Design of a Lunar Rover Utilizing Hydrogen-Oxygen Fuel Cell Technologies

Home , Apollo (spacecraft), Apollo 15, FLEX (satellite), Google Lunar X Prize, Lunar rover, Lunar Roving Vehicle

A Thesis

Presented in Partial Fulfillment of the Requirements for the Degree of Master of Science for Aeronautical and Astronautical Engineering in the Graduate School of The Ohio State University

Michael Phillip Snyder

Graduate Program in Aeronautical and Astronautical Engineering

The Ohio State University

2011

Master's Examination Committee:

Dr. Meyer J. Benzakein, Advisor

Dr. Gerald M. Gregorek

Michael Phillip Snyder

2011

Abstract

Future exploration of the solar system will depend on new designs and technologies that are efficient and versatile. Roving systems have explored the Moon and Mars but current means of supplying power are fragile and inefficient, or are considered hazardous to launch. NASA’s Glenn Research Center developed criteria necessary for the design of a robotic lunar rover with an extended exploration time. In order to satisfy these requirements a versatile rover equipped with a hydrogen-oxygen fuel cell with a 1 kilowatt nominal power output was designed to operate in the lunar environment for longer than 5 years continuously. Scaled testing of the rover was performed to predict the performance of the lunar rover. Testing was performed at the Ohio State University’s Aeronautical and Astronautical Research Laboratory in order to determined drawbar pull and sinkage of the rover. Regolith mitigation strategies were investigated in order to prolong the life of the rover by limiting and eliminating contamination caused by the lunar dust.

Dedicated to

To Diane, Walter, Andrea and everyone else who has helped me along to reach the starting point of my journey.

iii

Acknowledgments

I would like to thank my advisor, Dr. Meyer Benzakein for all of his help and guidance. I would also like to thank Dr. Paul Penko, Eric Joyce, and Joel Longo for their help with this project.

Vita

April 16, 1986………………………... Born, Sandusky Ohio, USA

June 2004…………………………….. Diploma, Bellevue Senior High School

June 14, 2009………………………… B.S., Aeronautical and Astronautical Engineering, The Ohio State University

October- December 2009………………. Graduate Research Assistant, Aeronautical and Astronautical Engineering, The Ohio State University

January- March 2010………………… Graduate Teaching Assistant, Aeronautical and Astronautical Engineering, The Ohio State University

September- December 2010……………. Graduate Teaching Assistant, Mechanical Engineering, The Ohio State University

January- March 2011………………… Instructor, Aeronautical and Astronautical Engineering, The Ohio State University

April- June 2011…………… Graduate Teaching Assistant, Mechanical Engineering, The Ohio State University

Publications

Snyder, M.P. and Joyce, E.R., Lunar Extra-Vehicular Activities and Colonization Strategies, AIAA SPACE 2008 Conference and Exposition, September 9-11, 2008, San Diego, California. Snyder, M.P. and Joyce, E.R., Optimization of Fuselage Design for a Sounding Rocket Using Composite Materials, 45th AIAA Joint Propulsion Conference and Exhibit, August 2-5, 2009, Denver, Colorado. Joyce, E.R. and Snyder, M.P., Solid Rocket Motor Design for a High Altitude Composite Rocket, 45th AIAA Joint Propulsion Conference and Exhibit, August 2-5, 2009, Denver, Colorado. Snyder, M.P. and Joyce, E.R., Robotic Lunar Rover Design Utilizing Fuel Cell Technologies and Regolith Mitigation Strategies, AIAA SPACE 2009 Conference and Exposition, September 14-17, 2009, Pasadena, California. Joyce, E.R. and Snyder, M.P., Lunar Legislation: Strategies for Developing and Protecting the Lunar Frontier, SPACE 2009 Conference and Exposition, September 14-17, 2009, Pasadena, California. Snyder, Michael and Eric Joyce. "Development of Active Rocket Guidance at The Ohio State University: a guidance system being constructed by undergraduate students." Ohio State Engineer Spring 2009: 06. Snyder, Michael and Eric Joyce. "Luna Plaustrum: building a test prototype lunar rover." Ohio State Engineer Spring 2009: 08-10. Snyder, M.P., Joyce, E.R. and Osborne, J.C., Component Propulsion System: A New Philosophy for Exploration, Space Propulsion 2010, May 3-6, 2010, San Sebastian, Spain. Snyder, M.P., et al, A Mars Utility Vehicle Design with Incorporated Regenerative Fuel Cell Technology and In-Situ Resource Utilization, 46th AIAA Joint Propulsion Conference and Exhibit, July 25-28, 2010, Nashville, Tennessee. Joyce, E.R., et al, Design of a Versatile Regenerative Fuel Cell System for Multi-Kilowatt Applications, SPACE 2010 Conference and Exposition, August 30- September 2, 2010, Anaheim, California. Snyder, M.P., et al, Mobile Instrument for Lunar Exploration Endeavors: The Design of a Fuel Cell Powered Lunar Exploration Vehicle, SPACE 2010 Conference and Exposition, August 30- September 2, 2010, Anaheim, California. Jedrey, R., et al, Preliminary Mars Ascent Rendezvous Study, SPACE 2010 Conference and Exposition, August 30- September 2, 2010, Anaheim, California. Dunn, J., et al, 3D Metal Printing in Space: Enabling New Markets and Accelerating the Growth of Orbital Infrastructure, Space Studies Institute Space Manufacturing 2010 Conference, October 29-31, Mountain View, California.

