Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Parametric characterization of an experimental vertical axis hydro turbine Rawlings, George William 2008

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2008_spring_rawlings_george.pdf [ 6.54MB ]
Metadata
JSON: 24-1.0068009.json
JSON-LD: 24-1.0068009-ld.json
RDF/XML (Pretty): 24-1.0068009-rdf.xml
RDF/JSON: 24-1.0068009-rdf.json
Turtle: 24-1.0068009-turtle.txt
N-Triples: 24-1.0068009-rdf-ntriples.txt
Original Record: 24-1.0068009-source.json
Full Text
24-1.0068009-fulltext.txt
Citation
24-1.0068009.ris

Full Text

PARAMETRIC CHARACTERIZATION OF AN EXPERIMENTAL VERTICAL AXIS HYDRO TURBINE by George William Rawlings BASc, University of British Columbia, 2005 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in The Faculty of Graduate Studies (Mechanical Engineering) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2008 © George William Rawlings, 2008 ABSTRACT The current research focuses on the design, fabrication, and testing of an experimental vertical axis tidal current turbine model to obtain first hand experimental data for use in validating numerical codes. In addition to obtaining repeatable experimental results using an entirely new system developed for the UBC towing tank, a parametric study was performed examining the effects of parasitic drag, tip losses, angle of attack, cambered blades, and shaft fairing on a free-stream device. The impacts on overall efficiency of each characteristic are quantified, leading to a prediction for the maximum efficiency of a free-stream device in the absence of losses. Upon the application of a venturi-style duct, significant gains were demonstrated in the shaft power acquired, as well as in the reduction of torque fluctuations. Application of downstream deflectors provided a further decrease in torque fluctuations with minimal decrease in efficiency, which is significant for structural considerations. A maximum Ck value of 0.473 was obtained for the ducted device compared to 0.272 for the free-stream case; however, the power produced was 12% less than what may be expected from a free- stream rotor of cross-sectional area equivalent to the duct capture area. An investigation into drag characteristics of a free-stream device further quantified the drag coefficient that may be expected, as well as the fluctuations of forces in parallel with the free-stream flow. Experimental results were then compared with a commercial RANS solver CFD model from a parallel study. This validation will enable further numerical refinement of the optimum tip-speed ratio and solidity values identified in previous research, as well as further advancements into angle of attack, airfoil profile, and ducting configurations. Lastly, a case study was presented using specif’ing a ducted 3.375m x 3.375m rotor operating in Quatsino Narrows on Vancouver Island capable of powering approximately 17 homes. 11 TABLE OF CONTENTS ABSTRACT.ii LIST OF TABLES v LIST OF FIGURES vi LIST OF SYMBOLS, NOMENCLATURES, AND ABBREVIATIONS xi ACKNOWLEDGEMENTS xiii 1 INTRODUCTION 1 1.1 Turbine Operating Principles 3 1.2 Previous Work / Motivation 6 1.3 Objectives / Scope of Work 9 2 EXPERIMENTAL SETUP AND PROCEDURE 11 2.1 Towing Tank and Carriage Overview 11 2.2 Baseline Model Parameters 13 2.3 Instrumentation 15 2.3.1 Instrumentation Components 15 2.3.2 Drive-train / Force Balance Configuration 15 2.4 Data Acquisition System 17 2.5 Calibration 18 2.6 Experimental Procedure 18 2.7 Data Processing Methodology 19 2.7.1 Data Selection and Averaging 20 2.7.2 Data Presentation 23 3 EXPERIMENTAL RESULTS 25 3.1 Angle of Attack and Revolution Angle Notation 25 3.2 Test Program Overview 26 3.3 Free-stream Turbine 28 3.3.1 Velocity and Reynolds Number Effects 30 3.3.2 Drive-train Comparison 34 3.3.3 Arm Profile Reduction 38 111 3.3.4 Single-blade .43 3.3.5 Angle of Attack 47 3.3.6 Cambered Blades 53 3.3.7 Blade End Plates 55 3.3.8 Shaft Fairing 59 3.3.9 Summary 63 3.4 Ducted Turbine 64 3.4.1 Venturi-type Ducting 66 3.4.2 Ducting with Deflectors 71 3.4.3 Summary 75 3.5 Drag Force 76 3.5.1 Summary 83 4 DISCUSSION 84 4.1 Measurement Accuracy 84 4.1.1 Instrumentation Uncertainty and Data Point Averaging 84 4.1.2 Run Repeatability 88 4.1.3 Revolution Speed Variation 95 4.2 Comparison with Numerical Predictions 99 4.2.1 Numerical Model Overview 99 4.2.2 Comparison of Results 101 4.3 Sources of Error 108 4.4 Sample Application 109 5 CONCLUSIONS AND RECOMMENDATIONS 112 5.1 Conclusions 112 5.2 Recommendations for Future Work 113 REFERENCES 116 APPENDIX A: Design Calculations 119 APPENDIX B: Component Drawings 132 APPENDIX C: Instrumentation and DAQ Components 159 APPENDIXD: Run Log 161 iv LIST OF TABLES Table 1-1: Available Davis et al. reports 7 Table 2-1: Principal model turbine parameters 14 Table 2-2: Degrees of revolution per sample for representative carriage speeds and TSR values 18 Table 3-1: Test program and corresponding parameters 26 Table 3-2: Reynolds numbers at varying velocities and TSR values for a free-stream device 31 Table 3-3: Expected and observed torque frequencies for gearbox and chains/sprockets drive-train 37 Table 3-4: Blade angle of attack at varying TSR and preset angle values at the 900 angle of revolution 48 Table 3-5: Blade angle of attack at varying TSR and preset angle values at the 270° angle of revolution 48 Table 3-6: Maximum Ck and percent increase over free-stream baseline 64 Table 3-7: Torque fluctuation coefficient for a free-stream and ducted turbine 70 Table 3-8: Maximum Ck and corresponding CTF for ducted turbine configurations (1.5 mis) 75 Table 3-9: Maximum Ck, percent change, and torque fluctuation coefficient 75 Table 3-10: Expected and observed experimental drag force frequencies 83 Table 4-1: Torque sensor and encoder uncertainty (percent of rated output) and absolute error 84 Table 4-2: Gearbox drive-train repeated run percent variation in Ck 90 Table 4-3: Sample chain/sprockets drive-train repeated run percent variation in Ck 91 Table 5-1: Maximum Ck, percent change, and torque fluctuation coefficient 113 v LIST OF FIGURES Figure 1-1: Distribution of Canada’s in-stream tidal current resource [3] 3 Figure 1-2: Vertical axis turbine schematic [] 4 Figure 1-3: Turbine driving force generation 5 Figure 2-1: Secondary carnage and turbine assembly drawing 12 Figure 2-2: Towing tank facility with main and secondary carriage 12 Figure 2-3: Turbine assembly with force balance and frame 13 Figure 2-4: Turbine rotor nomenclature (top view, inches) 14 Figure 2-5: Force balance and instrumentation configuration 16 Figure 2-6: Gearbox drive-train configuration 17 Figure 2-7: Typical run description (run duration 31.5 sec) 19 Figure 2-8: Matlab program interface 20 Figure 2-9: Range of data at steady-state for analysis 21 Figure 2-10: Torque vs. Angle of Revolution overlaid over one turbine revolution 22 Figure 2-11: Ensemble averaging 23 Figure 2-12: Example of Polar plot (counter-clockwise rotation) 24 Figure 3-1: Angle of attack notation 25 Figure 3-2: Flow direction relative to blade angular position 26 Figure 3-3: Free-stream turbine positioning (arm profiles A and B) 29 Figure 3-4: Arm profile C free-stream turbine positioning 30 Figure 3-5: Lift Coefficient vs. Angle of Attack using CFD for 634-021 at Re = 200 000, 500000 31 Figure 3-6: Cl! Cd vs. Angle of Attack for 634-021 at Re = 200 000, 500 000 32 Figure 3-7: Power coefficient (Ck) vs. tip-speed ratio (TSR) at varying velocities 33 Figure 3-8: Ck vs. TSR illustrating power loss due to parasitic drag from arm configuration A 34 Figure 3-9: Ck vs. TSR drive-train comparison (medium profile arms) 35 Figure 3-10: Torque vs. Angle of Revolution comparing chains!sprockets with gearbox drive at TSR 2.25, 2.5, 2.75, v=l.5 rn/s 36 vi Figure 3-11: Torque vs. Angle of Revolution comparing chains/sprockets with gearbox drive at TSR 2.25, 2.5, 2.75, v=2.0 mis 37 Figure 3-12: Torque data normalized frequency content for chains/sprockets and gearbox drive-train (free-stream, 1.5 mis, 2.5 TSR) 38 Figure 3-13: Arm profile cross-sections and connections 39 Figure 3-14: Ck vs. TSR for supporting arm comparison at 1.5 m/s 40 Figure 3-15: Ckvs. TSR of varying arm configurations (blades removed) at 1.5 mIs.... 41 Figure 3-16: Torque vs. Angle of Revolution for arm profiles B and C (ends and middle) at 1.5 mis and varying TSR 42 Figure 3-17: Torque vs. Angle of Revolution for 3 arms and end arms only at TSR=2.75, 3andv1.5mis 43 Figure 3-18: Ck vs. TSR for single and 3-bladed tests at 1.5 mis 44 Figure 3-19: Torque vs. Angle of Revolution at 1.5 rn/s for a single blade test 45 Figure 3-20: Torque vs. Angle of Revolution at 1.5 mis for a 3-blade test, single-blade test, and 3 superimposed single-blade tests 46 Figure 3-2 1: Torque vs. Angle of Revolution for a single-blade test with arm profiles B and C at TSR=3, v=l.5 rn/s 47 Figure 3-22: Ck vs. TSR for AoA = 0, 3, 5 degrees at 2 rn/s 49 Figure 3-23: Torque vs. Revolution Angle for AoA 0, 3, 5 deg at 2 mis, TSR = 2.25.50 Figure 3-24: Polar Plot of Torque vs. Revolution Angle for AoA 0, 3, 5 deg at 2 mis, TSR=2.25 51 Figure 3-25: Torque vs. Revolution Angle for AoA 0, 3, 5 deg at 2 mIs, TSR = 2.5. . 51 Figure 3-26: Polar Plot of Torque vs. Revolution Angle for AoA = 0, 3, 5 deg at 2 mis, TSR = 2.5 52 Figure 3-27: Torque vs. Revolution Angle for AoA = -3 deg at 1.75 mis, TSR 2.5.... 