Experimental Development of Orc

EXPERIMENTAL DEVELOPMENT OF ORC

TECHNOLOGY FOR WASTE HEAT-TO-

POWER CONVERSION

Chinedu Kingsley Unamba

Supervisor: Professor Christos N. Markides

Clean Energy Processes (CEP) Laboratory

Department of Chemical Engineering

Imperial College London

This dissertation is submitted for the degree of Doctor of Philosophy

August 2020

Dedicated to my loving parents.

DECLARATION

I hereby certify that, to the best of my knowledge, the content of this thesis is original and the results of my own work. All of the assistance I received and the sources I used have been properly acknowledged.

Unamba Chinedu Kingsley, April 2020

The copyright of this thesis rests with the author and is made available under a Creative Commons Attribution Non-Commercial No Derivatives licence. Researchers are free to copy, distribute or transmit the thesis on the condition that they attribute it, that they do not use it for commercial purposes and that they do not alter, transform or build upon it. For any reuse or redistribution, researchers must make clear to others the licence terms of this work.

iii

ACKNOWLEDGEMENTS

I would like to acknowledge Professor Christos Markides who has been a mentor, an inspiration and has given me so much help.

Also, Paul Sapin who contributed so much to experimental design and data analysis, Xi and Jian who helped a lot to experimental work and my colleagues James, Ahmad, Niyi, Xi, Jian, Kai, Alex, Mohammad, Michael who helped in their invaluable ways.

PUBLICATIONS

Journal publications

C.K. Unamba, P. Sapin, X. Li, J. Song, K. Wang, G. Shu, H. Tian and C. Markides, “Operational Optimisation of a Non-Recuperative 1-kWe Organic Rankine Cycle Engine,” Applied Sciences, vol. 9, p. 3024, 2019.

C.K. Unamba, M. White, P. Sapin, J.P. Freeman, S. Lecompte, O.A. Oyewunmi, C.N. Markides, “Experimental investigation of the operating point of a 1-kW ORC system,” Energy Procedia, vol. 129, pp. 875 – 882, 2017.

Conferences

J. Freeman, I. Guarracino, C. Unamba, O. Oyewunmi, N. Le Brun, M. White and C. Markides, “Developing a test bed for small-scale ORC expanders in waste- heat recovery applications,” in 3rd Engine ORC Consortium (EORCC) Workshop, Belfast, Northern Ireland, 14-16 September 2016.

C.K. Unamba, X. Li, J. Song, K. Wang, G. Shu, H. Tian, P. Sapin, C.N. Markides, “Off-design performance of a 1-kWe organic Rankine cycle (ORC) system,” in 32nd International Conference on Efficiency, Costs, Optimisation, Simulation and Environmental Impact of Energy Systems (ECOS), Wroclaw, Poland, 2019.

Poster

C.K. Unamba, A. Najjaran, J.P. Freeman, M. Herrando, C.N. Markides, “High- efficiency hybrid PV and solar-thermal combined cooling and power technologies,” Energy Future III (EF 3) Conference, 5-7 February, Sydney, Australia, 2018.

ABSTRACT

Several heat-to-power conversion technologies are being proposed as suitable for waste-heat recovery (WHR) applications, including thermoelectric generators, hot-air (e.g., Ericsson or Stirling) engines, and vapour-cycle engines such as steam or organic Rankine cycle (ORC) power systems. The latter technology has demonstrated the highest efficiencies at small and intermediate scales and low to medium heat-source temperatures and is considered a particularly suitable option for WHR in relevant applications. However, ORC systems experience variations in performance at part-load or off-design conditions, which needs to be predicted accurately by empirical or physics-based models if one is to assess accurately the techno- economic potential of such ORC-WHR solutions.

This thesis reports on the design and construction of a small-scale (1-kWe) ORC prototype and examines its operation over a range of conditions. It presents results from a series of experimental tests and related investigations into the part-load performance of the ORC engine, operated with R245fa as a working fluid, with the aim of producing high-fidelity steady-state and transient data relating to the operational performance of this engine. The experimental apparatus is composed of a rotary-vane pump, brazed-plate evaporator and condenser, and a scroll expander with an adjustable load. An electric heater is used to provide a hot oil-stream to the evaporator, supplied at different temperatures, from 100 °C to 140 °C.

The optimal operating conditions, i.e., pump speed and expander load, are determined at various heat-source conditions and the steady-state data points used to analyse the part-load performance of the engine. A performance map is drawn that captures a component scale performance

vii

and the overall system performance. An exergy analysis allows us to quantify the contribution of each component to the overall exergy destruction. The data can be used for the development and validation of advanced models capable of steady-state part-load and off-design performance predictions, as well as predictions of the transient/dynamic operation of ORC systems.

viii

CONTENTS

1 INTRODUCTION ...... 1

1.1 BACKGROUND...... 1

1.2 HEAT-UTILISATION TECHNOLOGIES ...... 3

1.2.1 Internal and external combustion engines ...... 3

1.2.2 Classification of heat-to-power systems ...... 4

1.2.3 Low-grade heat-to-power conversion ...... 8

1.3 ORC ENGINES ...... 8

1.4 AIM & OBJECTIVES ...... 10

1.5 THESIS CONTENTS AND LAYOUT ...... 11

2 LITERATURE REVIEW ...... 13

2.1 APPLICATIONS OF ORC TECHNOLOGY ...... 13

2.2 ORC DESIGN AND WORKING FLUID OPTIMISATION...... 15

2.3 ORC COMPONENTS ...... 21

2.3.1 Pump ...... 21

2.3.2 Heat exchangers ...... 22

2.3.3 Expanders ...... 26

2.4 RECIPROCATING-PISTON EXPANDERS...... 29

2.4.1 Gas spring testing facility ...... 31

2.4.2 Performance of reciprocating-piston expanders ...... 34

2.4.3 Loss mechanisms ...... 35

2.4.4 Work in progress ...... 40

3 ORGANIC RANKINE CYCLE MODELLING ...... 49

3.1 THE RANKINE CYCLE ...... 49

3.2 THERMODYNAMIC MODELLING ...... 52

3.2.1 Assumptions ...... 53

3.2.2 Component performance indicators ...... 53

3.2.3 Fluid properties ...... 54

3.2.4 Lumped-mass model governing equations ...... 54

3.3 SYSTEM AND COMPONENTS PERFORMANCE INDICATORS ...... 56

3.3.1 First- and second-law ORC performance indicators ...... 56

3.3.2 First- and second-law components performance indicators ...... 58

3.3.3 Pressure drops ...... 65

3.3.4 Net pressure suction head (NPSH) ...... 66

4 DESIGN SIMULATIONS ...... 69

4.1 THERMODYNAMIC DIAGRAMS ...... 70

4.2 INFLUENCE OF THE EXPANDER ISENTROPIC EFFICIENCY ...... 73

xii

4.3 INFLUENCE OF PINCH-POINT TEMPERATURE DIFFERENCE ...... 74

4.4 INFLUENCE OF THE WORKING FLUID MASS FLOWRATE ...... 75

4.5 INFLUENCE OF THE HEAT-SOURCE TEMPERATURE ...... 76

4.6 POTENTIAL OF LOW-GWP WORKING FLUIDS ...... 77

4.6.1 Hydrofluoroolefin R1233zd ...... 77

4.6.2 Hydrofluoroolefin R1234ze ...... 80

4.6.3 Hydrocarbon Isopentane ...... 82

5 EARLY-STAGE ORC TESTING FACILITY ...... 85

5.1 SYSTEM DESCRIPTION ...... 85

5.1.1 Pump ...... 87

5.1.2 Evaporator ...... 88

5.1.3 Expander ...... 90

5.1.4 Condenser ...... 92

5.2 EXPERIMENTAL PROCEDURES ...... 93

5.3 DATA ANALYSIS ...... 96

5.4 STEADY-STATE PERFORMANCE RESULTS ...... 97

5.4.1 Pump performance...... 97

5.4.2 Expander performance ...... 99

5.4.3 ORC overall efficiency ...... 101

xiii

5.4.4 Thermodynamic analysis of the optimum configurations ...... 103

5.5 TESTING FACILITY LIMITATIONS ...... 106

5.5.1 Cavitation ...... 106

5.5.2 Flow instability and calibration ...... 106

5.5.3 Pressure drops ...... 107

5.5.4 Corrosion ...... 107

5.5.5 Non-condensable gases ...... 108

5.5.6 Seal compatibility ...... 108

5.5.7 Wall heat losses ...... 109

6 IMPROVED FLEXIBLE ORC PROTOTYPE ...... 111

6.1 TESTING FACILITY IMPROVEMENTS ...... 112

6.1.1 Cavitation ...... 112

6.1.2 Flow instability and calibration ...... 113

6.1.3 Pressure drops ...... 114

6.1.4 Corrosion ...... 114

6.1.5 Non-condensable gases ...... 114

6.1.6 Seal compatibility ...... 115

6.1.7 Wall heat losses ...... 115

6.2 SYSTEM DESCRIPTION ...... 116

xiv

6.3 EXPERIMENTAL PROCEDURES ...... 120

6.3.1 Pressure testing ...... 120

6.3.2 Vacuum testing ...... 120

6.3.3 System filtration ...... 121

6.3.4 Steady-state operation ...... 121

6.4 DATA ANALYSIS ...... 122

7 EXPERIMENTAL RESULTS OF THE NON-RECUPERATIVE ORC PROTOTYPE ...... 125

7.1 OPERATING PARAMETER OPTIMISATION ...... 126

7.2 PART-LOAD PERFORMANCE MAPS ...... 126

7.3 THERMODYNAMIC ANALYSIS OF THE OPTIMUM CONFIGURATIONS ...... 134

7.4 EXPANDER PERFORMANCE ...... 139

7.5 DYNAMIC CHARACTERISTICS ...... 141

8 CONCLUSION ...... 145

8.1 SUMMARY OF FINDINGS ...... 147

8.2 FUTURE WORK ...... 150

9 BIBLIOGRAPHY ...... 153

10 APPENDIX ...... 173

xvi

LIST OF TABLES

TABLE 2.1: CHARACTERISTICS OF THE GAS SPRING TESTING FACILITY...... 31

TABLE 3.1 – BASELINE OPERATING CONDITIONS OF THE MODELLED ORC ENGINE...... 50

TABLE 4.1 – OPERATING CONDITIONS AND COMPONENTS PERFORMANCE INDICATORS USED

FOR THE NUMERICAL SIMULATIONS...... 70

TABLE 5.1: FLUID-O-TECH ROTARY VANE PUMP (TMFR 30-200 SERIES) SPECIFICATION...... 88

TABLE 5.2: FRIGOSYSTEM 18 KW(TH) ELECTRIC OIL HEATER SPECIFICATION...... 89

TABLE 5.3: SPECIFICATION FOR THE 1-KW SEMI-HERMETIC OIL FREE SCROLL EXPANDER ...... 91

TABLE 5.4: 22 KW ALFA LAVAL CONDENSER SPECIFICATION...... 93

TABLE 6.1: ORC TESTING FACILITY CHARACTERISTICS...... 118

TABLE 7.1: MAXIMUM PERFORMANCE INDICATORS FOR VARIOUS HEAT-SOURCE

TEMPERATURES...... 137

xvii

xviii

LIST OF FIGURES

FIGURE 1.1: WASTE HEAT TO ELECTRICITY CONVERSION MOTIVATION [1]...... 2

FIGURE 1.1.2: THERMAL EFFICIENCIES AND LIMITS OF SELECTED THERMODYNAMIC HEAT

ENGINES [12]...... 7

FIGURE 1.3: ORC SCHEMATIC SHOWING FLOW PROCESS IN A LOOP [15]...... 9

FIGURE 2.1: PRINCIPLE OF OPERATION OF A SCROLL EXPANDER [88]...... 28

FIGURE 2.2: RECIPROCATING-PISTON ENGINE WITH A SPARK PLUG [89]...... 30

FIGURE 2.3: COMPRESSION CHAMBER SHOWING THE ULTRASONIC EMITTER, RECEIVER AND

PRESSURE TRANSDUCER [95]...... 32

FIGURE 2.4: (A) EXPERIMENTAL SETUP FOR GAS-SPRING TESTING; (B) GAS SPRING

COMPRESSION CHAMBER...... 33

FIGURE 2.5: (A) LOSS MECHANISMS IN RECIPROCATING MACHINES SHOWING PISTON WALL

SECTION [95]; (B) PISTON RING [96]...... 36

FIGURE 2.6: (A) PRESSURE-VOLUME DIAGRAM OF THE PRESSURE DROP THROUGH THE

VALVE. ILLUSTRATION [101]; (B) VALVE OPENING [102]...... 38

FIGURE 2.7: LOSS MECHANISMS IN RECIPROCATING MACHINES SHOWING HEAT LOSS. ....39

FIGURE 2.8: PV DIAGRAMS FOR: (A) HEAVIER GASES [94]; (B) PV LIGHTER GASES [94]...... 41

FIGURE 2.9: (A) THERMAL LOSS DIAGRAM FOR HEAVY GAS (PROPANE) [94]; (B) THERMAL

LOSS DIAGRAM FOR LIGHT GAS (HELIUM, NITROGEN) [94]...... 43

xix

FIGURE 2.10: COMPARISON OF PRESSURE AS A FUNCTION OF VOLUME OVER A CYCLE

BETWEEN CFD MODEL WITH MASS LEAKAGE AND EXPERIMENTAL DATA [105]; (B)

COMPARISON OF PRESSURE AS A FUNCTION OF CRANK ANGLE OVER A CYCLE BETWEEN

CFD MODEL WITH MASS LEAKAGE AND EXPERIMENTAL DATA [105]...... 44

FIGURE 2.11: COMPARISON OF HEAT FLUX BETWEEN CFD MODEL WITH MASS LEAKAGE

AND EXPERIMENTAL DATA [105]...... 45

FIGURE 2.12: (A) MESH FOR TRANSLATION OF SLIDING VALVE; (B) MESH FOR

COMPRESSION AND EXPANSION OF THE CHAMBER...... 46

FIGURE 3.1: SCHEMATIC OF THE ARCHITECTURE OF A NON-RECUPERATIVE ORC ENGINE...... 51

FIGURE 3.2 - TEMPERATURE-ENTROPY DIAGRAM OF AN ORC ENGINE USING R245FA AS A

WORKING FLUID, EXTRACTING HEAT FROM A HIGH-TEMPERATURE WASTE-HEAT

SOURCE AT 120 °C UNDER THE OPERATING CONDITIONS STATED IN TABLE 3.1. ... 52

FIGURE 3.3: EXERGY FLOW DIAGRAM ACROSS THE PUMP...... 59

FIGURE 3.4: EXERGY FLOW DIAGRAM ACROSS THE EVAPORATOR...... 61

FIGURE 3.5: EXERGY FLOW DIAGRAM ACROSS THE EXPANDER...... 63

FIGURE 3.6: EXERGY FLOW DIAGRAM ACROSS THE CONDENSER...... 64

FIGURE 4.1: SCHEMATIC OF THE ARCHITECTURE OF A NON-RECUPERATIVE ORC ENGINE...... 69

FIGURE 4.2 - TEMPERATURE-ENTROPY DIAGRAM OF AN ORC ENGINE USING R245FA AS A

WORKING FLUID, EXTRACTING HEAT FROM A HIGH-TEMPERATURE WASTE-HEAT

SOURCE AT 120 °C UNDER THE OPERATING CONDITIONS STATED IN TABLE 4.1. ... 71

FIGURE 4.3 - PRESSURE-ENTHALPY DIAGRAM OF AN ORC ENGINE USING R245FA AS A

WORKING FLUID, EXTRACTING HEAT FROM A HIGH-TEMPERATURE WASTE-HEAT

SOURCE AT 120 °C UNDER THE OPERATING CONDITIONS STATED IN TABLE 4.1. ...72

FIGURE 4.4 - INFLUENCE OF EXPANDER ISENTROPIC EFFICIENCY ON ORC PERFORMANCE

USING R245FA AS A WORKING FLUID, TAKING HEAT FROM A WASTE-HEAT SOURCE AT

120 °C AND OPERATING CONDITIONS IN TABLE 4.1...... 73

FIGURE 4.5 - INFLUENCE OF PINCH-POINT ΔT IN EVAPORATOR AND CONDENSER ON ORC

PERFORMANCE USING R245FA AS A WORKING FLUID, EXTRACTING HEAT FROM A

WASTE-HEAT SOURCE AT 120 °C AND OPERATING CONDITIONS IN TABLE 4.1...... 74

FIGURE 4.6 – INFLUENCE OF WORKING FLUID FLOWRATE ON ORC PERFORMANCE USING

R245FA AS A WORKING FLUID, EXTRACTING HEAT FROM A WASTE-HEAT SOURCE AT

120 °C AND OPERATING CONDITIONS IN TABLE 4.1...... 75

FIGURE 4.7 - INFLUENCE OF HEAT-SOURCE TEMPERATURE ON ORC PERFORMANCE USING

R245FA AS A WORKING FLUID, EXTRACTING HEAT FROM A WASTE-HEAT SOURCE AT

120 °C AND OPERATING CONDITIONS IN TABLE 4.1...... 76

FIGURE 4.8 - INFLUENCE OF: (A) WORKING FLUID FLOWRATE; AND (B) HEAT-SOURCE

TEMPERATURE ON THE ORC PERFORMANCE USING R1233ZD AS A WORKING FLUID

UNDER THE OPERATING CONDITIONS STATED IN TABLE 4.1...... 79

FIGURE 4.9 - INFLUENCE OF: (A) THE WORKING FLUID MASS FLOWRATE; AND (B) THE

HEAT-SOURCE TEMPERATURE ON THE ORC PERFORMANCE USING R1234ZE AS A

WORKING FLUID UNDER THE OPERATING CONDITIONS STATED IN TABLE 4.1...... 81

FIGURE 4.10 - INFLUENCE OF: (A) THE WORKING FLUID MASS FLOWRATE; AND (B) THE

HEAT-SOURCE TEMPERATURE ON THE ORC PERFORMANCE USING ISOPENTANE AS A

WORKING FLUID UNDER THE OPERATING CONDITIONS STATED IN TABLE 4.1...... 83

xxi

FIGURE 5.1: P&I DIAGRAM OF THE PRINCIPLE PROTOTYPE ORC TESTING FACILITY [87]...... 86

FIGURE 5.2: (A) FLUID-O-TECH ROTARY VANE PUMP TMFR 30-200 [108]; (B)

VARIABLE FREQUENCY DRIVE [108]...... 87

FIGURE 5.3: (A) SWEP BRAZED PLATE HEAT EXCHANGER – EVAPORATOR [109]; (B)

FRIGOSYSTEM 18-KW ELECTRIC OIL HEATER [110]...... 89

FIGURE 5.4: 1-KW SEMI-HERMETIC OIL-FREE SCROLL EXPANDER (E15H022A-SH) [111]...... 90

FIGURE 5.5: PERFORMANCE MAP (AIRSQUARED) AT 1.25 BARG CONDENSING PRESSURE [111]...... 91

FIGURE 5.6: 22-KW ALFA LAVAL CONDENSER CB60-30H-F MODEL [112]...... 92

FIGURE 5.7: PICTURES OF THE EARLY-STAGE ORC TESTING FACILITY IN THE CLEAN

ENERGY PROCESSES LABORATORY...... 94

FIGURE 5.8: (A) EFFECT OF PUMP SPEED, HEAT-SOURCE TEMPERATURE AND

EXPANDER/GENERATOR LOAD ON MASS FLOW RATE; (B) EFFECT OF PRESSURE RATIO,

HEAT-SOURCE TEMPERATURE AND EXPANDER/GENERATOR LOAD ON PUMP

ELECTRICAL POWER...... 98

FIGURE 5.9: (A) EFFECT OF PRESSURE RATIO, HEAT-SOURCE TEMPERATURE AND

GENERATOR LOAD ON EXPANDER ELECTRICAL POWER; (B) EFFECT OF PRESSURE

RATIO, HEAT-SOURCE TEMPERATURE AND GENERATOR LOAD ON EXPANDER

EFFICIENCY...... 100

FIGURE 5.10: (A) EFFECT OF PRESSURE RATIO, HEAT-SOURCE TEMPERATURE AND

GENERATOR LOAD ON CYCLE THERMAL EFFICIENCY; (B) EFFECT OF PRESSURE RATIO,

HEAT-SOURCE TEMPERATURE AND GENERATOR LOAD ON TOTAL EXERGY

DESTRUCTION...... 102

xxii

FIGURE 5.11: (A) EFFECT OF PRESSURE RATIO, HEAT-SOURCE TEMPERATURE AND GENERATOR

LOAD ON EXPANDER EXERGY DESTRUCTION; (B) EFFECT OF PRESSURE RATIO, HEAT-

SOURCE TEMPERATURE AND GENERATOR LOAD ON EVAPORATOR EXERGY DESTRUCTION...... 104

FIGURE 5.12: (A) EXERGY DESTRUCTION PERCENTAGES OF DIFFERENT COMPONENTS AT

DIFFERENT PRESSURE RATIOS AT PART GENERATOR LOAD; (B) EXERGY DESTRUCTION

FRACTIONS OF DIFFERENT COMPONENTS AT DIFFERENT PRESSURE RATIOS AT CASE B

LOAD...... 105

FIGURE 6.1: ORC FUNCTIONAL PROTOTYPE SHOWING MAJOR COMPONENTS...... 112

FIGURE 6.2: CORIOLIS MFC 400 OPTIMASS FLOW METER [116]...... 113

FIGURE 6.3: VACUUM TEST SETUP DURING FUNCTIONAL ORC RIG COMMISSIONING. ... 115

FIGURE 6.4: ORC FUNCTIONAL RIG SHOWING SHEATHED TUBING ON HOT AND COLD SIDES...... 116

FIGURE 6.5: P&I DIAGRAM OF THE ORC RIG WITH RECUPERATOR AND SUPERHEATER...... 117

FIGURE 7.1: ORC PART-LOAD PERFORMANCE MAPS: (A) CONTOUR PLOTS OF THE NET

POWER OUTPUT; AND (B) OF THE THERMAL EFFICIENCY AS FUNCTIONS OF THE

EXPANDER SPINDLE SPEED, 휔EXP, AND OF THE WORKING FLUID MASS FLOWRATE,

푚WF, FOR A HEAT-SOURCE TEMPERATURE OF 100 °C...... 127

FIGURE 7.2: ORC PART-LOAD PERFORMANCE MAPS: (A) CONTOUR PLOTS OF THE NET

POWER OUTPUT; AND (B) OF THE THERMAL EFFICIENCY AS FUNCTIONS OF THE

EXPANDER SPINDLE SPEED, 휔EXP, AND OF THE WORKING FLUID MASS FLOWRATE,

푚WF, FOR A HEAT-SOURCE TEMPERATURE OF 120 °C...... 129

FIGURE 7.3: ORC PART-LOAD PERFORMANCE MAPS: (A) CONTOUR PLOTS OF THE NET

POWER OUTPUT; AND (B) OF THE THERMAL EFFICIENCY AS FUNCTIONS OF THE

xxiii

EXPANDER SPINDLE SPEED, 휔EXP, AND OF THE WORKING FLUID MASS FLOWRATE,

푚WF, FOR A HEAT-SOURCE TEMPERATURE OF 140 °C...... 131

FIGURE 7.4: ORC GROSS POWER OUTPUT PART-LOAD PERFORMANCE MAP AS FUNCTION OF

THE PUMP SPEED, 휔P AND THE EXPANDER RESISTIVE LOAD, 푅EXP (WHICH

REPRESENTS THE BRAKING TORQUE APPLIED TO THE EXPANDER SHAFT), FOR A HEAT-

SOURCE TEMPERATURE SET AT 140 °C...... 133

FIGURE 7.5: TEMPERATURE-ENTROPY DIAGRAMS PRESENTING THE STEADY OPERATION OF

THE ORC ENGINE WITH EFFICIENCY-OPTIMUM OPERATING CONDITIONS FOR HEAT-

SOURCE TEMPERATURES OF: (A) 120 °C, AND; (B) 140 °C. WORKING-FLUID STATE

REPORTED AT KEY LOCATIONS OF THE FACILITY: (1) PUMP INLET OR CONDENSER

OUTLET, (2) EVAPORATOR SATURATED LIQUID, (3) EVAPORATOR SATURATED

VAPOUR, (4) EVAPORATOR EXHAUST, (5) EXPANDER INLET, (6) EXPANDER OUTLET,

(7) CONDENSER SATURATED VAPOUR, (8) CONDENSER SATURATED LIQUID...... 134

FIGURE 7.6: EXERGY DESTRUCTION CONTRIBUTIONS OF VARIOUS COMPONENTS TO THE

OVERALL EXERGY DESTRUCTION IN THE ORC ENGINE, OPERATED WITH EFFICIENCY-

OPTIMUM PUMP SPEED AND EXPANDER LOAD AT THREE HEAT-SOURCE

TEMPERATURES...... 135

FIGURE 7.7: (A) EXPANDER ISENTROPIC EFFICIENCY EXPRESSED IN %, MAPPED AGAINST

THE SCROLL SPINDLE SPEED 휔P AND THE PRESSURE RATIO, 푟P. PERFORMANCE MAP

OBTAINED FROM CUBIC-SPLINE INTERPOLATION OF EXPERIMENTAL DATA, INDICATED

AS GREY DOTS ON THE MAP, OBTAINED FOR HEAT-SOURCE TEMPERATURES RANGING

FROM 100 °C TO 140 °C; (B) EXPANDER FILLING FACTOR AS A FUNCTION OF THE

EXPANDER SPINDLE SPEED...... 140

FIGURE 7.8: (A) DYNAMIC RESPONSE OF THE ORC SYSTEM TO A STEP IN PUMP VELOCITY

FROM 1100 TO 2000 RPM AT FIXED HEAT-SOURCE TEMPERATURE (푇HS = 120 °C)

AND EXPANDER LOAD (푅EXP = 25 Ω); (B) RISING AND FALLING STEP-RESPONSE

TIMES AS FUNCTIONS OF THE STEP AMPLITUDE IN EXPANDER POWER ...... 144

xxiv

xxv

xxvi

NOMENCLATURE

퐶v Flow coefficient

푐푣 Specific heat capacity at constant volume (J/kg/K)

푐푝 Specific heat capacity at constant pressure (J/kg/K)

퐷 (Hydraulic) diameter (m)

푓퐷 Darcy friction factor

퐺f Flow rate through valve (gal/min)

퐼 ̇ Exergy destruction rate (W)

퐾 Flow resistance coefficient

퐿 Pipe length (m)

푚̇ Mass flowrate (kg/s)

푁휓 Number of samples for 휓 measurement

푝 Pressure (Pa)

푞 Vapour quality

푄 Heat (J)

푄̇ Heat flow rate (W)

푟 Ratio (-)

xxvii

푟푝 Pressure ratio (-)

푟푣 Volume ratio (-)

푅 Radius (m)

SG Specific gravity

푇 Temperature (K)

푢 Uncertainty

푈 Internal energy (J)

v Velocity (m/s)

푣 Specific volume (m3/kg)

ℎ푠 Static head (m)

ℎ푓 total friction head (m)

푣푠 Dynamic head (m)

푉 Volume (m3)

푊 Work (J)

푊̇ Power (W)

푋̇ Exergy flow rate (W)

푋 Exergy (J)

xxviii

Greek characters

휂 Efficiency (%)

𝜌 Density

휙 Filling factor (-)

휓 Fictive time-resolved measurement

𝜎 Standard deviation

휏 Rising or falling time (s)

휔 Rotational speed (RPM)

훾 Specific heat-capacity ratio (or adiabatic constant)

Subscripts and superscripts cond Condenser cw Cooling water bw Back-work disp Displacement evap Evaporator exg Exergy exp Expander fin Final

xxix

hs Heat source

in Inlet/intake/input

init Initial

is Isentropic

out Outlet/exhaust/output

abs Absolute

NPSH Net positive suction head

NPSHA Net positive suction head, actual

NPSHR Net positive suction head, required

sen Sensor

RPM Revolution per minute

sub Subcooling

sup Superheat

HTX Heat exchanger

t Time

wf Working fluid

n Polytropic index

1 Pump inlet, condenser outlet

2 Evaporator inlet, pump outlet

xxx

3 Expander inlet, evaporator outlet

4 Condenser inlet, expander outlet

5 Oil into HTX

6 Oil out of HTX

7 Water in HTX

8 Water out of HTX

xxxi

xxxii Chapter 1: Introduction

1 INTRODUCTION

1.1 BACKGROUND

Global energy demand continues to increase at a fast pace. Fossil fuels, biomass and nuclear energy are currently the main ways for generating electrical power. Power generation from fossil fuel involves combustion, which leads to pollution, health problems, or consumption of a high-grade finite natural resource. There are numerous industrial processes that produce waste heat, e.g., the steel industry, refineries and cement industry. The waste heat generated by these industries as shown in Figure 1.1, often in the form of hot combustion gases, can then become a very valuable resource, as it can be recovered and partially converted to useful power at a reasonable cost using a range of waste-heat recovery (WHR) technologies.