Fields of Study

Major Field: Aeronautical and Astronautical Engineering

vii

Table of Contents

Abstract…………………………………………………………………………………... ii

Dedication……………………………………………………………………………….. iii

Acknowledgements……………………………………………………………………… iv

Vita………………………………………………………………………………………. v

List of Tables…………………………………………………………………………….. x

List of Figures…………………………………………………………………………… xi

Chapter 1: Background

1.1 Interplanetary Rover Exploration………………………………………... 1

1.1.1 Lunokhod…………………………………………………………... 1

1.1.2 Lunar Roving Vehicle……………………………………………… 2

1.1.3 Martian Rovers……………………………………………………... 3

1.1.4 Current Designs……………………………………………………. 4

1.2 Fuel Cell Technologies...... ………………………………………………. 5

Chapter 2: Requirements

2.1 Rover Design Requirements……………………………………………. 10

2.2 Mission Requirements………………………………………………….. 10

2.3 Regolith Mitigation……………………………………………………... 11

Chapter 3: Design and Testing

3.1 Design…………………………………………………………………... 13

viii

3.1.1 Base Station……………………………………………………... 13

3.1.2 Rover……………………………………………………………..16

3.2 Testing………………………………………………………………… 27

Chapter 4: Results……………………………………………………………………... 31

Chapter 5: Conclusions…………………………………………………………………37

References……………………………………………………………………………. 39

List of Tables

Table 1: Mass Break Down……………………………………………………………... 17

Table 2: Range, Payload, and Velocity…………………………………………………. 19

Table 3: Parameters for Rover and Scaled Rover………………………………………. 29

Table 4: Theoretical Results……………………………………………………………. 32

Table 5: Experimental Results………………………………………………………….. 33

List of Figures

Figure 1: General Fuel Cell Operation………………………………………………….. 6

Figure 2: Flow-Through Fuel Cell and Water Separation Schemes……………………...7

Figure 3: Non-flow-through PEM Fuel Cell……………………………………………...8

Figure 4: Base Station…………………………………………………………………... 14

Figure 5: Male Port Adapter with Nozzles……………………..………………………. 16

Figure 6: Power Allotment……………………………………………………………... 18

Figure 7: Rover Frame………………………………………………………………….. 20

Figure 8: Stress Analysis of Frame…………………………………………………...… 21

Figure 9: Track System…………………………………………………………………. 22

Figure 10: Motor Performance………………………………………………………….. 23

Figure 11: Tread Design………………………………………………………………... 24

Figure 12: Rover with Navigation System……………………………………………… 26

Figure 13: Assembled Rover Front View (Dimensions in meters)……………………... 27

Figure 14: Assembled Rover Side View (Dimensions in meters)……………………… 27

Figure 15: Scaled Tread Pod……………………………………………………………. 28

Figure 16: Regolith Test Bin……………………………………………………………. 29

Figure 17: Sinkage Test………………………………………………………………… 31

Figure 18: Pressure Contours for Helium………………………………………………. 34

Figure 19: Velocity Contours for Helium………………………………………………. 35

Figure 20: Pressure Contours for Nitrogen……………………………………………... 35

Figure 21: Velocity Contours for Nitrogen……………………………………………... 36

xii

Nomenclature

A Area b Track Width

ESA European Space Agency c Cohesion Factor

F Force k Parameter

GRC Glenn Research Center l Track Length

H Hydrogen m Mass

I Current n Valence Electrons

JPL Jet Propulsion Laboratory p Normal Pressure

LRV Lunar Roving Vehicle u Exponent of Deformation

M Molar Mass x Length

MCFC Molten Carbonate Fuel Cell v Velocity

MSL Mars Science Laboratory z Sinkage

NASA National Aeronautics and φ Internal Shearing Space Administration Resistance Angle

O Oxygen Subscripts PAFC Phosphoric Acid Fuel Cell C Pressure-Sinkage PEM Proton Exchange Membrane c Cell R External Resistance d Density RTG Radioisotope Thermoelectric Generator n Normal

SOFC Solid Oxide Fuel Cell t Track

Sol Martian Day φ Pressure-Sinkage

W Weight

xiii

Chapter 1: Background

The ability to explore and examine unknown terrain is critical to the future exploration of our solar system. Future mission plans require the ability to stay, for an extended period of time, on the surface of the planetary body that is being explored as well as gather as much information as possible for subsequent missions (1). In order to extend the overall time of exploration while simultaneously reducing the cost of a mission, robotic exploration is a vital method to examine the surface of a planetary body. Specifically, the exploration of the lunar surface cannot be fully completed without, at minimum, the aid of robotic rovers due to the extreme and hazardous environment. Currently, only one type of vehicle has remotely explored the Moon and several others have travelled to Mars. A new type of rover, capable of serving multiple types of missions, is needed. The National

Aeronautics and Astronautic’s Glenn Research Center (GRC) determined a list of requirements necessary to attain this goal. It is the intention of this thesis to satisfy all of the criteria developed in order to create the next generation of robotic lunar rover. Future manned and unmanned exploration programs will greatly benefit from information gathered by robotic operations and from the complimentary functions they can serve.

1.1 Interplanetary Rover Exploration

1.1.1 Lunokhod

In the early years of the race to land man on the Moon, the Soviet Union designed the first rover that would be used for remote exploration. This rover was designed initially in order to survey areas of the lunar surface destined for future Cosmonaut landings (2).

However due to several set-backs, including the failure of the launch of first Lunokhod mission, Lunokhod would land and would serve the purpose of being the only robotic, mobile explorer to land on the Moon (2). There were two Lunokhod rovers that landed and traversed the lunar surface. The first rover, Lunokhod 1, landed and operated for 11 lunar days (nearly three times the design life) and explored over 10 km of terrain (3).

Lunokhod 2 operated for four lunar days and explored 37 km of the surface (4).

According to Harvey, the eight wheeled rovers had a 756 kg and 840 kg mass, respectively, standing 4.42 m in length, 2.15 m in width, and 1.92 m in height. The rover utilized solid wheels and their wheel base was 2.22 m by 1.6 m. During the mission, the

Lunokhod rover was remotely controlled from Earth and could climb 20o slopes and could traverse side slopes of up to 45o. The anticipated top speed was only 2 km/hr and a radioisotope generated heated the rover during the lunar night, when the systems were in a hibernation mode (2). Lunokhod’s batteries were charged by photovoltaic cells that were covered by a dome in order to mitigate the lunar regolith contamination. The photovoltaic array was uncovered when the rover was stopped so that it could charge while the tires were not throwing the regolith into the sky and covering up on the photovoltaics (4).

Even though the speed of the rovers was comparative to a slow walk, the rovers carried scientific instruments, performed analysis, and most importantly demonstrated the ability to survive the lunar night. This would be the only robotic roving attempts to ever take place on the lunar surface. The Lunokhod missions prove that, while difficult, continual exploration is possible in an environment as harsh as the Moon. The Lunokhod roverwas limited by power since it was only meant to carry a specific payload and operate on a

1 specific mission platform. The use of solar power limited its’ exploration to lunar days, leaving a large portion of mission time in a hibernation mode in which the rover used its radioisotope generator for keeping the systems warm enough to survive.