53 Figure 3-28: Ck vs. TSR for cambered (0 and 5 deg) and symmetric (0 deg) blades at 1.5 mis 54 Figure 3-29: Torque vs. Angle of Revolution for symmetric (0 deg) and cambered (0 and 5deg)atl.5mIsandTSR=2.75 55 Figure 3-30: NACA 0012 profile and circular end plates 56 Figure 3-31: Ck vs. TSR for end plate comparison at 1.5 mis 57 vii Figure 3-32: Ck vs. TSR for end plate comparison at 2 mIs 57 Figure 3-33: Torque vs. Revolution Angle comparing end plates at 1.5 mis 58 Figure 3-34: Torque vs. Revolution Angle comparing end plates at 2 mIs 59 Figure 3-35: Shaft fairings 60 Figure 3-36: Ck vs. TSR with and without shaft fairing (1.5 and 2 mIs) 61 Figure 3-37: Torque vs. Revolution Angle with and without shaft fairing at 1.5 mis, TSR=2.75 61 Figure 3-38: Single blade with installed shaft fairing 62 Figure 3-39: Single Blade Torque vs. Revolution Angle with and without shaft fairing at 1.5 mis, TSR=2.75 63 Figure 3-40: Plan view of ducting (inches) 65 Figure 3-41: Cross-section of towing tank with ducting and turbine 66 Figure 3-42: Ck vs. TSR for the free-stream and ducted turbine at 1. 5 mIs 67 Figure 3-43: Extracted Power (W) vs. TSR for the tested ducted turbine and a free- stream turbine of equivalent capture area at 1.5 m/s 68 Figure 3-44: Torque vs. Revolution Angle for free-stream turbine at 1.5 m/s 69 Figure 3-45: Torque vs. Revolution Angle for ducted turbine at 1.5 rn/s 70 Figure 3-46: Ducting with deflectors 72 Figure 3-47: Ck vs. TSR for duct and deflector configurations 73 Figure 3-48: Torque vs. Angle of Revolution for ducted and deflector configurations. . 74 Figure 3-49: Polar plot of Torque vs. Angle of Revolution for ducted configurations. .. 74 Figure 3-50: Side view providing location of assumed centre of drag force 77 Figure 3-51: Drag Force vs. TSR for a free-stream turbine at varying velocity 78 Figure 3-52: Drag Coefficient vs. TSR with trend line for data at v=1.5, 1.75, 2mIs 79 Figure 3-53: Drag Force vs. Revolution Angle at 1.5 mIs, AoA=0 80 Figure 3-54: Drag Force vs. Revolution Angle at 2 mIs, AoA=0 80 Figure 3-55: Drag Coefficient vs. TSR for a single and 3-bladed device at 1 .5mIs, AoA=3 81 Figure 3-56: Drag Force vs. Revolution Angle for a single blade at 2 mIs, AoA=3 82 Figure 4-1: Standard Deviation and Torque vs. Revolution Angle for a free-stream device with gearbox drive-train at 1.5 rn/s and TSR=2.5 (N - 34) 85 viii Figure 4-2: Standard Deviation and Torque vs. Revolution Angle for a free-stream device with gearbox drive-train at 2 m/s and TSR2.5 (N 52) 86 Figure 4-3: Standard Deviation and Torque vs. Revolution Angle for a free-stream device with chains/sprockets drive-train at 2 mIs and TSR=2.25 (N - 33) 87 Figure 4-4: Standard Deviation and Torque vs. Revolution Angle for a ducted device with gearbox drive-train at 1.5 mIs and TSR3 (N — 45) 88 Figure 4-5: Torque vs. Revolution Angle for repeated runs with gearbox drive-train at 1.5 mIs, TSR=2.5 (arm profile C) 92 Figure 4-6: Torque vs. Revolution Angle for repeated runs with gearbox drive-train at 2 mis, TSR=2.5 (arm profile C) 92 Figure 4-7: Polar plot of Torque vs. Revolution Angle for repeated runs with gearbox drive-train at 2 mIs, TSR=2.5 (arm profile C) 93 Figure 4-8: Torque vs. Revolution Angle for ducted repeated runs with gearbox drive- train at 1.5 mIs, TSR=2.75 94 Figure 4-9: Torque vs. Revolution Angle for repeated runs with chains/sprockets drive- train at 1.5 and 2 mis, TSR=2.5 (arm profile B) 94 Figure 4-10: Torque (below) and RPM (above) vs. Revolution Angle for runs with chains/sprockets drive-train at 1.5 m/s 96 Figure 4-11: Torque (below) and RPM (above) vs. Revolution Angle for runs with gearbox drive-train at 1.5 mIs 97 Figure 4-12: Torque vs. Revolution Angle for ducted device at 1.5 mis 98 Figure 4-13: RPM vs. Revolution Angle for ducted device at 1.5 mis 98 Figure 4-14: Sample grid around the blades and shaft 100 Figure 4-15: Sample velocity contours for a simulation at 1 mis with TSR=2 100 Figure 4-16: Experimental Ck vs. TSR for arm profile C at 1.5 and 2 m/s 102 Figure 4-17: Ck vs. TSR for free-stream comparison of experimental and numerical results 102 Figure 4-18: Ck vs. TSR for ducted comparison of experimental and numerical results at 1.5 rn/s 103 Figure 4-19: Torque vs. Revolution Angle comparing free-stream experiments and Fluent at 1.5 rn/s and TSR2 104 ix Figure 4-20: Torque vs. Revolution Angle comparing free-stream experiments and Fluent at 2 rn/s and TSR=2.75 105 Figure 4-21: Torque vs. Revolution Angle for a ducted turbine at 2 ni’s and TSR=2.. 106 Figure 4-22: Torque vs. Revolution Angle for a ducted turbine at 1.5 m/s and TSR=2.75. 106 Figure 4-23: Drag Force and Torque vs. Revolution Angle for free-stream Fluent and experiments at 2 rn/s and TSR=2.75 107 Figure 4-24: Tidal current data 110 Figure 4-25: Power and torque output 111 Figure 4-26: Representative device configuration 111 x LIST OF SYMBOLS, NOMENCLATURES, AND ABBREVIATIONS A Turbine cross-sectional area (O.914m x O.686m) AoA Blade angle of attack (leading edge rotated outwards is positive) Bk Betz coefficient = 16/27 c Blade chord Cd Drag coefficient Ck Power coefficient CI Confidence Interval Cl Lift coefficient C Power coefficient accounting for Betz limit CTF Torque fluctuation coefficient CFD Computation fluid dynamics D Drag force DAQ Data acquisition deg, ° Degrees FFT Fast Fourier Transform HMCS Her Majesty’s Canadian Ship kWh Kilowatt-hour 1 Length m Metres MW Mega-watt n Number of blades N Number of observations in a sample (for standard deviation calculation) NI National Instruments NRC National Research Council of Canada r Turbine radius (centre of shaft to ¼ chord) p Density PA Extracted power = torque*angular frequency for current experiments xi RPM Revolutions per minute s Seconds Tavg Average torque Tmax Maximum torque Tmin Minimum torque TSR Tip-speed ratio p Viscosity UBC University of British Columbia V, v Free-stream velocity VAHT Vertical Axis Hydro Turbine Turbine angular velocity “, in Inches xii ACKNOWLEDGEMENTS Firstly, I would like to extend a sincere thank you to Dr. Sander Calisal for providing me with the opportunity to venture into the field of ocean energy. His openness to new ideas, guidance in all aspects related to fluid dynamics, and constant push to discover more is a first-class example of how research should be conducted. Secondly, thank you to Jon Mikkelsen for his almost daily consultation, as well as his demonstrated commitment to enhancing student experiences and providing new opportunities for every student showing an interest in marine engineering. To those in the lab, we’ve accomplished a large amount of work and have been fortunate to do it so well as a group. Specifically, Voytek Klaptocz has been a great example by simply getting stuff done when it matters and making sure everyone’s having a good time doing it, all the while being a valuable resource for ideas and general guidance. Yasser Nabavi and Mahmoud Alidadi have demonstrated immense dedication to the project and I am grateful for the time we have spent chasing ideas, as well as their repeated patience when explaining concepts. Similarly, thank you to Ye Li for his keenness to help out whenever possible. A large number of co-op and visiting students have also made valuable contributions to the project through work on design drawings, instrumentation, data acquisition, and data analysis. Those include Florent Cultot, Cameron Fraser, John Axerio, Robby Chen, Pierre Leplatois, Bo Zulonas, and Thomas Chabut. I would also like to extend my appreciation to Blue Energy, and in particular Jon Ellison, for both their financial contributions to the research as well as the good times we shared during the experiments. Additionally, thank you to Western Economic Diversification for the funding of experimental equipment and personnel to make the project possible. Lastly, thank you to my friends for their patience and to my family for their unwavering support in whatever I choose to do, and encouragement to do it right. xiii I INTRODUCTION The mounting evidence substantiating human-caused climate change [1], as well as the pending shortage of fossil fuels [2], is creating an increasing demand for clean, renewable sources of energy. Harnessing wind and photovoltaic energy is among the more traditional means of renewable energy capture; however, increasing attention is being turned to the world’s oceans as a resource for wave, tidal, and thermal energy extraction. Canada is fortunate to possess vast wave and tidal energy resources. The Canadian wave resource is estimated to be 146,500 MW, or more than double the current electricity demand, though it should be noted that only a fraction of this total may be extracted and converted to useful power due to power conversion, socio-economic factors, or technology limitations [3]. Similarly, Canada is endowed with abundant tidal current resources. Recent estimates put Canada’s tidal current resource at 42,240 MW based on examination of sites with over 1MW of in-stream power, again with only a fraction of that being extractable. Figure 1-1 below provides the distribution of this resource, equivalent to approximately 63% of Canada’s current electricity demand [3]. In addition to the significant resource available, tidal currents are advantageous in that they are highly reliable and predictable, and the extraction of this energy using low-head turbines is expected to be environmentally benign [4]. Tidal current energy extraction differs from tidal barrage type power plants (existing in France and Nova Scotia), which function primarily as dams and release water in a controlled manner after the water level on one side of the dam has dropped. Dr. Barry Davis, former Chief Hydrodynamic Designer for the HMCS Bras D’Or Hydrofoil Ship and Aerodynamic Loads Analyst for the Avro Arrow, was one of the first people to recognize the potential of tidal current energy extraction and began focusing his research here in 1978. Building upon the National Research Council of Canada’s (NRC) development of the Darrieus vertical axis wind turbine, he applied the technology to low head hyciro applications [4]. Dr. Davis’ research led to an extensive research program during the 1980’s developing the vertical