In developing nations like most African countries, one of the limiting factors of economic and technological progress is the availability of electricity, while, in the same time, there exists thousands of manufacturing plants whose rejected thermal energy can be converted to electricity. The organic Rankine cycle (ORC) engine is a WHR technology that can be used to efficiently convert thermal energy into useful power for electrifying homes, business and industrial facilities.

Chinedu Kingsley Unamba, August 2020 1 Experimental development of ORC technology for waste heat-to-power conversion

Figure 1.1: Waste heat to electricity conversion motivation [1].

Such heat-to-power conversion technologies are promising solutions to improve energy efficiency. These interests have been further sparked by escalating fuel prices and even stricter environmental regulations. Organic Rankine cycle systems are particularly promising technology for power generation from waste-heat sources, especially for low- to medium-grade heat sources [2]. Low-grade heat sources are typically below 300 °C and medium-grade heat sources between 300 °C and 500 °C, while heat streams higher than 500 °C are considered high-grade heat sources.

Several technologies therefore exist to generate power from thermal energy sources like the Brayton cycle used by gas turbines [3] and combined cycle plants, the steam Rankine cycle and the ORC, amongst others.

2 Chinedu Kingsley Unamba, August 2020 Chapter 1: Introduction

Interest in ORC technology stems from the desire to produce electricity from renewable or low-medium-grade waste heat and has grown considerably over the past few decades. Aside having an almost free energy input, it is considered environmentally friendly given the closed loop cycle in which it operates [4, 5]. In the context of waste heat, the heat source is usually considered “free” in that it is being rejected from other processes.

ORC technology can help curb our current dependence on fossil fuels, while reducing harmful emissions and pollution, promoting independence and conserving resources [6, 7]. A broad variety of processes can be found in industrial plants rejecting a significant amount of thermal energy in the form of combustion products into the atmosphere over a range of temperatures [8]. Studies across various global regions have confirmed that about 60% of industrial energy input is typically wasted [9].

1.2 HEAT-UTILISATION TECHNOLOGIES

There are several technologies employed to convert heat to power. Such technologies relying on heat engines are numerous and can be categorised based on the different thermodynamic processes undergone to achieve the desired thermal to power conversion.

1.2.1 Internal and external combustion engines

A heat engine is a thermodynamic device which converts thermal energy into useful work, while a heat pump uses work to transfer heat from one thermal reservoir to another. Heat engines can be categorised as internal or external combustion engines.

In internal combustion engine, fuel is delivered to the heat engine in form of chemical energy. Fuel combustion within the engine combustion

Chinedu Kingsley Unamba, August 2020 3 Experimental development of ORC technology for waste heat-to-power conversion

chamber/cylinder provides the heat required for power generation. Typical examples include diesel and gasoline engines.

External combustion engines, on the other hand, are heat engines operated with an internal working fluid to which heat is transferred from the combustion of a fuel outside the engine. By extension, the term “external combustion engine” designates all heat engines driven by an external heat source (e.g., combustion, waste heat, heat from renewable sources such as solar heat), with the heat transferred to the working fluid through the engine walls or heat exchangers. A well-known example is the beam engine, pioneering engine of the steam age. This thesis focuses on a specific kind of external combustion engine, namely the organic Ranking cycle (ORC) engine, which exhibits great thermodynamic potential for low-grade (waste) heat recovery or conversion, as detailed hereafter.

1.2.2 Classification of heat-to-power systems

Several external-combustion technologies can be used to convert heat to power. Examples include gas-phase heat engines (e.g., Stirling or Ericsson), thermo-acoustic engines (TA), thermofluidic oscillators (TFOs), thermoelectric generators (TEG) and Rankine engines.

Gas-phase heat engines refer to heat engines operated with gaseous working fluids (typically hot air) that do not undergo phase transition. In such engines, power is generated by the expansion of compressed and overheated gas. Common examples include the Stirling, Ericsson and Brayton engines. Stirling engines are closed-cycle engines where the working fluid ideally undergoes a Stirling thermodynamic cycle, made of an isothermal expansion, an isochoric heat removal (usually to ambient air), an isothermal compression and an isochoric heat addition (provided by the heat source). Closed- and open-cycle Ericsson engines differ from the Stirling engine in

4 Chinedu Kingsley Unamba, August 2020 Chapter 1: Introduction

that they ideally use isobaric heat addition and heat removal processes. Near- isothermal compression and expansion processes are typically achieved with positive-displacement devices operated a low spindle speed. Hence, Stirling and Ericsson engines usually exhibit low power density. Brayton engines, on the other hand, are closed- or open-cycle engines where the working fluid ideally undergoes a Brayton thermodynamic cycle (also called Joule or first Ericsson cycle), made of an adiabatic compression, an isobaric heat addition, an adiabatic expansion and an isobaric heat removal. Reversible Joule- Brayton cycle engines, where energy is stored as sensible heat in hot and cold thermal stores, are promising candidates for pumped thermal electricity storage applications.

Thermoacoustic (TA) engines are a class of heat engines that use high amplitude sound waves or the resonance of the standing wave to convert thermal energy first to acoustic energy, and then this acoustic energy is converted to generate power. Although the architecture sounds simple, optimising the design of TA engines is very challenging [10]. This makes it unsuitable for basic heat conversion technology where simplicity in controlling the operating parameters is paramount.

Thermofluidic oscillators (TFOs) are a class of heat engines that typically convert low-grade heat into useful work by setting a fluid into an oscillatory, periodic motion. Two-phase TFOs are known to be more advantageous than single phase heat engines like TAs or Stirling engines due to their reliance on phase change. Several TFO devices like the Non-Inertive-Feedback Thermofluidic Engine (NIFTE) and the liquid Stirling engine (Up-THERM) have been improved over the years and enjoy low capital and maintenance cost due to the reduced number of moving parts. However, at lower temperatures, <200 °C, the thermal and exergy efficiencies remain low in the region of 1-5% and 5-20%, respectively, making them less desirable in low-

Chinedu Kingsley Unamba, August 2020 5 Experimental development of ORC technology for waste heat-to-power conversion

temperature applications when compared to ORC systems [11], especially in applications that require high performance and power density.

Thermo-electric generators (TEGs) are solid-state semiconductors that convert heat flowing between two reservoirs at different temperatures into a direct current power source. The principle of operation relies on the Seebeck effect, which occurs when the junction of two dissimilar conductors maintained at different temperatures create an electric current in a loop by virtue of voltage difference and produce power. Most thermoelectric generators contain three main semiconductors: bismuth telluride, lead telluride and silicon germanium.

TEGs are quiet in operation, do not emit greenhouse gases and are highly scalable in terms of the power output with no maintenance cost. They are often compact and involve a direct conversion from heat to electricity unlike ORCs that require heat to be converted first to mechanical rotation of shaft and then to electricity in the armature of the electric generator. The application can be found in refrigeration and sensors. However, with limited amount of TEG module knowledge, the design expertise is often limited, and the materials used are sometimes inefficient. The thermal efficiency of TEGs are low relative when compared to other heat engines. Also, they have higher initial cost per Watt of electricity relative to other heat engines.

The Rankine engine, ideally based on the Rankine cycle (made of isentropic compression, isobaric heat addition, isentropic expansion and isobaric heat addition), is a two-phase heat engine. The working fluid in a Rankine machine undergoes phase change, unlike in Ericsson or Stirling cycles (hot air engines). The Rankine engine is a closed-loop external heat engine and is often categorised either based either: (i) on the type of working fluid used – water, organic compounds (organic Rankine cycle engines) or working-fluid

6 Chinedu Kingsley Unamba, August 2020 Chapter 1: Introduction

mixtures (e.g., in the Kalina cycle); or (ii) on the type of heat source (solar Rankine, nuclear Rankine), as shown in Figure 1.1.2.

Figure 1.1.2: Thermal efficiencies and limits of selected thermodynamic heat engines [12].

Most waste-heat recovery applications make use the ORC or the Kalina cycle for its power generation. The difference between the ORC, which uses an organic fluid, and the Kalina cycle, which uses a mixture of water and ammonia, is that there is more heat added (relative to the total heat addition) in the Kalina cycle during evaporation or boiling since the different binary mixtures have different evaporating temperatures. This leads to greater

Chinedu Kingsley Unamba, August 2020 7 Experimental development of ORC technology for waste heat-to-power conversion

efficiency in the Kalina cycle [13]. However, the use of water as a working fluid in the Kalina brings about a lot of problems from corrosion to requiring a higher evaporating temperature when compared to organic fluid with much lower evaporation temperature. This makes ORC a better choice for low- temperature applications.

1.2.3 Low-grade heat-to-power conversion

Figure 1.1.2 shows the different thermal efficiencies at different heat-source temperatures for different heat engines, as reported by Markides [12]. The author shows that at high temperatures (typically above 400 °C), all solar Rankine, nuclear Rankine, TEGs (thermoelectric generators), thermoacoustic heat engines (TAs) and Stirling engines exhibit thermal efficiencies greater than 20%. However, at lower temperatures, i.e., <200 °C, the ORC engine exhibits by far the highest efficiency. This makes the ORC the ideal candidate for low-grade heat recovery/utilisation applications and, hence, the principal focus of this thesis.

1.3 ORC ENGINES

The ORC engine is a thermodynamic heat engine, which means that it is a technology that converts heat into useful work. An ORC is basically a Rankine cycle using an organic fluid as a working fluid (typically with low boiling point) [14]. It employs the four basic processes of the Rankine cycle: an ideally isobaric evaporation, an ideally isentropic expansion, an ideally isobaric condensation and an ideally isentropic pumping process.

Rankine cycles can be trans- or sub-critical. A trans-critical cycle is one in which the working fluid is subjected to both sub- and super-critical states. In a sub-critical Rankine cycle, the working fluid is subjected to a maximum

8 Chinedu Kingsley Unamba, August 2020 Chapter 1: Introduction

pressure lower than its critical pressure. The sub-critical Rankine cycle forms the basis for ORC systems. Here, an organic compound (working fluid) is circulated by a pump. The recovered heat is transferred to the working fluid in an evaporator by a heat source.

The heat source could be from renewable energy, i.e. solar, geothermal or from industrial waste heat. The generated vapour is then expanded in an expansion device, thereby producing useful power when coupled to a generator, while the working fluid is condensed at the exhaust of the expander to feed the pump with liquid-phase fluid.

The organic Rankine cycle engine technology is particularly suitable for power generation from low-temperature heat sources. Low- to medium- grade heat-source streams are generated from renewable sources (such as geothermal or solar) or industrial waste-heat sources.

Figure 1.3: ORC schematic showing flow process in a loop [15].

Chinedu Kingsley Unamba, August 2020 9 Experimental development of ORC technology for waste heat-to-power conversion

Figure 1.3 shows the flow process of an ORC system loop. The working fluid is pumped from 1 to 2. The high-pressure fluid then goes into the evaporator where heat transfer takes place between the working fluid and the oil heater (2 - 3). The high-pressure, high-temperature working fluid then flows into the expander where it expands (3 - 4) and then condenses (4 - 1) in the condenser by exchanging het with a heat sink.

1.4 AIM & OBJECTIVES

The overarching aim of this research project is the experimental investigation of the steady and transient part-load performance of ORC engines. The objectives of this project are listed below.

1. Design, construct and gain practical appreciation of the challenges in deploying and operating ORCs while addressing common challenges phased in deploying and operating ORC systems like flow instabilities, inaccurate measuring of pressure and temperature by the pressure transducers and temperature sensors, corrosion of the inside and outside walls of the pipe, pressure drop reduction in pipes, valves and bends, non- condensable gases, seal compatibility. 2. Another objective was to provide system performance maps using different performance metrics like expander power output, thermal efficiency, expander isentropic efficiency exergy analysis to assess the contribution of each component to the overall exergy destruction in the system. 3. Finally, a dynamic characterisation is performed to understand how ORC systems respond to time varying control variables.

10 Chinedu Kingsley Unamba, August 2020 Chapter 1: Introduction

1.5 THESIS CONTENTS AND LAYOUT

In this thesis, an improvement is made to a small-scale ORC prototype, referred to as the early-stage prototype. The new testing facility is designed and constructed to gain practical appreciation of the numerous challenges involved in the deployment and operation of an ORC engine. High-fidelity data are generated for both steady-state and dynamic mapping of system performance of the functioning prototype.

Chapter 2 presents the literature review of ORC systems, where the applications of ORC engines are discussed, along with the optimisation of both the ORC design and working fluid, with an emphasis on the components.

In Chapter 3, thermodynamic modelling of the ORC is presented. This includes the energy Equations for analysing ORC performance and component performance indicators. An exergy analysis is performed to determine the individual contributions of each components to the overall exergy destruction.

In Chapter 4, the early-stage ORC prototype setup is presented, and the results discussed. The specifications of the components such as pumps condenser evaporator specifications used are shown. Experiments are carried out and the effects of the control parameters, i.e. heat-source temperature, pump speed and expander load (which are categorized in two cases: case A and case B) are analysed. The limitations of the ORC rig are also discussed. It also presents the performance map of the rig in addition to an exergy analysis to determine the contribution of each component to the overall exergy destruction in the system.

In Chapter 5, the functional prototype new rig is introduced with the modifications made from the previous rig. Modifications introduced include

Chinedu Kingsley Unamba, August 2020 11 Experimental development of ORC technology for waste heat-to-power conversion

flow stability through using Coriolis flow meter. increasing the NPSH, including a liquid receiver, using high temperature seals and better sensors that are calibrated. It presents the experimental procedures, data analysis, data acquisition and system description (of major components) as well as the control variables used.

Chapter 6 presents both steady-state and dynamic experimental results obtained on the functioning prototype. Experiments are carried out to generate 100s of steady state data points. An operational optimisation of the ORC performance is proposed based on the gathered data. The optimal operating conditions of the control variables i.e. pump speed and expander load are determined at various heat-source conditions (100 °C, 120 °C and 140 °C), thus resulting in a total of 124 steady-state data points used to analyse the part-load performance of the engine. An exergy analysis is performed to determine the contribution of each component to the overall exergy destruction in the system. Also, the dynamic characteristics of the system are mapped by varying the pump speed and reporting the response time of the system in response to this variation.

Chapter 7 presents the conclusion and summary of the findings from the experiment. It then suggests areas of further improvement.

12 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

2 LITERATURE REVIEW

2.1 APPLICATIONS OF ORC TECHNOLOGY

There are numerous applications of the ORC technology, including power generation from solar heat, use in internal combustion engines (ICEs) for exhaust waste-heat recovery, and waste-heat recovery from industrial processes (e.g., heavy-duty machinery [16], aluminium reduction cell [17]).

Unlike commercially demonstrated large-scale ORC power plants (commonly used in geothermal, biomass, solar and WHR applications [18]), the performance of micro- or small-scale ORCs (typically 100s-W to 10s-kW engines) relies on the design of cost-effective components and through optimizing the working fluid. For example, as shown by Freeman et al. [7], the thermal efficiency of a 1-kW solar-powered ORC engine can be increased by as much as 50% through a careful working-fluid selection and solar- collector optimisation.

Regarding the application to ICE-WHR, ORC engines can be used to harness two main streams of waste heat, i.e., the high-temperature exhaust gases (200 - 600 °C) and the low-temperature jacket cooling water (80 - 100 °C).

Tian et al. [19] investigated a basic ORC system for WHR from a diesel engine and compared 20 working fluids, giving specific optimal working fluids in

Chinedu Kingsley Unamba, August 2020 13 Experimental development of ORC technology for waste heat-to-power conversion

terms of different performance indexes, while Shu et al. [20] and Zhang et al. [21] conducted similar analyses using dual-loop ORC systems. Novel cycle architectures based on ORC systems have been explored [22, 23] such as using zeotropic mixtures [24] as ORC working fluids, adopting multiple evaporation pressures and adapting to cascade cycles. Working fluid selection is critical in these scenarios.

Also, solar power generation has been attracting increasing attention recently, and this presents an alternative application for ORC systems, especially since energy available from the sun has the greatest potential of all renewable sources [25]. A series of studies have focused on harnessing and maximizing the potential of solar ORC to power conversion systems.

Boyaghchi and Heidarnejad [26] conducted a thermo-economic analysis of a domestic solar combined cooling-heating-power cycle system based on an ORC engine, where thermodynamic and economic models were adopted for multi-objective optimisation using genetic algorithms. Garcia-Saez et al. [27] investigated solar-powered ORC systems for residential applications and demonstrated a payback period of about 4 years after considering different uses (corresponding to system operation mode) and different locations in Spain (corresponding to solar radiation).

Substantial improvement has been made in ORC technologies in waste-heat recovery, solar thermal power, biomass power plants and geothermal plants [28, 29, 30, 31, 32, 33, 34]. Commercial ORC systems have been successfully demonstrated for large-scale power plants by ORC manufactures such as Ormat, Turboden, BNI, UTC, Electratherm and Adoratec [27]. However, small-scale ORC systems are still in an embryo stage and no mature products appear on the market. The need to understand the potential (suitable system architectures and designs, optimal performance and operation) of ORCs systems in this context is imperative.

14 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

2.2 ORC DESIGN AND WORKING FLUID OPTIMISATION

Many studies have been performed based on combining first-law (energy) and second-law (exergy) thermodynamic cycle analyses [35], aimed at understanding the performance characteristics of ORC systems. Several theoretical investigations are reported in the open literature, while the experimental studies on small-scale ORC systems that could generate real power are still limited. Lemort et al. [36]constructed an ORC prototype with R123 as the working fluid and focused on the performance of an open-drive oil-free scroll expander.

In one such study, Baral et al. [37] investigated the role of the heat-source temperature, pressure ratio and expander load in determining the ORC system thermal efficiency, component efficiencies and system power output. At different stages of ORC system design, choices must be made concerning several operating parameters and component selection, namely the choice of expander, pump and refrigerant. Similar work was conducted by Chang et al. [38] while using R245fa as the working fluid. Pei et al. [2] reported preliminary dynamic testing results from the testing of a 3.75-kWe ORC system using R123 as a working fluid and a turbo-expander.

A rolling-piston expander was proposed for a solar ORC system using R245fa as a working fluid [39, 40]; testing results show an expander isentropic efficiency of 45.2% and an overall thermal efficiency of 4.2%. Zhang et al. [41] chose a single-screw expander for an ORC system for waste-heat recovery from a diesel engine and the maximum power output was 10.4 kWe. The viability of the ORC technology for small-scale energy systems is yet unproven. High-fidelity experimental data must be provided, both from laboratory-scale experiments and real-life systems, in order to assess the technical and economic potential of small-scale ORC systems.

Chinedu Kingsley Unamba, August 2020 15 Experimental development of ORC technology for waste heat-to-power conversion

ORC systems suffer a deterioration in performance in part-load or off-design conditions, and the high global warming potential (GWP) or flammability of common working fluids is an increasing concern. Highly efficient expanders suitable for small-scale ORC systems are also a bottleneck constraint.

Beyond traditional working fluid pre-selection method, White et al. [42, 43] preformed computer-aided molecular design (CAMD) of ORC working fluids within a CAMD-ORC framework that has been used by these and other authors to suggest simultaneously optimised working fluids and ORC system conditions that allow the best thermodynamic and thermo-economic performance from ORC systems in specified applications.

Mago et al. [44] used dry working fluids R113, R245ca, R123 and isobutane to show that R-ORC (recuperative ORC) engines exhibit higher efficiency than non-recuperative ORC, thus requiring less waste heat for a similar power output which ultimately influences the exergy efficiency [45, 46, 47, 48].

A key thermodynamic performance indicator is the thermal efficiency (that is the ratio of the useful power output to the heat input), which is theoretically limited by the Carnot cycle efficiency. For example, Declaye et al. [49] measured thermal efficiencies in the ORC systems ranging from 1.9% to 11.8% when using a 1.8-kW open drive scroll expander.

In addition, the properties of the working fluid have a great influence on overall ORC system performance; for example, amongst many other issues, some working fluids may require a large superheat which could be detrimental to cycle efficiency [50, 51, 52]. Component selection is also important, e.g., volumetric expanders are usually preferred at smaller-scale and lower-temperature applications where this research focuses due to their low-speed, two-phase fluid tolerance and low cost [53, 54].

16 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

The optimal operating conditions, i.e., pump speed, flow rate and expander load are usually determined for the basic architecture at various heat-source condition. The thermal efficiency of the system, and the effects of the expander load and net power output examined. These results will give references to future real-time model-based control.

In comparison to other power cycles that are suitable for waste-heat conversion at low temperatures, such as the steam cycle, the ORC cycle produces more power [55]. Even at that the power is also dependent on other things like the critical temperature of the working fluid. Fluids with critical temperature close to maximum temperature of the cycle produce highest cycle efficiency [56].

Higher cycle efficiency [57] was reported with a regenerative architecture for heat sink between 24 °C and 40 °C and heat source between 140 °C and 160 °C. An absolute 3.6% improvement of the engine thermal efficiency was demonstrated [58] over a road test through recovering waste heat with an ORC system operated with R245fa.

Zhao et al. [59] simulated an integrated ORC-ICE engine and stated that the ORC system brings an increase of 3.6% in engine thermal efficiency and a reduction of 10 g/kWh in engine brake specific fuel consumption at the engine rated condition although the findings need to be experimentally verified. Li et al. [60] carefully calibrated an integrated model of the engine and an ORC-based system against experimental data and demonstrated a 2.3% improvement in engine brake thermal efficiency over a cruise driving cycle, which could be further improved to 4.2% by component optimisation.

For ORC engines operating with low-temperature heat sources, Manolakos et al. [61] tested a small-scale ORC engine that exhibits up to 5% thermal efficiency using a 2-kWscroll expander and operated with R134a as a

Chinedu Kingsley Unamba, August 2020 17 Experimental development of ORC technology for waste heat-to-power conversion

working fluid. Li et al. [62] tested a basic ORC system using thermal oil at 135-160 °C as the heat source and R245fa as the working fluid. The expander isentropic efficiency was found to be less than 35%. Pei et al. [2] reported preliminary dynamic testing results of a 3.75-kWe ORC system with R123. Turbine efficiency of 65% and ORC thermal efficiency of 7% were obtained.

Li et al. [63] experimentally compared a recuperative and a non-recuperative ORC system with R123 where a 10-kWe axial-flow single-stage turbine was included. Zheng et al. [39] experimentally verified a rolling-piston expander for a solar ORC system using R245fa since results exhibit an expander isentropic efficiency of 45% and an overall thermal efficiency of 4%.

For ORC engines operating with medium or high-temperature heat sources, Yu et al. [64] constructed a cascade Rankine cycle-ORC test rig for a diesel engine waste-heat recovery and tested system performance at different heat- source conditions. The net power output was estimated to 12 kWe at an assumed expander efficiency of 85%.

Shu et al. [65] tested ORC systems with different working fluids and recommended R123 for long-haul trucks while R245fa for city buses. Shi et al. [66] proposed ORC-based WHR systems operated with CO2 and a 4.5 kW test bench was developed, using an expansion valve in lieu of a positive- displacement or turbo-expander. Li et al. focused on the effects of system layouts [67] and external perturbations [68] on the system dynamic responses of ORC-based waste heat recovery systems as well as the trends of application of supercritical CO2 [69].

More information about the current state-of-the-art of ORC system test rigs can be investigated further [70]. Several studies propose heuristic physics- based models for optimal system design.

18 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

After identifying the main losses (i.e., leakage losses, mechanical losses and the supply pressure drop) that affect the expander performance, Lemort et al. [36] proposed a semi-empirical model of an open-drive oil-free scroll expander using R123 in the EES software. Desideri et al. [71] developed physics-based dynamic modelling tools in the Dymola software, which has been validated against steady-state and transient experimental results from an 11-kWe stationary ORC engine, providing convenience for transient performance examination and control strategy development as reported [72] for comparison of adaptive model predictive control and gain-scheduled switching PID controller.

Landelle et al. [73] developed semi-empirical models of ORC diaphragm pumps considering energetic performance, volumetric efficiency and cavitation limits. Li et al. [68] developed physics-based models for ORC systems adopting CO2 as the working fluid in the GT-SUITE simulation environment and predicted the actual performance under a driving cycle.