1.1.2 Lunar Roving Vehicle (LRV)

During the later Apollo missions, Apollo 15, 16, and 17, the National Aeronautics and

Space Administration (NASA) wanted to expand the area that the astronauts were capable of conquering. One method, that was inevitably used, was to mobilize the explorers so that the rock and dust samples were obtained from a wider variety of regions. The LRV was implemented to provide the carrying capabilities required to hold two astronauts and scientific samples. Each of the four wheels was driven by independent motors and was comprised of a woven metal mesh with attached titanium chevrons which would flex and pull regolith for added traction (5). The vehicle’s power was supplied by two batteries containing 8.3 kW-hr of energy (6). The LRV was targeted to have a mass of only 181.8 kg and it was collapsible in order to fit in one of exterior quads of the lunar module that delivered it to the surface (5). The maximum payload the LRV could support was 440 kg which includes the mass of the astronauts that were necessary to operate the vehicle (6). The rover was manually driven and was incapable of remote operations. The LRV was designed as an exploration aid to ferry astronauts and lunar samples from the lunar module to an exploration site, and then back the lunar module. It was incapable of independent exploration of any type with the exception of a remote television camera that could be operated from the Earth (5). The LRV was equipped with a communications system that could transfer and receive information to and from Earth

(6). An environmental control unit enabled the rover’s subsystems to reject excess heat

2 generated while operating by radiating the heat to space when the LRV was stopped (6).

The LRV is a great example of how a large (1.83 m x 3.1 m x 1.12 m (6)) vehicle is capable of traversing the lunar environment with relative ease of operation.

1.1.3 Martian Rovers

There have been three rovers that have successfully landed and explored the Martian surface. Mars Pathfinder rover, or Sojourner, which was launched in 1996 and the Mars

Exploration Rovers (MERs), named Spirit and Opportunity that were launched separately in 2003.

Pathfinder

The 6-wheeled Mars Pathfinder rover, Sojourner, had a mass 10.6 kg and measured 68 cm long by 48 cm wide by 28 cm tall (7). A vehicle of this size was considered to be a micro-rover (7). According to Stone, the rover’s solar panel and battery system was capable of producing 1 W-hr of energy and the primary mission was only planned to have a duration of seven days. Mars Pathfinder proved that a low-cost, scientifically significant, exploratory rover could be developed and launched within three years. The rover could only reach a speed of .04 km/hr and could only traverse .5 km away from the lander (7). The rover survived for a period of approximately twelve weeks on the Martian surface (8). Pathfinder created the foundation for future programs that utilized rapid development methods by proving that the concept can work for space missions.

MERs

The six wheeled MERs require 900 W-hr per Sol or roughly 880 W-hr per Earth day to operate even though the time of operation was only a small fraction of a Sol (9). Each rover is powered by a battery recharged by photovoltaic cells (8). The rover Spirit was in

3 operation for 2010 Sols and traveled 7.73 km while Opportunity is currently on Sol 2750 and has traveled 33.93 km to date (10). The primary missions for the MERs were designed for 90 Sols and the vehicles were exposed to temperatures ranging from -100o to

0o C (8). Each MER stands at 1.4 long by 1.2 m wide by 1.5 m tall and are 176.5 kg (11).

The top speed of the rovers is 0.2 km/hr (8). The MERs proved that extended exploration is possible with no direct maintenance and prolonged exposure to harsh environments.

Methods and practices created for Mars can be used as a stepping stone for the development of a lunar rover.

1.1.4 Current Designs

Presently, there are no rovers designed specifically for use on the Moon. Several vehicles are being developed in an effort to win the Google Lunar X-Prize but unfortunately the majority of the details regarding these projects remain proprietary. However, based on the criteria of the contest, the rovers are most likely small in stature, incapable of traversing adverse terrain, and will not be on functioning on the surface for an extended period of time. This can be assumed because the main objective of the Google Lunar X-Prize is to have a robot travel 500 m across the surface while the funding has to be 90% private (12).

The funding limitation and the short range requirement indicate that a small rover with a limited amount of power will be utilized.

Current NASA and European Space Agency (ESA) efforts are focused on Martian exploration. NASA has launched the Mars Science Laboratory (MSL) and NASA and

ESA are working together to eventually launch the vehicle containing the ExoMars rover.

There are no known plans for a rover utilizing a fuel cell power source.

Mars Science Laboratory

The MSL mission features a large rover named Curiosity. According to NASA’s Jet

Propulsion Laboratory (JPL) the rover measures 3.05 m long, 2.74 m wide, and 2.13 m tall. The rover weighs 900 kg and contains a mobile geology lab, several cameras, and a laser capable of vaporizing rock particles for analysis. The primary mission is designed to last one Martian year and will explore the surface for signs of past microbial life and if conditions presently exist below the surface for life to be present today (13). The rover is planned to traverse 200 m per Martian day and be powered by a Radioisotope

Thermoelectric Generator (RTG) which was supplied by the United States’ Department of Energy (14). This generator produces a continuous 110 W to power the rover’s systems (14). If this mission is successful, it will mark the first time that a large, unmanned rover will have successfully explored a non-terrestrial body.

ExoMars

The ExoMars rover is a joint mission between NASA and ESA. The rover is currently in development and will be six wheeled, solar powered, and able to autonomously cover

100 m per Sol (15). Although the details of the mission are incomplete, it can be assumed that the rover will be of MER stature.

1.2 Fuel Cell Technologies

Fuel cells are devices that generate direct current electrical energy from chemical reactants (16). The convention is for two reactants to combine within a chamber that contains an anode and cathode which enables a flow of electrons, creating power (16).

Figure 1 shows the general operation.

Figure 1 (16): General Fuel Cell Operation

There are four major types of fuel cells that are currently being popularly used. They are:

Proton exchange membrane (PEM), Solid oxide fuel cell (SOFC), Molten carbonate fuel cell (MCFC), and Phosphoric acid fuel cell (PAFC) (16). Joyce et al affirm that PEM fuel cells have several advantages compared to other types of fuel cells. PEM fuel cells are simple in design and construction. Once a power demand is realized, the cells can easily be stacked on top of one another or placed in series to supply the required power. PEMs also have short start up times in comparison to the PAFC and the SOFC but longer start up times compared to the MCFC (16). The operating temperature of PEM fuel cells are lower than the other mentioned types (17), making integration into an overall system easier. Less mass and fewer design considerations would need to be devoted to thermal control and management when using PEM fuel cells. There are two design possibilities

6 with PEM fuel cells, flow through and non-flow through which are depicted in Figure 2 and Figure 3 respectively. Also depicted in Figure 2 are the two types of water separation for flow-through fuel cells, active and passive.