axis hydro turbine (VAHT) funded by over $1.3 million Canadian dollars. This work, completed as Nova Energy in collaboration with 1 the NRC, led to a number of demonstration projects, the publication of multiple reports, and several independent assessments validating the technology; however, due to the low cost of fossil fuels and the lack of political support for further development of tidal energy at the time, neither Nova Energy nor its successor Blue Energy could establish any major projects through the 1990s. In 2005, Blue Energy approached the University of British Columbia (UBC) to inquire about developing a computational fluid dynamics (CFD) model of the turbine to update their technology. Numerical models are a particularly useful tool in the field of tidal energy extraction as they: • Can be linked with an optimizer tool to efficiently conduct parametric studies and determine optimum turbine parameters • May evaluate designs at various scales, thus minimizing unknown scaling effects when changing turbine size • Can calculate blade loads used for mechanical calculations or incorporated directly into Finite Element Analysis software • Permit two-phase simulations that can predict cavitation inception • May incorporate site-specific current data, accurately predicting power output including cut-in and cut-out operating regimes • Enable examination of turbine interaction and provide insight into productive I destructive interference • Allow for flow visualization enabling prediction of environmental effects This need for numerical model development led to a collaborative research agreement and the ongoing research into the VAHT at UBC. In the meantime, since Dr. Davis’ research in the 1980’s, the market price of a barrel of oil had risen from $18 USD [5] per barrel in 1985 to over $100 USD in 2008, rendering tidal energy a feasible method of energy extraction. A number of tidal energy technology developers have also entered the market, attracted by current tidal energy cost estimates of 11 — 25 /kWh, and future estimates in the 5 —7 /kWh range [6]. 2 Tidal Energy Resources MW Points: Mean Potential Power <2 4”.7 7... 14 14...27 27... 53 53... 102 102... 198 198... 383 383...741 741 ... 1436 1436,.. 2780 2780... 5384 >5384 Figure 1-1: Distribution of Canada’s in-stream tidal current resource 13]. 1.1 Turbine Operating Principles — -%‘ I- 0 S 0 . 1000km The vertical axis turbine is a lift-driven device consisting of vertical foils (typically 3 or 4) mounted perpendicular to the flow, usually to a spinning central shaft as shown in Figure 1-2. This differs from a horizontal axis device, which is often similar to a wind turbine or ducted impeller or propeller mounted to the seabed. As the foils rotate, typically at 2-3 times the free-stream flow velocity, the free-stream flow inducess an angle of attack on the foil. The resultant of the lift and drag forces generated by the foil may be reduced to radial and tangential components, of which the tangential component drives the turbine rotation. Figure 1-3 illustrates this concept when a blade passes across the upstream side of the turbine. As the turbine continues to rotate, the relations between the vectors shift, and as a result tangential force is generated primarily in the regions upstream and downstream of the shaft. This causes torque fluctuations, or torque ripple, 3 of the turbine due to blades passing in and out of torque-generating regions. Similarly, the radial component of the force on the blades and the drag forces on the turbine fluctuate with blade position. These cyclic loads are of concern when designing for turbine reliability and longevity. Figure 1-2: Vertical axis turbine schematic 17]. TOP VIEW generaloc gearbox 4 These torque fluctuations are much less evident in horizontal-axis designs, and the primary arguments against the vertical-axis turbine are that the torque ripple is difficult to manage both for structural integrity and generator function, and that efficiency is lost given the turbine blades are only generating torque through select regions of each revolution. Conversely, there are a number of advantages unique to the vertical axis turbine, encouraging further examination: • Generators may be easily stored above the water surface and directly driven by the shaft • Only a single bearing is required underwater • Turbine rotates in same direction regardless of flow direction • The vertical design is conducive to stacking multiple turbines under bridges or other existing infrastructure Figure 1-3: Turbine driving force generation. 5 Until functional commercial units of both horizontal and vertical axis turbines are established and the cost per kWh is compared on a site-by-site basis, the design most suitable to tidal current applications remains unknown. 1.2 Previous Work I Motivation Prior to Davis’ work, Templin examined key parameters affecting Darrieus wind turbine operation by plotting power coefficient (Cp) as a function of tip-speed ratio (TSR) and solidity [8]. __ 1Cp = .— Equation 1 Bk 2 r.co TSR = — Equation 2 V n.c solidity = — Equation 3 r In the Cp calculation above, it is interesting to note the extracted power (PA) is divided by the power available in the free-stream passing through the turbine cross-sectional area, which would be the equivalent of efficiency for a free-stream device. This Cp value is then divided by the Betz coefficient (Bk = 16/27), which is the maximum theoretical efficiency for a free-stream turbine according to idealized wind theory [9], thus yielding the efficiency of the device compared to the theoretical maximum extraction possible. Davis then adapted Templin’s work to tidal turbines and generated a number of reports in collaboration with the NRC, upon which many of the initial turbine parameters and dimensionless coefficients were based for the UBC series of tests. Davis initiated the use of the power coefficient (Ck) to quantify turbine performance. This is similar to Cp above, though it is not divided by the Betz coefficient: 6 Ck= .A Equation 4 It should be noted that the Ck value is often used interchangeably with efficiency, though this is only appropriate when used in free-stream applications. This is because the addition of ducting, or operation in a confmed flume or tank, will enhance the turbine power output; however, the power output (PA) is still only being divided by an extractable power term that is a function of the free-stream velocity and cross-sectional area of the turbine, instead of a function the effective velocity through the turbine which is altered by the duct or confmed domain, or a function of the increased area affected by the duct cross-sectional area or the domain boundaries. As per Davis, power output data discussed below is presented in terms of Ck. The available Davis reports were as follows: Table 1-1: Available Davis et al. reports Report Title Synopsis NEL-002: Water Turbine Model Trials [10] Flume tank tests of vertical and horizontal axis water turbines. NEL-02 1: Ultra Low Head Hydroelectric Power Vertical axis water turbine flume tank Generation Using Ducted Vertical Axis Water tests with caissons, walls, and vane Turbines [11] duct configurations. NEL-022: Ultra Low Head Hydroelectric Power Continuation of NEL-021 with a Generation Using Ducted Vertical Axis Water more robust model. Turbines [12] NEL-03 8: Research and Development of a Installation of 70 kW turbine within a 50kW to 100kW Vertical Axis Hydro Turbine dam in Nova Scotia. for a Restricted Flow Installation [13] NEL-070: The Ducted Vertical Axis Hydro Investigates application of vertical Turbine for Large Scale Tidal Energy axis turbine in a 474 turbine tidal Applications [14] fence. NEL-08 1: Commissioning and Testing of a Examines repaired and enhanced 100kW Vertical Axis Hydraulic Turbine [15] version of model in NEL-038. Numerical model validation requires both power extraction data and torque data as a function of blade angle. Torque data as a function of blade angle, also known as a torque curve, is critical to provide insight into the regions where torque generation may be 7 enhanced to improve turbine performance, or may be altered to reduce torque ripple. Unfortunately, though discussed briefly by Davis et al. [12], the reports above did not contain sufficient torque curve data for model validation. Aside from Davis et a!., Gorlov patented a vertical axis turbine using helical blades to distribute the torque loading in 1994 (U.S. Patent 5451127) and continues development work in Korea [16]. Given the commercial nature of this venture, efficiency data and torque curve data is closely guarded. Similarly, research has been undertaken in Italy by the Ponte di Archimede S.p.A. Company and the University of Napoli on a turbine with a patented passive angle of attack adjustment mechanism [17,18]; though no publicly available torque curve data has been found. The United Kingdom is a leader in tidal energy technology given the active resource in Northern Scotland and generous government incentives promoting technology development. The former Department of Trade and Industry sponsored three reports on vertical-axis tidal turbines, though only one attempted experimental trials for numerical model validation and provided no useful quantitative data due to a number of factors, including excess friction in the gearbox and a less than ideal experimental flume facility [19]. Other recent efforts include a group from the University of Buenos Aires [201 that has looked into ducting effects, and a group from the University of Edinburgh [21] that has developed a number of numerical models and a conceptual design, though both are lacking experimental data for validation. Considering torque curve data that was able to be located, Shiono et al. [22] only provided torque curve data upon turbine start-up, and Highquest [23] obtained torque curve data limited to 2 or 3 turbine revolutions on a chart recorder in 1987, providing little accuracy for validation. Secondly, the literature search outlined above revealed no investigation into the drag forces on the turbine during operation, making mechanical design (particularly bearing specification) very difficult. Apart from the apparent lack of available turbine performance, torque ripple, and drag data, a number of factors affect one’s ability to properly use another researcher’s experimental data for model validation: 8 • Flume/towing tank blockage affects turbine performance and must be well documented • Drive-train losses may affect power output or dampen torque readings • Shaft and mounting arms affect turbine performance through interference effects and parasitic drag, and geometry and effects of each must be examined • Knowledge of revolution speed fluctuations is required as performance is highly dependent on TSR This lack of data and need for comprehensive first-hand knowledge of the experimental setup and parameters provided the motivation for the experimental investigation presented in this thesis. 