Simultaneous working-fluid and component optimisation is key for the design and development of efficient and economically viable micro- and small-scale ORC engines [74], which can be achieved with integrated computer-aided molecular design (CAMD) tools and ORC optimisation algorithms, such as those proposed by White et al. [42, 43] and van Kleef [75].

The confidence in the predicted techno-economic potential of the designed systems depends strongly on the capability of the thermodynamic model to capture engine performance in design and off-design conditions. In order to fully explore the market potential and examine the technical viability of small-scale ORC engines, high-fidelity experimental data must be provided both from laboratory-scale experiments and real-life systems, for performance assessment and simulation model development.

Chinedu Kingsley Unamba, August 2020 19 Experimental development of ORC technology for waste heat-to-power conversion

Although advancement is being made in conventional ORC, new technologies optimizing the net power output and minimizing exergy destruction in components are sort after. One of such involves eliminating components such as pumps to maximize the net power output. Pumpless ORCs are currently being experimented with and major advantages are already being reported experimentally [76, 77]. This is a promising research area.

Improvements in the form of reduced component scale exergy destruction through integrated models employing different components/materials of the ORC systems are also used to address areas of improvement [78, 36]. The most important parts of an ORC system are the working fluid and the expander. They represent component/material of the system subject to targeted significant incremental efficiency. The condenser and evaporator are limited in their contributions to the overall power output.

Relationship amongst these components, operating parameters, economic impacts and component/material choice, areas of application, heat source, power requirement and installation location are all factors to put into consideration during the deployment of ORC system.

When compared to other power cycles like the Kalina cycle studies show that ORC cycles have numerous advantages. For example, low optimum pressure in ORC application makes choice of component less expensive when compared to Kalina cycles with higher optimum pressure [79].

Exploiting low grade heat for power comes at a cost vis a vis making a trade- off between system architecture, design and acceptable mechanical and thermodynamic losses [80]. The exergy efficiency, the parameter that measures how efficient each component utilises the energy available to it to either perform work or transfer heat is depend ss on the architecture. A

20 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

RORC (recuperative ORC) produced 8.1% higher exergy efficiency than a BORC (basic ORC) as experimented by Feng et al. [81].

Once material choice and system architecture is optimized, losses and leakages often combated and minimized by using liquid receivers [82] and experimental and model iteration weigh in often manifesting in reduced net power output of the ORC system. Improving the overall and component scale efficiency through reduction in pinch point temperature, exegetic losses expander size and thermal conductance of the heat exchanger are important factors in optimizing the overall system efficiency or thermal efficiency of an ORC system [83].

The impedance for more commercial deployment of ORC has always been cost effectiveness and inefficiencies amongst others. This is usually reflected in different architecture with the aim of either getting more power, increasing cycle efficiency, reducing exergy destruction or optimizing heat addition to the system.

Although in reality, turbomachinery is more often deployed in commercial and industrial applications due to the large power scale (>100 MW) in which they operate, volumetric expanders on the other hand are readily used in laboratory as well as in small power operations.

2.3 ORC COMPONENTS

2.3.1 Pump

Pumps used in ORC systems vary from centrifugal pumps to positive- displacement pumps like the rotary type: vane, scroll or screw to the reciprocating type. The choice of pump would depend on the working fluid and other components especially the expander technology and the power

Chinedu Kingsley Unamba, August 2020 21 Experimental development of ORC technology for waste heat-to-power conversion

requirement. As organic Rankine cycles go from subcritical to supercritical where higher critical pressure is required for optimum pressure, the ratings of components making up the system of the system increases leading to larger pump requirement. This leads to an increased pump power consumption thus making supercritical systems more pump intensive and highly prone to a decreased net power.

Pump efficiencies are often of concern in supercritical power cycles. Pumps consume power and depending on the type of pump and the fluid being pumped, the power consumption might vary. In ORC applications, maintaining a low pump electrical power consumption is important because, like in all power cycles, this increases the net amount of electrical power generated by the system (after consumption by the pump).

2.3.2 Heat exchangers

In ORC systems, the choice or sizing of the evaporating heat exchanger is influenced by:

• the amount of heat transfer to the working fluid required; • the pressure drop; • the architecture of the ORC setup (presence or absence of recuperator or superheater).

This influences the number of plates as well as the plate area for heat transfer. The evaporator provides an area for heat exchange between the heat source (solar heat or industrial waste heat for example) and the working fluid. Here, the heat content or enthalpy of the working fluid is increased as the fluid changes phase form liquid to vapour state but this must be done with caution as efficiencies can also be improved by reducing the superheat [84], especially for dry or isentropic fluids.

22 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

A superheater may be added to the system after the evaporator to further increase the temperature or enthalpy of the working fluid. The extra heat provided by the superheater may be used to improve the quality of the fluid (to increase the dryness or vapour fraction) or add some degree of superheat.

Depending on the size of the heat exchanger, the thermodynamic property and the mass flow rate of the working fluid and the pinch point may vary. The pinch point is the temperature difference between the highest temperature in the ORC system (recorded from the heat source loop) and the temperature of the working fluid after exiting the evaporator. It is desirable to make the pinch point as low as possible. Larger mass flowrate increases the pinch point as there is insufficient time for heat transfer in the evaporator. The superheater can thus be used to compensate for this difference in temperature and in cases where the composition of working fluid and the mass flowrate produce a wet vapour upon exiting the evaporator.

The choice of heat exchanger is also known to influence the choice of expanders during off-design operations of an ORC engine. For example under variable heat source conditions for small and medium scale power generation, piston expanders are found to work best than a twin screw under the same condition with power losses increased from 25% to 33% [29].

The condenser provides an area for heat exchange between the working fluid and the cold fluid or heat sink. The sizing of the condenser is paramount to maximizing thermal efficiency. If heat is not adequately dumped in the condenser, the condensing pressure of the working fluid might be elevated to compensate for the increased temperature of the working fluid which will lead to a decreased pressure ratio across the expander. A decreased pressure ratio might lead to a decreased expander power output or thermal efficiency.

Chinedu Kingsley Unamba, August 2020 23 Experimental development of ORC technology for waste heat-to-power conversion

The position of the condenser with reference to the pump is also important if a NPSH (net positive suction head) is to be maintained in the pump. Having a NPSH prevents cavitation in the pump.

A liquid receiver is often place between the condenser and the pump to trap uncondensed vapour, non-condensable gas and prevent the two-phase fluid from circulating in the system. The liquid receiver also dampens the flow oscillation caused by the ratio of vapour/liquid of the working fluid between the condenser and the pump being circulated through the pump by storing and discharging more of the liquid state of the working fluid during such dynamic fluctuations. The addition of a liquid receiver is key for a sound operation of the ORC engine as it ensures that no bubble is sucked into the pump (acing as a tranquilisation chamber). This bubble entrainment in the pump together with the partial condensation in the condenser (caused by improper sizing of the condenser) could lead to flow instability in the system.

The liquid exiting the condenser should ideally be condensed to its saturated liquid state but in as much as condensation or cooling of the working fluid is required, the degree of sub-cooling (i.e., cooling the working fluid below its saturated temperature) is also important and must be properly negotiated. Lower condensation temperatures positively affect the performance of an ORC system [12]. The degree of sub-cooling gives an indication of the likelihood of cavitation occurring. If there is a degree of sub-cooling involved, cavitation could be prevented. Cavitation can be mitigated by either lowering the pump inlet temperature of the working fluid below its saturated liquid state (as done by sub-cooling) or increasing the pump suction pressure. The former can be ensured by installing a chiller between the condenser and the pump. Practically, this is avoided as chillers are expensive and also consume a lot of power, which in turn reduces the net power output of the ORC engine.

24 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

Working fluid selection is critical to the design of an ORC system. Issues such as toxicity, ODP (ozone depletion potential), GWP (global warming potential is a measure of the greenhouse potential of a compound relative to that of carbon dioxide – GWP of CO2 equals unity by definition), its compatibility with the seals and other components of the system are considered as well as the thermodynamic and thermophysical properties of the working fluid. The type of working fluid determines the internal pressure and temperature levels within the closed loop and the work potential; the thermodynamic properties of the working fluid must be investigated and chosen wisely.

The thermal stability of the working fluid is its heat resistance, its ability to all its physical properties during heating as the cyclical use of thermal fluids at high temperature make them susceptible to thermal degradation. This property indicates the quality of heat transfer a tendency that makes it resist chemical decomposition at high temperature. As temperature increases, thermal cracking and polymerisation will creep in breaking the molecular bonds that bind the molecules which will change or compromise the fluid chemical structure, decrease its thermal efficiency. Also, the molecular mass influences the performance of an ORC system. This benefit is seen in turbines as the molecular mass is inversely proportional to turbine enthalpy drop.

Due to thermodynamic constraints at lower heat-source temperature, the efficiencies of ORC systems in low-grade heat source applications are typically low and limited by either the Carnot efficiency or by a more practical efficiency metric referred to as the Chambadal-Novikov efficiency. The Chambadal-Novikov efficiency when used in analysing ORC performance considers the irreversibilities inherent in ORC operations: the exergy destruction in the system as well as the heat loss unlike the Carnot efficiency which considers the ORC engine as an idealized heat engine operating between two heat reservoirs (hot and cold) with no irreversibilities present.

Chinedu Kingsley Unamba, August 2020 25 Experimental development of ORC technology for waste heat-to-power conversion

Heat-to-power conversion from industrial waste heat often involves both relatively steady and fluctuating temperatures and mass flows. The larger the fluctuations of the heat source, the larger the deviations from steady-state (design) point performance predictions. Heat sources that may include solar thermal fluctuate at different times of the day, i.e. time-varying solar intensity. In a case like this, thermal storage technologies where heat is stored as sensible or latent heat are often employed.

Heat recovery from industrial waste heat is more challenging. Although relatively steady heat sources abound in industries such as glass manufacturing plants, utilizing heat form rapidly changing heat sources (e.g., truck or heavy-duty vehicles exhaust gases) involves highly transient behaviour of the ORC engine, which severely impacts its performance. The development of dynamic models is thus key to assess the potential of ORC engines in unsteady applications.

The dynamic response of ORC systems to temperature or mass flowrate changes in heat source provides an interesting outlook into a field that until now has been understudied. The system response time and the correlation between the step changes in the system control variable (parameters like the expander spindle speed, pump speed) and the system performance can be mapped through experimental research and the gains of deploying such system carefully debated.

2.3.3 Expanders

The expansion machine is a major component of the ORC system. It is a device that expands a working fluid by decreasing its pressure, while producing mechanical work. Its performance is very critical to the aim behind setting up an ORC system, which is the production of electrical power and it influences the performance of the whole system. Depending on the

26 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

application in which it is considered, expanders can be chosen with reference to the type of working fluid, the engine size and the power requirement. There are two main categories of expander:

1. Turbomachines or kinetic expanders, e.g., turbines; 2. Positive-displacement (or volumetric) expanders, which can be: • reciprocating expanders - e.g., reciprocating-piston; • rotary expanders - e.g., scroll, screw and rotary-vane expanders.

Turbomachines rely on the kinetic energy of the working fluid to generate a rotation of the blades. This rotary motion of the shaft being coupled to a generator induces an emf in the armature of the generator through electromagnetic induction thus providing electrical power to a load connected to it. They are suitable for medium- to high-power requirement, typically from 100s kW to 10s MW.

In volumetric expanders, the pressure of the working fluid is decreased by increasing the volume of the fluid. Most volumetric expanders differ in the way that the stator and the rotor are designed. They are suitable for medium- to low-power requirement, typically from 1 to 100s kW.

Volumetric expanders are a suitable choice when compared to turbomachines for medium- to low-grade heat-conversion technologies because they can handle two-phase expansion. Two-phase expansion is often prevalent at low temperatures. Kinetic expanders are designed to handle only gaseous phase as liquid droplets impinging on turbine blades moving at very high velocities tend to damage the turbine blade. The turbine blades are often the most critical but also most expensive part of a turbine due to the precision involved in their design and manufacture.

When comparing volumetric expanders, performances indicators such as filling factor and isentropic efficiencies are used to assess their performance.

Chinedu Kingsley Unamba, August 2020 27 Experimental development of ORC technology for waste heat-to-power conversion

In small power application <5 kW scroll expanders show a much higher isentropic efficiency 81% relative to the piston and screw expanders (53%) [85]. Leakages affect both the isentropic efficiencies and the filling factor and can affects the choice of expanders especially for rotary vane expanders [86].

Performance of expanders is very important in assessing the performance of the ORC system as they account for the second highest exergy destruction behind the evaporator [44] [87].

Scroll expanders are popular in laboratory-scale ORCs. Some of the advantages are that they have relatively high isentropic efficiency, can handle two-phase expansion and are rugged for small power scale applications. Typically, scroll expander ORC applications are in the power range of 1 to 20 kW. Expansion in this machine is caused by moving scrolls relative to each other. As the scrolls move circularly, the volume of space that houses the pocket of fluid at the centre of the expander is gradually increased until the fluid is pushed out at the tail end of the scroll, as shown in Figure 2.1.

Figure 2.1: Principle of operation of a scroll expander [88].

When volumetric expanders are used in laboratory in ORC systems, the knowledge gained in deploying, construction, troubleshooting and understanding the loss mechanisms can be extrapolated to a more practical and realistic industrial application involving turbomachinery.

28 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

More studies and experiments are being carried out in developing efficient ORC system as regards volumetric expanders as well as understanding the interplay amongst the loss mechanism inherent in them. The reciprocating piston expander has numerous advantages over its volumetric counterpart; scroll, vane, screw and so has been the subject of advanced research and work as presented in further section.

2.4 RECIPROCATING-PISTON EXPANDERS

Piston expanders are more robust, can achieve higher expander isentropic efficiencies and can accommodate wider range of pressure and volume ratios. As an alternative to the currently used scroll expander in the laboratory, reciprocating-piston expanders have shown great promise.

In order to harness the potential of reciprocating expander technologies, a proper understanding of inefficiencies arising from both mechanical and especially thermal processes is paramount. Therefore, a new experimental facility has been being developed to study the effects of the operational parameters on the performance of reciprocating expanders. This entails firstly a set of tests on a gas spring, which is a simplified reciprocating-piston expander model without gas intake or exhaust.

This allows us to focus on thermal loss and leakage mechanisms without the complexity of losses through the intake and exhaust valves. Reciprocating pistons working principle is similar to that of a petrol engine as in Figure 2.2.

Chinedu Kingsley Unamba, August 2020 29 Experimental development of ORC technology for waste heat-to-power conversion

Figure 2.2: Reciprocating-piston engine with a spark plug [89].

Previous work was done by Pfriem, Kornhauser and Smith as well as Lekic and Kok [90, 91, 92, 93] who developed models that can predict losses and the cyclic heat flux during compression and expansion of a gas spring.

Beyond lumped dynamic models, CFD is being used to predict the work loss in an expander as well as other thermodynamic and thermal (heat transfer) parameters of interest. Previous studies have been done by Taleb et al. [94] in this regard and they looked at the various loss mechanisms affecting operations of a reciprocating piston.

30 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

2.4.1 Gas spring testing facility

The experimental setup involves the use of a single stroke compressor with the compression chamber taken off as the gas spring chamber and connected to a motor. The compression chamber head is replaced by a sealed cylinder head. During operation, parameters such as P, V and T are recorded.

Table 2.1: Characteristics of the gas spring testing facility.

Connecting rod length (mm) 148.5

Bore (mm) 105

Stroke length (mm) 78

Dead volume height (mm) 14

Volume ratio 6.6

Frequency range (RPM) 27 to 300

The temperature in the chamber is measured by ultrasonic emitter and receiver. The pressure is measured by pressure transducers on the cylinder head while the volume of the gas is measured via shaft encoder’s crank angle. The gas used was nitrogen although preparations are underway to use argon. To promote understanding of reciprocating expanders, a proper knowledge of an existing similar system is needed. There is research on the study of gas springs which operates like a reciprocating expander but for the exclusion of valves. To understand reciprocating technologies and make improvements on them, the losses therein must be understood and mitigated.

Chinedu Kingsley Unamba, August 2020 31 Experimental development of ORC technology for waste heat-to-power conversion

Figure 2.3: Compression chamber showing the ultrasonic emitter, receiver and pressure transducer [95].

The losses in reciprocating expanders include heat transfer loss, friction, mass leakage and pressure losses. Considering this, gas springs can be described as reciprocating expanders without intake and exhaust valves but an enclosed space relying on a continuous expansion and compression cycle of the same volume of fluid.

32 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

(a)

(b)

Figure 2.4: (a) Experimental setup for gas-spring testing; (b) gas spring compression chamber.

The absence of valves in gas springs reduces the problem complexity and allow us to focus on other losses. But in design and study of reciprocating

Chinedu Kingsley Unamba, August 2020 33 Experimental development of ORC technology for waste heat-to-power conversion

expanders, valves losses play a major role and cannot be neglected. Advanced studied using CFD tools have simply added a system of valves unto existing gas spring models. Pressure drop across the valve during design may be further exacerbated through improper valve timing.

Gas springs also face other challenges like accumulation of debris in the expansion chamber over time which may not be present in reciprocating expanders that rely on a continuous intake and exhaust stream of working fluid. The prolonged accumulation of debris further reduces the compression or expansion inefficiency.

2.4.2 Performance of reciprocating-piston expanders

There are numerous advantages as already reported of having a reciprocating-piston expander in an ORC setup some of which are the high volume and pressure ratio, high expander isentropic efficiency. In order to explore or understand the performance of reciprocating expanders, a gas spring is being studied. The gas spring would be run to understand the relationship is performance and losses therein and to extrapolate these findings to expectations to a reciprocating expander.

Work has been done in the field of gas spring by Pfriem, Kornhauser and Smith as well as Lekic and Kok [90, 91, 92, 93] to develop models to understanding the heat flux mechanism during compression and expansion of gases. This represents one of the most challenging aspects in understanding the losses due to heat transfer in a gas spring and subsequently reciprocating expanders.

In ORC applications, friction and pressure drops can greatly affect the isentropic efficiencies of an expander and also shortening the intake period and filling factor that can affect power density [84].

34 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

2.4.3 Loss mechanisms

The inherent losses in reciprocating expanders are important as they account for how much work is done for any amount of input power. There are several ways in which the efficiency of reciprocating expander is compromised. The major losses are mechanical losses, thermal, thermodynamic and leakage in addition to pressure drop.

The pressure drop across a valve, the heat transfer to the walls, mass leakage and friction between the piston and the cylinder walls are all losses that should be minimized by good valve timing and design, better cylinder liner- piston seal contact to reduce mass leakage, good lubrication and a proper understanding of heat transfer along with solutions to minimize this.

In order understand and focus on the heat transfer losses, it was decided to investigate a gas spring apparatus (without intake and exhaust valves). Work in this field has carried by Kornhauser & Smith, Pfriem and Lekic & Kok, which led to lumped models based on heuristic parameters [90, 91, 92, 93].

The assumptions made in this earlier work were based on low volume ratios, no mass leakage and the use of light gases (hydrogen, nitrogen, helium and argon). These assumptions made the development of model less complicated but are not valid during intake and exhaust operation of an actual expander having valves and mass leakage as well. Also, other heavier gases like heavier organic compounds [94] may be tested over a larger range of volume ratio.

To address the restrictions in the assumptions made in previous studies (including concerning the gas types, the lack of mass leakage and absence of intake and exhaust valves, i.e. flow into/out of the compression and expansion space), we consider here different gas types thus comparing heavier and lighter gases in a gas spring, include mass leakage in the gas spring, and are currently adding intake/exhaust valves to a gas spring model.

Chinedu Kingsley Unamba, August 2020 35 Experimental development of ORC technology for waste heat-to-power conversion

2.4.3.1 Mechanical and leakage losses

One of the most common problems with flow devices is leaks and accounts for a significant loss in reciprocating expanders. Mechanical losses occur due to friction between the piston and the cylinder walls while mass leakage occurs when mass of the working fluid in the compression chamber escapes through the space between the piston and the cylinder walls as shown in Figure 2.5. This space is deliberate in piston design and its precision usually considered to be an important factor if not the most important in the efficiency of the expander.

(a) (b)

Figure 2.5: (a) Loss mechanisms in reciprocating machines showing piston wall section [95]; (b) piston ring [96].

The design usually gives a little clearance to reduce the friction between the piston and the wall as the piston slides up and down at very high speed. Although this friction cannot be eliminated, reducing the friction by allowing more clearance will inadvertently give rise to increased mass leakage. A trade-off must be reached which often translates to the clearance precision

36 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

and reflected economically. Effect of mass leakage results in both overall enthalpy drop and increased loss [97, 98, 99, 100].

2.4.3.2 Pressure losses

As in ORC operations, increasing the pressure of the system or the evaporating pressure is significant in improving the optimal performance of the system. However, pressure drops throughout the thermodynamic cycle adversely affect the overall system performance. The pressure losses in reciprocating-piston machines refer to the thermodynamic performance losses occurring during the intake and exhaust processes and are of two kinds: (i) pressure drops during intake/exhaust flows through the valve restrictions; and (ii) over- and under-compression losses.

The pressure drops through the intake/exhaust valves that occur during the working-fluid intake and exhaust processes into/out the expansion chamber are due to the flow resistance induced by the compressible flow through a nozzle (or restriction). The amplitude and share of this particular loss mechanism increase with increasing rotational speeds, as less time is left for the intake or exhaust flows to enter/exit the compression/expansion chamber.

Over- and under-expansion losses arise from non-ideal valve timings. For example, if the intake valve of a piston expander closes too early, the gas contained within the chamber will expand down to a pressure lower than the exhaust pressure. Hence, when the exhaust valve opens, the pressure within the expansion device exceeds that in the exhaust duct, thus creating a (reverse) flow into the chamber rather than out. Over- and under-expansion losses are particularly significant in part-load or off-design operating conditions. This can be addressed using variable valve timing.

Chinedu Kingsley Unamba, August 2020 37 Experimental development of ORC technology for waste heat-to-power conversion

(a)

(b)

Figure 2.6: (a) Pressure-volume diagram of the pressure drop through the valve. Illustration [101]; (b) valve opening [102].

38 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

2.4.3.3 Thermally induced thermodynamic losses

Heat transfer in reciprocating expanders has steady and unsteady components. The steady heat transfer component is easy to predict and to rectify with conventional insulation, but unsteady heat transfer is complex and not well understood on the other hand [103].

Unsteady gas-to-wall heat transfer can be described by a convective Equation like Newton’s law of cooling but while this Equation involves the use of a single constant convective heat transfer coefficient, this is not enough to represent the full heat transfer in a reciprocating piston. This is because the heat transfer coefficient varies over a cycle of operation, which is often ignored in the literature and usually generalize from steady heat transfer.

Figure 2.7: Loss mechanisms in reciprocating machines showing heat loss.

Chinedu Kingsley Unamba, August 2020 39 Experimental development of ORC technology for waste heat-to-power conversion

Also, there are two different temperature domains in an expanding compressing piston that can be modelled separately: the bulk volume temperature and the wall temperature. Although Newton’s law describes convective heat transfer, the out of phase heat transfer arising as a result of gas wall temperature difference during expansion/compression requires the introduction of a complex Nusselt number. Even when some studies account for the unsteady nature of the heat transfer, they do not account for the conjugate heat transfer. Conjugate heat transfer arises as a result of the temperature difference before and after compression/expansion [104].

2.4.4 Work in progress

Work is being done to gain an in-depth understanding of the effect of gases in heat transfer. A transient 3-D model is used to simulate the flow and gas- to-wall heat transfer in a piston gas spring with or without mass leakage, using different models for calculating the thermophysical fluid properties.

Earlier studies focused exclusively on light gases (e.g. hydrogen). The work done uses 3 gas models: an ideal gas model with constant thermodynamic properties, an ideal gas model with temperature dependent properties and a real gas model using Peng-Robinson Equation of state with temperature- and pressure-dependent properties. Figure 2.8 shows the effects of real gas behaviour on expansion and compression in a gas spring, as obtained from CFD simulations performed with the OpenFOAM platform.

40 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

(a)

(b)

Figure 2.8: PV diagrams for: (a) heavier gases [94]; (b) PV lighter gases [94].

Chinedu Kingsley Unamba, August 2020 41 Experimental development of ORC technology for waste heat-to-power conversion

The results from this analysis, shown in Figure 2.8a and Figure 2.8b reveal that the pressure-volume deviation from the ideal gas model is significant, overestimating the peak pressure by as much as 25% while showing a similar trend for all three models with lighter gases – nitrogen and helium. Regarding the thermal loss, for heavier gases, both ideal gas models underestimate the thermal losses – since they both underestimate the pressure losses [94].

Furthermore, Figure 2.9 shows that the thermal losses for all three models when using light gases is similar, but that with a heavier gas (propane) the temperature-dependent ideal gas model underestimates the losses as compared to other models.

42 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

(a)

(b)

Figure 2.9: (a) Thermal loss diagram for heavy gas (propane) [94]; (b) Thermal loss diagram for light gas (helium, nitrogen) [94].

Chinedu Kingsley Unamba, August 2020 43 Experimental development of ORC technology for waste heat-to-power conversion

(a)

(b)

Figure 2.10: Comparison of pressure as a function of volume over a cycle between CFD model with mass leakage and experimental data [105]; (b)

44 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

Comparison of pressure as a function of crank angle over a cycle between CFD model with mass leakage and experimental data [105].

The figure also shows peak losses up to 30% of the compression work, but losses that drop to low levels (of the order of a few % or lower) above Peclet numbers greater than 1000. As previous studies from literature assumed no mass leakage, an analysis was done to investigate how mass leakage influences heat flux and results shown in Figure 2.10.

Figure 2.10 also shows a comparison of the pressure as a function of volume and crank angle respectively over a cycle between CFD model with mass leakage, CFD model with no mass leakage and experimental values.

Figure 2.11: Comparison of heat flux between CFD model with mass leakage and experimental data [105].

Chinedu Kingsley Unamba, August 2020 45 Experimental development of ORC technology for waste heat-to-power conversion

These results suggest an overprediction of pressure by the CFD model including mass leakage and can be attributed to slight pressure loss to the crank case during experimental conditions. Figure 2.11 shows a comparison between CFD model with mass leakage and experimental values. This shows a slight match with an over prediction of the CFD model caused by overprediction of the pressure profile.

Further work was proposed on running a gas spring and including a system of valves. Due to the presence of valves, pressure drop is experienced throughout the valve operation. This usually results in lower pressure ratios and hence power.

(a) (b)

Figure 2.12: (a) Mesh for translation of sliding valve; (b) Mesh for compression and expansion of the chamber.