Figure 2 (16): Flow-Through Fuel Cell and Water Separation Schemes

The active and passive water separation methods both circulate a reactant, which is excess in the power producing reaction, along with the product to a separator. The remaining reactant is then put back through the cell. In the case depicted, hydrogen and oxygen are the two reactants. The excess reactant in a hydrogen-oxygen fuel cell is oxygen and the circulating oxygen removes water from the cathode surface (16).

Figure 3 (16): Non-flow-through PEM Fuel Cell

The water produced in a non-flow through PEM fuel cell is transferred by a hydrophilic membrane and stored with no circulating of the reactant (16). A non-flow through fuel cell is lighter and less complex due to the elimination of the water and reactant separator.

Non-flow through fuel cells tend to be more expensive, partially because most active research in fuel cells is in the automotive industry which use oxygen from the atmosphere are flow-through in design. Fuel cell technologies have been utilized in many space programs including project Gemini, project Apollo, and as recently as the Space Shuttle

(18). A regenerative fuel cell system has the capability of supplying electrical power to a sustained, long duration mission due to the ability to recycle products back into their

8 original reactants in order to reduce the amount of mass needed to be carried by the powered system (16).

Chapter 2: Requirements

2.1 Rover Design Requirements

The requirements set for this rover evolved from a study performed at NASA’s GRC detailing the challenges along with existing conceptual designs (19) and funded through a grant from the Ohio Aerospace Institute. The guidelines were developed during meetings with scientists at GRC and are as follows: 1 kilowatt nominal power supply without using batteries or RTGs, greater than 500 kilogram carrying capacity, minimum of 250 kilometer range, five year mission design life, and continuous operation. A five year design life requires that the rover survive for approximately 62 lunar days. Power will need to be available to supply a scientific payload that will be selected after the rover is designed and developed. The rover, therefore, will need to be a lunar mule, capable of carrying a variety of payloads to explore a wide range of possible landing sites. Another important constraint placed on the development is that the rover be designed utilizing current technologies. There can be no advanced research and development on any component involved with the construction. No current designs, already in operation or in development, utilize a hydrogen-oxygen fuel cell, making the off-the-shelf requirement significantly more difficult.

2.2 Mission Requirements

The primary mission requirement is that the rover be capable of traversing the lunar surface, both mare, or darker areas, and highland regions. The mare regions are flat, relative to the mountains and valleys of the highland areas however varying sized craters

10 are present in both (20). In order to navigate these terrains it is recommended all lunar roving vehicles be able to climb and descend a slope of 25o since craters commonly have only 5-10o slopes and this leaves a significant margin of mission flexibility (21). Any successful mission exploring that lunar surface will have to navigate and investigate these different regions effectively and without constant reroute to avoid obstacles that commonly occur. Another important aspect in the mission is to continually survive the lunar night and the temperature flux that occurs during the transition between night and day. During lunar night, which lasts for approximately 14 Earth days, the temperature can reach as slow as -150o C and can reach 100o C during the day (22). In addition to the electrical power dedicated to lighting the rover’s way, a significant amount of heat needs to be generated to keep vital components operating during night.

In order for this design to be feasible, the rover must be able to fit on a conceivable lunar lander. The recently cancelled Ares V rocket was designed to be compatible with a lander capable of supporting any future lunar landing attempt with the rover. The complete regenerative fuel cell system could be integrated on the rover, but in order to optimize performance, several components can be separated into a base station and operate in parallel with the rover systems. Solar activity will require protection to prevent damaging effects during the rover’s operation. Radiation, coronal mass ejections, and magnetic anomalies will have to be considered in order for the rover to survive its five year primary mission.

2.3 Lunar Regolith Mitigation

The top most layer, or regolith, of the lunar landscape is covered with a powder that can range in depth from several meters to a dusting. This layer is composed of pulverized

11 rock that has accumulated over billions of years from the constant impacts of projectiles on the lunar surface. The highland areas have smaller sized regolith particles when compared to the mare areas (23). The smaller sized particles make the risk of sinking into the surface and, consequently, allowing a vehicle’s wheels or tracks to become stuck more probable. Areas classified as “soft soil” create hazards for vehicles as demonstrated during the Apollo 15 mission where the LRV got stuck and, due to the low weight in the lunar environment, the astronauts simply moved the rover to fix the problem (24). This type of fix cannot be performed in a robotic mission so other mitigation factors must be used to counter this possibility.

Chapter 3: Design and Testing

3.1 Design

Robotic aids must be developed to conduct a wide variety of mission-related tasks if an initiative is taken to explore the lunar surface in the future. The mission’s objectives set requires a versatile vehicle with a regenerative fuel cell system. The regenerative system will be split up into two main components in order to maximize the rover’s performance.

The rover will contain the fuel cell system consisting of the following components: an oxygen storage tank, a hydrogen storage tank, and a water storage tank. The rover will be equipped with the primary communication system and all of the exploration and driving equipment with the exception of base station instrumentation and cameras.

3.1.1 Base Station

The split architecture of the regenerative fuel cell system requires that a base station contain the following components: electrolyzer, storage tanks for the generated hydrogen and oxygen, solar panels for powering the electrolyzer, and secondary communications systems. There will be a transfer dock on both the rover and base station that will enable the transfer of water generated from the rover’s fuel cell to the base station and the hydrogen and oxygen from the base station to the fuel cell storage tanks.

Power System

The base station, depicted in Figure 4, is 4.5 m long, 6m wide and 4m tall and will be powered by four strings of solar panels containing two sets of cells each. They are

13 extended from the core of the base station upon landing and the radiators, which are used to dissipate excess system heat, are located beneath the mounting positions of the strings.

BP SX3200 200 W photovoltaic cells were used for analysis. The photovoltaic cells are multi-crystalline silicon and are 0.16 x 0.16 m each. The cells produce 0.49 V at 8.16 A at their maximum power output capability. The cells have an 11.9 % efficiency and the solar arrays need to have a combined total of 32 m2 area in order to power the electrolyzer and environmental control systems (25). The accepted average solar flux at the lunar surface, which is approximately 1 astronomical unit away from the sun is 1366 W/m2 and varies by 3.3% due to the eccentricity of Earth’s orbit around the sun (26). This allows for the solar array area to produce up to 1801 kW-hr of energy, which is 29.5% greater than the 1270 kW-hr required for continuous surface operation. These power consumption values are based on the performance of the commercially available Giner

PEM electrolyzer (27). This allows the electrolyzer to be operated during the lunar night, at a range of solar flux conditions, and will power the systems after the solar cells’ performance degrades. Degradation occurs with prolonged operation, non-optimum sun incidence angles, and by contamination cause by dust particles settling on the surface of the panels.

