Events: Parachute Deployment, Heatshield Separation and Landing
The ballistic coefficients of the three assemblies following Aeroshell separation, i.e. front heatshield, lander assembly on main parachute and back cover assembly on drogue parachute, are subject to a number of variables. These include state of inflation and wake effects (itself a function of the separation distance between all the elements) and all change rapidly over a short period of time at different rates. Simply conducting a comparison of ballistic coefficients at any arbitrary snapshot in time is misleading. Consequently a mathematical model was created and a worst case prediction was made.
The model used to analyse Rear Cover separation, Inflation loads and Heatshield separation is described in the reference document 'Main Parachute Inflation Loads Software Detailed Definition Document - BEA2.SOF.00008.S.MMS Iss.0A Rev. D2 '.
The results of the modelling are described in ‘Main Parachute Deployment and Aerodynamic Performance Analysis BEA2.RPT.00053.S.MMS Iss. OA Rev. D1’.
I.1 Lander/Heatshield Separation
The Beagle 2 heatshield has been added into the entry model using data obtained from experiments carried out during the Viking program. This data suggests that the heatshield will be stable (it’s angle of attack will remain constant), and after separation it will move forward and laterally. The model takes into account wake effects between the heatshield and the lander, post separation. Separation is determined using the ballistic coefficients of the lander and heatshield (see the Main Parachute Inflation Loads Software Detailed Definition)
Figure I.1 shows the position of the lander, backcover, and heatshield during deployment, inflation, and for approximately 19s from the start of deployment. This is the original analysis prior to increasing the Front Heatshield diameter to aid airbag accommodation. The second chart is a close up of the trajectory of the parachute. Ticks on the plots occur at 0.5s intervals. Figure I.2 presents an update on this analysis with the revised heatshield diameter. Figure I.3 is a plot of the separation between the lander and heatshield. Figure I.4 shows the relative speed between the heatshield and lander as a function of time.
It can be seen that the trajectory of the heatshield remains essentially unchanged during inflation. At 19 seconds from the start of deployment the separation between the heatshield and lander is almost 900m.
It can be seen that the rate of separation increases rapidly as the lander slows down, due to the inflation of the main canopy, then gradually slows down, as the heatshield slows down, and the heatshield’s trajectory becomes more parallel to the lander trajectory. These trajectory paths then diverge rapidly as the velocity of the lander reduces.
Figure I.2 also shows the divergence of the drogue from the main chute whilst there is little drag being generated by the main parachute. Main inflation is not complete until there is clear separation of nearly 40m between the drogue and the main. Soon after separation the backcover drag becomes significant. It should be noted that the drag area of the back cover is equates to about 20% of that of the pilot chute and thus without the front shield in position the empty back cover generates significant drag as soon as it starts to clear the wake of the lander. Thus the rear cover contributes significantly to drogue separation. As the mainchute strips out of its bag completely, a tiebreak back to the structure further disturbs the back cover, introducing instability, increasing the rate of divergence further than that illustrated.
219 Beagle-2 Mission Report Annexes
Position of lander and backplate during deployment and inflation. Initial Altitude 1504.6 m 1600 Back cover Lander Heatshield 1400 0.5s timesteps for lander 0.5s timesteps for back cover 0.5s timesteps for heatshield 1200 Back cover pos. at full inflation Lander pos. at full inflation Canopy position at full inflation
1000
800 Altitude (m)
600
400
200
0 -300 -250 -200 -150 -100 -50 0 Position (m)
Position of lander, backcover, and heatshield during deployment and inflation. 1600
1500
1400
1300
1200
Back cover
Altitude (m) Altitude Lander 1100 Heatshield 0.5s timesteps for lander 0.5s timesteps for back cover 1000 0.5s timesteps for heatshield Back cover pos. at full inflation Lander pos. at full inflation Canopy position at full inflation 900
800
-100 -80 -60 -40 -20 0 Position (m)
Figure I.1 Position of the Lander, Backcover, and Heatshield During Deployment (Original Front Shield Diameter)
220 Beagle-2 Mission Report Annexes
Positions of lander, heatshield and drogue during separation
1900
1800
1700
1600 Position of Lander Altitude Position of Drogue 1500 Position of Heatshield 0.5s ticks for Lander 1400 0.5s ticks for Drogue 0.5s ticks for Heatshield Lander position at full Inflation 1300 Parachute position at full inflation Drogue position at full inflation Heatshield position at full inflation 1200 -90 -80 -70 -60 -50 -40 -30 -20 -10 0 Ground Postion
Positions of lander, heatshield and drogue during separation
1750
1700
Position of Lander Altitude 1650 Position of Drogue Position of Heatshield 0.5s ticks for Lander 1600 0.5s ticks for Drogue 0.5s ticks for Heatshield Lander position at full Inflation Parachute position at full inflation Drogue position at full inflation 1550 Heatshield position at full inflation -70 -65 -60 -55 -50 -45 -40 -35 -30 -25 Ground Postion
Figure I.2 Position of the Lander, Backcover, and Heatshield During Deployment (Final Front Shield Diameter)
221 Beagle-2 Mission Report Annexes
Lander-Heatshield separation 900
800
700
600
500
400
300 Lander-Heatshield separation (m)
200
100
0 2 4 6 8 10 12 14 16 18 20 Time from start of deployment (s) Figure I.3 Lander / Heatshield Separation
Lander-Heatshield separation rate 70
60 ) 1 - 50
40
30
20 Lander-Heatshield separation rate (ms
10
0 2 4 6 8 10 12 14 16 18 20 Time from start of deployment (s) Figure I.4 Lander / Heatshield Separation Rate
222 Beagle-2 Mission Report Annexes
Simple ballistic coefficient calculations showed that heat shield separation would occur during main inflation and thus it was essential to deal with the motion of the heatshield as an independent body right from the start and to take account of both the contact between heatshield and the lander prior to separation and the mutual aerodynamic interference between the lander and the heatshield for the whole of the inflation and beyond until any interference was negligible. In the same way as with the rear cover, continuing to monitor the heat shield beyond the point at which interference became negligible was trivial and was done to allay the concern that the heatshield might give a false signal to the RAT. Note that data from the Viking program showed that Heatshield separation was stable with no tendency to slide around the side of the Lander. Subsequent to the last minute change to the heatshield dimensions calculations were revisited using the new sizes and masses together with the revised Cd/q variation from flight test and established that separation occurred correctly. I.2 Drag Coefficients
As regards the parachute Cd the following is direct extract from BEA2.RPT.00053.S.MMS (Page 21):
"Subsequent testing suggested that the Cd under terminal conditions could be as high as 1.2 and that the inflation constant was 8 rather than 7. Based on this and assuming the same variation of Cd with dynamic pressure as shown by [fig] but factored by 1.2/0.96 the predicted peak load increased to 4.19 kN. More detail analysis revealed that the average Cd under terminal conditions was 1.042 and at inflation was 0.92. Thus the revised variation of Cd with dynamic pressure shown by [fig] was derived. This analysis also revealed that not only was the inflation constant 8 but the inflation profile was noticeably different from that originally assumed as shown by [fig]. With these final revisions based on test the nominal peak load reduced to 2.89 kN giving a fully factored design load including both variability (1.12) and the specified ultimate factor of 1.1 of 3.56 kN." Note that, during inflation the main Cd value is that for the appropriate dynamic pressure (predicted to be 68 Pa) so the highest main Cd considered was about 0.96
Summarising From Kistler we expected a Cd = 1.1 Our analysis was done at Cd = 0.96 Tests showed a Cd = 1.04 (mean)
I.3 Main Parachute material porosity
It was determined that the material comprising the canopy of the Main Parachute would be effectively non- porous due to the weave dimensions being smaller than the Martian upper atmosphere mean-free-path length and so all testing was done with a 'coated' fabric but at the correct dynamic pressure for Mars. This required low altitude testing on Earth. This was based on testing that was done during the Viking program on the porosity of materials at high altitude.