These pressure losses can be minimized by correct valve timing and ensuring maximum valve opening for intake and exhaust [106]. To understand the effect of the presence of valves, CFD model being built that adds the algorithm of valve operation to an already existing model. A CFD model using

46 Chinedu Kingsley Unamba, August 2020 Chapter 2: Literature review

OpenFoam, is being worked upon that will simulate the effects of valve opening and closing as shown in Figure 2.12.

In addition to the CFD simulations, an experimental facility is being constructed that introduces a valve system to a gas spring with the cylinder head being modified to accommodate two electromagnetic valves (intake and exhaust) and one pressure transducer. This arrangement is aimed to allow detailed measurements of the performance of the expander and a direct comparison of the loss mechanisms with the gas-spring results.

Chinedu Kingsley Unamba, August 2020 47 Experimental development of ORC technology for waste heat-to-power conversion

48 Chinedu Kingsley Unamba, August 2020 Chapter 3: Organic Rankine cycle modelling

3 ORGANIC RANKINE CYCLE MODELLING

This chapter presents a theorical model of ORC systems that has been developed to perform performance predictions, and the different modelling tools used in characterizing both the system and component performance. First-law and second-law indicators are derived to quantify the system and components performance, as well as the exergy destruction rate through all basic processes. Finally, a lumped predictive model derived from mass and energy conservation equations is built to: (i) explore the steady operation; (ii) investigate the influence of various control and component performance indicators on the overall ORC thermal efficiency; and (iii) provide us with reliable estimates of the pressure, temperature, heat and mass flowrate levels for a careful design and selection of the ORC components and sensors.

3.1 THE RANKINE CYCLE

The Rankine cycle is an ideal thermodynamic closed cycle composed of four processes undergone by a working fluid in a heat engine involving phase change: isentropic pressurisation, isobaric heat addition (evaporation), isentropic expansion and isobaric heat rejection (condensation).

Chinedu Kingsley Unamba, August 2020 49 Experimental development of ORC technology for waste heat-to-power conversion

An ORC engine is a Rankine heat engine operated with an organic working fluid. The four main processes involved in an ORC engine, further illustrated in Figure 3.1, are:

• Pressurisation of the working fluid, 1 – 2 • Evaporation of the working fluid, 2 – 3 • Expansion of the working fluid, 3 – 4 • Condensation of the working fluid, 4 – 1

The inlet/outlet states of the heat-source and heat-sink streams are also described: • heat addition from heat-source stream (to the working fluid), 5 – 6 • heat removal (from the working fluid) to heat-sink stream, 7 – 8

Figure 3.2 shows the temperature-entropy (T-s) diagram of an ORC engine using R245fa as a working fluid, extracting heat from a high-temperature waste-heat source at 120 °C under the operating conditions stated in Table 3.1. The governing equations solved to obtain the T-s diagram are described in Section 3.2.

Table 3.1 – Baseline operating conditions of the modelled ORC engine.

Description Value

Working fluid R245fa

Working-fluid mass flowrate 0.03 kg/s

Heat-source fluid Marlotherm oil

Heat-source temperature 120 °C

50 Chinedu Kingsley Unamba, August 2020 Chapter 3: Organic Rankine cycle modelling

Heat-source mass flowrate 0.1 kg/s

Heat-sink fluid Water

Heat-sink temperature 20 °C

Heat-sink mass flowrate 0.1 kg/s

Figure 3.1: Schematic of the architecture of a non-recuperative ORC engine.

Chinedu Kingsley Unamba, August 2020 51 Experimental development of ORC technology for waste heat-to-power conversion

8 7

Figure 3.2 - Temperature-entropy diagram of an ORC engine using R245fa as a working fluid, extracting heat from a high-temperature waste-heat source at 120 °C under the operating conditions stated in Table 3.1.

3.2 THERMODYNAMIC MODELLING

The thermodynamic analysis of an ORC engine provides us with useful insight into the underlying processes and their contribution to the overall performance.

52 Chinedu Kingsley Unamba, August 2020 Chapter 3: Organic Rankine cycle modelling

3.2.1 Assumptions

Steady-state thermodynamic models of ORC engines, such as the one presented in this section are commonly based on a set of widely accepted assumptions:

• A.1. no pressure losses in the piping (i.e., between the components) and in the evaporator and condenser units; • A.2. no heat losses to the surroundings; • A.3. no phase change occurring in the heat-sink and heat-source streams through the heat-removal and heat-addition processes; • A.4. no condensable gases trapped in the system; • A.5. the working fluid at the condenser outlet is saturated (no

subcooling, i.e., that the vapour quality is null: 푞1 = 0).

3.2.2 Component performance indicators

A set of performance indicators are used to capture the thermodynamic losses occurring in each component:

• The performance of the pump can be described by a component-scale indicator: its isentropic efficiency. The pump isentropic efficiency is defined in Equation (1) :

푊̇ pump,is 휂is,pump = . (1) 푊̇ pump,actual

• The performance of the expander is an important factor in the overall efficiency of the ORC system and can be described by a component- scale indicator: its isentropic efficiency. The expander isentropic efficiency is defined below:

Chinedu Kingsley Unamba, August 2020 53 Experimental development of ORC technology for waste heat-to-power conversion

푊̇ exp,actual 휂is,exp = . (2) 푊̇ exp,is

• The evaporator pinch-point temperature difference, ∆푇pp,evap , is defined as the minimum temperature difference between the working-fluid temperature profile in the evaporator and the heat- source temperature profile.

• the condenser pinch-point temperature difference, ∆푇pp,cond , is defined as the minimum temperature difference between the working-fluid temperature profile in the condenser and the heat-sink temperature profile.

3.2.3 Fluid properties

The working fluid, heat-sink and heat-source thermo-physical properties are obtained from the Reference Fluid Thermodynamic and Transport Properties (REFPROP) database provided by the National Institute of Standards and Technology (NIST) – which is a physical sciences laboratory and a non-regulatory agency of the United States Department of Commerce.

The equation of state (EoS) tabulated in the REFPROP database allows us to obtain both equilibrium and transport properties from two intensive thermodynamic properties, such as the pressure and temperature:

[ℎ, 푠, … ] = EoS (푇, 푃, fluid). (3)

3.2.4 Lumped-mass model governing equations

For a given working fluid, a non-recuperated subcritical organic Rankine cycle can be fully described by three thermodynamic variables that are:

54 Chinedu Kingsley Unamba, August 2020 Chapter 3: Organic Rankine cycle modelling

• the condensing pressure, 푃cond;

• the evaporating pressure, 푃evap;

• the evaporator outlet superheat, ∆푇sup = 푇3 − 푇sat(푃evap).

The pump inlet thermodynamic state is then fully defined from the knowledge of 푃cond and the zero-subcooling assumption (A.5.):

[푇1, ℎ1, 푠1] = EoS (푃cond, 푞1 = 0, fluid). (4)

At the pump outlet (at pressure 푃2 = 푃evap), the fluid thermodynamic state is obtained from: (i) the fictive enthalpy that would have been reached with an isentropic pumping process, ℎ2,is = EoS (푃2, 푠1, fluid) ; and (ii) the pump isentropic efficiency:

ℎ2,is − ℎ1 ℎ2 = ℎ1 + ; (5) 휂is,pump

[푇2, 푠2] = EoS (푃2, ℎ2, fluid) . (6)

The evaporator outlet state is obtained from the saturated evaporation temperature, 푇sat(푃evap), and the superheat, ∆푇sh:

푇3 = 푇sat(푃evap) + ∆푇sup ; (7)

[ℎ3, 푠3] = EoS (푇3, 푃3 = 푃evap, fluid). (8)

The expander state is obtained from:

• the fictive “isentropic” enthalpy, ℎ4,is = EoS (푃3, 푠3, fluid); and • the expander isentropic efficiency:

ℎ4 = ℎ3 − 휂is,exp(ℎ3 − ℎ4,𝑖푠) ; (9)

Chinedu Kingsley Unamba, August 2020 55 Experimental development of ORC technology for waste heat-to-power conversion

[푇4, 푠4] = EoS (푃4 = 푃cond, ℎ4, fluid) . (10)

At this point, all thermophysical properties of the working fluid are determined as functions of the three thermodynamic variables (condensing and evaporating pressure and evaporator superheat). Three closure equations are thus required. Two closure equations are provided by the pinch-point analysis of both the condenser and evaporator, while the third one is the overall energy balance derived from the first law of thermodynamics:

푄̇evap = 푄̇cond + 푊̇ net. (11)

This set of governing equations defines a non-liner problem (NLP) with 3 unknowns and 3 closure equations, which can be solved using NLP solvers. In this work, this problem is solved using the MATLAB fsolve function.

3.3 SYSTEM AND COMPONENTS PERFORMANCE INDICATORS

3.3.1 First- and second-law ORC performance indicators

Energy losses and thermodynamic irreversible processes hinder the performance of the ORC engine. System-scale performance indicators are derived in this section to quantify the deviation from an ideal and reversible operation.

First-law and second-law performance indicators are used to account for both energy losses (e.g., thermal losses) and the lost opportunities to produce useful work (e.g., through heat transfer over finite temperature differences). The latter can be quantified by the exergy destruction rates in each component (which are derived from exergetic analyses in Section 3.3), while the overall exergy efficiency (at the system scale) is a measure of how

56 Chinedu Kingsley Unamba, August 2020 Chapter 3: Organic Rankine cycle modelling

close the overall heat-to-power conversion thermal efficiency is to that of an ideal Carnot engine (which is a theoretical uppermost limit for the conversion efficiency).

The maximum thermal efficiency possible when operating between two thermal reservoirs is given by the Carnot efficiency in Equation (12):

푇c 휂Carnot = 1 − , 푇h (12)

where 푇ℎ is the temperature of the hottest reservoir and 푇푐 is the temperature of the coldest reservoir. This efficiency is ideal and does not consider the heat losses and exergy destruction due to heat transfer across finite temperature differences, inherent to the source-to-fluid (evaporator) and fluid-to-sink (condenser) heat transfer in ORC systems.

A more practical efficiency, which considers these irreversibilities, is the Chambadal-Novikov efficiency shown in Equation (13):

푇c 휂Novikov = 1 − √ . (13) 푇h

In reality, thermal efficiency is usually small as it captures both thermodynamic and mechanical losses arising from friction in the expander and pump and heat transfer to the environment. The thermal efficiency of the ORC system is defined as:

푊̇ net 휂thermal,ORC = , (14) 푄̇in where 푄̇in is the heat input to the system and the net power output is defined as:

Chinedu Kingsley Unamba, August 2020 57 Experimental development of ORC technology for waste heat-to-power conversion

푊̇ net = 푊̇ expander − 푊̇pump . (15)

The overall performance of the system can also be defined in terms of the system exergy efficiency, expressions of which are given in Equations (16) and (17):

휂thermal,ORC 휂exg = . (16) 휂Carnot

And also,

푋̇out 푊̇ net 휂exg = = , (17) 푋̇in 푚̇ hs(푥hs,out − 푥hs,in)

where 푋̇out is the total exergy flow out and 푋̇in is the total exergy flow into the system components.

3.3.2 First- and second-law components performance indicators

The compression, expansion, heat addition and removal processes in an actual ORC engine are not ideal. The performance of the four main components of the ORC engine is quantified through the definition of first- law and second-law indicators. This section presents a component-based performance analysis of the pump (isentropic efficiency, backwork ratio and exergy destruction rate), evaporator (isentropic efficiency and exergy destruction rate), expander (isentropic efficiency, filling factor and exergy destruction rate) and condenser (isentropic efficiency and exergy destruction rate).

58 Chinedu Kingsley Unamba, August 2020 Chapter 3: Organic Rankine cycle modelling

3.3.2.1 Pump

A component-scale indicator has already been defined to measure the pump performance: its isentropic efficiency. The pump isentropic efficiency is defined as:

푊̇ pump,is 휂is,pump = , (18) 푊̇ pump,actual while the back-work ratio is defined as:

푊̇pump 푏푤푟 = . (19) 푊̇exp

Figure 3.3: Exergy flow diagram across the pump.

Chinedu Kingsley Unamba, August 2020 59 Experimental development of ORC technology for waste heat-to-power conversion

The pump performance can also be measured through the exergy destruction rate in the pump, 퐼pumṗ , obtained from an exergy balance:

퐼pumṗ = 푊̇pump

− 푚̇ wf[ℎpump,out − ℎpump,in (20)

− 푇0(푠pump,out − 푠pump,in)].

From Figure 3.3, the exergy available to the pump is [ 푊̇ pump], the exergy used by the pump is [퐸2 − 퐸1] and the exergy destroyed is [푊̇pump − (퐸2 − 퐸1)].

The work across the pump is given by Equation (21):

푄̇pump = 푚̇ wf(ℎ2 − ℎ1). (21)

3.3.2.2 Evaporator

The performance of an evaporator in an ORC system can be quantified using the effectiveness and the exergy destruction rate. The effectiveness is defined as the ratio of the actual amount of heat transferred during evaporation to the maximum possible that can be transferred.

The exergy destruction rate across the evaporator is shown in Equation (22):

퐼evaṗ = 푇0[푚̇ wf(푠evap,out − 푠evap,in) − 푚̇ hs(푠hs,evap,out − (22) 푠hs,evap,in)].

From

Figure 3.4, the exergy available to the evaporator is [퐸5 − 퐸6], the exergy used in evaporation is [퐸3 − 퐸2] while the exergy destroyed is [(퐸5 − 퐸6) −

(퐸3 − 퐸2)].

60 Chinedu Kingsley Unamba, August 2020 Chapter 3: Organic Rankine cycle modelling

Figure 3.4: Exergy flow diagram across the evaporator.

The heat input during the evaporation process is given by Equation (23):

푄̇evap = 푚̇ wf(ℎ3 − ℎ2) . (23)

This is also equal to:

푄̇evap = 푚̇ hs푐p,hs(푇5 − 푇6) . (24)

3.3.2.3 Expander

The performance of the expander is an important factor in the overall efficiency of the ORC system and can be described by a component-scale indicator such as its isentropic efficiency, its filling factor and the exergy

Chinedu Kingsley Unamba, August 2020 61 Experimental development of ORC technology for waste heat-to-power conversion

destruction rate across it. The expander isentropic efficiency is shown in Equation (25):

푊̇ exp,actual 휂is,exp = , (25) 푊̇ exp,is

where 푊̇ exp,actual is the expander work (read form the power meter connected to the generator) and in the absence of heat loss:

푚̇ wf(ℎexp,in − ℎexp,out) 휂is,exp = , (26) 푚̇ wf(ℎexp,in,is − ℎexp,out,𝑖푠)

where ℎexp,in is the enthalpy at the expander inlet, ℎexp,out is the enthalpy at the expander outlet, while ℎexp,out,is is the enthalpy that would have been obtained at the expander exhaust if the gas had undergone an isentropic expansion from the inlet conditions to the outlet pressure level.

For a real positive-displacement expander, Equation (26) above cannot be used due to heat losses. A more suitable Equation (25) is thus used.

The exergy destruction rate across the expander is shown in Equation (27):

퐼exṗ = 푚̇ wf[ℎexp,in − ℎexp,out − 푇0(푠exp,in − 푠exp,out)] − 푊̇exp . (27)

In Figure 3.5, the exergy available to the expander is [퐸4 − 퐸3], that used by the expander is [푊̇exp] and the exergy destroyed is [(퐸4 − 퐸3) − 푊̇exp].

The filling factor which denotes the fraction of the theoretical mass flow rate that exists in the given operating condition is shown in Equation (28):

푚̇ wf 휙exp = , (28) 𝜌exp,in휔exp푉exp,disp

62 Chinedu Kingsley Unamba, August 2020 Chapter 3: Organic Rankine cycle modelling

where 𝜌exp,in is the expander inlet density, 휔exp is the scroll spindle speed and 푉exp,disp is the expander built-in displacement.

Figure 3.5: Exergy flow diagram across the expander.

The work done by the expander is given by Equation (29) in the absence of heat loss:

푊̇exp = 푚̇ wf(ℎ3 − ℎ4) . (29)

3.3.2.4 Condenser

The performance of a condenser in an ORC system can be quantified using the effectiveness and the exergy destruction rate. The effectiveness is defined as the ratio of the actual amount of heat transferred during condensation to the maximum possible that can be transferred.

The exergy destruction rate across the condenser is given by Equation (30):

Chinedu Kingsley Unamba, August 2020 63 Experimental development of ORC technology for waste heat-to-power conversion

퐼conḋ = 푇0[푚̇ wf(푠cond,out − 푠cond,in) (30) − 푚̇ cw(푠cw,cond,out − 푠cw,cond,in)] .

Figure 3.6: Exergy flow diagram across the condenser.

From Figure 3.6, the exergy available to the condenser is [퐸8 − 퐸7], the exergy used by the condenser is [퐸4 − 퐸1] and the exergy destroyed is [(퐸8 − 퐸7) −

(퐸4 − 퐸1)].

The heat output during condensation process is given by Equation (31):

푄̇cond = 푚̇ wf(ℎ4 − ℎ1) . (31)

This is also equal to:

푄̇cond = 푚̇ cw푐p,cw(푇8 − 푇7) . (32)

64 Chinedu Kingsley Unamba, August 2020 Chapter 3: Organic Rankine cycle modelling

3.3.3 Pressure drops

The need to minimize pressure drop in ORC cannot be overemphasized. Pressure drop in ORC systems is exacerbated by wrong component design, too long pipe length and inadequate bore size.

Longer pipe length than necessary leads to increased pressure drop. Pressure drop is sensitive to pipe bore sizes especially at sections of the system where the velocity is expected to be high.

Expander exit is more likely to have greater velocity due to the decreased fluid density. Since mass is conserved (continuity), lower density translates to increased velocity and since pressure drop is proportional to the square of the velocity, such section would cause the greatest pressure drop.

Also, if smaller diameter of pipe used at expander outlet is small, the pressure drop would be more significant according to Darcy-Weisbach Equation:

∆푝 𝜌 v2 = 푓 . (33) 퐿 D 2 퐷

Valves with high 퐶v are preferred as they allow for lower pressure drop:

푆퐺 퐶 = 퐺 √ , (34) v f ∆푃

where 퐶v is the flow coefficient expressed in US gal at 60 F, 퐺f is the flow rate in US gal/min, SG is the specific gravity and ∆p is the pressure drop in psi.

For fittings, the Equation below can be applied:

퐾 푓D = ; (35) 퐿eq⁄퐷 or for elbows:

Chinedu Kingsley Unamba, August 2020 65 Experimental development of ORC technology for waste heat-to-power conversion

퐾 푓 = , (36) D 푅⁄퐷

where 퐿eq⁄퐷 is the ratio of the equivalent length to the pipe diameter, and

푅⁄퐷is the ratio of the elbow radius to the pipe diameter. 푓D is the friction factor and K is the resistance coefficient, which can be used alongside the Darcy-Weisbach Equation to get the pressure drop across fittings and elbows.

3.3.4 Net pressure suction head (NPSH)

The NPSHA is the difference between the liquid pressure at the suction of the pump (suction head) and the vapour pressure of the liquid at the prevailing temperature (liquid vapour head).

To avoid cavitation, the available net positive suction head, NPSHA, must be greater than the required net positive suction head supplied by the manufacturer (NPSHR), i.e. (푁푃푆퐻퐴 > 푁푃푆퐻푅).

The difference between NPSHA and NPSHR as shown in (37) is the net positive suction head (NPSH) that should always remain +ve (푁푃푆퐻 > 0).

푁푃푆퐻 = 푁푃푆퐻퐴 − 푁푃푆퐻푅 . (37)

For open loops like pumping from an open reservoir:

2 푃 푉 푃vapour 푁푃푆퐻퐴 = [ atm + ℎ − s − Δℎ ] − [ ] , (38) 𝜌푔 s 𝜌푔 f 𝜌푔 where:

푃 atm is the atmospheric pressure head. 휌푔

66 Chinedu Kingsley Unamba, August 2020 Chapter 3: Organic Rankine cycle modelling

푃 vapour is the vapour pressure head. 휌푔

ℎs is static head of the fluid over the pump centre line.

푉 2 s is dynamic head due to pressure drop. 휌푔

Δℎf is the total head from friction loss.

To prevent cavitation, it is desirable that the temperature of the working fluid be reduced. If the fluid is cold, NPSHA and consequently the NPSH will be higher because the vapour pressure (at the now reduced temperature) will be lower.

For closed loops like the ORC system, the absolute pressure is used to determine the net pressure suction head:

2 푃 푉 푃vapour 푁푃푆퐻퐴 = [ abs + ℎ − s − Δℎ ] − [ ] , (39) 𝜌푔 s 𝜌푔 f 𝜌푔 where:

푃 abs is the absolute pressure head (푃 = 푃 + 푃 ). 휌푔 a gauge atm

ℎs is static head of the fluid over the pump centre line.

푉 2 s is dynamic head due to pressure drop. 휌푔

푃 vapour is the vapour pressure head. 휌푔

Δℎf is the total head friction loss.

Chinedu Kingsley Unamba, August 2020 67 Experimental development of ORC technology for waste heat-to-power conversion

68 Chinedu Kingsley Unamba, August 2020 Chapter 4: Design simulations

4 DESIGN SIMULATIONS

The resolution of the system of non-linear equations presented in Section 3.2 allows us to investigate the ORC performance dependence upon: (i) the component performance indicators; and (ii) the operating conditions. The fluid-state numbers used in this Chapter refer to those defined in Figure 4.2.

Figure 4.1: Schematic of the architecture of a non-recuperative ORC engine.

Chinedu Kingsley Unamba, August 2020 69 Experimental development of ORC technology for waste heat-to-power conversion

4.1 THERMODYNAMIC DIAGRAMS

The operation of an ORC engine can be further described by using thermodynamic diagrams. In this section, the performance and operation of an ORC engine are explored.

Table 4.1 – Operating conditions and components performance indicators used for the numerical simulations.

Description Value

Working fluid R245fa

Working-fluid mass flowrate 0.03 kg/s

Heat-source fluid Marlotherm oil

Heat-source temperature 120 °C

Heat-source mass flowrate 0.1 kg/s

Heat-sink fluid Water

Heat-sink temperature 20 °C

Heat-sink mass flowrate 0.1 kg/s

Evaporator pinch point ∆푇pp,evap = 8 K

Condenser pinch point ∆푇pp,cond = 8 K

Pump isentropic efficiency 휂is,pump = 90 %

Expander isentropic efficiency 휂is,exp = 70 %

70 Chinedu Kingsley Unamba, August 2020 Chapter 4: Design simulations

With the conditions stated in Table 4.1, the thermal efficiency predicted by the lumped model is 7.74% (against a theoretical Carnot efficiency of 25.4%, which corresponds to an exergy efficiency of 30.4%), with a net power output of 506 W and heat input of 6.5 kW. The corresponding temperature-entropy diagram shown in Figure 4.2 provides further insight into the thermodynamic processes. In particular, the area contained between the heat-source profile (dotted line) and the working-fluid profile in the evaporator (2 - 3) is proportional to the exergy destroyed in this component, while that contained between the heat-sink profile (dashed-dotted line) and the working-fluid profile in the condenser (4 - 1) is proportional to the exergy destroyed in the condenser. It appears that the latter is lower than the exergy destroyed in the evaporator.

Figure 4.2 - Temperature-entropy diagram of an ORC engine using R245fa as a working fluid, extracting heat from a high-temperature waste-heat source at 120 °C under the operating conditions stated in Table 4.1.

Chinedu Kingsley Unamba, August 2020 71 Experimental development of ORC technology for waste heat-to-power conversion

The exergy analysis of the condenser – based on the use of Equation (30) shown previously – and of the evaporator – using Equation (22) – indicates indeed that the exergy destruction rate in the evaporator (1.39 kW) is 16% higher than that in the condenser (1.1 kW).

Figure 4.3 shows an alternative representation of the thermodynamic cycle undergone by the working fluid. The pressure-enthalpy diagram exhibits the heat and work into/out the heat engine and helps identify:

• the ratio between the heat input – proportional to (풉ퟑ − 풉ퟐ) – and

gross power output – proportional to (풉ퟒ − 풉ퟑ) (in other words, the P-h diagram provides a visual estimate of the thermal efficiency); • the pressure ratio applied to the expander.

Figure 4.3 - Pressure-enthalpy diagram of an ORC engine using R245fa as a working fluid, extracting heat from a high-temperature waste-heat source at 120 °C under the operating conditions stated in Table 4.1.

72 Chinedu Kingsley Unamba, August 2020 Chapter 4: Design simulations

4.2 INFLUENCE OF THE EXPANDER ISENTROPIC EFFICIENCY

The performance of the expansion machine powered by the high-pressure, high-enthalpy vapour flow leaving the evaporator is a key parameter in the overall ORC performance. The metrics used to quantify the latter, the expander isentropic efficiency, typically varies between 30% and 90% depending on the expander type and operating conditions (e.g., off-design operation leads to over- or under-expansion thermodynamic losses [107]).

Figure 4.4 shows the influence of the expander performance on both the thermal efficiency and net power output, which both linearly increase with increasing the isentropic efficiency.

Figure 4.4 - Influence of expander isentropic efficiency on ORC performance using R245fa as a working fluid, taking heat from a waste-heat source at 120 °C and operating conditions in Table 4.1.

Chinedu Kingsley Unamba, August 2020 73 Experimental development of ORC technology for waste heat-to-power conversion

4.3 INFLUENCE OF PINCH-POINT TEMPERATURE DIFFERENCE

Evaporator and condenser performance is difficult to predict. It requires advanced two-phase moving-boundary models to determine the effective heat transfer. Alternatively, a pinch-point constraint can be applied to determine the effectiveness of the heat-addition and heat-removal processes.

Figure 4.5 - Influence of pinch-point ΔT in evaporator and condenser on ORC performance using R245fa as a working fluid, extracting heat from a waste- heat source at 120 °C and operating conditions in Table 4.1.

In Figure 4.5, the influence of the pinch-point temperature difference in both evaporator and condenser on the ORC performance is shown. Increasing the pinch-point constraint logically decreases the net power output and thermal

74 Chinedu Kingsley Unamba, August 2020 Chapter 4: Design simulations

efficiency. In other words, increasing the heat exchanger effective area or heat transfer coefficient (through the addition of porous coatings for example) leads to significant improvements in the ORC capability.

4.4 INFLUENCE OF THE WORKING FLUID MASS FLOWRATE

An important control parameter in ORC engines is that of the pump speed, which can be easily changed during operation. Increasing the latter leads to increasing the working fluid mass flowrate. A given ORC system is expected to perform best at given – often so-called design – operating conditions.