Figure 4: Base Station

The solar panels will charge a lithium-ion polymer battery that has a capacity of 11 kW- hr. This will enable the battery to be capable of storing all of the energy required to run the vital systems during a lunar night. The battery has a wide range of operational temperatures (213.15 K to 253.15 K), weighs approximately 81.5 kg, and occupies a volume of 0.032 m3 inside the base station (28).

Electrolyzer

Electrolyzers are devices that reduces a compound into component elements (16). An electrolyzer is needed to form a regenerative power system using a hydrogen-oxygen fuel cell. According to Joyce, during electrolysis, a solution is subjected to direct current and is decomposed. The direct current is supplied by the batteries that have been charged by the solar panels on the base station. The hydrogen and oxygen combine in the fuel cell to create energy and the byproduct of this reaction is water (16). In the electrolyzer, the water is decomposed such that at the negative electrode, an external electrical supply provides the electrons and the protons are removed from the source such that:

+ - 4H + 4e 2H2 and at the positive electrode:

+ - 2H2O O2 + 4H + 4e which completes the reaction and the oxygen (O) and the hydrogen (H) are collected and pressurized into storage tanks (29). Due to the high operating pressure of the electrolyzer, once the gases are decomposed, they can be slowly pumped up to the storage tanks’ pressure in a reservoir prior to being fed into the tanks. The operating pressure of the electrolyzer is 60% of the pressure of the rover and base storage tanks (27).

Docking Station

The base station must contain a docking station that supports the refueling of the rover and discharge of the rover’s water tank. This docking station must contain three ports, one each for the hydrogen, oxygen, and water, and these ports must be protected from contamination due to the regolith. In order to mitigate this contamination, a series of nozzles will be placed along the male docking mechanism as depicted in Figure 5. An inert gas stored in the rover, will be blown through the nozzles in an effort to clear any dust that may be present on both the male and female port adapters before insertion and transfer begins.

Figure 5: Male Port Adapter with Nozzles

3.1.2 Rover

The rover is designed to operate with a 1 kW hydrogen-oxygen fuel cell and have a mass of 750 kg. Due to the mass of the vehicle, tracks are used for propulsion and this enables a range of 550 km. The rover must operate continuously for five years and have the capability of performing scientific experiments and conducting extensive exploration of

16 the lunar surface. The 750 kg mass will be broken down as shown in Table 1. All masses were taken from CAD and commercially available components of the rover.

Table 1: Mass Break Down

Item Mass (kg) Structure 208.5 Fuel Cell System 6.9 Storage Tanks (30) 135 Reactants 128.24 Drive System 201.5 Nav/Comm Systems 50 Environmental Protection 20 Total Dry Mass 750.14

Fuel Cell

The rover is designed to operate nominally at 1 kW and have a maximum range of 550 km. This distance allows for the rover to travel over 250 km from the base station, as set in the design requirements, and gives a 10% performance margin. The fuel cell design constraints were determined by using an ElectroChem 1 kW nominal, 6 kW peak, hydrogen-oxygen fuel cell that is commercially available. The fuel cell will be non-flow through in order to reduce the amount of complexity in the system. This fuel cell’s stack

2 contains 45 cells that have an area, Ac, of 0.023 m and a current density, Id, of 0.1 A per cm2 (16). The total mass flow rate of the reactants, m , can be determined by Equation 1 which uses a modified version of Faraday’s Law (25).

Where F is Faraday’s constant, M is the molar mass, and n is the reaction valence electron number. Using a reaction efficiency of 50% (16) the mass flow rate for the

17 reactants was found to be 2.22 x 10-4 kg/s which yields an oxygen mass flow rate of 2.08 x 10-4 kg/s and a hydrogen mass flow rate of 1.35 x 10-5 kg/s. The power allotment for the rover’s systems is shown in Figure 6.

Figure 6: Power Allotment

The total mass of reactants that will be used is determined by multiplying the mass flow rate by the total time the rover will be in use. The maximum velocity of the rover must be determined in order to find the total time the rover will take to traverse its maximum range. Using the kinetic energy, combined with the energy required to overcome friction, the velocity can be determined using Equation 2.

Where m is the mass of the rover, v is the velocity, Fn is the normal force, and xt is the length of contact of a track. In order to overcome surface roughness and soil compaction, the rover is assumed to be going up a 1.5o instead of a flat terrain (24). The energy that goes towards driving is limited to only 750 W and that yields a velocity of 0.95 m/s. The amount of time it would take to cover the 550 km range is therefore 5.789 x 105 s. Table

2 compares maximum velocity and range of the rover for varying payloads up unto the design payload of 500 kg, making the total rover mass 1250 kg.

Table 2: Range, Payload, and Velocity

Range (km) Payload (kg) Velocity (m/s) 550 0 0.95 515 100 0.89 486 200 0.84 463 300 0.80 440 400 0.76 423 500 0.73

Totals of 120.42 kg of oxygen and 7.82 kg of hydrogen are required for the fuel cell to operate over this time. Composite tanks can store gases at pressures exceeding 34.52

MPa making them an acceptable choice for the storage of the hydrogen and oxygen (25).

The operating pressure of the PEM is approximately 2.75 MPa allowing for the reactant storage tanks to bleed into the fuel cell without having to increase the pressure (31).

Using the storage pressure of 34.52 MPa, the oxygen tank has a volume of 0.24 m3 and the hydrogen tank has a volume of 0.25 m3. These volumes were determined using the

19 perfect gas law at an operating temperature of 270 K. Using the standard density of water of 1 g/cm3, the water storage tank is 0.13 m3.

Structure

The primary structure of the vehicle is 6011T aluminum. Figure 7 is the basic frame used for structural analysis. Using ANSYS finite element analysis software, loads were placed

Figure 7: Rover Frame

on the frame from all of the component’s mass at their points of attachment to the frame.