I.4 Main Parachute/Airbag re-contact after bouncing
A simplified analysis of the first bounce scenario was done during the design phase of the EDLS Probe using the LS DYNA software tool, but because of the large number of variables (descent velocity, wind velocity and direction, terrain slope and roughness, surface type, rock abundance and types etc.) it was not possible to carry out a proper Monte-Carlo analysis at the time. It had been concluded that with the original 'gliding' main parachute design and a strop length of 31.5 m to the point of confluence of the rigging the probability of re-contact was small enough to be discounted. When the Airbag pressure was reduced to avoid dynamic overpressure and consequently the main parachute was re-designed to be larger and therefore have a lower terminal velocity but zero glide, it was concluded that these factors would reduce the probability of re- contact even further than with the original design and so no increase in strop length was necessary. However it has subsequently been shown that using an improved simulation of the bounce scenario a strop length is marginal when all other factors ie. wind, terrain slope, surface type are worst-case ie. in line. Visual evidence from videos of main parachute landings show that there is a period of at least 1 second after ground contact when the canopy hovers before descending at < 4m/sec. Further an 'optimiser' tool for LS
223 Beagle-2 Mission Report Annexes
DYNA has subsequently become available and it is possible to perform a Monte-Carlo assessment of the first bounce scenario. However it is still extremely difficult to analyse the complete landing sequence consisting of any number of bounces between 12 -20 because the dispersions of the variables rapidly increase the number of solutions towards infinity. This is the inherent disadvantage of the bouncing Airbag design.
Tables I.1 and I.2 and Figure I.5 show the re-contact calculation carried out for a canopy descending at 4 m/s and traversing in a 11 m/s wind. The characteristics of the airbag bounce depend on the value of coefficient of restitution (a measure of the elasticity) selected. Three cases, each with different values of coefficient of restitution for the vertical component are considered; “1” a perfectly elastic collision, “0.65” a value representative of airbag damping originally used by Martin Baker before test data was available (and the basis of the MBA design) and “0.80” airbag damping used in airbag bounce analysis derived from drop test performance.
Coefficient of Restitution (Vertical) 1 0.8 0.65 Coefficient of Restitution (Lateral) 1 0.9 0.9 First Bounce Max Height / m 34.6 22.1 14.6 Time to Max Height T / s 4.3 3.5 2.8 Translation at Max Height / m 95 68 56
Table I.1 First Bounce Characteristics
The time of flight T and the location of the lander at the peak of the first bounce has been calculated for each case. The position of the canopy at T, assuming a strop length of 31.5m and a 41.5m canopy height are given in Table I.2.
Time T / s 5.4 3.5 4.3 Canopy Traverse Distance / m 48 38 31 Canopy Altitude Loss / m 17.3 13.8 11.2 Canopy Altitude / m 24.2 27.7 30.3
Table I.2 Canopy Drift Characteristics
34.6
e=0.8 e=0.65 e= 1 22.1
83 68 95
Figure I.5 Canopy and Airbag Motion
While the analysis demonstrates conditions whereby re-contact is possible a number of factors have been excluded. The airbag is likely to either lose or pickup lateral velocity, dependent upon any local slope and slope direction. The analysis assumes that both airbag and parachute continue in same straight path. This is unlikely, since the first impact will be imperfect and strop drag will occur - neither is predictable.
An additional 10m on the strop length would have reduced the probability of re-contact.
224 Beagle-2 Mission Report Annexes
I.5 Overall Summary
From the analysis and testing of the Main Parachute together with all the associated issues of back cover and front shield separation, there is no reason to suppose that the combined front/back separation using one set of pyro activated bolt cutters is a credible failure mode; that the parachute did not inflate correctly or behave abnormally during descent. The only point that could be marginal and is worth a further study is the Airbag re-contact with main parachute after first bounce using a strop length to the rigging confluence point of 31.5 m
225 Beagle-2 Mission Report Annexes
This page left intentionally blank.
226 Beagle-2 Mission Report Annexes
J Lovell Telescope Observations of the Beagle 2 Landing Site
Ian Morison, Jodrell Bank Observatory The University of Manchester
J.1 Summary of Observations The Lovell radio telescope was equipped with a specially built cryogenic receiver for the Beagle 2 uplink frequency of 401 MHz. It incorporated high-temperature super-conducting filters to reject the significant terrestrial interference at adjacent frequencies but allow a signal at 401 MHz to pass virtually un-attenuated and was almost certainly the best receiver ever built for use at this frequency.
When we learnt that no signals had been received by the Mars Odyssey on the morning of Christmas Day we decided to observe during the whole time that the Beagle 2 landing site was visible from the Jodrell Bank Observatory.
Observations were carried out on the evenings of December 23rd to 28th, 2003. After the 28th the landing site was on the far side of Mars whilst Mars was above our horizon. Further observations were made on the 23rd to 25th January, 2004.
The system was fully tested on the ground with a Beagle 2 transmitter module prior to the observations, and we built our own test transmitter to precisely simulate the expected Beagle 2 transmissions to give a full system test prior to each observation.
Given the system noise temperature of 80K, a Beagle 2 signal should have reached a maximum of ~15 dBs above noise. Real time displays allowed us to see signals barely above the noise floor of the receiver and, with appropriate folding of the data carried out in post analysis, a Beagle 2 signal would have been detected even if somewhat below the noise floor. Sadly, no signals from Beagle 2 were detected.
227 Beagle-2 Mission Report Annexes
J.2 The Receiver
Figure J.1 - The 401 MHz Receiver cryostat in the development laboratory during testing with a spare Beagle 2 transmitter
The receiver was contained within a cryostat cooled by a compressed Helium refrigeration system to a temperature of ~ 15K. This low temperature minimizes the thermal noise inherent in the low noise amplifiers (LNAs) but, even more important allowed the incorporation of high-temperature super-conducting filters between the feed and the LNAs.
The band around 401 MHz contains many very strong terrestrial signals which will be picked up by the feed and enter the LNAs. Even though they are not at the observed frequency they can cause inter-modulation products that can mask a weak signal by substantially increasing the receiver noise floor. If a filter made of normal materials is placed in front of the LNAs, it can remove the interfering signals and thus reduce the noise floor but will, due to resistive losses in the material making the filter, also attenuate the desired signal so reducing the effective sensitivity of the receiver.
The University of Birmingham Engineering Department have pioneered the use of super-conducting materials in the construction of narrow band pass filters. Due to the lack of resistive loss in the material, these filters can greatly attenuate out of band signals without attenuating those in band. The result of incorporating these specially made filters is a receiver that could perform close to the theoretical performance even in the presence of strong signals nearby in frequency.
The receiver was built to receive both left and right circularly polarized signals, both of which were fully analyzed, so that we could not, in error, be observing on the opposite polarization to that transmitted by Beagle 2.