Figure 4.6 – Influence of working fluid flowrate on ORC performance using R245fa as a working fluid, extracting heat from a waste-heat source at 120 °C and operating conditions in Table 4.1.

Chinedu Kingsley Unamba, August 2020 75 Experimental development of ORC technology for waste heat-to-power conversion

Figure 4.6 shows an example of how the ORC performance varies with the operating pump speed. For constant component performance indicators (i.e., expander and pump isentropic efficiencies, constant pinch-point constraints in the heat exchangers), the net power output is found to peak for a working- fluid mass flowrate of 0.035 kg/s under the operating conditions stated in Table 3.1, while the thermal efficiency monotonically decreases with increasing the working fluid mass flowrate.

4.5 INFLUENCE OF THE HEAT-SOURCE TEMPERATURE

Figure 4.7 - Influence of heat-source temperature on ORC performance using R245fa as a working fluid, extracting heat from a waste-heat source at 120 °C and operating conditions in Table 4.1.

ORC performance, similar to all other thermodynamic heat engines, is directly dependent on the source-to-sink temperature difference, as shown theoretically by the definition of Carnot efficiency. In real endo- and exo- irreversible applications, the dependence of the net power output and

76 Chinedu Kingsley Unamba, August 2020 Chapter 4: Design simulations

thermal efficiency upon the boundary conditions imposed on the system is less straightforward. Figure 4.7 shows an example of the expected variation of Rankine-type heat-engine performance with heat-source temperature.

4.6 POTENTIAL OF LOW-GWP WORKING FLUIDS

The organic compound used in the experiments presented in this thesis is R245fa. The main reasons why this particular fluid has been chosen are: (i) safety reasons – R245fa is a non-flammable and non-toxic organic fluid; (ii) practical reasons – the R245fa saturation pressure at room temperature (ranging from 25 to 20 °C) is above atmospheric pressure, which prevents non-condensable gases (mainly nitrogen and oxygen) or humidity to be sucked into the testing facility once filled; and (iii) environmental reasons – R245fa is a non-ozone depleting fluid, however, it is a hydrofluorocarbon (HFC) with a large global warming potential (GWP = 1030).

In this section, the potential of alternate, low-GWP working fluids is explored – including hydrofluoroolefins (HFOs) and hydrocarbons (HCs).

4.6.1 Hydrofluoroolefin R1233zd

R1233zd is a hydrofluoroolefin (HFO) typically used as a low-GWP replacement of R123 in vapour-compression chillers. It is non-flammable and exhibits a GWP as low as 4.5, which makes it a great candidate for use in low-temperature ORC applications (critical temperature of 166.4 °C).

Figure 4.8 (a) shows how the performance of the ORC engine operated with R1233zd as the working fluid depends on the pump speed as this is varied (resulting in working-fluid mass flowrate variations). For constant component performance indicators (i.e., expander and pump isentropic efficiencies, constant pinch-point constraints in the heat exchangers), the net

Chinedu Kingsley Unamba, August 2020 77 Experimental development of ORC technology for waste heat-to-power conversion

power output is found to peak for a working-fluid mass flowrate of 0.033 kg/s under the operating conditions stated in Table 3.1, i.e. with a heat source at 120 °C: 506 W power output and 7.2% thermal efficiency.

Figure 4.8 (b) shows the dependency of the ORC performance (thermal efficiency and net power output) on the heat-source temperature for a fixed mass flowrate of 0.0033 kg/s. For a 160 °C heat source, nearly 1 kW is generated by the heat engine operated with R1233zd, with a 11.7% thermal efficiency. In comparison with R245fa, the ORC engine operated with R1233zd exhibits similar thermal efficiency levels but proves to produce a higher net power output.

78 Chinedu Kingsley Unamba, August 2020 Chapter 4: Design simulations

(a)

(b)

Figure 4.8 - Influence of: (a) working fluid flowrate; and (b) heat-source temperature on the ORC performance using R1233zd as a working fluid under the operating conditions stated in Table 4.1.

Chinedu Kingsley Unamba, August 2020 79 Experimental development of ORC technology for waste heat-to-power conversion

4.6.2 Hydrofluoroolefin R1234ze

R1234ze is another promising hydrofluoroolefin (HFO), which exhibits a GWP as low as 7. It is however flammable, so specific safety features must be put in place for its use, especially in large scale systems. It is yet a great candidate for use in low-temperature ORC applications (with a critical temperature of 109.4 °C).

Figure 4.9 (a) shows how the performance of the ORC engine operated with R1234ze as a working fluid varies with the pump speed (i.e., mass flowrate). For constant component performance indicators (i.e., expander and pump isentropic efficiencies, constant pinch-point constraints in the heat exchangers), the net power output is maximum at a working-fluid flowrate of 0.045 kg/s under the operating conditions stated in Table 3.1, i.e. with a heat source at 120 °C: 511 W power output and 6.8% thermal efficiency. Figure 4.9 (b) shows the dependency of the ORC performance (thermal efficiency and net power output) on the heat-source temperature for a fixed mass flowrate of 0.0045 kg/s. Due to the low critical temperature of R1234ze (109.4 °C), the performance is reported only for heat-source temperatures lower than 130 °C, at which 680 W is generated with a 7.7% thermal efficiency. In comparison with R245fa, the ORC engine operated with R1234ze exhibits similar thermal efficiency levels but is limited to lower temperature applications.

80 Chinedu Kingsley Unamba, August 2020 Chapter 4: Design simulations

(a)

(b)

Figure 4.9 - Influence of: (a) the working fluid mass flowrate; and (b) the heat-source temperature on the ORC performance using R1234ze as a working fluid under the operating conditions stated in Table 4.1.

Chinedu Kingsley Unamba, August 2020 81 Experimental development of ORC technology for waste heat-to-power conversion

4.6.3 Hydrocarbon Isopentane

Isopentane (R601a) is a hydrocarbon and promising candidate for waste-heat recovery applications using ORC engines. Even though it exhibits a low GWP (5) and is non-ozone depleting, isopentane is to handle with great care as it is highly flammable (safety group A3). It exhibits a critical temperature of 187.2 °C and can thus be used over a large range of heat-source temperatures.

Figure 4.10 (a) shows how the performance of the ORC engine operated with R1234ze as a working fluid varies with the pump speed (i.e., mass flowrate). For constant component performance indicators (i.e., expander and pump isentropic efficiencies, constant pinch-point constraints in the heat exchangers), the net power output is found to peak for a working-fluid mass flowrate of 0.02 kg/s under the operating conditions stated in Table 3.1, i.e. with a heat source at 120 °C: 510 W net power output with a 7.3% thermal efficiency.

Figure 4.10 (b) shows the dependency of the ORC performance (thermal efficiency, power output) on the heat-source temperature for a fixed flowrate of 0.02 kg/s. For a 160 °C heat source, 1.1 kW is generated by the heat engine operated with isopentane, with a 10.7% thermal efficiency. In comparison with other working fluid investigate in the current Chapter, the ORC engine operated with isopentane proves to produce the highest net power output.

82 Chinedu Kingsley Unamba, August 2020 Chapter 4: Design simulations

(a)

(b)

Figure 4.10 - Influence of: (a) the working fluid mass flowrate; and (b) the heat-source temperature on the ORC performance using isopentane as a working fluid under the operating conditions stated in Table 4.1.

Chinedu Kingsley Unamba, August 2020 83 Experimental development of ORC technology for waste heat-to-power conversion

84 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

5 EARLY-STAGE ORC TESTING FACILITY

In this Chapter, results are drawn from tests performed on a small-scale (1 kWe) early-stage ORC testing facility, made of a rotary-vane pump, a brazed- plate evaporator and a brazed-plate condenser, a scroll expander with a built-in volume ratio of 3.5 and using R245fa as the working fluid. An electric oil-heater acted as the heat source, providing hot oil at temperatures in the range 120 - 140 °C. Based on the generated data, relationships were drawn between the operating conditions of the ORC engine and changes in the heat- source temperature, pump and expander speeds leading to working fluid flow rates between 0.0088 kg/s and 0.0337 kg/s, from which performance maps are derived.

5.1 SYSTEM DESCRIPTION

The principle prototype in the ORC facility was used to determine the optimal operation conditions for the ORC. The setup comprised the basic ORC architecture; pump, evaporator, expander and condenser.

Chinedu Kingsley Unamba, August 2020 85 Experimental development of ORC technology for waste heat-to-power conversion

Figure 5.1: P&I diagram of the principle prototype ORC testing facility [87].

The piping and instrumentation (P&I) diagram shown in Figure 5.1 describes the placement of the major components and of the control instrumentation and sensors. It also shows the flow path of the heat source and heat sink.

86 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

5.1.1 Pump

The pump used in the ORC testing facilities is a positive-displacement rotary vane pump consisting of vanes rotating inside a cavity. Rotary vanes are usually lubricated with oil which not only lubricates the bearings but also provides hydrodynamic seal between the high- and low-pressure sides. The consideration for selecting the size of rotary vane pump was based on knowledge a priori of various parameters like pressure limit of the expander (14 bar abs) to be used, the pump cost and maximum flowrate of the working fluid. The specification of the pump is shown in Table 5.1.

The pump was placed between the condenser and the evaporator. The vane pump was connected to a speed controller which was used to change the speed of the pump. The pump speed was adjusted in steps of 1100, 1500, 1750, 2000, 2500, 2750, 3000 and 3500 RPM. Attached to the pump discharge line is a sonic flow meter that recorded the volume flowrate of R245fa.

(a) (b)

Figure 5.2: (a) Fluid-O-Tech rotary vane pump TMFR 30-200 [108]; (b) variable frequency drive [108].

Chinedu Kingsley Unamba, August 2020 87 Experimental development of ORC technology for waste heat-to-power conversion

Table 5.1: Fluid-O-Tech rotary vane pump (TMFR 30-200 series) specification.

Rotary vane pump TMFR 30-200 series

Maximum differential pressure 16 barg

Static pressure 20 barg

Minimum flowrate 80 L/h

Maximum flowrate 550 L/h

Speed range 1100 RPM to 3500 RPM

Actual power 250 W

5.1.2 Evaporator

The evaporator provides an area for heat exchange between the heat source and the working fluid. Here the heat content is increased as the fluid changes phase form liquid to vapour. The sizing of the heat exchange based on the amount of heat to be transferred, the pressure drop. From this requirement, the number of plates as well as the plate area was determined.

The refrigerant passes from the pump through the evaporator. The heat source is an electric oil heater simulating waste heat from industrial process whose oil temperature range is from 0 - 200 °C. The heating of the oil takes place in the oil heater chamber and is transported via a flexible loop to the heat exchanger - evaporator.

88 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

A superheater is added to the system after the evaporator to optionally improve the quality of the fluid or provide an extra degree of superheat. The specifications of the oil heater is shown in Table 5.2.

(a) (b)

Figure 5.3: (a) SWEP Brazed plate heat exchanger – evaporator [109]; (b) Frigosystem 18-kW electric oil heater [110].

Table 5.2: Frigosystem 18 kW(th) electric oil heater specification.

Cooling capacity 52 kW

Heating capacity 18 kW

Flow capacity 4800 L/h

Electric power 20.2 kW

Chinedu Kingsley Unamba, August 2020 89 Experimental development of ORC technology for waste heat-to-power conversion

5.1.3 Expander

The choice of the scroll expander for the experiment was influenced by the cost of the equipment, its isentropic efficiency and nominal power. Scroll expanders have good isentropic efficiency, can handle two phase expansion and are rugged for small power scale applications.

Figure 5.4: 1-kW semi-hermetic oil-free scroll expander (E15H022A-SH) [111].

90 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

Table 5.3: Specification for the 1-kW semi-hermetic oil free scroll expander

(E15H022A-SH). Nominal output 1 kW(e)

Volume ratio 3.5

Displacement 14.5 cm3

Max speed 3500 RPM

Maximum inlet pressure 13.8 bar(a)

Maximum inlet temperature 175 °C

The performance of the scroll expander as indicated by its isentropic efficiency at various revolution supplied by the manufacturer in Figure 5.5.

Figure 5.5: Performance map (Airsquared) at 1.25 barg condensing pressure [111].

Chinedu Kingsley Unamba, August 2020 91 Experimental development of ORC technology for waste heat-to-power conversion

5.1.4 Condenser

The condenser provides an area for heat exchange between the cold fluid and the working fluid. In doi ng so, it changes the working fluid from the vapour phase to liquid phase while rejecting the heat to cooling water (tap water). Figure 5.6 shows a 22-kW Alfa Laval condenser as used in this work.

Figure 5.6: 22-kW Alfa Laval Condenser CB60-30H-F Model [112].

The working fluid exiting of the expander while in vapour or two-phase state is condensed as it passes the condenser. It then exits into a liquid receiver. The condenser is a water-cooled counter flow heat exchanger.

92 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

Table 5.4: 22 kW Alfa Laval condenser specification.

Condenser model CB60-30H-F (Alfa Laval)

Condenser type Brazed-plate

Condenser area 1.62 m2

The condenser cools the working fluid from vapour state to its liquid state. At the exit of the condensers is a liquid receiver that traps he uncondensed or non-condensable gases.

5.2 EXPERIMENTAL PROCEDURES

The early-stage ORC testing facility was operated over a range of pump speed, heat-source temperature and expander resistive load. The aim was to determine the optimal performance of the system by adjusting the operating conditions and to generate a performance map. The performance indicators were the expander power output, cycle efficiency and expander isentropic efficiency. The performance map was derived by changing some operational parameters or control variables like the heat-source temperature, pump speed and generator resistive load.

The ORC rig comprises a rotary vane pump (TMFR SS051A), a 10.83 kW SWEP brazed plate heat, a 1 kW scroll expander (E15H022A-SH) magnetically coupled to a 2.5 kW Voltmaster generator and a 22 kW Alfa Laval brazed plate condenser (CB80-30H-F).

Chinedu Kingsley Unamba, August 2020 93 Experimental development of ORC technology for waste heat-to-power conversion

Figure 5.7: Pictures of the early-stage ORC testing facility in the Clean Energy Processes Laboratory.

94 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

The working fluid R245fa flows through the pump and is pressurized in the pump. The pressurized fluid then goes into the evaporator where it is evaporated to its saturated vapour state or slightly superheated.

The heat source was from an 18 kW(th) electric oil heater. This simulated the waste heat the evaporating temperature was adjusted via the electric thermal oil heater that uses Marlotherm as the oil. The oil industrial oil has a thermal decomposition temperature of 360 °C. The heat source setpoint was kept at 120 °C, 130 °C, 140 °C. The Alpha Laval plate heat exchanger condenser was water cooled with water from the tap as the cold stream. The cooling water entered the condenser at 19 °C and 2 bar (absolute) and exited at approximately 21 °C at a flowrate of 22 L/min.

The frequency of the expander was not imposed by an inverter or the electricity grid but depended directly on the attached generator load or the expander resistive load. The expander is a volumetric type, a 1-kW semi hermetic scroll expander magnetically coupled to a generator. The generator is rated 2.4 kW.

For a given heat source temperature, the pump speed was varied from 1100 RPM to 3500 RPM (this caused a change in mass flow rate and increase in pump outlet pressure and expander torque).

The system was allowed to reach steady-state conditions before values were recorded. The pressures and the temperatures at the inlet and exit of each component were measured via pressure transducers and thermocouples.

Two expander resistive load banks were used to adjust the torque applied on the rotating expander shaft. Case A called “partial load” where two 100-W light bulbs were connected in parallel with an effective resistance of 50 – 70 Ω. Case B called “full load” where two 100 W light bulbs and two 100 Ω resistors were connected in parallel with an effective resistance of 25 – 30 Ω.

Chinedu Kingsley Unamba, August 2020 95 Experimental development of ORC technology for waste heat-to-power conversion

5.3 DATA ANALYSIS

The system was allowed to reach steady-state conditions before values were recorded. The pressure and the temperature values at the inlet and exit of each component were measured via pressure transducers and thermocouples. There were 8 thermocouples and 8 pressure transducers at the inlet and exit of each component.

Using the temperature and pressure of the fluid, the other fluid properties like the enthalpy were looked up using the Reference Fluid Thermodynamic and Transport Properties (REFPROP) database provided by the National Institute of Standards and Technology (NIST) – which is a physical sciences laboratory and a non-regulatory agency of the U.S. Department of Commerce.

The working-fluid flow rate was measured using an ultrasonic flow meter at the pump inlet, while the expander-generator electrical power output and ORC pump electrical input power were measured using digital power meters.

The uncertainty associated with the processed experimental results (for example on the thermal efficiency) arises from uncertainty in the raw measurements that propagate through the data processing. It is determined by an uncertainty analysis, described in Appendix A.

The highest uncertainty was in the measurement of the flow meter by the ultrasonic device. Error propagation was accounted for in the sensors and measuring/recording devices. Time averaging was done, and data was collected using Keysight DAQ. System temperatures were measured using 1.5mm T-type thermocouples with a stated accuracy of ±1 °C, and system pressures were measured using analogue pressure transducers (PXM209) with a full-scale accuracy of ±0.25%.

96 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

5.4 STEADY-STATE PERFORMANCE RESULTS

5.4.1 Pump performance

In Figure 5.8a as the pump speed increases, the pressure ratio increases as expected, resulting in an increase in the working-fluid mass flow rate.

Hence, the electrical power required by the pump at these higher pressure ratios, or pump speeds, increases as shown in Figure 5.8b. There is no significant change in the pumping power requirement with a change in the heat-source temperature, or in the generator load. The steep increase in pump electrical power at the later part of the pressure ratio in Figure 5.8b is due to the large pressure drop with high pump rotational speed.

The outliers seen in Figure 5.8a is due to flow instability in the mass flowrate measurements believed to be caused by flow oscillations generated by instabilities in the condenser unit. As the condenser outlet is aligned vertically with the pump inlet, it is very likely that, in under certain operating conditions, a two-phase mixture is sucked into the pump (as gravity is not helping to drive the liquid downwards). When this happens, the pump displaces a lower amount of working fluid for a given RPM (because the equivalent density of a two-phase mixture is less than that of pure liquid) and the actual mass flowrate drops, thus leaving more time for the vapour to condense in the condenser unit. It follows that more liquid is sucked into the pump, which then displaces a larger amount of working fluid, thus increasing the mass flowrate again. The amount of vapour leaving the condenser increases and the mass flowrate drops again. Unlike the Coriolis flow meter, the ultrasonic flow meter used here is sensitive to two phase flow thereby producing flow measurements that were limited in accuracy.

Chinedu Kingsley Unamba, August 2020 97 Experimental development of ORC technology for waste heat-to-power conversion

(a)

(b)

Figure 5.8: (a) Effect of pump speed, heat-source temperature and expander/generator load on mass flow rate; (b) effect of pressure ratio, heat- source temperature and expander/generator load on pump electrical power.

98 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

5.4.2 Expander performance

Corresponding performance results relating to the operation of the expander are shown in Figure 5.9. The electrical power generated by the expander- generator combination increases as the pressure ratio increases, as expected. When the Case B load is engaged, the system is able to generate a maximum electrical output of 600 W at a pressure ratio of 4.3 (corresponding to 60% of the stated nominal expander-generator output).

For the partial-load configuration, a maximum electrical power output of 200 W is achieved, again at a pressure ratio of 4.3. At Case B load, an increasing heat-source temperature results in a slightly increased expander power output and an increase in the pressure ratio measured across the expander. At Case A load, increasing the heat source temperature showed little or no advantage in terms of increasing electrical power output, particularly at higher pressure ratios. As can be seen from Figure 5.9a, the divergence of magnitude of power between the two Cases (A and B) is due to the increase in pressure drop with pump rotational speed. This increased pressure drop together with the increased mass flowrate increases the power output.

Chinedu Kingsley Unamba, August 2020 99 Experimental development of ORC technology for waste heat-to-power conversion

(a)

(b)

Figure 5.9: (a) Effect of pressure ratio, heat-source temperature and generator load on expander electrical power; (b) effect of pressure ratio, heat-source temperature and generator load on expander efficiency.

100 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

The isentropic efficiency of the expander attains a maximum (around ~80%) at pressure ratios between 3.2 and 3.8 at Case B load, as shown in Figure 5.9. This is higher than expected from the manufacturer (~75%). This can be explained by the error propagation from the measurement of the mass flow rate as well as the temperature. There are errors associated with temperature measurements and these errors are often exacerbated when the data analysis involves temperature difference.

Interestingly, the intermediate heat-source temperature appears to give the maximum overall isentropic-efficiency value, at a pressure ratio of 3.7, which this is considered to be a suitable pressure ratio across the expander. Nevertheless, this observation must be seen in the context of the relative errors in the measurement of the efficiency indicated by the bars in Figure 5.9, which are of the order of ±10%.

5.4.3 ORC overall efficiency

The cycle thermal efficiency is plotted in Figure 5.10, and this increases generally with pressure ratio at both Case A and Case B load. The overall trend is a result of the expander power output increasing, shown in Figure 5.9a. At Case B load, the cycle efficiency increases with pressure ratio also because the generator load can absorb more of the power generated by the expander-generator compared to the part-load conditions, reaching a maximum of ~6% at a pressure ratio of 3.6. At part generator load, the thermal efficiency shows a slight decrease as the pressure ratio increases beyond 3.1. This shows that in Case B, the system has potential for higher cycle efficiency. As seen earlier in Figure 5.9a, Case B results in higher expander power and thus consequently higher cycle efficiency. This is as a result of more torque available to rotate the expander spindle in Case B than in case A resulting in more expander power.

Chinedu Kingsley Unamba, August 2020 101 Experimental development of ORC technology for waste heat-to-power conversion

(a)

(b)

Figure 5.10: (a) Effect of pressure ratio, heat-source temperature and generator load on cycle thermal efficiency; (b) effect of pressure ratio, heat- source temperature and generator load on total exergy destruction.

102 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

5.4.4 Thermodynamic analysis of the optimum configurations

The exergy analysis was performed on the four main system components and the total exergy destruction is plotted in Figure 5.11. It is observed that the exergy destruction increased at higher pressure ratios but was insensitive to the heat-source temperature at which the tests were performed and to the generator load used on the expander-generator.

A further analysis of the exergy breakdown within the system (normalized by the total) at Case A and Case B load are shown in Figure 5.12a and Figure 5.12b respectively. The results have been plotted for six different operating condition Cases. Cases 1, 2 and 3 refer to minimum pressure-ratios and Cases 4, 5 and 6 to maximum pressure ratios, at heat-source temperatures of 120 °C, 130 °C and 140 °C respectively.

Aside optimisation of ORC systems by proper selection of working fluid [113], the results suggest that the expander accounts for the second highest exergy destruction after the evaporator, which agrees well with previous studies [51, 114, 87, 115].

The exergy destruction in the condenser is slightly lower than the expander, whilst the exergy destruction in the pump is negligible.

At both Case A and Case B load, an increase in the heat-source temperature results in higher exergy losses, however, these results are normalized by the total, and indicate that at higher pressure ratios and Case B load the losses shift from the evaporator more towards the expander (up to 20-25%, depending on the temperature).

Chinedu Kingsley Unamba, August 2020 103 Experimental development of ORC technology for waste heat-to-power conversion

(a)

(b)

Figure 5.11: (a) Effect of pressure ratio, heat-source temperature and generator load on expander exergy destruction; (b) effect of pressure ratio, heat-source temperature and generator load on evaporator exergy destruction.

104 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

(a)

(b)

Figure 5.12: (a) Exergy destruction percentages of different components at different pressure ratios at part generator load; (b) exergy destruction fractions of different components at different pressure ratios at Case B load.

Chinedu Kingsley Unamba, August 2020 105 Experimental development of ORC technology for waste heat-to-power conversion

5.5 TESTING FACILITY LIMITATIONS

ORC systems are designed to produce power and yet are often characterized by inefficiencies. Some of these inefficiencies arise during design, installation, non-optimal operating conditions and from wrong material selection. This section presents the limitation of the proof of principle ORC rig and why high-fidelity data could not be generated.

5.5.1 Cavitation

The conditions at pump inlet are quite important and ideally should be set to provide a temperature and pressure enough to prevent the working fluid from boiling and collapsing thus damaging the pump material. Ideally a higher pressure than the saturated vapour pressure of the working fluid or a lower temperature than the saturated vapour temperature is required to prevent cavitation. The ORC rig produced a lot of noise which is believed to be a consequence of bubbles formation and collapsing - cavitation. The positioning of the condenser was almost at the same level as the pump resulting in the suction head being less by the gravity head.

5.5.2 Flow instability and calibration

Instrument error arose from poor temperature sensor and pressure transducer calibration, vibrational disturbance from pump and expander, thermal drift for the pressure sensors at temperatures higher than its design temperature.

The instability in the mass flowrate measurements is believed to be caused by flow oscillations generated by instabilities in the condenser unit due to the generation of two-phase fluid flows. The ultrasonic flow meter used here

106 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

is sensitive to two-phase flow and does not capture such density changes accurately.

5.5.3 Pressure drops

The need to minimize pressure drop in ORC system cannot be overemphasized. Proper sizing of the evaporator as well as using right valves and fittings.

Some of the valves and elbows used caused large pressure drop. The rig was quite compact, and the fluid had to change direction several times resulting in sharp bends.

The gains made from shorter pipe length was annulled by sharp bends. The pipe diameter at the expander exit accounts for one of the largest pressure drop as pressure drop is proportional to the square of the velocity. The diameter was not large enough and resulted in even more pressure drop.

5.5.4 Corrosion

Corrosion of pipes occurs during ORC operation on the external or internal wall. Corrosion can be mechanical resulting in pitting or electrochemical resulting in oxidation.

The copper tubes used had a lot of copper carbonate formed on the external walls. It was difficult to tell which debris was from corrosion when the inside walls were inspected during decommissioning as there was a lot of debris from decomposed nitrile seals, the effect can result in heat transfer reduction as well as pressure drop inside the tubing.

Chinedu Kingsley Unamba, August 2020 107 Experimental development of ORC technology for waste heat-to-power conversion

5.5.5 Non-condensable gases

Venting is not permitted in most ORC operations as the fluid may not just be toxic, it could be hazardous or flammable, damage the ozone layer and increase global warming. Non-condensable gas in the system remains trapped in while in circulation with the working fluid. Examples include nitrogen, carbon dioxide, air. They are detrimental to ORC operations as they run the risk of creating a negative pressure upon condensation. This effect can result in air being sucked in from the surrounding atmosphere through minute spades between fittings into the system.

This leak through joints between fittings can result in the adulteration of organic fluid over time and can also be seen in the measurement of the flow rate as this creates a lot of flow instability. This could have been averted by a proper initial vacuum testing of the rig before charging the system with the refrigerant.