1.3 Objectives I Scope of Work The primary purpose of this thesis is to acquire baseline power output and torque ripple data for both a free-stream and ducted vertical axis current turbine for the purpose of validating numerical models, which are currently being developed by two other graduate students. These tests will also serve to enhance understanding of work completed by previous researchers, as well as investigate a number of turbine parameters and quantifr their corresponding effects on performance. More specifically: • Acquire power coefficient data for both a free-stream and ducted vertical axis turbine in the UBC campus towing tank • Acquire torque fluctuation data for both a free-stream and ducted vertical axis turbine over the course of a turbine revolution • Investigate effects of TSR, blade angle of incidence, cambered blades, and various ducting configurations on turbine performance and torque fluctuations. Effects of shaft fairings, arms, and foil end plates are also examined • Experimentally investigate magnitude of forces parallel to the free-stream flow on the turbine for future design applications (referred to as drag forces) 9 Chapter 2 below outlines the entirely new system developed for conducting tests in the UBC towing tank. This includes the requirement for a secondary carriage to accommodate the turbine testing. An overview of the data acquisition (DAQ) software, instrumentation, experimental procedure, and data analysis program is also provided as well as the baseline model parameters. Chapter 3 presents the experimental power coefficient and torque curve results from the three experimental test programs and discusses their significance. An overview of the recorded drag data is also provided. Chapter 4 examines experimental errors and compares select power output, torque curve, and drag curve results with theory. These results are then used to develop a case study specifying a sample unit capable of powering 17 homes in Quatsino Narrows on Vancouver Island. Chapter 5 contains conclusions and recommendations for future work. 10 2 EXPERIMENTAL SETUP AND PROCEDURE All instrumentation, data acquisition equipment and software, experimental equipment, and data analysis software was purchased, built, or written specifically for this research program and is described below. 2.1 Towing Tank and Carriage Overview Experimental testing was conducted in the UBC campus towing tank, which is a 200’ long by 12’ wide by 8’ (7’ of water) deep fresh water tank. The main cantilevered carriage, typically used for ship model testing, runs on rails alongside the tank. The tank is oriented in the east-west direction and runs were performed traveling both towards the wave-maker (due east) and towards the dock (due west). A secondary carriage spanning the width of the tank was constructed and attached to the main carriage and used as the testing platform for the turbine, as shown in Figure 2-1 and Figure 2-2. The use of the secondary carriage was necessary to accommodate the large turbine device: • Support increased weight and drag force compared to typical ship hull model tests • Facilitate turbine installation and removal • Provide easy access for adjustments • Serve as a platform for the large amount of instrumentation including motor and drive-train The secondary carriage was fabricated of welded aluminum c-channel in two halves that were then bolted together. Two rubber wheels rested on both the outer rail and the side of the tank opposite the main carriage, while two v-grooved wheels ran along the rail closest to the water. The entire secondary carriage was bolted to the front of the main carriage, with a diagonal brace providing added support. 11 Turbine assembly Secondary Main carriagecarriage location / / Figure 2-1: Secondary carriage and turbine assembly drawing. Figure 2-2: Towing tank facility with main and secondary carriage. 12 2.2 Baseline Model Parameters The three turbine blades are attached to a central shaft that is supported at both ends by ball bearings. The top bearing is mounted on the force balance, while the bottom bearing is constrained to a horizontally-mounted bottom plate supported by two vertical rectangular beams forming a u-shaped frame. These beams are bolted to the secondary carriage and stiffened using 3 guy wires each; two extending in the plane of the flow direction (one forwards and one backwards) up to the secondary carriage, and the third extending in the plane perpendicular to flow direction and out towards the side up to the secondary carriage. The turbine assembly with arms supporting the blades at the ¼ span locations is shown in Figure 2-3 below, along with the supporting frame and force balance for mounting the instrumentation. Figure 2-3: Turbine assembly with force balance and frame. 13 Principal model parameters are provided in Table 2-1 and Figure 2-4: PARAMETER DIMENSION / CHARACTERISTIC Diameter (across foil chord) 36 in Number of blades 3 Blade span 27 in Blade profile NACA 634-02 1 and 634-42 1 Chord length 2.70 in ideal; 2.57 in manufactured Shaft outer diameter 1.9 in Arm profiles supporting the blades varied between test programs and therefore are not listed above. Appendix A and Appendix B contain component sizing calculations and part drawings respectively. Specific turbine and ducting position within the towing tank is discussed in Section 3 for each case presented. Table 2-1: Principal model turbine parameters. /7 ,1 ‘V / / 01.90 120° 036.00 — 2.57 CHORD / /‘ 27 BLADE SPAN ,, // (TYP) - Figure 2-4: Turbine rotor nomenclature (top view, inches). 14 2.3 Instrumentation 2.3.1 Instrumentation Components The components used for measuring drag force, torque, turbine angle, and for driving the turbine were as follows: • 31W Micro Max motor 182TCZ TEFC from Marathon Electric with Parker SSD AC vector drive controller and braking resistor kit (may be used for both driving and braking turbine) (7/8” shaft; 230V, 4.6A, 5400 max. safe rpm) • 2 of PT-Global SG-PT4000-500 lb s-type load cells • Futek Torque Sensor, 0 - 369 ft lb, 0.2% accuracy, aluminum, 2mV/Voutput, 7” length (TRS300) • Accu-Coder 776-B-S-2048-R-PP-E-P-A-N 1-7/8” through-bore encoder (2048 increments per revolution) • U.S. Digital encoder digital-analog converter (used with encoder) • CONEX gearbox B091020.LAAJU, TEXTRON fluid and power. Ratio 20:1, SHC 634 lubrication, helicoidal gear geometry Additional specifications on the components above may be found in Appendix C. Carriage speed was monitored using a pre-existing system on the towing carriage. 2.3.2 Drive-train I Force Balance Configuration Model revolution speed was controlled using an AC motor, and for the first two test programs chains and sprockets drove the turbine shaft, as well as provided the ratios necessary to scale the revolution speeds between the turbine and motor shafts. The motor, chains and sprockets, and lay-shaft (consisting of the torque sensor) all mounted to the bottom plate of the force balance as shown in Figure 2-5. This lower plate was hung from the top plate using two pairs of hinged arms and was thus free to translate relative to the top plate; additionally, large holes were cut in the top plate to allow the main turbine shaft and lay-shaft to pass through without contact. Two load cells (one on each side of the force balance) were then used to ground the bottom plate relative to the top, and thus measure the forces on the bottom plate. To accelerate the turbine to the desired rotation speed, the motor drove the lay-shaft, which consisted of the torque sensor 15 and adaptive couplings mounted vertically on two bearings, at a 14:72 ratio. The lay- shaft then drove the main turbine shaft at a 20:3 6 ratio. Alternately, when the motor was acting as a brake to slow the turbine rotation, the system drove in the reverse direction. This chain and sprocket system was used to facilitate drag force measurement using this force balance design, as well as to allow for flexibility to change the sprocket ratios should the motor or torque sensor not performed as anticipated. Lastly, the encoder was mounted directly around the main turbine shaft above the top bearing. For the third test program, the chain and sprocket drive-train was replaced with a 20:1 gearbox, and the force balance plates were rigidly joined using a plate and aluminum channel (Figure 2-6). This was an attempt to reduce revolution speed fluctuations (discussed in Section 4.1.3 below) by using a more rigid system with the 90° worm gear drive, and thus drag measurements were no longer recorded. A second bearing was Torque sensor \Bottom plate —I Turbine shall Figure 2-5: Force balance and instrumentation configuration. 