A 4 MHz band centred on 401 MHz of the left and right circularly polarized signals were down-converted to an intermediate frequency (IF) of 30 MHz and brought down to the Lovell Observing Room.
228 Beagle-2 Mission Report Annexes
J.3 The Digital Data Acquisition System The IF came into the Observing Room with a bandwidth of ~4 MHz centred on 30 MHz. This centre frequency was determined by the local oscillator, which we chose to be 431573000 Hz, i.e., an observing frequency of 401.573 MHz.
The signals were passed though some power control circuits including two manually set attenuator/amplifier units, and a 24 dB amplifier module. Both polarizations of the IF were fed directly into sampling modules. These first band-limited the 30 MHz IF to 5 MHz bandwidth. The signal was then down-converted to baseband using MCL demodulator devices. These used another LO set at 30 MHz, to produce in-phase and quadrature components of the signal.
At the output of the demodulator units for each of the four signals, there were two gain stages and a low- pass filter to provide an anti-aliasing function. This filter was tuned to roll off at around 13 kHz, and attenuate by 40 dB at 50 kHz. As we were analysing a complex signal containing both positive and negative frequencies, that corresponded to an overall bandwidth of 26 kHz. Finally the signals were applied to a pair of Analogue to Digital Converters - one for each polarization. These devices sampled at a rate of 5.0 MHz - so highly over sampling the signals. They provided 8-bit outputs. The external attenuators were set to maintain a good non-saturated level that was monitored manually using an in-built power meter. The two data streams were then decimated to reduce the data rate from 5.0 MHz to 100 kHz. The sampled data was then transferred by direct memory access into the memory of a 2 GHz processor PC.
J.4 Software Processing and Real-time Data Display A sequence of 131072 samples for each polarization were then Fast-Fourier Transformed to give power spectra every 1.3 seconds; each having a resolution of ~0.7 Hz. Each spectrum was saved to disk for archiving and post processing. In addition a live "waterfall display" was generated for each polarization. The two computer screens showed a rolling sequence of 100 power spectra; time as the vertical axis and frequency the horizontal axis. The power across the spectrum was encoded using a combination of colour and brightness that was determined in tests to give the greatest contrast between the background noise floor and the weak CW signals.
Whereas a single spectrum would need a signal significantly above the noise floor to be visible, by displaying a sequence of such spectra in a "waterfall" plot allows the easy detection by eye of CW signals only just above, or at, the noise floor of the receiver.
To improve the real-time signal to noise ratio we were able to average a number of successive spectra. During the on-line observations we concentrated our efforts and display on the region of the band where the Beagle 2 signal was expected. The centre of this band was determined by the transmit frequency and the calculated Doppler shift of about -20 kHz. The width of the displayed band was determined by the specified tolerance of the Beagle 2 transmitter. However during post processing we examined the full 25 kHz sampled bands.
229 Beagle-2 Mission Report Annexes
J.4.1 Post Processing As for the first series of observations we were expecting a sequence of 10 seconds on and 50 seconds off, it was possible to fold the data to sum sequential 60 seconds periods of data in such a way that noise within the 10 sec on periods would be integrated. This increased the detectability limit of a weak CW with this duty cycle by about a factor of 2 to 3. Waterfall plots for the full period of observations were then produced and examined.
Figure J.2 - A post-processed Waterfall Display showing 100 minutes of data across 12 kHz of the recorded bandwidth. Several weak signals can be perceived which are marginally above level of the noise floor. J.5 Test Observations In the week prior to the first scheduled observations on Christmas day, a Beagle 2 test transmitter was brought to JBO so that we could run an end-to-end test of the full system. This was successful but we became very aware of the drifts in the transmit frequency as the transmitter warmed up during the test transmissions and this enabled us to plan for the appropriate bandwidths that would need to be displayed during the observations.
We also built a test transmitter which was computer controlled to give the expected 10 sec on/ 50 second off duty cycle at the nominal Doppler shifted Beagle 2 signal. This was activated for a test transmission prior to each observing period.
230 Beagle-2 Mission Report Annexes
Figure J.3 - The "Beagle in a Box" test transmitter installed in the receiver testing facility.
Figure J.4 - A waterfall display showing 10 minutes of data prior to observations with the JBO test transmitter switched on.
231 Beagle-2 Mission Report Annexes
J.6 Telescope Pointing The Lovell Telescope is not normally used to follow planets such as Mars. A computer program was thus developed to track Mars using ephemeris routines supplied by JPL. During observations the demanded Azimuths and Elevations provided by the control computer to point the telescope were compared with those computed by two other ephemeris programs running on off-line computers. All three agreed. The beam width of the Lovell Telescope at 401 MHz was vastly greater than the angular size of Mars so that signals from any part of the observed disc would have been received. J.7 Observation Planning We used the Starry Night Pro program to show when the longitude of the Beagle 2 landing site would be visible from JBO. This determined the start time of our observing sessions. The observations on each day had to be terminated when Mars dropped below the south-western horizon.
Figure J.5 - Two Mars graphics showing Beagle 2 landing site coming into view (left) and when Mars set (right) on December 25th 2003 J.8 Observations Observations were made on the 25th to 28th December 2003 and 23rd to 25th January 2004. We observed whenever Mars was above the horizon and the Beagle 2 landing site was visible. This included significant observing time outside the windows when Beagle 2 was expected to be transmitting. All times are UTC.
Day Mars Above Horizon Observing Period Dec 25 17:00 - 00:20 Dec 26 17:45 - 00:30 12:00 - 00:30 Dec 27 18:00 - 00:25 Dec 28 20:20 - 00:30 Jan 23 15:45 - 23:45 Jan 24 10:45 - 00:15 14:45 - 00:00 Jan 25 15:20 - 00:30
232 Beagle-2 Mission Report Annexes
We were very impressed with the cleanness of the band. The filters had worked to great effect. Though many weak terrestrial signals were seen for periods during the observations, only one satellite signal passed briefly through the band that could have obscured a signal from Beagle 2 - this was outside the nominal transmission periods.
Observations of radio sources gave a system noise temperature of ~ 80 K. Calculations by ESOC of the expected signal strength from Beagle 2 indicated that, with the antenna at right angles to us, the signal would be expected to be ~ 15 dBs above our noise floor - falling as the orientation changed. We believe that we would have been able to detect a signal over at least a range of ±60° from this ideal condition as Mars rotated on its axis, so changing the antenna orientation. Sadly, no signals that matched the Beagle 2 transmission signature were seen.
Figure J.6 - The Lovell Observing Room December 25th 2003 The central screens show the waterfall displays of the left and right circular polarization data.
J.9 The JBO Beagle 2 Team University of Manchester, Jodrell Bank Observatory Ian Morison Project Scientist Tim O’Brien Assistant Project Scientist – Offline Data Analysis Angela Bayley Lovell Telescope Control Software Christine Jordan Real Time Data Analysis Software Tim Ikin Data Acquisition Hardware David Glynn and Receiver Design, Construction and Testing Neil Roddis Anthony Holloway Computing Facilities David Stannard Observations
University of Birmingham, Emerging Device Technology Resource Centre Fred Huang Design and Construction of the High Temperature Superconducting Filters
233 Beagle-2 Mission Report Annexes
This page left intentionally blank.