The maximum predicted payload of 500 kg was also added to the top of the rover frame for the analysis. A distributed load for the internal components was placed on the bottom section of the frame to account for the floor plate that will distribute the load on the actual rover. The loads under lunar gravity were applied with a factor of safety of five.

This factor of safety will ensure that the rover can operate without failure under any driving condition and possible drops from short distances. Figure 8 is the stress analysis

20 performed on the frame. The analysis shows that the structure does not fail under the expected loading conditions. External shielding, in the form of aluminum and lead

Figure 8: Stress Analysis of Frame

plating, is added to the frame to limit the amount of radiation exposure to the internal components of the rover. This plating also further strengthens and stiffens the structure by their addition. An additional non-structural element, is multiple layers of nickel foil with a KEL-F coating. KEL-F is a heat resistant material capable of absorbing temperature of over 477 K, which is above any temperature to which the rover is anticipated to be exposed (32). The multiple layers of aluminum, lead, and nickel foil should also protect against space weather damage.

Drive System

The drive system for the rover consists of four “tread pods” that are mounted to each corner of the frame. Each tread pod contains two separate track systems as shown in

Figure 9. The track system has one drive wheel in which the sprocket that will drive the

21 actual track is internal to the radius of the wheel. This is a regolith mitigation strategy enabling the exterior covering of the tread pods to extend past the tracks in the upper portion, limiting exposure. Tension rollers are on the top and bottom of the track assembly and their role is to keep tension in the track to minimize slippage in the track belt. Reducing this slippage will increase the overall efficiency of the vehicle as the

Figure 9: Track System

vehicle loses power in the transmission from track to soil when slip increases (33). The road wheels are used to transfer the majority of the vehicles weight through the track belt and onto the surface while the idler wheels keep the shape of the track and freely rotate.

The entire drive train is enclosed to mitigate contamination. The drive wheels are each driven by a motor comparable to the MB-EMU-75-2Q produced by Allied Motion. Using

Figure 10 and the diameter of the drive wheel, 0.254m, the gear ratios needed for the

22 transmission can be determined. Gearing the motor’s speed only affects the efficiency of the drive system by a 2% reduction (33). Based on Figure 10, the gear ratios are calculated to be ten, eight, seven, and six. These gear rations allow for the entire envelope of operations. The transmission will be directly mounted onto the motor inside the enclosed drive assembly.

Figure 10 (34): Motor Performance

Tracks were chosen for the drive system because they are more efficient at transferring load to soil than a wheel and are less probable to get stuck in soil when compared to a wheel based system (35). The increased complexity and, therefore, a higher risk of failure has been countered in the design by robustness and redundancy. Each track system in the tread pod is capable of detachment if it would bind or fail, allowing a decrease in friction with the surface and not making the remaining functional track systems drag the disabled system.

The rover will be steered by both skid steering, where one side’s tracks operate in one direction while the opposite side moves in the reverse direction, and by linear actuators moving the tread pods. An active suspension system is unnecessary due to the low speed and unlikelihood of shocks from drops. In order for drops to occur that would justify the need for an active suspension system, the navigation system would have to suffer a primary and secondary failure, resulting in an immediate mission failure.

The track’s treads, Figure 11, were designed to be non-binding by using rounded

Figure 11: Tread Design (Dimensions in meters)

attachment joints. The tread pattern was iteratively developed through experimentation.

Various tread patterns were constructed and then tested using JSC-1a lunar regolith simulant. The designs were ranked based upon the angle of tilt that the treads would need to obtain before the regolith fell off and the ease with which the tread broke the surface of the regolith. The initial design that ranked the highest was then modified until an optimum design was determined.

Navigation and Communication

The rover will need to have an on-board navigation and communication system to allow continuous exploration operations. Since the command and feedback time from the Earth to the Moon is very small, approximately three seconds, a completely remote operation is possible. However in order to utilize all of the potential exploration capabilities, the rover must also be able to travel autonomously. The lack of both a substantial atmosphere and objects that could be used as reference for distances on the Moon, a navigation system for manual and autonomous operations must include a means to measure obstacles and terrain. A LiDAR system that enables the mapping of terrain for a full 360o field of view will be implemented to generate an interface that can be used to drive the rover (36).

Using a LiDAR system coupled with a Pathfinder type navigation system, the rover can be navigated using only 95 Watts of power (7) (36). There will be two complete and identical navigation systems for the sake of redundancy. This system contains three video cameras, two located on the front of the rover and one on the back to enable driving in both forward and reverse. The assembled rover complete with navigation systems

(LiDAR represented by white cubes) is shown in Figure 12.

Figure 12: Rover with Navigation System

Communications with Earth will be transmitted directly to and received directly from the rover. This eliminates the need for a relay system and prevents the line of sight restrictions that plagued the Pathfinder system (7). The communication system will contain high and low gain antennas to maximize data transfer and efficiency. A system with similar power as the MSL is more than sufficient for communicating with Earth

(13).

In order to navigate during lunar night or at low lighting conditions, high intensity lights are needed. ARC HID lights will be utilized and placed on the back and front of the rover. Four lights will be placed on the front and two on the back of the rover. Each light uses 10 W and produces 550 lumens while only weighing 0.19 kg each (37). One light is capable of providing sufficient illumination to support visual driving. Figure 13 and 14 are dimensioned CAD drawings of the completely assembled rover.

Figure 13: Assembled Rover Front View (Dimensions in meters)

Figure 14: Assembled Rover Side View (Dimensions in meters)

3.2 Testing

Manufacturing this rover on a full scale would be an expensive and difficult task.

Budgetary restraints alone prohibit this from being a reality during this project. In order to test the fundamental drive concepts, a scale model must be used. Since the availability

27 of a lunar environment, approximately 1/6 Earth’s of gravity, is not feasible for the testing required, all performance characteristics were calculated for Earth conditions as well. The dimensions of the rover were reduced by 84% for the scale model. The scaled mass was determined by taking the full scaled rover’s mass to tread area ratio (normal pressure) and keeping it consistent. A single tread pod was constructed out of

Lynxmotion Tri-Track chassis and was used for testing. All values that were determined were simply multiplied by four to obtain values for a full rover. Figure 15 is a picture of the tread pod that was tested. The power used for the scale model was then determined by maintaining a constant power to mass ratio between the full scale rover and the scale model. Due to the structural loading limitations of the scaled tread pod, only the empty condition could be tested.