16 added to the top plate to minimize shaft deflections, and a flexible coupling was used to couple the torque sensor and gearbox. 2.4 Data Acquisition System The following National Instruments (NI) data acquisition hardware components were used for these trials: • 1 cDAQ-9 172 8-slot USB Chassis with rail mounting kit • 1 NI 9205 32-Channel +1- 1OV 250 ks/s 16-bit analog input module used with encoder and carriage speed • 1 NI 9237 4-Ch 50 ks/s per channel 24-bit analog input module used with torque sensor Supplementary DAQ hardware information may be found in Appendix C. Labview software was developed to take 100 samples on each channel (angle, torque, carriage speed, and load cell 1 and 2 where necessary). Each set of 100 samples was then averaged and written to an output file, and this sequence was performed at a frequency of Flexible coupling— (torque sensor hidden) Shaft beaiingsEncoder// Turbine shaft Figure 2-6: Gearbox drive-train configuration. 17 approximately 240 Hz, or every 0.00406 seconds. Table 2-2 below provides number of degrees of revolution per data point for representative velocity and TSR values. Table 2-2: Degrees of revolution per sample for representative carriage speeds and TSR values. Number of Degrees per Sample Velocity (m/s) TSR = 1.5 TSR = 2 TSR = 2.5 TSR = 3 1.5 1.14 1.53 1.91 2.29 2 1.53 2.04 2.54 3.05 2.5 Calibration Calibration of the instrumentation components was performed as required. The torque sensor utilized a manufacturer supplied constant that was verified in the lab. Routine checks using the shunt resistor were then performed validating the 0-500Nm range. Similarly, routine checks were used to verify that the angular encoder was accurate over 0-360°. Lastly, each load cell was connected one at a time and calibrated by applying a force (typically up to 16 lb) to the lower force balance plate using a rope and pulley system. 2.6 Experimental Procedure For each test run, a standard procedure was followed: 1. The carriage and turbine were stopped while the waves dissipated on the water surface and the vortices dissipated in the tank 2. The turbine was manually rotated such that a blade was in the 180° position and the encoder was reset to 180° (0° corresponds to when a blade is heading directly into the oncoming flow as discussed Section 2.7 below) 3. The DAQ system and motor driving the turbine were started a. If drag data was being recorded, then the DAQ system was started and allowed to run for a few seconds to record values at zero velocity before starting the turbine and allowing it to reach the desired revolution speed b. If no drag data was being recorded, the turbine was started and allowed to reach the desired revolution speed; the DAQ system then started to record 4. The carriage accelerated up to speed while the motor maintained the turbine at the desired revolution speed 18 5. The carriage ran down the tank at the desired speed for the maximum allowable distance 6. The carriage was decelerated to a stop 7. The DAQ system was stopped and the motor driving the turbine was turned off, allowing the turbine to come to rest Figure 2-7 below illustrates this procedure (case 2.b) using a plot of torque measurement vs. cumulative angle of turbine rotation for a typical run; the duration of the recorded data period was 31.5 seconds. 2500 5000 Figure 2-7: TypicaL run description (run duration 31.5 sec). 2.7 Data Processing Methodology A Matlab program was developed to first read the raw data files output from the DAQ program, then format the data, and subsequently facilitate “on-the-spot” data analysis. This analysis primarily consisted of plotting loads recorded by each individual load cell, Torque vs. Cumulative Angle 120 140 ---- --- 100 80 .jI ill_i t_ 11.1 60 z 4) 0 I- 40 - II 20 0 .20 .40 -60 — ;—-;;—; ‘;;; 1 Turbine rotation at arriag Steady state operation at desired rpm acceleration desired carriage speed ‘I 20000 Carriage deceleration Angle (deg) 19 the total load, the torque values, or the turbine revolution speed versus either time or revolution angle (either cumulative or reduced to over 1 revolution). The raw data files were processed such that the recorded parameters from the different test programs could be plotted on the same plots, enabling comparison. Figure 2-8 displays the primary Matlab program user interface. — cnuor L,e ane (easig 0 Decreasing Sort Figure 2-8: Matlab program interface. 2.7.1 Data Selection and Averaging The Matlab program was written to select the range of data at outside of the carriage acceleration and deceleration periods and thus suitable for analysis. Examining the carriage velocity data column, the beginning and end of the range of data at the desired carriage velocity was specified. 10% of the length of this specified range was further eliminated from either end, leaving the middle 80% of the data at the constant velocity to I , - ‘T — DataBasev2 Please select the Its(s) that you woiid Ie to lead: ___________ C:lDocumerits end Sefti &OdlOesktop2(t)7 Experimertel Data’FromTow O II.w 5rm RunOl 3_vi .5_2.Stsr-M.txt :RunO13v1 .5_2.5tsr.txt RiinOl4-v1 -2JStsr-Mlxt ____ Ren014_vl 5_27sr-M1xt Rue0141.52.7ttsr.txt SaveAs: I c___. n. [j Save Database Llrs: Metric (Nm) Abscissa (x): Theta CWiate (y) - Torque Legend Filename •1 ______ E OiApiA data to excel (wi 0L*pt to screen if ftenne left blat) ® L0z E Pr* Black and l.Me aFils Er Excel lienwie to wrte to: Q P01W Grh Polar Zero: __________ P: Crne Despey FF701 ‘dirrate I I Plow Tiae Turbulence Mezh Pilezsae RPM Vel. TSP. S Wells Step Model S Tnt. Ord. Size Airfoil Coexe.nt i.-9so/ Z.49 ö.OO - N/A N7A N/A Ni RunOl4—vl.5—2.7Stsr—M.txt 85.6 lSOe/s 2.73 0.004 N/A N/A N/A N/A Select All j [Select None lop ci Files Update Selected DataItes Stsit >1 FftSit: 0 20 be written to a new Matlab file (with a “-M” extension to the file name) for further analysis as shown in Figure 2-9. This method of selecting the steady-state range was tested during experiments and provided consistent torque profiles at either end of the range. Columns written in the “-M” file included Time, Theta, Torque, RPM, Carriage Velocity, Time-step, as well as Load Celli, Load Cell 2, and Total Load where applicable. Calculations were performed as necessary to complete the columns above: • Shaft revolution speed ratios were applied to the torque values when the lay-shaft experimental setup was used. • Moment arm ratios were applied as needed to the drag force calculation (further discussed in Section3.5). • Instantaneous angular velocity, and subsequently RPM, for a given point was the average of the 12 closest points to minimize data spikes from the small interval (change in theta over time-step) used for calculation. f Figure 2-9: Range of data at steady-state for analysis. 140 Torque vs. Cumulative Angle 120 100 80 E z 0 I. 60 40 20 0 8( -20 ‘0 9000 10000 11000 12000 13000 14000 15000 16000 Angle (deg) 17000 18 00 21 An ensemble averaging technique was then used to collapse the data onto one turbine revolution. The torque values over each revolution for a single run were plotted over 360 degrees, or overlaid on each other, as shown by the small points in Figure 2-10. The data was then isolated into 4 degree increments, as demonstrated in Figure 2-11, in which an average (cross) is obtained from the overlaid data points for each increment. Figure 2-10 also displays the resulting average torque curve over one revolution. A A I Figure 2-10: Torque vs. Angle of Revolution overlaid over one turbine revolution. 120 140 -----.-.-- - 100 80 11 E z 0 I 60 i1 40 Overlaid Data Points $ —.—Averaged Data 20 0 -20 50 100 150 200 250 300 Angle (deg) 350 400 22 ++ + + + ±+ + 4 ++ Figure 2-11: Ensemble averaging. 2.7.2 Data Presentation Data is typically presented in three forms. Firstly, plots are often given as power coefficient vs. tip-speed ratio, demonstrating the capability of the device to extract power from the free-stream current. The two other plots are used to enhance understanding of the turbine operation, and provide the parameter of interest (typically torque) vs. angle of rotation in both Cartesian (ie. Figure 2-10) and Polar (ie. Figure 2-12) coordinates. Because Polar plots typically distort the plots and don’t easily display negative values, they are primarily used as a visualization tool for highlighting the regions of turbine revolution that could benefit from flow adjustment to enhance turbine performance as well as reduce torque fluctuations. Figure 2-12 illustrates the torque generated by a three-bladed turbine oriented such that at 0 degrees a blade is headed directly into the flow. Flow enters the turbine from the top of the image (90°) and rotation is counter clockwise. The 3 peaks are created as torque is generally produced by each blade as it passes through approximately 90°-120° in the region upstream of the shaft. 23 Torque (Nm) vs. Angle (deg) Flow direction 120 60 1W / 33 -S S• S 210 240 270 Figure 2-12: Example of Polar plot (counter-clockwise rotation). 24 3 EXPERIMENTAL RESULTS The specific setup and results for each test program conducted are discussed in the Sections 3.2 through 3.4. Experimental errors and measurement accuracy are later discussed in Section 4.1. 3.1 Angle of Attack and Revolution Angle Notation Blade incidence angle (commonly referred to as angle of attack — A0A) was investigated in a number of the experiments, and is considered positive when the leading edge of the blade was rotated outwards from the main shaft as shown in Figure 3-1 below. Figure 3-1: Angle of attack notation. Blade position over the course of a revolution is also of importance when reading plots and understanding turbine operation. For the results presented below, a blade is considered to be at 0 degrees when it is headed directly into the flow, and is at 180 degrees when it is moving in the same direction as the flow. This is illustrated in Figure 25 3-2 below, with a blade generally producing torque at approximately the 90 degree position and 270 degree position, as it passes perpendicular to the free-stream flow. Figure 3-2: Flow direction relative to blade angular position. 