234 Beagle-2 Mission Report Annexes
K Assessment of the Cruise Phase Thermal Performance
B. M. Shaughnessy, RAL Adapted from document BGL2-GEN-RAL-TN-033 Issue d01 Revision 0
K.1 Introduction
K.1.1 Scope This document provides an assessment of the Cruise phase thermal performance of Beagle 2. Specifically it considers whether there are any implications for the anomaly encountered following separation from Mars Express. For much of the Cruise phase, telemetry was limited to a few key parameters monitored by Mars Express. Of thermal interest were the Lander battery temperature, the temperature at the interface with Mars Express, and the heater circuit currents. The Lander was activated for checkouts a number of times during Cruise and for these periods the data from temperature sensors on the Lander were recorded. Other data from Mars Express implies that there was measurable outgassing in the vicinity of Beagle 2. The implications of this with respect to thermal performance are also considered.
This document (dated 16/02/04) is at a draft stage and presents an initial assessment of the Cruise phase thermal performance. It will be refined and formally issued in due course.
K.1.2 Acronyms AAM Aeroshell Assembly Model (Probe) APS Auxiliary Power Supply ARM Aeroshell Release Mechanism CE Common Electronics MLI Multi-layer Insulation VDA Vacuum deposited aluminium
235 Beagle-2 Mission Report Annexes
K.2 Reference Documents The following REFERENCE DOCUMENTS (RD) provide further background information or have been used as a source of information repeated herein.
RD Title Number Issue Revision RD1 Chester, E., various emails, 31/12/03 – 06/02/04 - - - RD2 As-Built Configuration List for Beagle 2 Heater BGL2-HTC-RAL -PTL-001 1 0 Circuits RD3 Contamination Control Engineering Design NASA-CR-4740 May - Guidelines for the Aerospace Community 1996 RD4 Sims, M., various emails 11/03 - - - RD5 Ross, R. G., Cryocooler Load Increase due to 12th International - - External Contamination of Low-ε Cryogenic Cryocooler Conference, Surfaces Cambridge, MA, USA, 18- 20/06/02 RD6 Beagle 2 AAM Probe Thermal Balance and BGL2-GEN-RAL-TR-004 1 0 Verification Test Report RD7 Beagle 2 AAM Probe Thermal Mathematical Model BGL2-GEN-RAL-TN-032 1 0 Correlation RD8 Beagle 2 Probe Flight Thermal Predictions BGL2-GEN-RAL-TN-030 2 0
Table K.1 - Mars Express monitored thermal telemetry for days 183 and 349
Day Days since launch Sun distance, km Solar Aspect Angle 183 30 1.55E+08 152 349 196 2.19E+08 118
Heater Duty Cycle, % Day SUEM Backplate ARM Battery 183 0 0 100 12 349 0 27 100 8.5
Average Heater Power, W Day SUEM Backplate ARM Battery 183 0.0 0.0 5.0 1.0 349 0.0 1.3 5.0 0.7
Temperatures, oC Day Interface ARMa Battery 183 8.5 -10 -6.3 349 1 -13 -6.5 a – ARM temperature from checkout telemetry
236 Beagle-2 Mission Report Annexes
K.3 Cruise Phase Thermal Performance Summaries of Mars Express monitored data on days 183 (just before first checkout) and 349 (just before final checkout) of 2003 are given in Figure K.1 and Table K.1 [RD1]. The minimum and maximum temperatures recorded during all times that the Lander was activated for checkouts are presented in Table K.2 [RD1].
The heater duty cycles presented in Table K.1 have been estimated from the heater current plots shown in Figure K.1. The mean power dissipations at these heaters have been estimated from the heater resistances measured during installation [RD2]. It is noted that the current data shown in Figure K.1a is slightly lower than that observed at the Lander Operations Control Centre (the lowest current in Figure K.1a is about 0.15 A whilst the lowest observed current was about 0.19 A). From knowledge of the resistance of the individual heater circuits this implies that the bus voltage was ~ 21 V instead of ~ 28 V. However, for the purposes of this investigation the bus voltage is taken as 28 V and it is assumed that there was an error in the storage, conversion, or extraction of the Mars Express telemetry.
Day 183 to 185: NPWD2471 Day 349 to 351: NPWD2471 Battery Heater 0.4 0.6 Battery Heater 0.35 Backplate Heater 0.5 0.3
0.4 0.25
0.2 0.3 Current, A 0.15 Current, A 0.2 0.1 ARM Heater 0.1 0.05 ARM Heater 0 0
Figure K.1 - Heater current data for days 183 and 349
The Beagle 2 thermal subsystem performed very close to expectations for the duration of the Cruise phase. However, two observations can be made from the telemetry, and these are expanded upon below:
• During checkouts the Common Electronics warmed at a rate quicker than expected and thus achieved higher temperatures than predicted. With the exception of the APS, all temperatures remained within applicable limits. During a Lander operation on 21/11/2003 the APS was allowed to exceed its upper operational temperature by 2°C.
• With the exception of brief periods during checkouts, the Aeroshell Release Mechanism (ARM) heaters remained at a 100% duty throughout Cruise. Temperatures of the heated components therefore remained about 5°C colder than would have been attained if the heaters had cycled as required.
APS Temperature The maximum temperature of the APS during all periods of Lander activity is given in Table K.3. The Common Electronics showed no indication of degraded performance through the course of Cruise. Indeed, there is no reason to suspect that the APS would be damaged by exceeding its upper operational temperature by only 2°C (on 21/11/2003).