5.5.6 Seal compatibility

Thermal decomposition of the seal arises when the seal used in between fittings to prevent leak decomposes at temperatures higher than its design temperature. There is also chemical decomposition of the seals due to its incompatibility with the working fluid. The seals used here were made from nitrile. Nitrile decomposes at ~125 °C, this was lower than the highest temperature in the system which was ~140 °C. The result was noticed as black debris accumulated at various sections in the pipe when the rig was decommissioned.

108 Chinedu Kingsley Unamba, August 2020 Chapter 5: Early-stage ORC testing facility

5.5.7 Wall heat losses

An important performance indicator is the thermal efficiency of the ORC. It tells how much electrical power is being extracted from the thermal energy input. Heat can be wasted through poor insulation.

There was a lot of heat loss form the copper tubes used as they were not properly insulated. Copper is highly thermally conductive (385 W/mK) than steel (50 W/mK) and is prone to dissipate heat to the surrounding at a much faster rate.

Not protecting tubing or pipes from heat loss to the surrounding can result in large temperature drops along tubing which can further complicate data analysis.

There was absence of adequate tube insulation at the high temperature side especially between the evaporator and expander which meant that there was less heat available to the expander. Also, with poor insulation around the heating oil tubing, there was less heat available to the evaporator from the electric oil heater.

Chinedu Kingsley Unamba, August 2020 109 Experimental development of ORC technology for waste heat-to-power conversion

110 Chinedu Kingsley Unamba, August 2020 Chapter 6: Improved flexible ORC prototype

6 IMPROVED FLEXIBLE ORC PROTOTYPE

The thermal efficiency of ORC engines suffers from inadequate component sizing (e.g., expander built-in volume ratio that does not correspond to that imposed by the working fluid and source-to-sink temperature ratio). Even when components are appropriately chosen, the operating conditions may deviate from the design conditions. Irreversibilities (both thermal and mechanical) also cause further inefficiencies. All these make optimizing an ORC through design operation invaluable.

The ORC functional protype is a designed improvement on the functional prototype. it addresses cavitation in the pump, flow instability and calibration/sensor errors, data acquisition, reduction or complete elimination of debris in the fluid system, compatible of seal. The facility was designed with an optional recuperator and superheater and provision was made for quick connect and disconnect of various components including expander and pump in the event of a maintenance or repair. Figure 6.1 shows the newly improved testing facility.

Chinedu Kingsley Unamba, August 2020 111 Experimental development of ORC technology for waste heat-to-power conversion

Figure 6.1: ORC functional prototype showing major components.

6.1 TESTING FACILITY IMPROVEMENTS

This section details the different methods that were employed to overcome that limitations on the early stage ORC rig.

6.1.1 Cavitation

Although the NPSH was not calculated, the suction head was increased by increasing the static head (0.6m above the pump centre line) which prevented cavitation. The effect was seen as the noise level in the pump dropped (quieter) significantly.

112 Chinedu Kingsley Unamba, August 2020 Chapter 6: Improved flexible ORC prototype

6.1.2 Flow instability and calibration

The rig was fitted with more modern version of the pressure transducer than the earlier rig (PXM 309) and more sensitive 1 mm T type thermocouple. The thermocouples were calibrated against cold water at 0 °C and boiling water at 100 °C while the difference was compensated for in the data acquisition system. A more recent version of the pressure transducer was chosen and calibrated. A more modern version of the Keysight DAQ was used with better scanning speed.

Figure 6.2: Coriolis MFC 400 Optimass flow meter [116].

The Coriolis flow meter with much higher measurement accuracy capable of handling two phase fluid was also chosen as shown in Figure 6.2.

Chinedu Kingsley Unamba, August 2020 113 Experimental development of ORC technology for waste heat-to-power conversion

6.1.3 Pressure drops

The valves and fittings were chosen after calculations with the right (high) coefficient of flow and K values. Shorter pipes were used where necessary and sharp bends were avoided. Minimal use of valves and bends were all implemented to significantly reduce the pressure drop.

Also, large diameter pipe at the expander exit prone to high velocity and consequently larger pressure drop was used.

6.1.4 Corrosion

Copper tubes were replaced with stainless steel tubes. The use of copper also comes with oxidation as formation of carbonates and oxides are found over time in inner and outer wall of the tube. The use of stainless steel ensures that corrosion is reduced if not eliminated.

6.1.5 Non-condensable gases

A good pressure test was done on the followed by a vacuum test. The vacuum test was carried out held at 4 mbar and the setup is shown in Figure 6.3. This was to ensure that all unwanted gas was expelled. The rig was fitted with a liquid receiver (Figure 6.1) to trap the unwanted gas that may be present.

114 Chinedu Kingsley Unamba, August 2020 Chapter 6: Improved flexible ORC prototype

Figure 6.3: Vacuum test setup during functional ORC Rig commissioning.

6.1.6 Seal compatibility

The nitrile seal which decomposes at 125 °C was replaced with a more thermally stable seal - Viton (fluorocarbon rubber) which can withstand even higher temperatures up to 200 °C.

6.1.7 Wall heat losses

The walls were insulated using high quality as shown in Figure 6.4. Tubing are protected with different materials Rockwool was used on the high

Chinedu Kingsley Unamba, August 2020 115 Experimental development of ORC technology for waste heat-to-power conversion

temperature side while nitrile rubber was used on the low temperature side. The wall heat loss through tubing. Protecting tubing or pipes from heat loss to the surrounding can reduce large temperature drops along tubing and improve data analysis and reduce errors.

Figure 6.4: ORC functional rig showing sheathed tubing on hot and cold sides.

6.2 SYSTEM DESCRIPTION

The ORC engine design incorporates a pump, evaporator, expander, condenser, superheater, recuperator and a liquid receiver. Changes in evaporating temperature, pressure ratio, mass flowrate, expander load was investigated, and performance maps drawn.

116 Chinedu Kingsley Unamba, August 2020 Chapter 6: Improved flexible ORC prototype

Figure 6.5: P&I diagram of the ORC rig with recuperator and superheater.

Figure 6.5 shows the piping and instrumentation diagram show the ORC rig piping and the different processes the working fluid takes. It also shows the flow path of the path of the heat source and heat sink. On the P&ID diagram, the ORC process involving the basic components (pump, evaporator, expander and condenser) is shown with bold colours.

The optional components which include the superheater, recuperator and metering valve are greyed out. The optional components are used depending on the need of the extra requirement of the experiment. For example, the superheater is used increase the superheated state of the working fluid or to heat the working fluid to a state of either saturated vapour or superheat if it is in a two-phase region. The recuperator is used to preheat the working fluid before it is evaporated in order to get the working fluid closest to the desired superheated state as possible.

Chinedu Kingsley Unamba, August 2020 117 Experimental development of ORC technology for waste heat-to-power conversion

The metering valve serves as a bypass for the working fluid. In order to increase or decrease the mass flow rate, the metering valve is used which varies the mass flowrate without the need to change the speed of the pump.

A liquid receiver is placed in the pump suction line, downstream of the condenser to trap non-condensable gases (such as air) and potentially non- condensed vapour, so that only liquid-phase fluid is sucked into the pump. The expander shaft is connected to a single-phase asynchronous generator with magnetic couplings, so as to avoid leakage from the expander case.

Table 6.1: ORC testing facility characteristics.

Working fluid

Refrigerant R245fa

Mass flowrate [20 - 60] g/s

Heat source

Thermal fluid Marlotherm oil

Mass flowrate 1.4 kg/s

Available supply temperature [80 - 150] °C

Heat sink

Cooling fluid Water

Mass flowrate 0.4 kg/s

Supply temperature 18 °C

118 Chinedu Kingsley Unamba, August 2020 Chapter 6: Improved flexible ORC prototype

Pump

Model TMFR2 (Fluid-o-Tech)

Type Magnet-driven rotary vane pump

Pump rotating speed [1100 - 3000] RPM

Heat exchangers

Condenser model CB60-30H-F (Alfa Laval)

Condenser type Brazed-plate

Condenser area 1.62 m2

Evaporator model B12Lx18 (SWEP)

Evaporator type Brazed-plate

Expansion machine

Model E15H22N4.25 (Air Squared)

Type Scroll expander

Displacement 14.5 cm3/rev

Nominal power 1 kWe

Built-in volume ratio 3.5

Chinedu Kingsley Unamba, August 2020 119 Experimental development of ORC technology for waste heat-to-power conversion

6.3 EXPERIMENTAL PROCEDURES

This section presents the procedures adhered to during commissioning of the rig to ensure that results subsequently generated are within the limits of experimental error.

6.3.1 Pressure testing

The pressure testing was carried out in the same way as the leak test but instead of the vacuum pump, nitrogen bottle at 200 bar was used. The rig was pressurized in different section isolating other parts that were not required. Ideally pressure testing should be done up to 1.5 times the maximum operating pressure. The maximum operating pressure is at 12 barg limited by the expander. The ORC rig was pressure tested and held at 20 barg and held but a bypass was made at the inlet and exit of the expander to prevent it from being damaged. Once the integrity of various parts was confirmed, the entire rig is filled with pressured gas while the DAQ monitors the pressure drop. The rig is left for a day and the pressure drop on the DAQ monitor. This was repeated until there is no pressure drop in the system.

6.3.2 Vacuum testing

A vacuum test is was done as shown in Figure 6.3 test that involves reducing or removing the content of a sealed chamber or system until the pressure drops below atmospheric and to absolute zero. In this state there is neither the presence of air nor contaminant. In addition to a system reaching such a low vacuum pressure, it must be able to hold for a considerable amount of time say 1 hour as many systems fail vacuum leak test because the initially vacuum state simply fails to hold as the system leaks after certain time due

120 Chinedu Kingsley Unamba, August 2020 Chapter 6: Improved flexible ORC prototype

to the presence of water vapour. The ORC rig was vacuumed in stages (as the pressure test) down to 14 mbar and held for 1 hour.

6.3.3 System filtration

System filtration involves filtering the organic fluid to remove debris that may have accumulated during operation. The cleaning of ORC fluid is a requirement in major operations. The filter installed was Swagelok stainless steel Tee-Type particulate filter. A 440 μm pore size and 1/2-inch port size, part number (SS-8TF-440). This type of Swagelok filter can be replaced without removing the body ensuring that the highly volatile working fluid is not exposed to both the personnel and the environment during the filtering process. In order to filter, the upstream bypass filter valve is opened. This bypasses the mass flowmeter and continues the normal system route to the evaporator. Debris added to the system will give false flowrate reading. In addition, they will not expand in the expander and over time will clog sensors and build up in components. If not checked, will reduce volumetric capacity of fluid holding components as well as reduce the accuracy of sensors. Debris arise from installation wrong procedures as they might be introduced during building of the system.

6.3.4 Steady-state operation

The rig is run by starting the electric oil heater with a set point of either 100 °C, 120 °C or 140 °C. It was left to run for an hour to heat up the system in order to achieve a quasi-steady equilibrium with the system. The pump is turned on while the variable frequency drive is set to provide a pump rotational speed of 1100, 1500, 1750, 2000, 2500, 2750, 3000 and 3500 RPM. The expander initially bypassed (to prevent two phase fluid from entering the expander during start-up) is then opened allowing the circulation of the

Chinedu Kingsley Unamba, August 2020 121 Experimental development of ORC technology for waste heat-to-power conversion

working fluid through the system. For each heat source temperature, a particular pump speed is chosen and an array the resistive loads connected to the expander via the generator is varied from 10 to 55 Ω. The system is allowed run for 1 hour to achieve steady state. Then the data acquisition system is activated to collect 70 data points. This is repeated for different combinations of resistive load, pump speed and heat source temperature.

6.4 DATA ANALYSIS

The ORC rig was connected to data acquisition system (DAQ, Keysight DAQ970) and relevant power supplies for data collection. It acquired data from all the sensors and flowmeters on the rig. The electronic connections on the rig were in three categories: the pressure sensors (Omega PXM 209 and 309, 4-20 mA output, 28 V excitation voltage), thermocouples (T-type) and flow meters.

A total of 21 pressure transducers (Omega PXM 309 with 4-20 mA output and 28-V excitation voltage) and 21 thermocouples (T-type) are installed at various positions, mainly at the inlet and outlet of each component in the rig, of which 17 sets of sensors (one pressure transducer and one thermocouple) are used to monitor the state parameters of the working fluid, while the remaining 4 sets are used to record the temperature and pressure of the heat source and the heat sink

The data acquisition system receives all electronic signals conditions it and converts into analogue signals for monitoring, management, modification and display via a computer. It thus collects the current output from the strain on the pressure transducer and the potential difference between the thermocouple ends as well as the voltage and current from all the flow meters and converts it into electrical signal and sends it into 3 different

122 Chinedu Kingsley Unamba, August 2020 Chapter 6: Improved flexible ORC prototype

module slots built into the rear of the DAQ. The analogue signals are 4-20 mA from the pressure transducers, voltage from the thermocouple, and both current and voltage signals from the flow meters.

Each of the three modules (DAQ901A) receives analogue signal from the sensors and flowmeters on ORC rig. Each module consists of a 20-channel solid state multiplexer card. The cards are solid state standardized instead of vacuum tube to ensure faster signal travel time. Each multiplexer card can receive up to 20 signals.

Three flow meters, a high-precision Coriolis flowmeter (Krohne Optimass 6000), a turbine flowmeter and an Omega RTD flowmeter, are used to measure the working-fluid, the heat-source (oil) and the cooling-water mass flowrates, respectively. A power analyser (Rohde & Schwarz HMC 8015) is connected to the generator to measure the voltage, currency, frequency and the consumed electricity by the resistance. The gross power output of the ORC system, i.e., the power generated by the scroll expander, is obtained. The power consumed by the pump is measured by a power meter (Brennenstuhl PM 231E), which is subtracted from the gross power output to determine the ORC net power output

All aforementioned sensors/measurements were connected to a DAQ system for data acquisition. In addition, the changes of the parameters such as pressure, temperature and other thermodynamic or transport properties can be captured and used for detailed analysis and dynamic investigation of component and system performance, which is of particular importance under transient conditions. The DAQ has a scan speed of 450 channels per second. A single scan entails data acquisition from the 45 channels.

The accuracy of the data was limited by the errors being propagated and the method in which efforts amplify the signal to noise ratio was employed.

Chinedu Kingsley Unamba, August 2020 123 Experimental development of ORC technology for waste heat-to-power conversion

The method assumes that signals become stable or quasi stable over several measuring cycles. The more measurements are taken and averaged out, the better the result. The first step in time averaging involved data measurement and acquisition. This is done by a Keysight data acquisition (DAQ970A) system and in 2 stages.

The measurements are taken repeatedly over a pre-set time frame and stopped after 100 samples were taken. The number of measurements is pre- set with no limit of time frame for approximately 2 minutes or 120 seconds, which corresponds to about 72 data points.

Then a simple averaging was done using MATLAB and the properties of the fluid were looked up using REFPRO software. The performance of the ORC system and of its components was then assessed through the analysis of the time-resolved data recorded with the DAQ system. The uncertainty on the temperature, pressure, volumetric and mass flowrates data acquired during the experimental runs comes from the uncertainty on the sensor’s readings, derived from the sensors accuracy as given by the manufacturers and from the standard deviation of the time-resolved data around their mean value. The combined uncertainty on a single measurement is obtained from the summation in quadrature of both contributions to provide a 95% confidence interval on the steady value. This is further explained in Appendix A.

124 Chinedu Kingsley Unamba, August 2020 Chapter 7: Experimental results of the non-recuperative ORC prototype

7 EXPERIMENTAL RESULTS OF THE NON-RECUPERATIVE ORC PROTOTYPE

This chapter presents results from an experimental investigation of the part- load performance of a 1-kWe ORC engine, operated with R245fa as a working fluid, with the aim of producing high-fidelity steady-state and transient data relating to the operational performance of this system.

However, ORC systems experience variations in performance at part-load or off-design conditions, which needs to be predicted accurately by empirical or physics-based models if one is to assess accurately the techno-economic potential of such ORC-WHR solutions. The experimental apparatus is composed of a rotary-vane pump, brazed-plate evaporator and condenser, and a scroll expander with an adjustable load. An electric heater is used to provide a hot oil-stream to the evaporator, supplied at three different temperatures: 100 °C, 120 °C and 140 °C.

The optimal operating conditions (pump speed, expander load), are determined at various heat-source conditions, resulting in 124 steady-state data points used to analyse the part-load performance of the engine.

Chinedu Kingsley Unamba, August 2020 125 Experimental development of ORC technology for waste heat-to-power conversion

A maximum thermal efficiency of 4.2% is reported for a heat-source temperature of 120 °C, while a maximum power output of 508 W is obtained for a temperature at 140 °C. An exergy analysis allows us to quantify the contribution of each component to the overall exergy destruction.

The share of the evaporator, condenser and expander components are all significant for the three heat-source conditions, while the exergy destroyed in the pump is negligible by comparison (below 4%). The data can be used for the development and validation of advanced models capable of steady- state part-load and off-design performance predictions, as well as predictions of the transient/dynamic operation of ORC systems.

7.1 OPERATING PARAMETER OPTIMISATION

The optimal values of the operating – or control – parameters (that are the resistive load, 푅exp , connected to the expander generator and the pump speed, 휔p) vary with the external conditions, i.e., with the cooling-water and heat-source conditions. They are determined for various heat-source temperatures, ranging from 100 °C to 140 °C, through a parametric optimisation that allows for the determination of the optimum-load and to investigate the part-load performance of the ORC engine. The pump RPM and expander load are varied through the available ranges of 1100 RPM to 3500 RPM and 10 ohm to 55 ohm respectively and the steady-state ORC performance is recorded systematically.

7.2 PART-LOAD PERFORMANCE MAPS

The working fluid mass flowrate, 푚̇ wf, and the expander spindle speed, 휔exp are intelligible variables that can be readily used in comprehensive or heuristic models, together with other system parameters.

126 Chinedu Kingsley Unamba, August 2020 Chapter 7: Experimental results of the non-recuperative ORC prototype

(a)

(b)

Figure 7.1: ORC part-load performance maps: (a) contour plots of the net power output; and (b) of the thermal efficiency as functions of the expander spindle speed, 휔exp, and of the working fluid mass flowrate, 푚̇ wf, for a heat- source temperature of 100 °C.

Chinedu Kingsley Unamba, August 2020 127 Experimental development of ORC technology for waste heat-to-power conversion

Figure 7.1, Figure 7.2 and Figure 7.3 show the part-load performance maps for three investigated temperatures (100 °C, 120 °C and 140 °C), using both the net power output and the thermal efficiency as performance indicators.

The contour plots are made of isometric curves ranging from the lowest to the highest value measured for a set heat-source temperature which allows to visually ascertain the optimal working conditions. With the heat-source temperature set at 120 °C, a maximum thermal efficiency of 4.2±0.1% (that corresponds to a power output of 399±5 W and 16.2±0.3% exergy efficiency) is reached for a mass flowrate of 35.3±0.2 g/s, according to Figure 7.2b, while a maximum net power of 470±7 W (which corresponds to 3.9±0.1% thermal efficiency) is generated with a working fluid mass flowrate of 45.1±0.3 g/s and a resistive load of 36 Ω according to Figure 7.2a.

Unlike the latter, the conditions found to maximize the thermal efficiency constitute a set of global optimum variables, as changing further the pump

RPM or expander load would not lead to higher 휼퐭퐡 values. Higher power production could however be obtained with the current apparatus in similar conditions, if higher working fluid flowrates were investigated – which unfortunately was not possible with the current equipment.

Moreover, the operating conditions corresponding to the maximum net power differ from those that correspond to the maximal thermal efficiency, as a higher mass flowrate also increases the heat input. Table 7.1 summarizes the results obtained for the three heat-source temperatures investigated.

128 Chinedu Kingsley Unamba, August 2020 Chapter 7: Experimental results of the non-recuperative ORC prototype

(a)

(b)

Figure 7.2: ORC part-load performance maps: (a) contour plots of the net power output; and (b) of the thermal efficiency as functions of the expander spindle speed, 휔exp, and of the working fluid mass flowrate, 푚̇ wf, for a heat- source temperature of 120 °C.

Chinedu Kingsley Unamba, August 2020 129 Experimental development of ORC technology for waste heat-to-power conversion

Two objective functions are considered to determine the optimum results in

Table 5.2, shows net power output (max 푾̇ 퐧퐞퐭) and thermal efficiency (max

휼퐭퐡). Maximising the latter is equivalent to maximising the exergy efficiency,

휼퐞퐱퐠 , as the Carnot efficiency is fixed for given heat-source and heat-sink temperatures, which is why only two optimisation scenarios are listed.

For a single heat-source temperature, the mass flowrate found to maximize the efficiency is lower than that maximizing the power output, as a higher flow leads to a higher heat input.

For 푻퐡퐬 = 100 °C, however, the same operating parameters are found to maximize both the power and efficiency of the system, with a steady cycle that exhibits a relatively low superheat at the exhaust of the evaporator,

∆푻퐬퐮퐩 = 7.2 °C. Increasing further the working fluid flowrate led to a “negative” superheat, i.e., a two-phase mixture at the expander intake, and a drop in thermal efficiency. This limit could however be exceeded with a higher evaporating capacity, either with an increased heat-source mass flowrate or increased evaporator area.

130 Chinedu Kingsley Unamba, August 2020 Chapter 7: Experimental results of the non-recuperative ORC prototype

(a)

(b)

Figure 7.3: ORC part-load performance maps: (a) contour plots of the net power output; and (b) of the thermal efficiency as functions of the expander spindle speed, 휔exp, and of the working fluid mass flowrate, 푚̇ wf, for a heat- source temperature of 140 °C.

Chinedu Kingsley Unamba, August 2020 131 Experimental development of ORC technology for waste heat-to-power conversion

The operating optimum locus, which designates the set of operating parameters that minimize (or maximize) an objective function, is found to vary with the heat-source temperature, as observed in Figure 7.1, Figure 7.2 and Figure 7.3.

Regardless of the chosen performance indicator, the higher 푇hs, the higher the working fluid mass flowrate that favours the ORC performance. In other words, the higher the heat-source temperature, the higher the optimum power potential. Yet, the maximum thermal efficiency reached at 140 °C, 휂th = 4.1±0.1%, remains low compared either: (i) to the theoretical limit of the Carnot efficiency that is 29.5% with a heat-sink temperature at 18 °C, or (ii) to the Chambadal-Novikov efficiency that is a theoretical limit derived from endo-reversible thermodynamics, equal to 16% in the current conditions.

Proceeding further, the pump RPM and expander load were varied over the available ranges and the steady-state ORC performance was recorded systematically. Figure 7.4 presents an example of a map thus obtained, where the measured gross power output is presented as a function of the control variables, for a heat-source temperature fixed at 140 °C.

The value of the optimum expander load does not vary significantly with the imposed pump speed (as it remains around 30 Ω), while a maximum expander power output of 670 ± 2 W is obtained for a pump rotating velocity of 3000 RPM, as seen from the contours plotted at the base of Figure 7.4.

The second-law indicator that is the exergy efficiency is equal to 13.9±0.4% for the optimal configuration with 푇hs = 140 °C. This relatively low value is explained by the large superheat at the inlet of the expander unit, ∆푇sup = 37.5 °C, which denotes a lost potential.

132 Chinedu Kingsley Unamba, August 2020 Chapter 7: Experimental results of the non-recuperative ORC prototype

Figure 7.4: ORC gross power output part-load performance map as function of the pump speed, 휔p and the expander resistive load, 푅exp (which represents the braking torque applied to the expander shaft), for a heat- source temperature set at 140 °C.

Indeed, the high superheat at the exhaust of the evaporator reflects that a higher evaporating pressure could have been reached, in turn leading to higher power output for a similar heat input.

In the present scenario, it is believed that this limit is due to the lack of expansion vessel in the testing facility, which imposes a constraint on the thermodynamic cycle as the overall mass distributed in the different components and piping is constant. The effects of the mass load and the development of charge-sensitive models are discussed by Dickes et al. [117].

Chinedu Kingsley Unamba, August 2020 133 Experimental development of ORC technology for waste heat-to-power conversion

7.3 THERMODYNAMIC ANALYSIS OF THE OPTIMUM CONFIGURATIONS

Control parameters were identified to maximize simultaneously the thermal and exergy efficiencies at various heat-source conditions. This section aims at further analysing the thermodynamic losses in these optimum configurations, both from a first- and second-law point of view.

(a) (b)

Figure 7.5: Temperature-entropy diagrams presenting the steady operation of the ORC engine with efficiency-optimum operating conditions for heat- source temperatures of: (a) 120 °C, and; (b) 140 °C. Working-fluid state reported at key locations of the facility: (1) pump inlet or condenser outlet, (2) evaporator saturated liquid, (3) evaporator saturated vapour, (4) evaporator exhaust, (5) expander inlet, (6) expander outlet, (7) condenser saturated vapour, (8) condenser saturated liquid.

134 Chinedu Kingsley Unamba, August 2020 Chapter 7: Experimental results of the non-recuperative ORC prototype

The steady operation and working-fluid states at key locations of the ORC engine are presented in Figure 7.5 that shows temperature-entropy (T-s) diagrams of the optimum (maximizing efficiency) configurations, as measured for heat-source temperatures of 120 °C and 140 °C. The working fluid circulated by the pump enters the evaporator in subcooled liquid state (1). The heat transferred from the thermal oil in the evaporator preheats the liquid-phase fluid (1-2), before this undergoes an isothermal evaporation (2- 3) and superheating in the last section of the heat exchanger (3-4). Thermal losses in the pipes lower the vapour superheat at the entrance of the expander (5), where the refrigerant expands (5-6). The work done on the scroll shaft is dissipated in the resistive load bank. The working fluid is finally condensed (6-8) to a subcooled liquid state before returning to the pump (1).

Figure 7.6: Exergy destruction contributions of various components to the overall exergy destruction in the ORC engine, operated with efficiency- optimum pump speed and expander load at three heat-source temperatures.

The exergy destroyed in the processes undergone by the working fluid appear graphically in the T-s diagrams. First, the vapour expansion process (5-6) is a non-isentropic process, due both to the thermal losses to the

Chinedu Kingsley Unamba, August 2020 135 Experimental development of ORC technology for waste heat-to-power conversion

surroundings and to the irreversibility of heat- and mass-transfer occurring within the device. The performance of the scroll expander is further analysed in the expander section below. The mismatch observed between the temperature profiles of the working fluid and heat-source and cooling water streams in the heat exchangers is another source of exergy loss.