Figure 15: Scaled Tread Pod

Table 3 is a list of parameters for the scaled and non-scaled, non-loaded rovers. The normal pressure value has the contact factor parameter in the calculation. This factor is the number of tracks systems multiplied by the area of the track and then multiplied by the number of track links in contact with the ground. There are 10 links in contact with the terrain for both the rover and scaled rover.

Table 3: Parameters for Rover and Scaled Rover

Parameter Rover Scaled Rover Mass (kg) 750 48 Weight(kN) 7.36 0.41 Track Contact Length (m) 1.52 0.25 Track Contact Width (m) 0.31 0.05 Track Area (m^2) 3.77 0.13 Normal Pressure (kN/m^2) 0.41 0.41 Power (kW) 0.75 0.041 Power to Weight Ratio 0.10 0.10

A 1 m long x 1 m wide x 0.25 m deep sand pit was constructed at the Ohio State

University’s Aeronautical and Astronautical Research Laboratory in order to complete testing. Sand is an inexpensive regolith simulant. It had many of the same soil-mechanics characteristics and is ideal for testing the scaled rover (25). A bin of lunar regolith simulant, Figure16, was used to investigate how regolith contamination affects components and materials.

Figure 16: Regolith Test Bin

The test pit verified aspects of the rover that had been predicted analytically. Drawbar pull and sinkage both were determined experimentally in the pit. Several analytic results

29 were unable to be verified experimentally include: work performed compacting terrain, towage work, and traction effort. A 222.4 N maximum load digital scale was used to determine drawbar pull on the rover.

Computational work was performed with the structure (see previous section) and the mitigation nozzles. The nozzles were analyzed using computational fluid dynamics to determine if the effect was great enough to disturb dust that had settled on the surface of the docking mechanisms.

Chapter 4: Results

The sinkage, z, was found by Equation 3 (33).

( ) (3

Where p is the normal pressure, b is the width of track, u is the exponent of terrain deformation, and both kc and kφ are pressure-sinkage parameters. The pressure-sinkage parameters are working medium specific and dry sand was used in all cases. Kc and kφ are given the constant values of .99 and 1528.43 respectively. The sinkage depth was tested by placing the scaled tread pod on the surface, then removing it and measuring the depth of the impression left in the sand. Figure17 depicts one of the tests of sinkage.

Figure 17: Sinkage Test

The work done to compact the terrain so that the tracks make a rut through the sand is given by Equation 4 (33), where l is the length of the track.

( ) ( ) (4

The tractive effort of the rover was determined by Equation 5 (33).

( ) (5

Where c is the cohesion factor, 1.04, and φ is the angle of internal shearing resistance which is 28o. Once the tractive effort was determined, the drawbar pull was calculated using Equation 6 (33).

R is the external resistance of the rover which is negligible due to the low weight and slow speed of the scaled rover. The scaled rover must move at the slowest possible speed so that inertial influences are minimized (38). Finally the work done by towing was determined using Equation 7 (33).

( ) ( ) (7

( )( )

The results from the calculations are shown in Table 4.

Table 4: Theoretical Results

Theoretical Result Rover on Moon Rover Scaled Rover Drawbar Pull of Rover (N) 308.53 1848.7 19 Sinkage (mm) 3.03 15.48 6.71 Work to Compact (J) 533.54 16320.4 453.81 Towing Work (J) 810.98 24807.04 213.69

The experimental results from the testing of the scaled rover are found in Table 5. The sinkage depth was estimated as no precision method for determination was available.

Table 5: Experimental Results

Run 1 2 3 4 5 Average Sinkage (mm) 5 3 5 5 5 4.6 Drawbar Pull (N) 16.68 15.79 16.37 14.46 15.97 15.85

The error between the theoretical and the average experimental results for sinkage was

32%. This can be attributed to no source of precision measurement and no way of determining if the depression distance on the sides, where it was measured, was comparable to the rest of the depression. Shear forces in the sand on the edges would be greater and would resist deformation if the sides of the pit are not far enough away from the test space. The large space needed to verify this would require facilities and resources outside of the possibility of this project.

The error between the theoretical and the average experimental results for drawbar pull was 16.6%. The sources for error in this experiment can be attributed to a wide range of sources. The power inputted to the rover and transmitted to the drive motor was likely lower than anticipated. There was also assumed no track slip, which, even in minor instances of approximately 1% could cause significant loss of available power. The other probable source of error would be the assumptions made on the constants involving the working medium. The calculations were based on dry sand, which, with any environment with any humidity would be impossible to keep dry without a sophisticated environmental control system.

The nozzles for the docking ports were analyzed using the CFD software Fluent.

Nitrogen and helium were used as a working fluid for this regolith mitigation strategy.

Both gases were analyzed at the same boundary and operating conditions. The reservoir pressure was 34 MPa with the exit pressure being a vacuum. The initial mass flow rate was set to .001 kg/s. The total pressures and the velocities at the exit were examined in

Figures 18- 21.

Figure 18: Pressure Contours for Helium

Figure 19: Velocity Contours for Helium

Figure 20: Pressure Contours for Nitrogen

Figure 21: Velocity Contours for Nitrogen

Based on the Fluent CFD simulations, the helium gas has a higher exit pressure and higher exit velocity than the nitrogen gas. The higher exit pressure and higher exit velocity are desirable for use in a regolith mitigation system. This will displace the maximum amount of regolith for a given volume of inert gas. The rover will utilize helium as the inert gas. This should provide the necessary force needed to clear the connections while not adding significant weight to the system.

Chapter 5: Conclusions

The rover designed can navigate the lunar surface environment for at least five years using technologies that have already been developed. The 1 kW of power supplied by the hydrogen-oxygen fuel cell can drive the rover, communicate with Earth, navigate, and supply power for environmental controls and payloads while operating nominally. The non-flow through fuel cell requires masses of hydrogen and oxygen that are capable of being carried by the rover to allow it to carry out its 550 km required maximum range.

The rover can carry an additional 500 kg of payload on top of the 750 kg dry mass.

Regolith mitigation strategies were implemented in order to prolong the life of the rover and allow more desirable operating conditions of the systems.

Results from the experiments indicate that the lunar rover scaling is possible but research must be continued into finding accurate parameters to be used in the theoretical calculations. A larger testing facility would have also benefitted the testing.

Theoretical results compared the lunar rover with a scaled prototype and the performance required by the scaled rover can be accomplished using commercial off-the-shelf components with minor modifications. The CFD works verifies that the inert gas nozzle mitigation strategy is feasible and should be implemented in systems that care likely to be contaminated by regolith.