237 Beagle-2 Mission Report Annexes
Table K.2 - Maximum and minimum temperatures during checkouts
Measured Limit Limit Margin Sensor Location Min Max Min Max Type Min Max MT0001 Battery Unit -7 31 -30 45 OP 23 14 MT0002 Battery Unit -8 34 -30 45 OP 22 11 MT0003 Battery Unit -9 30 -30 45 OP 21 15 MT0004 Battery Unit -7 33 -30 45 OP 23 12 MT0005 Battery Unit -8 31 -30 45 OP 22 14 MT0006 Battery Unit -7 34 -30 45 OP 23 11 MT0007 CEP -9 41 -50 55 OP 41 14 MT0008 CEP -10 40 -50 55 OP 40 15 MT0009 CEM -8 39 -50 55 OP 42 16 MT0010 CEM -8 38 -50 55 OP 42 17 MT0011 APS -8 57 -50 55 OP 42 -2 MT0012 Transceiver BBU -10 34 -65 55 OP 55 21 MT0013 Lander Innershell -9 22 125 NON-OP 103 MT0014 Lander Innershell -12 16 125 NON-OP 109 MT0015 Lander Innershell -14 -2 125 NON-OP 127 MT0016 Lander Innershell -15 -5 125 NON-OP 130 MT0017 Lander Innershell -15 -5 125 NON-OP 130 MT0018 ARM Bracket -11 -4 -50 80 OP/NON-OP 39 84 MT0019 Lander Innershell -14 -3 125 NON-OP 128 MT0020 Instrument arm -14 -3 -100 125 NON-OP 86 128 MT0021 Instrument arm -14 -4 -100 125 NON-OP 86 129 MT0022 Instrument arm -13 0 -100 125 NON-OP 87 125 MT0023 XRS BEE -14 5 -65 20 NON-OP 51 15 MT0024 PPS -8 25 -50 55 OP 42 30 MT0025 XRS BEE -14 4 -65 20 NON-OP 51 16 MT0026 Transceiver RFU -10 29 -65 55 OP 55 26 MT0027 Transceiver RFU -11 29 -65 55 OP 54 26 MT0028 Transceiver RFU -11 29 -65 55 OP 54 26 MT0029 Lid hinge mechanism -11 11 -95 50 NON-OP 84 39 MT0030 GAP source -13 -5 -65 40 NON-OP 52 45 MT0031 GAP MS -16 21 -65 40 NON-OP 49 19 MT0038 CEP -9 41 -50 55 OP 41 14
238 Beagle-2 Mission Report Annexes
Table K.3 - Maximum APS temperatures recorded during Lander checkouts
Date of Lander Temperature Margin Operation 05/07/2003 49 6 13/07/2003 49 6 01/09/2003 30 25 07/10/2003 48 7 09/10/2003 49 6 21/11/2003 57 -2 17/12/2003 53 2
Table K.4 - Predictions for day 183 (predicted values shown in bold type)
Heaters modelled with heated nodes as boundaries at stated temperatures
Model Day Average Heater Power, W Temperatures, oC SUEM Backplate ARM Battery Interface ARMa Battery CRU301 ~183 0.0 1.0 3.0 1.2 9 -6 -6 a – corresponds to heated node. Adjacent node 2345 corresponds to sensor MT18 and thermostat location
Heaters modelled with heated nodes as diffusion with power input as stated
Model Day Average Heater Power, W Temperatures, oC SUEM Backplate ARM Battery Interface ARMa Battery CRU402 183 0.0 0.0 5.0 1.0 9 4 2 a – node 2345 corresponds to sensor MT18 and thermostat location
12 Battery heater 10 8 ARM heater 6 C o 4 2 0 -2 00.250.50.751
Temperature, Temperature, -4 -6 -8 -10 Emissivity
Figure K.2 - Impact of backshell MLI surface emissivity on battery and arm heater temperatures (with fixed power to heaters)
239 Beagle-2 Mission Report Annexes
ARM Heater Duty The ARM heater circuit was designed intentionally to have no redundancy (in the form of additional thermostats or heater circuits). Therefore slightly reduced margins were applied to the sizing of heater dissipations to reduce risk in the event of a closed-circuit (i.e. heater ON) failure.
An increased duty-cycle implies either a greater heat loss than expected locally at the ARM, or a greater heat loss than expected from the Probe as a whole (i.e. via MLI).
Table K.4 compares predictions made for day 183 with heaters modelled as (i) boundary nodes at a fixed temperature and (ii) diffusion nodes using the estimated mean heater dissipation from Table K.1. With heaters held at a fixed temperature, the average heater power is correctly predicted for the battery, but under-predicted for the ARM. When the in-flight power dissipations are used, both the battery and ARM temperatures are somewhat warmer than those measured during Cruise.
Figure K.2 shows the predicted effect of increasing the emissivity of the multi-layer insulation (MLI) on the Probe backshell (i.e. with an aluminised outer surface) for the case with heaters modelled using the in-flight power dissipations. Although temperatures are reduced with increased emissivity, the observations (i.e. Table K.1) cannot be replicated.
Together these predictions imply that the modelled thermal links from the ARM are not consistent with the flight-build (i.e. the heat transfer from the ARM to the Lander is too great and that to the aeroshell is too small).
A simple sensitivity analysis has been made to assess whether changing thermal links at the ARM could in principle achieve the observed results. Interface conductances between the Lander ARM bracket and the frontshield were increased by a factor of ten and the conductance of the frontshield inner skin was doubled (see Table K.5). It therefore appears likely that the Cruise phase observations could be replicated by careful adjustment of thermal links at the ARM. The impact of this on the Coast phase is considered in the next Section, but is shown to be minimal.
Table K.5 - Predictions with increased heat loss at ARM
Heaters modelled with heated nodes as diffusion with power input as stated
Model Day Average Heater Power, W Temperatures, oC SUEM Backplate ARM Battery Interface ARMa Battery CRU412 183 0.0 0.0 5.0 1.0 9 -7 -5 CRU413 349 0.0 1.3 5.0 0.7 1 -6 -6 a – node 2345 corresponds to sensor MT18 and thermostat location
240 Beagle-2 Mission Report Annexes
Figure K.3 - Residence time of molecules as a function of surface temperature [RD3]
Contamination of MLI surface finishes Data from Mars Express indicates that there was measurable outgassing in the vicinity of Beagle 2 [RD4]. The radiative properties of the external surfaces would be affected if outgassed products were to condense on them. The outgassed products are likely to be water or organics. Deposition of water ice on a low- emissivity surface tends to increase emissivity whilst the solar absorptivity remains relatively low. Deposition of organic contaminants on a low-emissivity surface tends to increase both emissivity and solar absorptivity at a similar rate – tending towards an emissivity/absorptivity (α/ε) ratio of unity.
The residence time of contaminants is related to surface temperature (see Figure K.3). Contaminants will condense on a cold surface but would sublimate if the surface temperature was then raised beyond a certain level. If it were to occur, water ice would probably be the most critical contaminant because the low α/ε would inhibit warming of the surface when the Probe is exposed to sunlight during Coast.
Water-ice to metal bonds are likely to form readily on surfaces colder than 0 °C, however, the residence time for water-ice bonding to water-ice remains short (< 1s) for surfaces warmer than about -100°C [RD5]. Pressures of 10-6 torr were recorded in the vicinity of Beagle 2 [RD4]. At this pressure, ice will form on surfaces colder than about -110°C [RD5].
Organic contamination will adhere to surfaces more readily than water ice. The residence time for organic contamination remains short for surfaces warmer than about -30 °C (Figure K.3).
Table K.6 gives the predicted minimum external surface temperatures of the MLI during Cruise. With reference to Figure K.3 it is unlikely that water ice could have formed on the backshell MLI , however it could potentially have formed on other MLI surfaces. Organic contamination could potentially have formed on any of the MLI surfaces. As shown earlier in Figure K.2, increased emissivity alone cannot replicate the observed performance. Therefore, it is most likely that if there was contamination, levels were low enough not to significantly alter the radiative properties of the aluminised backshell MLI, or were confined to the colder surfaces with existing high emissivity values.
241 Beagle-2 Mission Report Annexes
Table K.6 - Predicted minimum external surface temperatures of MLI during Cruise (no margins applied).