3.2 Test Program Overview Three programs were performed in August 2006, November 2006, and Aug/Sep 2007. Table 3-1 provides details on model configuration and parameters examined during each test program. It should be noted that for each test program the arm profiles were subsequently reduced, while specific arm profiles, end plate specifications, and other turbine parameters may be found in Appendix B. A detailed run log may be found in Appendix D. Table 3-1: Test program and corresponding parameters. TEST PROGRAM PROGRAM DETAILS August 2006 • Chain and sprocket drive-train (approx. 575 runs) Flow into Turbine 111111 9 or-’ / — — () 180° I!’ — — 270° 26 TEST PROGRAM PROGRAM DETAILS • High-profile arms (configuration A) supporting blades at Y4 chord • Symmetric blade profile 634-021 • Parameters tested: o Blade angles of attack -5, 0, 3, 5, 10 o Carriage velocities 1,1.25,1.5, 1.75,2 mIs o TSR values 1.25 —3.5 at 0.25 increments o Single blade o Arms without blades attached November 2006 • Chain and sprocket drive-train (approx. 460 runs) • Medium-profile arms (configuration B) supporting blades at ¼ chord • Symmetric blade profile 634-02 1 • Parameters tested: o Carriage velocities 1 — 2 mIs at 0.25 increments o TSRvalues 1.25—3.5 o Free-stream turbine at AoA -3,0,3,5 deg o Single blade at AoA = 3 deg o Ducted turbine with open ends at AoA 0,3,5 deg o Medium profile arms without blades Aug/Sep 2007 • Gearbox drive-train (approx. 340 runs) • Parameters tested: o TSR values 1.5 — 3.5 at 1.5 mIs carriage speed, and 1.5 — 2.75 at 2 mIs carriage speed o Medium-profile arms at ¼ locations vs. low-profile (NACA 0012) arms at ends and middle of blades o Medium-profile arms with circular and foil end plates o 2 vs. 3 arms (foils end supported with removable 27 TEST PROGRAM PROGRAM DETAILS arm at centre) o Symmetric blade 634-021 at AoA = 0, and cambered blade 634-421at AoA = 0, 5 deg o Single blade o Duct with end covers and deflectors at varying positions o Shaft fairing with single blade, 3 blades, and ducted turbine o Low-profile arms without blades 3.3 Free-stream Turbine Figure 3-3 below illustrates turbine positioning within the tank for both the high and medium profile arms (profiles A and B discussed in Section 3.3.3) supporting the blades at the ¼ chord locations. Figure 3-4 highlights the change in turbine position to accommodate ducting with end caps when the low-profile (NACA 0012) supporting arms were used at the ends, and usually middle, of each blade. The following tests and parameters are discussed in Sections 3.3.1 through 3.3.8: • 3.3.1 Velocity / Reynolds Number Effects • 3.3.2 Drive-train Comparison • 3.3.3 Arm Profile Reduction • 3.3.4 Single-blade • 3.3.5 Angle of Attack • 3.3.6 Cambered Blades • 3.3.7 Blade End Plates • 3.3.8 Shaft Fainng Lastly, Section 3.3.9 summarizes these results. 28 Figure 3-3: Free-stream turbine positioning (arm profiles A and B). Tank width = 144” 29 Figure 3-4: Arm profile C free-stream turbine positioning. 3.3.1 Velocity and Reynolds Number Effects Reynolds number, and as a result free-stream velocity and tip-speed ratio, affect turbine performance. Table 3-2 below illustrates the range of Reynolds numbers observed at the primary velocities and TSR values examined. As these values range between 32 600 and 522 000, the foil is in a transition region and the lift coefficient will be significantly affected as the turbine velocity is increased. Figure 3-5 and Figure 3-6 provide lift coefficient and lift/drag coefficient respectively vs. angle of attack for a NACA 634-02 1 foil at Re = 200 000 and Re = 500 000 [24]. These results were obtained using CFD software, as it is very difficult to find experimental data for such coefficients at the range of angles of attack needed for turbine analysis at the Reynolds numbers of interest. At Re 500 000, Cl / Cd may be 35% larger than for Re = 200 000, greatly affecting turbine performance. These effects are evident in Figure 3-7, demonstrating improved turbine 30 O< m I’ ) - I z CD CD CD - ;w 0 0 0 0 4 0 0 0 0 0 q O )k JW - - - J ( J — - C Y iL J C D - J I’ -3 -- O ) W U’ I 0 ) - J CJ i 0 ) W C ) m rn m m m m m m m + + + + + + - , - + + c D c D 4 .c Y i 3 1 U I 4 C 3 iC 3 1 - k ) W C D k )k ) 0 ) - ” W C IC D — - C 0 (J 1 C D - 3 ’ J C D C D C )0 )0 ) m m m m m m m m m + + + + + + + + + 0 )C y I0 ) 4 3 U i . C J 1 U 1 - - I (I) - C ) - l’ J C ) ( - N ) • C 0 0 )0 ) 0 ) N ) - - - C, J C 0 ) 0 ) m m m m m m m m m + + + + + + + + + 0 )0 )0 ) 0 )0 )0 ) 0 )0 )0 ) 0 )C JI O I 0 )0 )0 ) J U 1 C Y I C) CD o - C) C D D CD C)-t C) • • CD II . — C) D CD - CD CD . D C) — 0 CD . c j 0 I - CD - CI ) D 0 (M - CD 0 DD CD.-. C) . C D D Cl ) j 0 (D Cl ) . , 0 D Q CD -t N )C 3 — C .J U ) - N )N ) 0 )- N ) c o — o L J b O - ü )N J 0 )0 )N 3 - 0 ) - m m m m m m m m m + + + ÷ + + + + + 0 )0 )0 ) 0 )0 )0 ) 0 )0 )0 ) 0 )0 )0 ) 0 1 0 )0 ) (3 10 10 1 40 35 30 25 Re=50U 000 20 7/ Re=200000 15 p / 10• / 5 0 I I 0 5 10 15 20 25 30 35 40 45 Angle of Attack (deg) Figure 3-6: Cl! Cd vs. Angle of Attack for 634-021 at Re = 200 000, 500 000. 32 0.1 Upon removing the airfoils and testing the supporting arms to investigate parasitic drag, at all velocities the power coefficient as a function of TSR is quite consistent (Figure 3-8). This indicates the Reynolds number effects are having a more significant impact on the lift characteristics of the foil than on the drag characteristics of the supporting arms (supporting arm effects are discussed in more detail in Section 3.3.3, along with connections between arm and foil). The supporting arms operate at lower Reynolds numbers, primarily due to the majority of the arm length is at a shorter radius leading to lower velocities, and thus are further from the sensitive transition region. 0.08 0.06 0.04 C) 0.02 0 -0.02 -0.04 -0.06 TSR Figure 3-7: Power coefficient (Ck) vs. tip-speed ratio (TSR) at varying velocities. 33 0.00 0.5 1 1.5 2 2.5 3 3.5 -0.02 004 -0.06 -0.08 ____________________ 00 + Velocity 1 .0 -0.10 Velocity 1.25 - . D Velocity 1.5 -0.14 Velocity1.75 Velocity 2.0 -0.16 o Velocity 2.25 -0.18 -0.20 TSR Figure 3-8: Ck vs. TSR illustrating power loss due to parasitic drag from arm configuration A. 3.3.2 Drive-train Comparison It is important to compare similar turbine configurations using the two different drive- trains to ensure that the data from each program was reasonably similar, given turbine operating efficiency should be the same regardless of the drive-train used to drive or break the turbine; however, one may expect minor differences in the efficiency and torque curve plots, primarily due to the fact that in the chain/sprockets drive-train the torque sensor also served as a lay shaft and was not linked directly in-line with the turbine shaft as it was with the gearbox. Figure 3-9 provides the power coefficient vs. TSR for runs using the different drive-trains at both 1.5 m/s and 2 rn/s at the optimum operating TSR values of a free-stream turbine. The higher efficiency at 2 m/s, is attributed to Reynold’s number effects, as discussed in Section 3.3.1. 34 0.160 ::: 0.100 F 1 ./ .7 // 0.090 —a— Chains/sprockets_i .5m/s 0.060 -—Gearbox_1.5m/s -*- Chains/sprockets_2m/s 0.040 —-Cearbox 2m/s 0.020 0.000 1.50 1.75 2.00 2.25 2.50 2.75 3.00 TSR Figure 3-9: Ck vs. TSR drive-train comparison (medium profile arms). The efficiencies above show percent differences typically on the order of 10%, though less agreement is observed at 2 m/s and with a TSR of 2. Apart from measurement accuracies, differences in the curves may result from: • With the layshafi, power is transmitted through a chain and additional bearings before being registered by the torque sensor, so one may expect this drive system to have lower power, as is the case at higher TSR values, while flexing in the chain/sprocket system could also have an effect. • Fly-wheel effects of the sprockets about the torque sensor and flexing in the system may also serve to minimize the tendency of the chain/sprocket configuration to require/receive driving torque from the motor, thus artificially increasing the apparent efficiency. Figure 3-10 and Figure 3-11 illustrate the torque curves at the optimal TSR values (2.25, 2.5, 2.75) at 1.5 rn/s and 2 rn/s respectively. It is evident that the chains/sprockets drive- train configuration has lower, wider torque peaks observed by the torque sensor at both velocities due to flexing in the chains absorbing shock in the system, and inertial effects of the sprockets. Alternately, the flexible coupling used with the gearbox drive-train 35 allowed for a small amount of backlash, leading to the flattening of the curve observed as torque magnitude passes through zero. This backlash likely also produced a slamming effect once the coupling re-engaged, leading to sharper, higher peaks than what may actually be observed in an ideal system. Figure 3-10: Torque vs. Angle of Revolution comparing chains/sprockets with gearbox drive at TSR = 2.25,2.5, 2.75, v=1.5 rn/s. 36 Figure 3-11: Torque vs. Angle of Revolution comparing chains/sprockets with gearbox drive at TSR = 2.25, 2.5, 2.75, v2.O rn/s. Frequencies of torque input are also masked by the chains/sprockets drive-train. Table 3-3 provides the expected frequencies of torque ripple based on blade position, as well as the observed frequencies which were obtained by running a Fast Fourier Transform on the torque data for runs at 1.5 rn/s and TSR=2.5. Figure 