Hydrogen and oxygen fuel cell technologies are a viable option for future rover designs.

Their performance is comparable to solar and RTG power systems that currently are

37 used. The regenerative fuel cell concept will enable long duration exploration and influence the development of future manned and unmanned mission architecture.

Bibliography

1. Exploration Systems Architecture Study. s.l. : NASA-TM-2005-214062, 2005.

2. Harvey, Brian. Soviet and Russian Lunar Exploration. Chichester : Praxis, 2007.

3. Luna 17/ Lunokhod 1. NASA.gov. [Online] NASA. [Cited: March 15, 2011.] http://nssdc.gsfc.nasa.gov/nmc/masterCatalog.do?sc=1970-095A.

4. Luna 21/Lunokhod 2. NASA.gov. [Online] NASA. [Cited: March 15, 2011.] http://nssdc.gsfc.nasa.gov/nmc/masterCatalog.do?sc=1973-001A.

5. Young, Anthony. Lunar and Planetary Rovers. Chichester : Praxis, 2007.

6. Boeing. Lunar Roving Vehicle Operations Handbook. Huntsville : NASA, 1971.

7. Stone, H. W. Mars Pathfinder Microrover a Small, Low-Cost, Low-Power Spacecraft,

NASA ID-20060036290, 1995.

8. NASA (January 2004). "Mars Exploration Rover Landings". Press Kit.

http://marsrover.nasa.gov/newsroom/merlandings.pdf. Retrieved 2010-1-23.

9. Snyder, M.P., et al, A Mars Utility Vehicle Design with Incorporated Regenerative Fuel Cell Technology and In-Situ Resource Utilization, 46th AIAA Joint Propulsion

10. Mars Exploration Rovers. Update: Spirit and Opportunity. [Online] NASA, October 19, 2011. [Cited: October 21, 2011.] http://marsrover.nasa.gov/mission/status.html.

11. Lindemann, Randel A. Mars Exploration Rover Mobility Assembly Design, Test and Performance. s.l. : Jet Propulsion Laboratory, 2005. 05-1950.

12. Prize Details. Google Lunar X-Prize. [Online] Google, 2010. [Cited: September 25, 2011.] http://www.googlelunarxprize.org/prize-details.

13. Mars Science Laboratory. Mission. [Online] NASA Jet Propulsion Laboratory, 2011. [Cited: September 29, 2011 .] http://marsprogram.jpl.nasa.gov/msl/mission/rover/.

14. Mars Science Laboratory Fact Sheet. s.l. : NASA, 2011. JPL 400-1416.

15. ExoMars Rover. Robotic Exploration of Mars. [Online] ESA, October 7, 2011. [Cited: October 15, 2011.] http://exploration.esa.int/science- e/www/object/index.cfm?fobjectid=45084.

16. Joyce, E.R., et al. Design of a Versatile Regenerative Fuel Cell System for Multi- Kilowatt Applications, SPACE 2010 Conference and Exposition, August 30- September 2, 2010, Anaheim, California.

17. Department of Energy. Multi-Year Research, Development and Demonstration Plan-Fuel Cells. 2007.

18. Halpert, Gerald, et al. Batteries and Fuel Cells in Space. The Electrochemical Society Interface. Fall, 1999.

19. Zakrajsek, James J., et al. Rover design concepts and challenges.1st Space Exploration Conference, January 30- February 1, 2005, Orlando Florida. AIAA– 2005–2525.

20. United States Geological Survey. Lunar Terrain Mapping and Relative-Roughness Analysis. 1971. 599-G.

21. Kring, D.A., et al.. A Rover Based Strategy for the Roboitc and Human Phases of the Lunar Exploration Initiative. 2007.

22. Schmitt, Harrison H., Return to the Moon: Exploration, Enterprise, and Energy in the Human Settlement of Space, Praxis Publishing Ltd, New York, 2006.

23. Maps of Maturity-Correlated Parameters of the Lunar Regolith. Kaydash, V. et al. s.l. : Microsymposium 38, 2003. MS043.

24. Lunar Mobility Review. Kring, David A. s.l. : Lunar Exploration Initiative, 2006.

25. Snyder, M.P. and Joyce, E.R., Robotic Lunar Rover Design Utilizing Fuel Cell Technologies and Regolith Mitigation Strategies, AIAA SPACE 2009 Conference and Exposition, September 14-17, 2009, Pasadena, California. 26. Pisacane, Vincent L. The Space Environment and Its Effects on Space Systems. Reston : AIAA, 2008.

27. Military and Aerospace Electrolyzers (OGP/LPE). Products. [Online] Giner Electric, 2011. [Cited: May 2, 2011.] http://www.ginerinc.com/products.php?a=OGP.

28. Lithium Polymer Batteries. Products. [Online] PowerStream Technologies, 2011. [Cited: January 6, 2011.] http://www.powerstream.com/li-pol.htm.

29. Spiegel, Colleen. Designing & Building Fuel Cells. New York : McGraw-Hill, 2007.

30. Hydrogen Composite Tank Program. Sirosh, Neel. s.l. : U.S. DOE Hydrogen Program Review, 2002. NREL/CP-610-32405.

31. Burke, Kenneth A. High Energy Density Regenerative Fuel Cell Systems for Terrestial Applications. s.l. : NASA. 1999-209429.

32. Schultz, A.W. Thermal Conductivityof Teflon, KEL-F, and Duroid 5600. Watertown : Watertown Arsenal, 1958. 951661.

33. Wong, J. Y. Theory of Ground Vehicles. Hoboken : Wiley, 2008.

34. Allied Motion Technologies. Brushless DC Motor-Drive EnduraMax 75s Series. [Specifications Sheet] 2011.

35. Turnage, G. W. Performance of Soils Under Track Loads Report 1. Springfield : U.S. Army Material Command, 1971. M-71-5.

36. Velodyn. High Definition LiDAR HDL-32E. [Data Sheet] 2001.

37. Light & Motion ARC NiMH Lights. mtbr. [Online] 2011. [Cited: September 5, 2011.] http://www.mtbr.com/cat/accessories/lights/light-and-motion/arc- nimh/prd_358006_130crx.aspx.

38. Scale-Model Testing of Soil-Vehicle Systems. Hetherington, John G. 2, s.l. : Journal of Battlefield Technology, 2004, Vol. 7. 1440-5113.