MLI Location Temperature, °C Mars Express Top Floor -126 Collar -120 Backshell -75 Frontshield -170
Table K.7 - Summary of flight predictions for Coast
CASE : CO1 Item Temperature, °C Minimum Maximum Aeroshell Backplate -2 3 Aeroshell Backshell -10 -1 Aeroshell Frontshield -36 -30 Mortar -3 Main Parachute -5 Aeroshell Release Mechanism -14 -13 Airbags -13 -8 Airbag Gassing System -13 -12 Battery Sector Common Electronics -10 Battery -10
Table K.8 - Coast phase temperature requirements and predicted margins
Prediction + Requirements Uncertainty Margin Min Max Min Max Min Max Aeroshell Backplate -58 0 -12 13 46 -13 Aeroshell Backshell -58 40 -20 9 38 31 Aeroshell Frontshield -58 -23 -46 -20 12 -3 Mortar -50 37 -13 7 37 30 Main Parachute -23 47 -15 5 8 42 Aeroshell Release Mechanism -50 50 -24 -3 26 53 Airbags -80 50 -23 2 57 48 Airbag Gassing System -29 5 -23 -2 6 7 Battery Sector Common Electronics -100 125 -20 0 80 125 Battery -30 30 -20 0 10 30
242 Beagle-2 Mission Report Annexes
Table K.9 - Coast predictions with increased heat loss at ARM (Model COA311)
Temperature, °C Item Minimum Maximum Aeroshell Backplate -3 3 Aeroshell Backshell -12 -2 Aeroshell Frontshield -32 -23 Mortar -5 Main Parachute -7 Aeroshell Release Mechanism -20 -19 Airbags -14 -9 Airbag Gassing System -14 -14 Battery Sector Common Electronics -12 Battery -13
K.4 Coast, Entry, and Descent Phases During Coast, battery power must be conserved for entry, descent, and initial landed operations. Therefore, no power was available for thermal control of the Probe during the Coast phase. Instead the thermal design relied upon heating from the Sun, which is at an almost constant aspect angle on the Probe during Coast. The ability of the Probe thermal design to meet Coast phase requirements was verified though thermal balance testing on the AAM Probe [RD6; RD7]. A sensitivity study estimated an uncertainty of ± 5 °C as a result of uncertainties in the thermal balance test set-up [RD7]. Tables K.7 and K.8 summarise the original flight predictions for Coast [RD8]. It is clear that there were substantial margins between raw predictions and the applicable temperature limits.
The impacts of increased heat loss at the ARM and contamination of external surface finishes are considered further below:
Increase in heat loss from ARM Table K.9 shows the effects of changing thermal links to increase heat loss from the ARM (as described in the previous Section). Although the ARM temperature falls in comparison with the original flight predictions, substantial margins remain between the raw predictions and the applicable temperature limits. The changes to the ARM thermal links have not been correlated with the in-flight data, however these predictions indicate strongly that increased heat loss from the ARM would not cause the Probe to fall below applicable temperature limits.
Contamination of MLI surface finishes From examination of the Cruise telemetry it is unlikely that there was significant build-up of contaminants on the MLI during Cruise. For completeness, however, the implications for the Coast phase of such contaminants are considered below:
• With the backshell MLI external emissivity increased to 0.9 (absorptivity remains 0.08) to simulate severe contamination by water ice (Model COA310) its surface temperature is predicted to be -125 °C. At this temperature, the residence time of water ice is about 100 s (see Figure K.3). Therefore, under such conditions the ice would sublime and the backshell MLI would return towards its natural finish. The time constant of the sublimation would be much quicker than that of the Probes’ thermal response and so there would be no noticeable impact on the temperatures within the Probe. For information, the steady-state battery and Airbag Gassing System temperatures are predicted to be about -28 °C for this case.
243 Beagle-2 Mission Report Annexes
• With the backshell MLI external properties changed to α = ε = 0.9 to simulate organic contamination (Model COA312) its surface temperature is predicted to be -15°C. Under such conditions the contaminants would sublime very readily and the backshell MLI would return towards it natural finish. The steady-state battery and Airbag Gassing System temperatures are predicted to be about -20°C for this case.
Finally, it is worth noting that the Probe has a long thermal time constant. Following ejection, it is predicted that the Probe takes two or three days to reach its nominal steady-state temperatures (similar durations were observed during test). Considering that the Coast phase duration is a little less than six days, this time constant provides additional margin on the lower temperature limits.
K.5 Landed Phase There is nothing in the Cruise thermal data that suggests concern for the Landing phase.
244 Beagle-2 Mission Report Annexes
K.6 Conclusions This document has provided an assessment of the available Cruise phase thermal telemetry and has considered implications for the mission phases following separation from Mars Express (namely Coast, Entry, Descent, and Landed).
The Beagle 2 thermal subsystem performed very close to expectations for the duration of the Cruise phase. There were, however, some anomalies in the thermal data which have been discussed and are summarised in Table K.10.
This assessment demonstrates that there was no indication in the Cruise phase thermal telemetry of problems that could have contributed to the anomaly encountered with Beagle 2 following separation from Mars Express.
Table K.10 - Cruise thermal performance summary
Observation Assessment APS exceeded operational upper The Common Electronics and in particular the APS warmed limit by 2 °C on 21/11/2003 quicker than expected and thus reached higher temperatures.
The Common Electronics showed no sign of degraded performance.
There is no reason to suspect that the APS would be damaged by exceeding its upper operational temperature by 2°C.
ARM heater duty at 100% for Only a few degrees drop in temperature compared with much of Cruise phase expectations.
Most likely due to difference between as-built and modelled thermal links at the ARM.
Highly unlikely to affect Coast phase thermal performance.
Measurable outgassing reported No evidence that there was significant contamination of the in the vicinity of Beagle 2 external surface of the Probe.
However if contamination had occurred during Cruise it would very probably sublime when the Probe is exposed to the Sun during Coast.
In the highly unlikely case that contaminants occurred and did not sublime during Coast, the Probe would run colder, but critical items would probably not cool much below lower temperature limits.
245 Beagle-2 Mission Report Annexes
This page left intentionally blank.
246 Beagle-2 Mission Report Annexes
L Prediction of Size of a Beagle 2 Impact Crater
M. Hannington, NSC LOCC, University of Leicester
In February and March 2004, as part of the search for evidence of Beagle 2 on the surface of Mars, a number of high resolution images of the Beagle 2 landing ellipse were returned from the Mars Orbiter Camera on Mars Global Surveyor and provided to the Beagle 2 team by Malin Space Science Systems Limited. In support of the analysis of these images an estimate was required of the size of crater that Beagle 2 would have made on impacting with the surface if all or part of the Entry Descent and Landing sequence failed.
Determination of crater size was based on the method described by K Holsapple in reference [A]. Crater size is calculated given information about the impactor and the impacted surface using a technique developed by Schmidt and Holsapple, scaling from empirical data of terrestrial cratering.
Impact velocity and flight angle were determined from EDL modelling incorporating a range of available atmospheric density profiles.
L.1 Crater Size Equation: Schmidt – Holsapple Pi scaling
For a Simple crater Holsapple gives
α β Piv = K1 ⋅ (Pi2 + K2 Pi3 ) Eq. 1
µ − 3µ where α = 1+ and β = 2 2 + µ
K1, K2 and µ are constants based on target characteristics and the terms of the equation are as described following. All units are SI.
Pi2 is the inverse Froude number; a ratio of inertial force to gravitational force.
gL Pi2 = Eq. 2 200v2 g = Acceleration due to gravity L = Linear dimension of impactor v = Impact velocity
The Pi3 term introduces target strength and density:
Y Pi = Eq. 3 3 2 10 ⋅ dtar ⋅ v
Y is a measure of target strength dtar = Density of target v = Impact velocity
247 Beagle-2 Mission Report Annexes
Crater volume and diameter are then calculated:
⎛ Piv ⋅ m ⎞ −6 CV = ⎜ ⎟ ×10 Eq. 4 ⎝ dtar ⎠
3 Cd = 2Kr ⋅ CV Eq. 5
Cv = crater volume Cd = crater diameter m = impactor mass Kr is the radius scaling constant and is based on target characteristics
Cd is the internal diameter of the crater. The diameter of the crater rim is assumed to be a factor of 1.3 larger.
L.2 Beagle 2 Specifics for Crater Size Calculation
The Schmidt – Holsapple equation was implemented with constants selected to be specific to the Beagle 2 impact case. The mass of the Beagle 2 probe is 68.4 kg. The longest dimension of the probe is 0.934 m.