3-12 provides the frequency content of these runs, and it is evident that the higher frequencies have a greater influence with the gearbox drive train. Table 3-3: Expected and observed torque frequencies for gearbox and chains/sprockets drive-train. Expected Experimental Frequencies (radisec) Primary Observed Run # Drive-train radlsec 1 puiselblade 2 puiseslblade 3 pulses/blade 4 puiseslbiade Frequencies from FFT Runi 045a Chains/sprockets 8.12 24.36 48.72 73.08 97.44 24.32 48.6 72.95 — RunOiG Gearbox 8.30 24.90 49.80 74.70 99.60 24.69 49.4 74.14 98.77 E z II, 0 I- Theta (degrees) 37 1 .0 L Figure 3-12: Torque data normalized frequency content for chaiiis/sprockets and gearbox drive- train (free-stream, 1.5 mIs, 2.5 TSR). Recognizing these differences in the drive-trains, it is reasonable to have confidence in the efficiencies obtained in using either drive-train; however, one must recognize that the chains/sprockets configuration masks the peak torque values. Alternately, the play in the flexible coupling of the second configuration leads to a bucketing of the torque curve, and potentially sharper, higher peaks due to impact in the coupling when it re-engages. It is reasonable to expect that the true torque curve in an ideal system would lie between the two, likely closer to the gearbox drive-train case. 3.3.3 Arm Profile Reduction Figure 3-13 illustrates the various arm profiles examined during the test programs. It is important to notice the clamping mechanism allowing for adjustable angle of attack used for profiles A and B. Upon removing the blades to examine the power absorbed by the arms, a large portion of the clamping mechanism was also removed, greatly reducing the 0.9 -e—Run 1045a Chains:: E.RunOi6Zarbox 0.61 >. ° 05 ci, 0.4 U 0.3 0.2 0.1 0.0 0 -0.1 Angular Frequency (radls) 38 parasitic drag compared to when the blade was mounted. The ends and middle connections used for profile C are also shown. Profile Cross-section (inches) Connection Type Arm profile A Quarter-chord 0.1 0 Arm profile B —1MH 0.38 Ends and middle Arm profile C (NACA 0012) 0.32 CZEZzE: A Figure 3-13: Arm profile cross-sections and connections. Figure 3-14 below provides Ck vs. TSR of the turbine model using the various arms illustrated in Figure 3-13 above. Efficiency significantly increases with each subsequent decrease in arm profile. The most significant jump comes when changing from arm profile B to C, even though configuration C has a third central arm. This is primarily due to reduced drag, but also due to the end-plate effect gained from mounting the arms at the ends of the blades, as well as the increased working span of the foil compared to the ¼ chord mounted configurations. The more stream-lined design of configuration C also performs better at a higher TSR, indicating the foil provides better performance at TSR 39 closer to 2.75 or 3, but the trade-off with parasitic drag from the bulkier arms lowers the optimal TSR ratio with configuration B. A further significant increase in performance is gained when removing the middle arm and running with the blade mounted using arms only at the ends. The Ck value of the two arm configuration decreases much more slowly at TSR values of 3 — 3.5, indicating better performance at a larger range of TSR values, which is beneficial for performance over a range of current speeds. The large difference in performance between 2 and 3 arms at TSR > 2.25 may be explained by the v2 dependency of arm drag having a larger relative impact at higher rotation speeds, and thus removal of the middle arm creates a significant drop in resistance. Additionally, the middle arm does not improve lift characteristics about the end of the foil as the end arms do, so its removal is purely reducing parasitic drag and not reducing lift generated by the foil. Figure 3-14: Ck vs. TSR for supporting arm comparison at 1.5 mIs. To facilitate comparison with theory, which typically ignores arm effects or requires an empirical formulation, the parasitic drag induced by the arms must be known. Figure 3-15 presents Ck vs. TSR of the various arm configurations when running the turbine 0.35 0.3 0.25 0.2 0.15 C., 0.1 0.05 0 -0.05 -0.1 TSR 40 model with the blades removed. Though this plot provides insight into what Ck losses are occurring due to the drag on the arms, simple subtraction of these Ck values from those in the plot above does not simulate an ideal case without parasitic drag for the following reasons: • ¼ span mounting of the foils reduces span of the blade working as an airfoil • Positioning the arms at the ends of the foil will affect tip losses • Upon removing the blades for these tests, bolt heads, etc. are also removed and thus in the assembled case parasitic drag will be larger. This was particularly the case for arm configurations A and B, where their mounting configuration incorporated a clamping mechanism about the arm, which added much drag but was removed with the blade (Figure 3-13 above). 0.5 1 1.5 2 2.5 3 3.5 . - •-.. •N N N •••. N N US\ \\ — —ArmA \ ---ArmB \ —.— Arm C (ends and middle) Figure 3-15: Ck vs. TSR of varying arm configurations (blades removed) at 1.5 mIs. 0 -0.02 -0.04 -0.06 -0.08 . -0.1 -0.12 -0.14 -0.16 -0.18 -0.2 TSR 41 Torque curves comparing arm profiles B and C (ends and middle) are provided in Figure 3-16 below. Arm profile C has significantly higher torque peaks transmitted to the shaft due to the reduced drag from the arms, though it is interesting to note that profile C demonstrates more negative torque readings at TSR = 2.25 and 2.5. Figure 3-17 provides the torque curves at 1.5 mIs comparing the three arms (profile C) for each blade vs. the case when just the end arms were supporting the blades at TSR = 2.75 and 3. As one might expect, the removal of the middle arm leads to significantly higher torque peaks (hollow data points), which is reflected in the increased Ck value in Figure 3-14. Figure 3-16: Torque vs. Angle of Revolution for arm profiles B and C (ends and middle) at 1.5 mIs and varying TSR. 42 The plots above (primarily Figure 3-14) demonstrate an improvement in turbine performance by a factor of four simply when going from arm profile A to C; Section 3.3.7 further examines the effect of tip losses on turbine performance. 3.3.4 Single-blade Figure 3-18 below provides Ck vs. TSR for a 3-bladed test (arm configuration C), and two single-bladed tests (arm configurations B and C) at 1.5 m/s using the gearbox drive- train. It is apparent that interference and flow disruption play a significant role in reducing the power output of the 3-bladed configuration. At the highest Ck value for the 3-bladed test (TSR = 2.5), the single blade efficiency is 55.5% that of the 3-bladed design. Beyond this TSR value, the 3-bladed efficiency drops, while the single bladed efficiency continues to climb. E z 60 0 I- Theta (degrees) Figure 3-17: Torque vs. Angle of Revolution for 3 arms and end arms only at TSR2.75, 3 and v1.5 rn/s. 43 0.300 1 Blade. Arm Config C - 0.250 -*— 1 Blade, Arm Config B / —— 3 Blades, Arm Config c 0200 • /•• 0.150 ,1 / 0.100 0 0.050 0.000 I I I I 0. 10 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4. 0 -0.050 TSR Figure 3-18: Ck vs. TSR for single and 3-bladed tests at 1.5 mIs. Figure 3-19 illustrates the torque output of the single-blade test over a revolution at 1.5 rn/s for TSR = 2.5, 3, 3.5. The double peak at the primary torque-producing region (near 900) is believed to be caused by flow separation on the blade. When the flow separates due to the large angle of attack induced by the free-stream flow, the drag increases and the turbine produces less torque until the flow re-attaches. Meanwhile, near 270°, a double peak in torque creation due to a loss in lift caused by vortices shed by the shaft may be observed. 44 Figure 3-19: Torque vs. Angle of Revolution at 1.5 mIs for a single blade test. Figure 3-20 superimposes three sets of torque data (TSR = 2.5 at 1.5 m/s) from the single-blade tests phased at 120° and compares them to the 3-bladed experimental test. The 3-bladed experimental result varies greatly from the superimposed single-blade result, and this is likely due to a combination of a number of factors: • Interference and vortex shedding disrupts the flow at the downstream blades, reducing ability to cleanly create lift • The additional power being extracted by the multiple blades changes pressure distribution at the front of the turbine, affecting the amount of torque available for extraction at the 90° position • The phase shift observed between the peaks is believed to be caused by the fluctuating turbine revolution speed due to the larger forces involved (discussed further in Section 4.1.3) E z 0 0 F- Theta (degrees) 45 Lastly, an interesting result was obtained when comparing tests done using arm profiles B and C, as shown in Figure 3-21 at 1.5 m/s with a TSR = 3 (both using the gearbox drive- train). Similar shaft interference is obtained in the vicinity of 2700, though surprisingly the larger arm profile (B) shows higher torque values. Meanwhile, across the 90° position, a near opposite torque profile is created. An explanation for this is that across 900 the added drag from the arms and clamping mechanism, as well as tip losses, reduce the torque generated, while just before 270° vortex interactions with the anns or clamping mechanism may be acting on the arms to enhance performance. Lastly, the dual peak observed in the 90° position with the lower profile arm is likely due to flow separation on the blade. With profile B, it is hypothesized that the large clamping mechanism at the ¼ chord locations, as well as tip losses, result in a much less pure observation of flow separation characteristics and instead yield a single pulse. Similar results were obtained at different TSR values. Figure 3-20: Torque vs. Angle of Revolution at 1.5 mIs for a 3-blade test, single-blade test, and 3 superimposed single-blade tests. 