136 Chinedu Kingsley Unamba, August 2020 Chapter 7: Experimental results of the non-recuperative ORC prototype

Table 7.1: Maximum performance indicators for various heat-source temperatures.

푻퐡퐬 푾̇ 퐧퐞퐭 휼퐭퐡 휼퐞퐱퐠 휼퐞퐱퐠,퐢퐬 풎̇ 퐰퐟 흎퐞퐱퐩 푷퐞퐯퐚퐩 푷퐜퐨퐧퐝 휟푻퐬퐮퐩 휟푻퐬퐮퐛

(°C) (W) (%) (%) (%) (g/s) (Hz) (bar) (bar) (°C) (°C)

Max 푾̇ 퐧퐞퐭

100 359±12 4.1±0.2 18.7±0.7 37±1 35.7±0.2 30.1±0.3 9.0±0.06 1.5±0.06 7±3 3.4±2.8

120 470±7 3.9±0.1 15.4±0.3 36.4±0.5 45.1±0.3 34.5±0.1 11.2±0.06 1.6±0.06 18±2.8 3.7±2.8

140 508±3 3.9±0.1 13.5±0.2 36.6±0.3 46.1±0.3 35.5±0.1 11.8±0.06 1.75±0.06 35±2.8 4.4±2.8

Max 휼퐭퐡 (= 휼퐞퐱퐠)

100 359±12 4.1±0.2 18.7±0.7 37±1 35.7±0.2 30.1±0.3 9.0±0.06 1.5±0.06 7.2±3 3.4±2.8

120 399±5 4.2±0.1 16.2±0.3 37.3±0.4 35.3±0.2 31.8±0.1 9.6±0.06 1.5±0.06 26±2.8 4.2±2.8

140 493±10 4.1±0.1 13.9±0.4 36.8±1 43±0.3 34.5±0.2 11.3±0.06 1.73±0.06 37.5±3 4.3±2.8

Chinedu Kingsley Unamba, August 2020 137 Experimental development of ORC technology for waste heat-to-power conversion

The area contained between the curve (1-4) and the heat-source profile (as well as between (6-8) and the heat-sink profile) is directly proportional to the exergy destruction in the evaporator (or condenser). From Figure 7.5, it is observed that the exergy destroyed in the evaporator: (i) is larger than that destroyed in the condenser, and (ii) increases significantly with the heat- source temperature, while that destroyed in the condenser remains fairly constant between the two diagrams plotted.

The contributions of the different components of the ORC engine to the overall exergy destruction is presented using pie charts in Figure 7.6. It presents the share of the condenser, evaporator, expander and pump in the thermodynamic losses throughout the cycle, for the three heat-source temperatures investigated, as taken from the efficiency-optimum configurations listed Table 7.1

It is first observed that the losses incurred in the pump represents a constantly small amount compared to those in other components. The thermal losses and irreversibility in the evaporator and expander appear to generate the most significant part of the exergy destruction in all Cases, while the exergy destruction rate in the condenser remains under 20% of the total, gradually increasing from 15% to 19% for heat-source temperatures ranging from 100 °C to 140 °C.

The lost availability in the evaporator increases significantly with 푇hs from 36% to 46%, and represents the main source of loss in the configuration selected at 140 °C. This confirms the observations made on the T-s diagram. The contribution of the expander, on the other hand, decreases significantly with an increase in heat-source temperature, from 46% down to 32%. The performance of the expansion unit depends on various factors, which are further described in the next section.

138 Chinedu Kingsley Unamba, August 2020 Chapter 7: Experimental results of the non-recuperative ORC prototype

7.4 EXPANDER PERFORMANCE

The performance of a scroll expanders is adversely affected by four main loss mechanisms that are:

• radial and flank internal leakage; • heat-transfer losses; • friction losses; • under or over-expansion losses that occur due to a mismatch between the built-in and actual pressure ratios across the expander [118].

Two performance indicators are presented in Figure 7.7, using all steady- state measurements gathered through the parametric optimisation study.

Chinedu Kingsley Unamba, August 2020 139 Experimental development of ORC technology for waste heat-to-power conversion

(a)

(b)

Figure 7.7: (a) Expander isentropic efficiency expressed in %, mapped against the scroll spindle speed 휔p and the pressure ratio, 푟p. Performance map obtained from cubic-spline interpolation of experimental data, indicated as grey dots on the map, obtained for heat-source temperatures ranging from 100 °C to 140 °C; (b) expander filling factor as a function of the expander spindle speed.

140 Chinedu Kingsley Unamba, August 2020 Chapter 7: Experimental results of the non-recuperative ORC prototype

A part-load performance map of the scroll expander is shown in Figure 7.7a, where the contours of the (electrical) isentropic efficiency are presented as functions of the pressure ratio imposed across the device, 푟p, and the scroll spindle speed, 휔p.

First, it is worth noting that neither the pressure ratio nor the expander rotation speed have independent impact on the efficiency. The overall drop of the expander performance observed when increasing the pressure ratio is due to increased pressure losses and under-expansion phenomena, while the drop of efficiency at high RPMs is mainly due to increased pressure losses through the intake and exhaust ports and increased friction. At low RPMs, the decrease in efficiency is due to increased thermal and mass-leakage losses.

As observed in previous studies [119, 120], the filling factor, 휙exp, plotted in Figure 7.7b, is found to decrease monotonically with the expander rotation speed, with values close to those measured in the cited references.

The effect of the inlet temperature is also shown, as the filling factors reported for higher temperatures are greater than those at lower temperatures for a fixed RPM. For clarity reasons, error bars are not shown in Figure 7.7b; the computed filling factors exhibit relative uncertainties below 2%.

7.5 DYNAMIC CHARACTERISTICS

In real-life applications, the working conditions of the ORC engine vary dynamically. The performance variation of the system can be predicted with advanced transient modelling or by integrating time constants in quasi- steady models. Experimental time-resolved data and step-response characteristics are key for developing dynamic system models.

Chinedu Kingsley Unamba, August 2020 141 Experimental development of ORC technology for waste heat-to-power conversion

The time response of the expander power output to rising or falling steps in pump rotating velocity is investigated for a heat-source temperature set and held constant at 120 °C and for an expander resistive load of 25 Ω. Once a steady operation is attained – with 휔p = 1100 RPM – the data collection is initiated before increasing the pump speed up to 2000 RPM, and later decreased back to the initial value.

The dynamics of the ORC system is investigated by recording the time response of the power generated in the expander to a rising or falling step in pump rotational speed. The time-resolved data, φ(t), are reported using non- dimensional transient parameters, 휑̅(푡) defined as:

휑(푡) − 휑 휑(푡) = init . (40) 휑fin − 휑init

The influence of the step amplitude on the rising and falling times is using the expander power step amplitude as a reference is defined as:

훥푊̇exp = |푊̇ exp,fin − 푊̇ exp,init| . (41)

The dimensionless step change for the pump is given by Equation (42):

휔p(푡) − 휔p,init 휔̅̅̅p̅ = . (42) 휔p,final − 휔p,init

The dimensionless system response is given by Equation (43)

̇ ̇ ̅̅̅̅̅̅ 푊exp(푡) − 푊exp,init 푊̇exp = . (43) 푊̇ exp,final − 푊̇ exp,init

The rising-step response is reported in Figure 7.8a using non-dimensional parameters, as defined in Equation (40) . The rising (or falling) time, 휏, that is the time taken for the response to rise (or fall) from 10% to 90% of the

142 Chinedu Kingsley Unamba, August 2020 Chapter 7: Experimental results of the non-recuperative ORC prototype

steady-state response, is thus determined from the analysis of the transient response.

The dynamic test is repeated with various steps, namely between 1100 RPM and 1500 RPM and between 1100 RPM and 2500 RPM. The influence of the step amplitude on the rising and falling times is reported in Figure 7.8b, using the expander power step amplitude as a reference ass defined in Equation (41).

All system response time to rising steps in the pump velocity are higher than the corresponding falling steps, by 7% for the largest step tested and by up to 30% for the smallest. In other words, the ORC system reacts more rapidly to a step down in power than to a step up, as far as only the pump speed is varied.

For both rising and falling steps, it is also observed that the response time decreases with increasing step amplitude. This denotes that the ORC-engine inertia has a more significant impact for smaller power steps. Care must then be taken for predicting the dynamic behaviour of ORC engines using quasi- steady models that estimate the time-varying performance based on steady- state results.

Chinedu Kingsley Unamba, August 2020 143 Experimental development of ORC technology for waste heat-to-power conversion

(a)

(b)

Figure 7.8: (a) Dynamic response of the ORC system to a step in pump velocity from 1100 to 2000 RPM at fixed heat-source temperature (푇hs = 120

°C) and expander load (푅exp = 25 Ω); (b) rising and falling step-response times as functions of the step amplitude in expander power

144 Chinedu Kingsley Unamba, August 2020 Chapter 8: Conclusion

8 CONCLUSION

Series of experimental tests were made on a small-scale (1 kWe) ORC testbed facility (proof-of-principle prototype), comprising of a rotary-vane pump, a brazed-plate evaporator and a brazed-plate condenser, a scroll expander with a built-in volume ratio of 3.5, and using R245fa as the working fluid.

In order to meet some of the objectives earlier outlined in this research of maximizing power and thermal efficiency, certain parameters were controlled: the heat source temperature, pump speed and expander resistive load. The heat source was provided by an 18-kW thermal electric oil heater at a mass flowrate of 1.13 kg/s. Its temperature was adjusted to 120 °C, 130 °C and 140 °C. The pump speed was varied by a VFD (variable frequency drive connected to it) from 1100 RPM through 3500 RPM. The expander resistive load were in two Cases. Case A called “partial load” where two 100 W light bulbs were connected in parallel with an effective resistance of 50 – 70 Ω. Case B called “full load” where two 100 W light bulbs and two 100 ohm resistors were all connected in parallel with an effective resistance of 25 – 30 Ω.

Chinedu Kingsley Unamba, August 2020 145 Experimental development of ORC technology for waste heat-to-power conversion

The operating points were varied, resulting in performance characteristics variations that were investigated. Energy and exergy analyses were performed based on the resulting data.

Motivated by the need to extract more power and improve efficiency on the initial prototype ORC rig as would be driven by the market potential of small- scale waste-heat recovery (WHR) systems, the functional prototype was designed and constructed. It would also address the numerous challenges on the original prototype and provide practical appreciation of the challenges in deploying and operating ORC systems: flow instability, measurement errors from temperature sensors and pressure transducers, pipe/tube corrosion, thermal decomposition of the nitrile seals and provide flexibility to the possibility of a different ORC architecture (superheater and recuperator) as well as perform a dynamic performance characterisation of the system. The modified ORC rig would be capable of generating high fidelity data for model prediction.

The design of technically and economically efficient ORC systems is challenging, especially for converting low- to medium-grade thermal energy into useful power. The simultaneous optimisation of the working fluid and of the system components, accounting for part-load and off-design operating conditions requires accurate physics-based models, able to predict the system performance over a wide range of working conditions. High-fidelity experimental measurements are key to inform and validate semi-empirical and comprehensive models.

Although the new functional rig was expected to meet the principal objective of higher expander power output and thermal efficiency, there were no guarantees aside generating high quality data. Larger and better built ORC systems might suffer from more pressure drops than a smaller and compact system thereby offsetting some of its gains.

146 Chinedu Kingsley Unamba, August 2020 Chapter 8: Conclusion

The design and operation of the functional ORC presents the operational optimisation of an experimental non-recuperative 1-kWe ORC engine, made of a rotary-vane pump, brazed-plate evaporator and condenser, and a scroll expander with an adjustable load.

Heat is supplied to the prototype at various temperatures (ranging from 100 to 140 °C) using an 18-kWth heater that circulates thermal oil in the evaporation unit, while excess heat is rejected to a cooling water stream supplied at 18 °C.

The optimal operating (or control) parameters, i.e., the pump speed and expander load, are determined for different heat-source temperatures, while the part-load performance of the engine is explored with a total of 124 steady-state data points while an array of rheostats acting as the resistive load on the expander (10 to 55 ohm) was used.

The net power output, thermal efficiency and exergy efficiency was reported as functions of the working-fluid mass flow rate and expander-scroll spindle speed, thus providing useful part-load performance maps that can be readily used to inform numerical models, together with the detailed expander performance maps provided – whereby the filling factor and electrical isentropic efficiency of the expansion device are reported as functions of the spindle speed and pressure ratio.

8.1 SUMMARY OF FINDINGS

Analysis of the experimental results showed that over the range of investigated heat-source temperatures (120 - 140 °C), higher pressure ratios led to higher pump power consumptions and higher expander outputs, with expander isentropic efficiencies reaching values up to ~80%. This was slightly above the maximum isentropic efficiency of the expander from the

Chinedu Kingsley Unamba, August 2020 147 Experimental development of ORC technology for waste heat-to-power conversion

manufacturer (~75% when R245fa is used). This overestimation was caused by errors from the measurements of pressure transducer, temperature sensor and the mass flowrate (due to the flow instability) and further propagated in the data analysis.

Overall ORC system thermal efficiencies of up to 6% were attained at an intermediate heat-source temperature (130 °C), pressure ratio (3.6) and at Case B load. Both the pump power and expander power, and therefore cycle thermal efficiency were very sensitive to the generator load, but relatively insensitive to the heat-source temperature within the investigated envelope of experimental conditions.

An exergy analysis was also performed at chosen operating conditions (minimum/maximum pressure ratios at each heat-source temperature for Case A/Case B load) and showed that the expander accounted for the second largest exergy destruction after the evaporator.

For the functioning prototype, a maximum net power output (i.e. the electrical power generated through the vapour expansion minus the pump consumption) of 508±3W is reported for a heat-source stream supplied at 140 °C, while a maximum 4.2±0.1% thermal efficiency is reached with a heat- source temperature at 120 °C as against 6% at 130 °C for the proof-of- principle prototype, yet obtained with an undefined uncertainty.

The maximum exergy efficiency, 18.7±3%, is reached for the lowest heat- source temperature investigated in this study, 100 °C for the functional prototype.

The contribution of the main components to the exergy destruction in the overall heat-to-power conversion process is determined through a steady- state exergy analysis of the three optimum configurations identified, i.e., maximising both the thermal and exergy efficiency. It is found that a

148 Chinedu Kingsley Unamba, August 2020 Chapter 8: Conclusion

negligible amount of the thermodynamic losses occur in the pump, which represents 3% of the overall availability loss, while the two major exergy destruction rates are measured in the expansion and evaporation units, which represent 32 to 46% and 36 to 46%, respectively.

The results from the exergy analysis of the functional prototype correlate with that of the proof-of-principle protype to suggest that the evaporator accounts for the largest exergy destruction, followed by the expander.

Although there was a decrease in the thermal efficiency from approximately 6% at 130 °C for the principle prototype to 4.2% at 140 °C for the functional prototype, this shows the contributions of several factors to overall system performance. In particular, the proof-of-principle prototype is more compact than the functioning engine investigated in this study, reducing both pressure and heat losses in the pipes, and is built using fewer elbows and valves, which inherently affect performance due to increased pressure drops.

However, this additional complexity brings more flexibility (notably with the addition of a recuperator) and allows an easy replacement of all components.

As expected, the maximum expander output power recorded for the functional prototype was approximately 700 W in comparison to 600 W in the proof-of-principle prototype due to the increased stability of the working fluid flow rate by the addition of a liquid receiver.

The dynamic characteristics of the prototype have also been investigated by measuring the key time constants, as derived from the analysis of the expander power-output time response to a rising or falling step in pump rotating velocity. It is found that the rise time is 7 to 30% higher than the fall time i.e. the systems responds faster in falling timesteps than in rising time steps. Both step-response times are also found to decrease with the amplitude of the command step.

Chinedu Kingsley Unamba, August 2020 149 Experimental development of ORC technology for waste heat-to-power conversion

It also shows that at larger step changes, the system reacts or responds faster. Although this is not what is expected (as larger step changes are often accompanied by larger inertia), this provides an interesting basis for further investigation and analysis into ORC dynamic models.

8.2 FUTURE WORK

Beyond the obtained results, the testing facility is an outcome in itself. This apparatus will be used in the future for testing multiple ORC system configurations (recuperative and non-recuperative), using multiple working fluids (even flammable working fluids), with easy-to-replace components and high-fidelity measurement capability.

An Investigation of the performance increase using porous-coated heat exchangers or reciprocating-piston expanders with advanced variable valve timing controls investigating the dynamic performance of the ORC engine with time-varying heat-source conditions and advanced dynamic controls of the key operating parameters of the system.

The ORC rig is highly flexible and allows for easy addition or substitution of different components of the system. For example, reciprocating expander can be used in place of the scroll expander currently used.

Automation of the ORC system involving building a microcontroller and models to respond to time varying heat source by automatically optimizing the pump speed and expander load to either have as objective function: maximum power or maximum thermal efficiency is currently ongoing.

The heat source conditions can be modified to allow for time-varying conditions by having a hot and cold oil reservoir that flow intermittently or by modifying the programmable logic unit of the oil heater to alternate the

150 Chinedu Kingsley Unamba, August 2020 Chapter 8: Conclusion

temperature at a certain frequency. In either Case, the ORC setup has been designed to be flexible and to allow for much advanced remote control.

Chinedu Kingsley Unamba, August 2020 151 Experimental development of ORC technology for waste heat-to-power conversion

152 Chinedu Kingsley Unamba, August 2020 Chapter 9: Bibliography

9 BIBLIOGRAPHY

[1] Volumatrix, "Waste heat to electricity," [Online]. Available: http://volumatrixgroup.com/?tag=heat-energy-into-electricity. [Accessed 20 June 2020].

[2] G. Pei, J. Li, Y. Li, D. Wang and J. Ji, "Construction and dynamic test of a small-scale organic rankine cycle," Energy, vol. 36, no. 5, pp. 3215-3223, 2011.

[3] E. W. Beans, "Comparative thermodynamics for Brayton and Rankine cycles," Journal of Engineering for Gas Turbines and Power, vol. 112, no. 1, pp. 94-99, 1990.

[4] O. A. Oyewunmi, A. J. Haslam and C. N. Markides, "Towards the computer-aided molecular design of organic rankine cycle systems with advanced fluid theories," in SusTEM International Conference, Newcastle, UK, 2015.

Chinedu Kingsley Unamba, August 2020 153 Experimental development of ORC technology for waste heat-to-power conversion

[5] C. Kirmse, O. A. Oyewunmi, A. J. Haslam and C. N. Markides, "Comparison of a novel organic-fluid thermofluidic heat converter and an organic Rankine cycle heat engine," Energies, vol. 9, no. 7, p. 479, 2016.

[6] J. Freeman, K. Hellgardt and C. N. Markides, "An assessment of solar- powered organic Rankine cycle systems for combined heating and power in UK domestic applications," Applied Energy, vol. 138, pp. 605- 620, 2015.

[7] J. Freeman, K. Hellgardt and C. N. Markides, "Working fluid selection and electrical performance optimisation of a domestic solar-ORC combined heat and power system for year-round operation in the UK," Applied Energy, vol. 186, pp. 291-303, 2017.

[8] C. N. Markides, "The role of pumped and waste heat technologies in a high-efficiency sustainable energy future for the UK," Applied Thermal Engineering, vol. 53, no. 2, pp. 197-209, 2013.

[9] F. Vélez, J. J. Segovia, M. C. Martín, G. Antolín, F. Chejne and A. Quijano, "A technical, economical and market review of organic Rankine cycles for the conversion of low-grade heat for power generation," Renewable and Sustainable Energy Reviews, vol. 16, no. 6, pp. 4175-4189, 2012.

[10 J. A. Mumith, T. Karayiannis and C. Makatsoris, "Design and ] optimization of a thermoacoustic heat engine using reinforcement learning," International Journal of Low-Carbon Technologies, vol. 11, no. 3, pp. 431-439, 2016.

[11 C. N. Markides, "Low-concentration solar-power systems based on ] organic rankine cycles for distributed-scale applications: Overview and

154 Chinedu Kingsley Unamba, August 2020 Chapter 9: Bibliography

further developments.," Frontiers in Energy Research, vol. 3, no. 47, pp. 1-16, 2015.

[12 C. N. Markides, "Low-concentration solar-power systems based on ] organic Rankine cycles for distributed-scale applications: overview and further developments," Frontiers in Energy Research, vol. 3, p. 47, 2015.

[13 H. Jouhara, N. Khordehgah, S. Almahmoud, B. Delpech, A. Chauhan and ] S. A. Tassou, "aste heat recovery technologies and applications," Thermal Science and Engineering Progress, vol. 6, pp. 268-289, 2018.

[14 E. Macchi, "Theoretical basis of the organic Rankine cycle," in Organic ] Rankine Cycle (ORC) Power Systems: Technologies and Applications, 2017, pp. 3-24.

[15 C. Kirmse, A. I. Taleb, O. Oyewunmi, A. Haslam and C. Markides, ] "Performance comparison of a novel thermofluidic organic-fluid heat converter and an organic Rankine cycle heat engine," in 3rd International Seminar on ORC Power Systems, Brussels, 2015.

[16 F. Galuppo, P. Dufour, M. Nadri, T. Reiche and V. Lemort, "Experiment ] design for waste heat recovery modelling in heavy duty trucks," IFAC- PapersOnLine, vol. 51, no. 31, pp. 726-731, 2018.

[17 Z. Wang, N. Zhou and G. Jing, "Performance analysis of ORC power ] generation system with low-temperature waste heat of aluminium reduction cell," Physics Procedia, vol. 24, pp. 546-553, 2012.

[18 S. Kang, "Design and experimental study of ORC (organic Rankine cycle) ] and radial turbine using R245fa working fluid," Energy, vol. 41, no. 1, pp. 514-524, 2012.

Chinedu Kingsley Unamba, August 2020 155 Experimental development of ORC technology for waste heat-to-power conversion

[19 H. Tian, G. Shu, H. Wei, X. Liang and L. Liu, "Fluids and parameters ] optimization for the organic Rankine cycles (ORCs) used in exhaust heat recovery of internal combustion engine (ICE)," Energy, vol. 47, pp. 125- 136, 2012.

[20 G. Shu, L. Liu, H. Tian, H. Wei and G. Yu, "Parametric and working fluid ] analysis of a dual-loop organic Rankine cycle (DORC) used in engine waste heat recovery," Applied Energy, vol. 113, pp. 1188-1198, 2014.

[21 H. Zhang, E. Wang and B. Fan, "A performance analysis of a novel system ] of a dual loop bottoming organic Rankine cycle (ORC) with a light-duty diesel engine," Applied Energy, vol. 102, pp. 1504-1513, 2013.

[22 S. Lecompte, H. Huisseune, M. Van Den Broek, B. Vanslambrouck and M. ] De Paepe, "Review of organic Rankine cycle (ORC) architectures for waste heat recovery," Renewable and Sustainable Energy Reviews, vol. 47, pp. 448-461, 2015.

[23 L. Shi, G. Shu, H. Tian and S. Deng, "A review of modified organic Rankine ] cycles (ORCs) for internal combustion engine waste heat recovery (ICE- WHR)," Renewable and Sustainable Energy Reviews, vol. 92, pp. 95-110, 2018.

[24 O. A. Oyewunmi and C. N. Markides, "Effect of working-fluid mixtures ] on organic Rankine cycle systems: Heat transfer and cost analysis," in 3rd International Seminar on ORC Power Systems, Brussels, Belgium, 2015.

[25 R. Bet Sarkis and V. Zare, "Proposal and analysis of two novel integrated ] configurations for hybrid solar-biomass power generation systems:

156 Chinedu Kingsley Unamba, August 2020 Chapter 9: Bibliography

Thermodynamic and economic evaluation," Energy Conversion and Management, vol. 160, pp. 411-425, 2018.

[26 F. Boyaghchi and P. Heidarnejad, "Thermo-economic assessment and ] multi objective optimization of a solar micro CCHP based on Organic Rankine Cycle for domestic application," Energy Conversion and Management, vol. 97, pp. 224-234, 2015.

[27 I. Garcia-Saez, J. Méndez, C. Ortiz, D. Loncar, J. Becerra and R. ] Chacartegui, "Energy and economic assessment of solar organic Rankine cycle for combined heat and power generation in residential applications," Renewable Energy, vol. 140, pp. 461-476, 2019.

[28 D. Ibrahim, O. A. Oyewunmi, A. J. Haslam, A. Pantaleo and C. N. Markides, ] "Computer-aided working fluid design and optimisation of organic Rankine cycle (ORC) systems under varying heat-source conditions," in 32nd International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, Wroclaw, Poland, 2019.

[29 M. A. Chatzopoulou, S. Lecompte, M. D. Paepe and C. N. & Markides, "Off- ] design optimisation of organic Rankine cycle (ORC) engines with different heat exchangers and volumetric expanders in waste heat recovery applications," Applied Energy, vol. 253, p. 113442, 2019.

[30 M. A. Chatzopoulou, M. C. Simpson, P. Sapin and C. N. Markides, "Off- ] design optimisation of organic Rankine cycle (ORC) engines with piston expanders for medium-scale combined heat and power applications," Applied Energy, vol. 238, pp. 1211-1236, 2019.

Chinedu Kingsley Unamba, August 2020 157 Experimental development of ORC technology for waste heat-to-power conversion

[31 A. Ramos, M. A. Chatzopoulou, J. Freeman and C. N. Markides, ] "Optimisation of a high-efficiency solar-driven organic Rankine cycle for applications in the built environment," Applied Energy, vol. 228, pp. 755-765, 2018.

[32 E. Barbier, "Geothermal energy technology and current status: An ] overview," Renewable and Sustainable Energy Reviews, vol. 6, no. 1-2, pp. 3-65, 2002.

[33 J. B. Obi, "State of art on ORC applications for waste heat recovery and ] micro-cogeneration for installations up to 100kWe," Energy Procedia, vol. 82, pp. 994-1001, 2015.

[34 B. F. Tchanche, G. Lambrinos, A. Frangoudakis and G. Papadakis, "Low- ] grade heat conversion into power using organic Rankine cycles - A review of various applications," Renewable and Sustainable Energy Reviews, vol. 15, no. 8, pp. 3963-3979, 2011.

[35 E. Cihana and B. Kavasogullari, "Energy and exergy analysis of a ] combined refrigeration and waste heat driven organic Rankine cycle system," Thermal Science, vol. 21, no. 6, pp. 2621-263, 2017.

[36 V. Lemort, S. Quoilin, C. Cuevas and J. Lebrun, "Testing and modelling a ] scroll expander integrated into an organic Rankine cycle," Applied Thermal Engineering, vol. 29, pp. 3094-3102, 2009.