Parameter Value Units m 68.4 kg L 0.934 m
Table L.1 - Impactor Parameters for Crater Size Calculation
248 Beagle-2 Mission Report Annexes
Holsapple provides empirically derived values of K1, K2, Kr, Y and µ for targets including ice, dry sand, dry soil, hard soil / soft rock, hard rock and lunar regolith. Densities provided for these targets are shown in Table .2.
Target Surface Density kg/m3 Ice 983 Dry Sand 1700 Dry Soil 1700 Hard Soil / Soft Rock 2100 Hard Rock 3200 Lunar Regolith 1500
Table L.2 - Target Surface Densities
A Hard Soil / Soft Rock target was chosen as most representative of the Martian Surface at Isidis. Table L.3 gives the corresponding parameters for the model.
Parameter Value Units Y 1E7
K1 0.22
K2 0.215
Kr 1.1 3 dtar 2100 kg / m
Table L.3 - Target Parameters for Crater Size Calculation
Acceleration due to gravity on Mars is taken as 3.7m/s2.
L.3 EDL Modelling It was elected to determine impact characteristics under the EDL failure case that both the drogue and main parachutes fail to deploy. The existing model of the EDL, developed by Fluid Gravity Engineering during the design of the EDL system, was not available for the analysis and a simplified dedicated model was therefore developed at the LOCC. The core model function is based on details provided at Reference B and the output shows acceptable correlation with results obtained from the Fluid Gravity model. Events occur at similar altitudes and times. Peak levels of acceleration are slightly higher overall.
Entry into Martian atmosphere occurs at 5.4km/s at an angle of 16.5o to the horizontal. Design phase EDL modelling predicted a nominal EDL resulting in velocity reduced to 327.8m/s before drogue chute deployment (PDD) and further reduced to 92.6m/s at main parachute deployment.
The probe’s expected velocity and flight angle on impact were determined for varying EDL scenarios. Increased initial velocity and flight angle on entry into the atmosphere and the effect of varying the atmospheric density profile were considered. The six different density profiles modelled are labelled; MGS Nominal, Unblended SPICAM, MGS 90%, MGS 80%, MGS 70% and MGS 50%.
Figures L.1 to L.3 show the acceleration, velocity and flight angle profiles for the EDL modelled with the MGS Nominal atmospheric density profile and assuming no parachute deployment.
249 Beagle-2 Mission Report Annexes
Beagle 2 Deceleration vs Altitude (MGS Density Profile)
25
20
15
10 Deceleration / g
5
0 -20000 0 20000 40000 60000 80000 100000 120000 140000 Altitude / m
Figure L.1 - Beagle 2 Deceleration Profile Using MGS Atmospheric Density Values
Beagle 2 Velocity vs Altitude (MGS Density Profile)
6000
5000
4000
3000 Velocity / m/s Velocity
2000
1000
0 0 20000 40000 60000 80000 100000 120000 140000 Altitude / m
Figure L.2 - Beagle 2 Velocity Profile Using MGS Atmospheric Density Values
250 Beagle-2 Mission Report Annexes
Beagle 2 Flight Angle (MGS Density Profile)
0 -20000 0 20000 40000 60000 80000 100000 120000 140000
-10
-20
-30
-40 Angle / Degrees
-50
-60
-70 Altitude / m
Figure L.3 - Beagle 2 Flight Angle Using MGS Atmospheric Density Values
The origin of each altitude axis is at Mars datum (3km above the surface at Isidis). Figure L.1 shows the deceleration experienced without pyro firings and parachute deployment with the peak of the profile at 20g. The curve crosses the Parachute Deployment Device (PDD) trigger, a falling g level reducing to 0.7g, at around 6.5km above the surface.
The second EDL case modelled incorporated data provided by the Mars Express SPICAM instrument team. The data was returned from SPICAM measurements taken on January 13, 2004 and includes derived density readings for altitudes from 33.4 to 145km.
There is a significant divergence between the SPICAM data and those of the standard models of Mars’ atmosphere which includes the MGS density profile. SPICAM data indicates a much less dense atmosphere over the range available. As described at Reference C a polynomial fit to the data allowed construction of a density profile from top of atmosphere to the surface. The EDL was remodelled with the SPICAM atmosphere producing the deceleration profile shown in Figure L.4.
251 Beagle-2 Mission Report Annexes
Beagle 2 Deceleration vs Altitude (SPICAM Unblended Density Profile)
25
20
15
10 Acceleration / g
5
0 -20000 0 20000 40000 60000 80000 100000 120000 140000 Altitude / m
Figure L.4 - Beagle 2 Deceleration Profile Using SPICAM Atmospheric Density Values
EDL was additionally remodelled with density profiles based on reduced versions of the MGS Nominal. The MGS 90% profile, for example, is the MGS Nominal profile with each density data point value reduced by 10%. The velocity and flight angle at the surface for the MGS Nominal, SPICAM fit and reduced MGS profiles are given in Table L.4.
Impact Velocity Impact Angle Density Profile m/s degrees MGS Nominal 219.13 64.9 SPICAM Unblended Fit 1334.26 21.5 MGS 90% 230.6 60.3 MGS 80% 246.04 54.9 MGS 70% 267.61 48.4 MGS 50% 409.07 32.2
Table L.4 - Impact Velocity and Angle for Varying Atmospheres
Figure L.4 shows that the SPICAM Fit EDL is clearly non-survivable. The PDD trigger g level is not reached before impact with the surface. However there are limitations to the method adopted of fitting a polynomial to the SPICAM data. Densities calculated at the lower altitudes, where critical EDL events occur, are increasingly reduced compared to other models and the expected general shape of the profile at these altitudes is not exhibited in the fit. The approach taken elsewhere in this document is to “blend” the SPICAM data to the standard profile with the blend commencing at differing altitudes. It is emphasised that the Unblended SPICAM Fit is not considered a probable density profile. The analysis is included as it provides a useful upper bound to possible impact parameters.
252 Beagle-2 Mission Report Annexes
Varying atmospheric entry parameters by a few km/s in velocity or a few degrees of flight angle had minimal effect of final impact parameters. Increasing entry angle by 10 degrees pushes the EDL to the limit of survivability in terms of achieving PDD. Increasing entry angle by more than a degree probably places the impact site outside of the landing ellipse (dimensions 57 m by 7.6 m). No consideration has been given to survivability in terms of the thermal regime endured at higher entry angles. Table L.5 shows a cross section of the cases modelled.
Entry Velocity Entry Angle Impact Velocity Impact Angle m/s degrees m/s degrees 7000 16.5 219.71 64.42 5555 25 232.5 56.85 5555 30 248.79 52.54 5555 45 417.12 50.65 7000 60 787.15 61.21
Table L.5 - Impact Velocity and Angle for Varying Entry Parameters
L.4 Beagle 2 Crater Size Calculations
L.4.1 General Calculation for Beagle 2 Impactor Figures L.5 and L.6 and Table L.6 show a representative selection of the range of crater sizes determined for a Beagle 2 sized impactor onto a Hard Soil / Soft Rock target. Figure L.5 demonstrates the effect of varying impact angle while velocity remains constant at 219 m/s. Figure L.6 shows crater size for impacts at 45° for increasing impact velocity.
Beagle 2 Impact Crater Size. Impact Velocity = 219.13 m/s
2.5
2
1.5
1 Crater Rim Diameter / m / Diameter Rim Crater
0.5
0 25 35 45 55 65 75 85 Impact Angle / Degrees
Figure L.5 - Beagle 2 Impact Crater Size. Impact Velocity = 219.13 m/s.
253 Beagle-2 Mission Report Annexes
Beagle 2 Impact Crater Size. Impact Angle = 45 Degrees
14
12
10
8
6 Crater Rim Diameter Diameter Rim / m Crater 4
2
0 012345678910 Impact Velocity / km / s
Figure L.6 - Beagle 2 Impact Crater Size. Impact Angle = 45 Degrees.
Impact Angle / Impact Velocity km/s Crater Diameter / m degrees 20 1 2.60 45 1 3.85 90 1 4.64 20 3 4.72 45 3 6.92 90 3 8.31 20 7 7.38 45 7 10.80 90 7 12.93
Table L.6 - Crater Size Predictions. Impact Velocities 1, 3 and 7 km/s
The same cases were modelled for different target characteristics. Changing the target to Dry Sand or Lunar Regolith increased crater size by factors of 1.5 to 2 respectively. A Hard Rock target reduces crater size by factor of 0.5.
254 Beagle-2 Mission Report Annexes
L.4.2 Incorporating Impact Results from EDL Modelling Complete parachute failure during the MGS Nominal EDL results in an impact with the surface of velocity 219.13 m/s and angle 64.9o. A crater of rim diameter 1.94 m would be created.
Unblended SPICAM Fit gives impact velocity and angle of 1334 m/s and 21.5o respectively. For these parameters a crater of rim diameter 3.16m is calculated.
Table L.7 shows the crater rim size results for impact characteristics corresponding to the EDL models carried out for the six atmospheres.
Impact Velocity Impact Angle Crater Rim Density Profile m/s degrees Diameter / m MGS Nominal 219.13 64.9 1.93 SPICAM Unblended Fit 1334.26 21.5 3.16 MGS 90% 230.6 60.3 1.95 MGS 80% 246.04 54.9 1.95 MGS 70% 267.61 48.4 1.94 MGS 50% 409.07 32.2 2.04
Table L.7 - Beagle 2 Impact Crater Size for Differing EDL Scenarios
The most probable Beagle 2 impact crater would have a rim diameter of approximately 2 meters.
L.5 Beagle 2 Impact Feature Size Any feature created by a Beagle 2 impact will naturally be larger than the diameter of the crater size itself, the feature size must include the size of the ejecta field that surrounds the crater.
In the first instance a very simple calculation was carried out to give radius of ejecta field. Values of ejecta volume and lip height were calculated. Assuming that the ejecta is neatly stacked in a circular spread of triangular cross section, with one side of the cross section given by the crater lip height, it is simple to calculate the radius of the field.
The results from this simple calculation were compared with an approach based on ejecta velocity. The spread of ejection velocity against percentage of ejecta expelled can be determined. Assuming an ejection angle of 30o the distance travelled for each percentage is calculated. 30o is chosen as an average value (45 degrees would result in the greatest ranges) and atmospheric resistance is neglected. A diagram of concentric circles can be built up, each circle representing the range that a certain percentage of the ejecta travels beyond.
This approach can lead to inherently unphysical results. Reduce the percentage of ejecta and the radius of the ejecta field can continue to rise indefinitely. To constrain the calculation the volume of ejecta material that lies between two range boundaries was determined and the thickness of the ejecta field at each range obtained. A scale factor is introduced to adjust the blanket thickness to the calculated lip height at the crater rim. The limit of the ejecta field is taken to be when blanket thickness reaches sub millimetre values.
To bound the ejecta blanket thickness to lip height it was required to reduce calculated ejecta velocities by a factor of 1/3.
Figure L.7 shows the ejecta blanket thickness variation for the MGS Nominal impact calculated in this manner.
255 Beagle-2 Mission Report Annexes
Ejecta Field Profile. Nominal EDL Impact
70
60
50
40
30 Ejecta Height / mm
20
10
0 0123456 Range from Crater Center / m
Figure L.7 - Beagle 2 Crater Ejecta Field Thickness Variation.
Ejecta field radius for the MGS Nominal case is calculated at between 1.4 and 2 m by these two methods.
Ejecta field radius for the SPICAM fit EDL case is around 0.5 m to 1m larger than the MGS Nominal.
The total size of the likely Beagle 2 crater feature is predicted to be between 5 and 6 m in diameter.
L.6 Conclusions Beagle 2 crater feature size for a failed EDL has been calculated. Crater rim diameter is calculated at approximately 2 m. An ejecta field radius 1.4 to 2 m, measured from the crater rim is estimated.
Predicted total crater feature size is 5 to 6m diameter.
This is the most probable feature size, and is representative of 5 EDL failure cases with decreasingly dense atmospheres modelled.
An upper bound on feature size might be given at 9m provided by an EDL based on SPICAM density profile.
Two caveats require mention.
1. Holsapple notes that the limit of the equations’ ability to model crater size is approached at impact velocities less than 1m/s.
2. At Reference [E] Melosh provides an algorithm for calculating crater size based on an earlier published version of Holsapple’s scaling method. This technique was initially appropriated for the Beagle 2 calculation and provided an estimated crater rim size of 6m. At impact velocities between 1km/s and 7km/s the analysis predicts crater size in the range 10 to 30m, craters 3 times the size of those calculated using Holsapples updated scaling method. The same basic algorithm is employed in both methods and the discrepancy is perhaps accounted for by the difference between the values for constants chosen by Melosh.
It is considered that the analysis provided in this document represents a pessimistic (from the point of view of locating a Beagle 2 impact crater) estimate on crater size.
256 Beagle-2 Mission Report Annexes
L.7 References Ref [A] Crater Database and Scaling Tools http://keith.aa.washington.edu/craterdata/
Ref [B] MEX Lander Delivery Module: Beagle 2 ESOC Interface. BEA2.ICD.00005.MMS Issue 1 10 October 2003
Ref [C] Tenuous Analysis of a Tenuous Atmosphere, Ed Chester, University of Leicester
Ref [D] ENTRY & DESCENT MODELLING Jim Clemmet, 25/04/04
Ref [E] Computing Crater Size from Projectile Diameter http://www.lpl.arizona.edu/tekton/crater_c.html
257 Beagle-2 Mission Report Annexes
This page left intentionally blank.
258 Beagle-2 Mission Report Annexes
M Beagle 2 Landing Ellipse Evolution
Figure M.1 - Beagle 2 Landing Ellipse Location Isidis
1,2,3 show pre-launch landing ellipses based on various entry angles defined in the initial landing site selection process, 4 is the selected landing ellipse with the dark central area the actual landing ellipse achieved when delta differential one way range (DDOR) tracking and ejection data was taken into account.
The following diagram M.2 illustrates the different sizes of the ellipses.
259 Beagle-2 Mission Report Annexes
Figure M.2 - Beagle 2 Landing Ellipse Evolution
It can be seen that if DDOR tracking had been baselined a smaller ellipse (114 x 76 km) could have been used leading potentially to a number of alternative landing sites. The post-ejection ellipse although smaller than this is a result of the navigation and ejection data.
DDOR tracking is a method where as well as tracking the spacecraft, a background radio-source eg a quasar is used as a reference and spacecraft already in orbit around Mars are used as additional position references.
260