[37 S. Baral, D. Kim, E. Yun and K. C. Kim, "Energy, exergy and performance ] analysis of small-scale organic Rankine cycle systems for electrical power generation applicable in rural areas of developing countries," Energies, vol. 8, no. 2, pp. 684-713, 2015.

158 Chinedu Kingsley Unamba, August 2020 Chapter 9: Bibliography

[38 J. Chang, C. Chang, T. Hung, J. Lin and K. Huang, "Experimental study and ] CFD approach for scroll type expander used in low-temperature organic Rankine cycle," Applied Thermal Engineering, vol. 73, no. 2, pp. 1444-1452, 2014.

[39 N. Zheng, L. Zhao, X. Wang and Y. Tan, "Experimental verification of a ] rolling-piston expander that applied for low-temperature organic Rankine cycle," Applied Energy, vol. 112, pp. 1265-1274, 2013.

[40 X. Wang, L. Zhao, J. Wang, W. Zhang, X. Zhao and W. Wu, "Performance ] evaluation of a low-temperature solar Rankine cycle system utilizing R245fa," Solar Energy, vol. 84, no. 3, pp. 353-364, 2010.

[41 Y. Zhang, Y. Wu, G. Xia, C. Ma, W. Ji, S. Liu, K. Yang and F. Yang, ] "Development and experimental study on organic Rankine cycle system with single-screw expander for waste heat recovery from exhaust of diesel engine," Energy, vol. 77, pp. 499-508, 2014.

[42 M. T. White, O. A. Oyewunmi, M. Chatzopoulou, A. Pantaleo, A. Haslam ] and C. N. Markides, "Computer-aided working-fluid design, thermodynamic optimisation and thermoeconomic assessment of ORC systems for waste-heat recovery," Energy, vol. 161, pp. 1181-1198, 2018.

[43 M. T. White, O. A. Oyewunmi, A. J. Haslam and C. N. Markides, "Industrial ] waste-heat recovery through integrated computer-aided working-fluid and ORC system optimisation using SAFT-g Mie," Energy Conversion and Management, vol. 150, pp. 851-869, 2017.

Chinedu Kingsley Unamba, August 2020 159 Experimental development of ORC technology for waste heat-to-power conversion

[44 P. J. Mago, L. M. Chamra, K. Srinivasan and C. Somayaji, "An examination ] of regenerative organic Rankine cycles using dry fluids," Applied Thermal Engineering, vol. 28, no. 8-9, pp. 998-1007, 2008.

[45 S. Safarian and F. Aramoun, "Energy and exergy assessments of ] modified organic Rankine cycles (ORCs)," Energy Reports, vol. 1, pp. 1- 7, 2015.

[46 Ö. Kas, "Energy and exergy analysis of an organic Rankine for power ] generation from waste heat recovery in steel industry," Energy Conversion and Management, vol. 77, pp. 108-117, 2014.

[47 J. Li, G. Pei, Y. Li, D. Wang and J. Ji, "Energetic and exergetic investigation ] of an organic Rankine cycle at different heat source temperatures," Energy, vol. 38, no. 1, pp. 85-95, 2012.

[48 C. Fatma, A. Kilic and C. Selman, "Energy and exergy analysis of an ] organic Rankine cycle in a biomass-based forest products manufacturing plant," Turkish Journal of Electrical Engineering & Computer Sciences, pp. 5100-5112, 2016.

[49 S. Declaye, S. Quoilin, L. Guillaume and V. Lemort, "Experimental study ] on an open-drive scroll expander integrated into an ORC (Organic Rankine Cycle) system with R245fa as working fluid," Energy, Vols. 173- 183, p. 55, 2013.

[50 O. A. Oyewunmi and C. N. Markides, "Thermo-economic and heat ] transfer optimization of working-fluid mixtures in a low-temperature organic Rankine cycle system," Energies, vol. 9, p. 448, 2016.

160 Chinedu Kingsley Unamba, August 2020 Chapter 9: Bibliography

[51 O. A. Oyewunmi, A. I. Taleb, A. J. Haslam and C. N. Markides, "An ] assessment of working-fluid mixtures using SAFT-VR Mie for use in organic Rankine cycle systems for waste-heat recovery,," Computational Thermal Sciences, vol. 6, pp. 301-306, 2014.

[52 O. A. Oyewunmi, S. Lecompte, M. De Paepe and C. N. Markides, ] "Thermodynamic Optimization of Recuperative Sub- and Transcritical Organic Rankine Cycle Systems," in 30th International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, San Diego, California, 2017.

[53 M. Imran, M. Usman, B. Park and D. Lee, "Volumetric expanders for low ] grade heat and waste heat recovery applications," Renewable and Sustainable Energy Reviews, vol. 57, pp. 1090-1109, 2016.

[54 G. Carraro, S. Rech, A. T. G. Lazzaretto and P. Danieli, "Dynamic ] simulation and experiments of a low-cost small ORC unit for market applications," Energy Conversion and Management, vol. 197, p. 111863, 2019.

[55 S. Quoilin, M. V. D. Broek, S. Declaye, P. Dewallef and V. Lemort, "Techno- ] economic survey of organic rankine cycle (ORC) systems," Renewable and Sustainable Energy Reviews, vol. 22, pp. 168-186, 2013.

[56 I. Encabo Cáceres, R. Agromayor and L. O. Nord, "Thermodynamic ] optimization of an organic Rankine cycle for power generation from A low temperature geothermal heat source," in Proceedings of the 58th Conference on Simulation and Modelling (SIMS 58), Reykjavik, Iceland, September 25th-27th, 2017.

Chinedu Kingsley Unamba, August 2020 161 Experimental development of ORC technology for waste heat-to-power conversion

[57 J. Navarro-Esbrí, F. Molés, B. Peris, A. Mota-Babiloni and K. Kontomaris, ] "Experimental study of an organic Rankine cycle with HFO-1336mzz-Z as a low global warming potential working fluid for micro-scale low temperature applications," Energy, vol. 133, pp. 79-89, 2017.

[58 S. Francisco, O. Delgado, N. Lutsey, D. Rotz, K. Sisken, D. Kayes, K. ] Damon, G. Salemme, D. Koeberlein, R. Mihelic, W. de Ojeda, P. Amar, D. Kodjak, F. Kamakaté, R. Muncrief, B. Sharpe, V. Franco and N. Clark, "The U.S. supertruck program expediting the development of advanced heavy-duty vehicle efficiency technologies," 2014.

[59 R. Zhao, H. Zhang, S. Song, Y. Tian, Y. Yang and Y. Liu, "Integrated ] simulation and control strategy of the diesel engine–organic Rankine cycle (ORC) combined system," Energy Conversion and Management, vol. 156, pp. 639-654, 2018.

[60 X. Li, H. Tian, G. Shu, M. Zhao, C. N. Markides and C. Hu, "Potential of ] carbon dioxide transcritical power cycle waste-heat recovery systems for heavy-duty truck engines," Applied Energy, vol. 250, pp. 1581-1599, 2019.

[61 D. Manolakos, G. Kosmadakis, S. Kyritsis and G. Papadakis, ] "Identification of behaviour and evaluation of performance of small scale, low-temperature Organic Rankine Cycle system coupled with a RO desalination unit," Energy, vol. 34, pp. 767-774, 2009.

[62 L. Li, Y. Ge, X. Luo and S. Tassou, "Experimental investigations into ] power generation with low grade waste heat and R245fa organic Rankine cycles (ORCs)," Applied Thermal Engineering, vol. 115, pp. 815- 824, 2017.

162 Chinedu Kingsley Unamba, August 2020 Chapter 9: Bibliography

[63 M. Li, J. Wang, W. He, B. Wang, S. Ma and Y. Dai, "Experimental ] evaluation of the regenerative and basic organic Rankine cycles for low- grade heat source utilization," Journal of Energy Engineering, vol. 139, pp. 190-197, 2013.

[64 G. Yu, G. Shu, H. Tian, Y. Huo and W. Zhu, "Experimental investigations ] on a cascaded steam-/organic-Rankine-cycle (RC/ORC) system for waste heat recovery (WHR) from diesel engine," Energy Conversion and Management, vol. 129, pp. 43-51, 2016.

[65 G. Shu, M. Zhao, H. Tian, Y. Huo and W. Zhu, "Experimental comparison ] of R123 and R245fa as working fluids for waste heat recovery from heavy-duty diesel engine," Energy, vol. 115, pp. 756-769, 2016.

[66 L. Shi, G. Shu, H. Tian, G. Huang, T. Chen, X. Li and D. Li, "Experimental ] comparison between four CO2-based transcritical Rankine cycle (CTRC) systems for engine waste heat recovery," Energy Conversion and Management, vol. 150, pp. 159-171, 2017.

[67 X. Li, G. Shu, H. Tian, G. Huang, P. Liu, X. Wang and L. Shi, "Experimental ] comparison of dynamic responses of CO2 transcritical power cycle systems used for engine waste heat recovery," Energy Conversion and Management, vol. 161, pp. 254-265, 2018.

[68 X. Li, H. Tian, G. Shu, C. Hu, R. Sun and L. Li, "Effects of external ] perturbations on dynamic performance of carbon dioxide transcritical power cycles for truck engine waste heat recovery," Energy, vol. 163, pp. 920-931, 2018.

Chinedu Kingsley Unamba, August 2020 163 Experimental development of ORC technology for waste heat-to-power conversion

[69 J. Sarkar, "Review and future trends of supercritical CO2 Rankine cycle ] for low-grade heat conversion," Renewable and Sustainable Energy Reviews, vol. 48, pp. 434-451, 2015.

[70 B. Park, M. Usman, M. Imran and A. Pesyridis, "Review of organic ] Rankine cycle experimental data trends," Energy Conversion and Management, vol. 173, pp. 679-691, 2018.

[71 A. Desideri, A. Hernandez, S. Gusev, M. van den Broek, V. Lemort and S. ] Quoilin, "Steady-state and dynamic validation of a small-scale waste heat recovery system using the ThermoCycle Modelica library," Energy, vol. 115, pp. 684-696, 2016.

[72 A. Hernandez, A. Desideri, S. Gusev, C. Ionescu, M. Den Broek, S. Quoilin, ] V. Lemort and R. De Keyser, "Design and experimental validation of an adaptive control law to maximize the power generation of a small-scale waste heat recovery system," Applied Energy, vol. 203, pp. 549-559, 2017.

[73 A. Landelle, N. Tauveron, R. Revellin, P. Haberschill, S. Colasson and V. ] Roussel, "Performance investigation of reciprocating pump running with organic fluid for organic Rankine cycle," Applied Thermal Engineering, vol. 113, pp. 962-969, 2017.

[74 L. Tocci, T. Pal, I. Pesmazoglou, B. Franchetti, L. Tocci, T. Pal, I. ] Pesmazoglou and B. Franchetti, "Small Scale Organic Rankine Cycle (ORC): A techno-economic review," Energies, vol. 10, p. 413, 2017.

[75 L. van Kleef, O. A. Oyewunmi and C. N. Markides, "Multi-objective ] thermo-economic optimization of organic Rankine cycle (ORC) power systems in waste-heat recovery applications using computer-aided

164 Chinedu Kingsley Unamba, August 2020 Chapter 9: Bibliography

molecular design techniques," Applied Energy, vol. 251, p. 112513, 2019.

[76 P. Gao, L. W. Wang, R. Z. Wang, L. Jiang and Z. S. Zhou, "Experimental ] investigation on a small pumpless ORC (organic rankine cycle) system driven by the low temperature heat source," Energy, vol. 91, pp. 324- 333, 2015.

[77 P. Gao, Z. X. Wang, L. W. Wang and H. T. Lu, "Technical feasibility of a ] gravity-type pumpless ORC system with one evaporator and two condensers," Applied Thermal Engineering, vol. 145, pp. 569-575, 2018.

[78 S. Quoilin, V. Lemort and J. Lebrun, "Experimental study and modelling ] of an organic Rankine cycle using scroll expander," Applied Energy, vol. 87, no. 4, pp. 1260-1268, 2010.

[79 A. Nemati, H. Nami, F. Ranjbar and M. Yari, "A comparative ] thermodynamic analysis of ORC and Kalina cycles for waste heat recovery: A case study for CGAM cogeneration system," Case Studies in Thermal Engineering, vol. 9, pp. 1-13, 2017.

[80 D. Ziviani, A. Beyene and M. Venturini, "Advances and challenges in ORC ] systems modeling for low grade thermal energy recovery," Applied Energy, vol. 121, pp. 79-95, 2014.

[81 Y. Feng, Y. Zhang, B. Li, J. Yang and Y. Shi, "Comparison between ] regenerative organic Rankine cycle (RORC) and basic organic Rankine cycle (BORC) based on thermoeconomic multi-objective optimization considering exergy efficiency and levelized energy cost (LEC)," Energy Conversion and Management, vol. 96, pp. 58-71, 2015.

Chinedu Kingsley Unamba, August 2020 165 Experimental development of ORC technology for waste heat-to-power conversion

[82 S. Wang, C. Liu, C. Zhang and X. Xu, "Thermodynamic evaluation of leak ] phenomenon in liquid receiver of ORC systems," Applied Thermal Engineering, vol. 141, pp. 1110-1119, 2018.

[83 M. Khennich and N. Galanis, "Optimal design of ORC systems with a low- ] temperature heat source," Entropy, vol. 14, no. 2, pp. 370-389, 2012.

[84 J. L. Bouvier, V. Lemort, G. Michaux, P. Salagnac and T. Kientz, ] "Experimental study of an oil-free steam piston expander for micro- combined heat and power systems," Applied Energy, vol. 169, p. 788– 798, 2016.

[85 O. Dumont, R. Dickes and V. Lemort, "Experimental investigation of four ] volumetric expanders," Energy Procedia, vol. 129, pp. 859-866, 2017.

[86 V. Vodicka, V. Novotny, J. Mascuch and M. Kolovratnik, "Impact of major ] leakages on characteristics of a rotary vane expander for ORC," Energy Procedia, vol. 129, pp. 387-394, 2017.

[87 C. K. Unamba, M. White, P. Sapin, J. Freeman, S. Lecompte, O. A. ] Oyewunmi and C. N. Markides, "Experimental Investigation of the Operating Point of a 1-kW ORC System," Energy Procedia, vol. 129, pp. 875-882, 2017.

[88 V. Lemort, S. Declaye and S. Quoilin, "Experimental characterization of ] a hermetic scroll expander for use in a micro-scale Rankine cycle," Journal of Power and Energy, vol. 226, no. 1, pp. 126-136, 2012.

[89 Shutterstock, "Basic engine design," [Online]. Available: ] https://www.shutterstock.com/image-vector/basic-engine-design- vector-illustration-590799575. [Accessed 20 June 2020].

166 Chinedu Kingsley Unamba, August 2020 Chapter 9: Bibliography

[90 A. Kornhauser and J. Smith Jr., "Application of a complex Nusselt ] number to heat transfer during compression and expansion," Journal of Heat Transfer, vol. 116, pp. 536-542, 1994.

[91 H. Pfriem, "Periodic heat transfer at small pressure fluctuations," NACA ] Technical Memorandum, 1943.

[92 A. Kornhauser and J. Smith Jr., "The effects of heat transfer on gas spring ] performance," Journal of Engineering Resource Technology, vol. 115, pp. 70-75, 1995.

[93 U. Lekic and J. Kok, "Heat transfer and fluid flows in gas springs," Open ] Thermodynamic Journal, pp. 13-26, 2010.

[94 A. Taleb, P. Sapin, C. Barfuß and D. Fabris, "CFD analysis of thermally ] induced thermodynamic losses in the reciprocating compression and expansion of real gases," in 1st International Seminar on Non-Ideal Compressible-Fluid Dynamics for Propulsion & Power, 2017.

[95 P. Sapin, M. Simpson, A. J. White and C. N. Markides, "Lumped dynamic ] analysis and design of a high-performance reciprocating-piston expander," in 0th International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, San Diego, California, 2017.

[96 MachineDesign, "Mechanical seal," [Online]. Available: ] https://www.machinedesign.com/mechanical-motion- systems/hydraulics/article/21833677/the-dos-and-donts-of-seal- engineering. [Accessed 20 June 2020].

Chinedu Kingsley Unamba, August 2020 167 Experimental development of ORC technology for waste heat-to-power conversion

[97 N. Zheng, L. Zhao, X. Wang and Y. Tan, "Experimental verification of a ] rolling-piston expander that applied for low-temperature Organic Rankine Cycle," Applied Energy, vol. 112, pp. 1265-1274, 2013.

[98 M. C. Simpson, G. Rotolo, P. Sapin, P. De Palma, A. J. White and C. N. ] Markides, "Thermodynamic performance maps of reciprocating-piston expanders for operation at off-design and part-load conditions," in 13th International Conference on Heat Transfer, Fluid Mechanics and Thermodynamics, 2017.

[99 M. C. Simpson, A. M. Chatzopoulou, O. A. Oyewunmi, N. Le Brun, P. Sapin ] and C. N. Markides, "Technoeconomic analysis of internal combustion engine - organic Rankine cycle systems for combined heat and power in energy-intensive buildings," Applied Energy, vol. 253, pp. 1-13, 2019.

[10 M. C. Simpson, S. Schuster, D. Ibrahim, O. A. Oyewunmi, P. Sapin, A. T. 0] White and C. N. Markides, "Small-scale, low-temperature ORC systems intime-varying operation: Turbines or reciprocating-piston expanders," in 32nd International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, Wroclaw, Poland, 2019.

[10 P. Stouffs, M. Tazerout and P. Wauters, "Thermodynamic analysis of 1] reciprocating compressors," International Journal of Thermal Sciences, vol. 40, no. 1, pp. 52-66, 2001.

[10 K. Boutsen, Critical Flow Analysis and Understanding of Automotive 2] Exhaust Ports [dissertation], Miami: University of Miami, 2016.

168 Chinedu Kingsley Unamba, August 2020 Chapter 9: Bibliography

[10 R. Mathie, C. N. Markides and A. J. White, "A framework for the analysis 3] of thermal losses in reciprocating compressors and expanders," Heat Transfer Engineering, vol. 35, no. (16–17), pp. 1435-1449, 2014.

[10 M. Yang and R. Yeh, "Thermodynamic and economic performances 4] optimization of an organic Rankine cycle system utilizing exhaust gas of a large marine diesel engine," Applied Energy, vol. 149, pp. 1-12, 2015.

[10 C. Barfuß, Numerical Simulation of Compression and Expansion 5] Processes in Gas Springs for Innovative Electricity Storage [thesis], London: Imperial College London, 2016.

[10 C. Willich and A. J. White, "Heat transfer losses in reciprocating 6] compressors with valve actuation for energy storage applications," Journal of Energy Storage, vol. 14, no. 2, pp. 322-328, 2017.

[10 Y. M. Kim, D. G. Shin and C. G. Kim, "Optimization of design pressure 7] ratio of positive displacement expander for vehicle engine waste heat recovery," Energies, vol. 7(9), pp. 6105-6117, 2014.

[10 Fluid-o-Tech, "TMFR 30-200 Series Vane Pumps," [Online]. Available: 8] https://www.fluidotech.it/en/products/technologies/rotary-vane- pumps/tmfr-30-200- series/#:~:text=The%20TMFR%2030%2D200%20series,with%20a% 20long%20service%20life.. [Accessed 08 June 2020].

[10 SWEP, "B12 Plate heat exchanger," [Online]. Available: 9] https://www.swep.net/products/b12/. [Accessed 08 June 2020].

Chinedu Kingsley Unamba, August 2020 169 Experimental development of ORC technology for waste heat-to-power conversion

[11 Frigosystem, "18 kW electric oil heater," [Online]. Available: 0] http://www.frigosystem.it/en/thermorefrigeration. [Accessed 08 June 2020].

[11 Airsquared, "1 kW semi-hermetic scroll expander for lubricated or oil- 1] free expansion with gases other than air (E15H022A-SH)," [Online]. Available: https://airsquared.com/products/scroll- expanders/e15h022a-sh/. [Accessed 08 June 2020].

[11 Alfa-Laval, "22-kW Alfa Laval condenser CB60-30H-F Model," [Online]. 2] Available: https://www.alfalaval.co.uk/products/heat-transfer/plate- heat-exchangers/brazed-plate-heat-exchangers/. [Accessed 08 June 2020].

[11 J. G. Andreasen, U. Larsen, T. Knudsen, L. Pierobon and F. Haglind, 3] "Selection and optimization of pure and mixed working fluids for low grade heat utilization using organic rankine cycles," Energy, vol. 73, pp. 204-213, 2014.

[11 C. K. Unamba, P. Sapin, X. Li, J. Song, K. Wang, G. Shu, H. Tian and C. N. 4] Markides, "Operational optimisation of a non-recuperative 1-kWe organic Rankine cycle engine," Applied Sciences, vol. 9, p. 3024, 2019.

[11 J. Freeman, I. Guarracino, C. Unamba, O. Oyewunmi, N. Le Brun, M. White 5] and C. Markides, "Developing a test bed for small-scale ORC expanders in waste-heat recovery applications," in 3rd Engine ORC Consortium (EORCC) Workshop, Belfast, Northern Ireland, 14-16 September 2016.

[11 Krohne, "MFC 400 OPTIMASS Coriolis flow meter," [Online]. Available: 6] https://krohne.com/en/products/flow-measurement/components-

170 Chinedu Kingsley Unamba, August 2020 Chapter 9: Bibliography

and-auxiliary-equipment-for-flow-measurement/signal- converters/mfc-400/. [Accessed 20 June 2020].

[11 R. Dickes, O. Dumont, L. Guillaume, S. Quoilin and V. Lemort, "Charge- 7] sensitive modelling of organic Rankine cycle power systems for off- design performance simulation," Applied Energy, vol. 212, pp. 1262- 1281, 2018.

[11 V. Lemort and A. Legros, "Positive displacement expanders for organic 8] Rankine cycle systems," in Organic Rankine Cycle (ORC) Power Systems, 2017, pp. 361-396.

[11 B. Woodland, J. Braun, E. Groll and W. Horton, "Experimental testing of 9] an organic Rankine cycle with scroll-type expander," in International Refrigeration and Air Conditioning Conference, 2012.

[12 O. Dumont, A. Parthoens, R. Dickes and V. Lemort, "Experimental 0] investigation and optimal performance assessment of four volumetric expanders (scroll, screw, piston and roots) tested in a small-scale organic Rankine cycle system," Energy, vol. 165, pp. 1119-1127, 2018.

[12 P. Sapin, A. Taleb, C. Barfuss, A. J. White, D. Fabris and C. N. Markides, 1] "Thermodynamic losses in a gas spring: Comparison of experimental and numerical results," in 12th International Conference on Heat Transfer, Fluid Mechanics and Thermodynamics, Malaga, 2016.

[12 P. Sapin, M. Simpson, A. J. White and C. N. Markides, "Performance maps 2] for reciprocating-piston compressors or expanders," in ECOS, 2017.

Chinedu Kingsley Unamba, August 2020 171 Experimental development of ORC technology for waste heat-to-power conversion

172 Chinedu Kingsley Unamba, August 2020 Chapter 10: Appendix

10 APPENDIX

Chinedu Kingsley Unamba, August 2020 173 Experimental development of ORC technology for waste heat-to-power conversion

174 Chinedu Kingsley Unamba, August 2020 Chapter 10: Appendix

APPENDIX A: UNCERTAINTY ANALYSIS

Experimental results are only as good as the personnel conducting it, the components involved and most importantly the accuracy or sensitivity of the sensors used. Even when sensors are calibrated accurately, external stimuli like temperature effect and pressure may need to be compensated for.

Calibration errors, temperature coefficient errors (zero coefficient and gain coefficient error) accuracy: hysteresis, repeatability and linearity, resolution error and long term drift.

The errors in instrumentation and the subsequent propagation in the experimental results acquired during the experimental runs are fundamentally due to the uncertainty in recording the pressure, temperature, volume flowrate and mass flowrates. They arise mainly from the uncertainty on the sensor readings, usensor and the standard deviation of the time resolved data around their mean.

The errors comprise linearity, hysteresis and repeatability and was calibrated to be usensor = ±0.25% FS. If the uncertainty in single measurements 퐴 and 퐵 are 푢퐴 and 푢퐵 , respectively, and if the operation is described by Equation (44):

퐶 = 퐴 (±) 퐵 , (44) then the error propagated is:

2 2 푢C = √(푢A) + (푢B) . (45)

If the operation is described by Equation (46):

퐶 = 퐴 (×÷) 퐵 , (46) then the error propagated is:

Chinedu Kingsley Unamba, August 2020 175 Experimental development of ORC technology for waste heat-to-power conversion

푢 푢 2 푢 2 C = √( A) + ( B) . (47) 퐶 퐴 퐵

If the operation is described by Equation (48):

퐶 = 푓(퐴, 퐵) , (48) then the error propagated is:

휕퐶 2 휕퐶 2 푢 = √( 푢 ) + ( 푢 ) . (49) C 휕퐴 A 휕퐵 B

Sequel to this, if the thermal efficiency or the overall ORC cycle efficiency is:

푊̇ net 휂th = , (50) 푄̇evap then the error propagation is analysed for both the numerator and denominator.

For the numerator,

푊̇ net = 푊̇exp − 푊̇pump , and, (51) from Equation (51), the associated error is:

2 2 √ (52) 푢푊̇ net = (푢푊̇ exp) + (푢푊̇ pump ) ,

where 푢푊̇ exp and 푢푊̇ pump can be obtained from the expander power meter and pump power meter, respectively.

For the denominator, using Equation (53):

176 Chinedu Kingsley Unamba, August 2020 Chapter 10: Appendix

푄̇evap = 푚̇ wf(ℎ3 − ℎ2) , (53) the associated error propagation is:

푢 2 푢 2 √ 푚̇ wf (ℎ3−ℎ2) (54) 푢푄̇evap = ( ) + ( ) , 푚̇ wf ℎ3 − ℎ2

while 푢푚̇ wf is given by the mass flowrate manufacturer, 푢(ℎ3−ℎ2) can be calculated using:

√ 2 2 (55) 푢(h3−h2) = (푢h3) + (푢h2) .

Since enthalpy is a function of pressure and temperature, and the error associated with calculating the enthalpy can be analysed using Equation (56):

휕ℎ 2 휕ℎ 2 푢 = √( 푢 ) + ( 푢 ) , (56) h 휕푃 P 휕푇 T

휕ℎ 휕ℎ where both and can be obtained from REFPROP software of 휕푃 휕푇 thermodynamic properties., while the uncertainty on the temperature and pressure measurements are obtained from the sensors calibration curves.

Chinedu Kingsley Unamba, August 2020 177