46 3.3.5 AngIe of Attack Blade set angles of attack of-5, -3, 0, 3, 5, and 10 degrees were tested throughout the test programs. 0, 3, 5 and -3 degrees provided the most insightful results and are discussed below. -3 degrees reduced turbine performance by almost 50%, while -5 and 10 degrees were highly ineffective. Figure 3-22 presents Ck vs. TSR at 2.0 mIs for AoA 0, 3, and 5 degrees. It is interesting to note that at TSR < approximately 2.35, an AoA = 3 yielded the best performance, while at TSR > approximately 2.35 an AoA = 5 provided better performance. In the vicinity of 90° of the turbine revolution (ie. directly upstream of the shaft), having a positive preset angle of attack decreases the angle of attack observed by the blade; meanwhile, in the vicinity of 270° (directly downstream of the shaft), a preset angle of attack on the blade increases the observed angle of attack on top of that caused by the free-stream flow. These are the most significant angles of attack experienced by Figure 3-21: Torque vs. Angle of Revolution for a single-blade test with arm profiles B and C at TSR=3, v=1.5 rn/s. 47 the blades Table 3-5. (not accounting for vortex interactions) and are provided in Table 3-4 and Table 3-4: revolution. Blade angle of attack at varying TSR and preset angle values at the 900 angle of Angle of Attack (degrees) at 90° Position TSR Flow-induced Angle Net Angle (3° preset) Net Angle 5° preset) 2 26.6 23.6 21.6 2.25 24.0 21.0 19.0 2.5 21.8 18.8 16.8 2.75 20.0 17.0 15.0 TSR Flow-induced Angle Net Angle (3° preset) Net Angle 5° preset) 2 26.6 29.6 31.6 2.25 24.0 27.0 29.0 2.5 21.8 24.8 26.8 2.75 20.0 23.0 25.0 Table 3-5: revolution. Blade aagle of attack at varying TSR and preset angle values at the 270° angle of Angle of Attack (degrees) at270° Position 48 018 1 4/ —*- A0A = 00.1 f I—A0A=3 1/ L..AoA=5 0.08 J/ / 006. 0.04 0.02 / 0• I I I I 0 0.5 1 1.5 2 2.5 3 3.5 TSR Figure 3-22: Ck vs. TSR for AoA =0,3,5 degrees at 2 mIs. Figure 3-23 illustrates the torque generated at a TSR = 2.25, which is generating higher peaks than at TSR = 2.5 (Figure 3-25). This is because larger angles are being experienced at 900 at the lower TSR value, generating more lift. A contributing factor to this is the dynamic stall effect, which tends to delay stall [25] that typically occurs near AoA = 8° for the 634-02 1 airfoil at these Reynolds numbers (2.92E05 to 3.82E05 for 2 m/s). Also at TSR = 2.25, the 5° angle of attack generates larger peaks due to reduced stall upstream of the turbine, while it also creates similar, or slightly worse, low torque values downstream of the shaft due to an increased tendency to stall. At TSR = 2.5, the peak values in general are lower, though the turbine performance is better. This is due to the fact that the low-points in the torque curve are higher than at TSR = 2.25. This is caused by less stalling around the back of the turbine since the angle induced by the free-stream flow is smaller at the higher TSR value. Comparing the 3° and 5° preset angles of attack, 5° is creating substantially higher peaks due to reduced stall upstream of the turbine. Downstream of the shaft, both angles create similar 49 negative torque peaks. One should note that this is a simplified assessment of the situation, as dynamic stall and vortex shedding onto the downstream blades play a significant role; however, flow visualization capturing these phenomena is extremely difficult. A key conclusion from this examination is that optimal angles of attack likely lie in the vicinity of 2 to 5 degrees, and are dependent on operating tip-speed ratio. Polar plots (Figure 3-24 and Figure 3-26) are also provided to aid with visualization. Figure 3-23: Torque vs. Revolution Angle for AoA =0,3, 5 deg at 2 mIs, TSR = 2.25. 70 60 50 j—.——Run1084--A0A=0 I —4——-Tun1284—AoA3 [—Run1384_AoA=5 — 40 E z e 30 0 i—. 20 10 0 -10 I I I I I 0 50 100 150 200 250 300 350 400 Theta (degrees) 50 90 120 ._— T—... 60 ——RunlOS4--AOA=O .Lec / N Run12--AoA3/ 7 Run1384-.AoAS / // 207A’. 9’/ A 180 I-- “---- -----“- --—--- -“-. “-. -.-. ‘---“. ..-. .-.1 0 210 / 339 /N: 240 ..__.__r 270 Figure 3-24: Polar Plot of Torque vs. Revolution Angle for AoA =0,3,5 deg at 2 mIs, TSR = 2.25. ti_i I I I I I I I I I—.-—Run1085AoA=0 A —‘—Run1285--AOA=3 60 1 4r [z—Run1385AoAr5I II f 44 50 1/ (/, \: E40 47/ 14 1+ 30 20 UI 10 it -1 -_-‘ I I I I I I I -50 0 50 100 150 200 250 300 350 400 Theta (degrees) Figure 3-25: Torque vs. Revolution Angle for AoA =0,3, 5 deg at 2 mIs, TSR = 2.5. 51 go Lastly, Figure 3-27 illustrates the torque curves generated with a preset AoA -3 and 0 deg at 1.75 mIs. The reduced torque peak at AoA = -3 indicates that this angle is significant enough to increase the angle of attack observed by the blade past its stall point, reducing its ability to produce torque in the vicinity of 90°. The more negative lows in the torque curve indicate that this was also effective at reducing torque generated in the vicinity of 270° by reducing the observed angle of attack. 180 Figure 3-26: Polar Plot of Torque vs. Revolution Angle for AoA = 0, 3,5 deg at 2 mIs, TSR 2.5. 52 50 I I I I I I I I Figure 3-27: Torque vs. Revolution Angle for AoA = -3 deg at 1.75 mIs, TSR = 2.5. 3.3.6 Cambered Blades Investigation using cambered blades was performed using a cambered version (634-421) of the symmetric blade tested above. Power coefficient vs. TSR at 1.5 mIs is plotted in Figure 3-28 for the cambered blade at AoA = 00 and 5°, as well as for the symmetric blade case. It is apparent that the cambered blade offers a substantial increase in efficiency, especially at 5°. 4 I — Run1485 — A0A -3 —--—Run1065_AoA=0 40 30 E 20 z 4, 10 0 -10 0 50 100 150 200 250 300 350 400 Theta (degrees) I I I I I I I 53 Examining the torque curves, Figure 3-29 provides torque vs. angle of revolution for the symmetric blade (AoA = 0) and the cambered blade (A0A = 0 and 5) for the optimum TSR = 2.75 at 1.5 mIs. As expected, the symmetric blade produces higher peaks as a blade passes near the 90°, because the cambered blade is effectively flying upside down. However, the cambered blade at 5° is effective in reducing this upside down angle of attack, and produces greater torque than the 0° cambered blade case. Additionally, as the cambered blade passes downstream of the turbine near 270°, the cambered blade is better suited to producing lift in this location, increasing the minimum torque values observed. This also appears to produce torque over a greater range, as indicated by the wider peaks, leading to improved turbine performance. Figure 3-28: Ck vs. TSR for cambered (0 and 5 deg) and symmetric (0 deg) blades at 1.5 rn/s. 54 3.3.7 Blade End Plates Proof of concept tests were performed using end plates on the blades to examine the possibility of reducing tip losses when supporting arms were mounted at the ¼ chord positions. Riley examined the use of end plates [261 and demonstrated that end plates with a foil-shaped cross-section were advantageous. Therefore, rectangular end plates with length equal to the chord and width of 1.5” with a NACA 0012 cross-section profile were applied, as suggested by Klaptocz [27]. Additionally, disc shaped end plates [0.25” thick] with a rounded edge and diameter equal to the foil chord were also tested given the circular path the turbine blade travels. Figure 3-30 displays the NACA 0012 (with flattened edge to sit flush on the foil) end plate and the circular end plate mounted to the blade. Figure 3-29: Torque vs. Angle of Revolution for symmetric (0 deg) and cambered (0 and 5 deg) at 1.5 rn/s and TSR = 2.75. 55 Figure 3-31 and Figure 3-32 below provide Ck vs. TSR for the end plate configurations compared to the case without end plates at 1.5 mIs and 2 m/s respectively. At both speeds the NACA 0012 end plates provided the best results, increasing the Ck value by 16% (at 1.5 mIs) and 12% (at 2 mIs). The circular end plates also demonstrated an improvement over the case without. Figure 3-30: NACA 0012 profile and circular end plates. 56 0.160 0.140 0.120 0.100 0.080 0.060 0.040 0.020 0.000 0. -0.020 -0.040 C) 0.020 0.000 0.00 0.50 iN —n--No end plates 7/ t NACA 0012 end plates -*— Circular end plates v/i / I 0 0.50 1.00 1.50 2.00 2.50 3.00 3. ‘0 TSR Figure 3-31: Ck vs. TSR for end plate comparison at 1.5 mIs. 0.180 0.160 0.140 0.120 0.100 0.080 0.060 0.040 c No end plates NACA0012endplates if — Circular end plates I/1/,/ 7/ •// V V Figure 3-32: Ck vs. TSR for end plate comparison at 2 mIs. 1.00 1.50 2.00 2.50 3.00 TSR 57 Considering the torque curves, at 1.5 m/s (Figure 3-33) the NACA 0012 appears to produce increased torque peaks, while the circular end plates produces smaller and slightly wider torque peaks which rarely enters a negative torque region. Conversely, at 2 rn/s (Figure 3-34) the NACA 0012 end plates produce lower, wider torque curves while the circular end plates produce higher torque peaks. Lastly, it is possible to create thinner disc end plates, which would reduce associated drag and improve performance. E z a, 0 I- Theta (degrees) Figure 3-33: Torque vs. Revolution Angle comparing end plates at 1.5 mIs. 58 3.3.8 Shaft Fairing Given the interference observed in the single blade tests, fairings were fabricated and placed around the shaft as an attempt to minimize the shaft vortices (Figure 3-35). Figure 3-36 below provides Ck vs. TSR for runs with and without the shaft fairing at 1.5 and 2 rn/s. Tests were conducted using arm configuration C, and a fairing was placed each between the upper/middle arms and the middle/lower arms. For both speeds, the fairings either reduced performance or had negligible effect. Figure 3-34: Torque vs. Revolution Angle comparing end plates at 2 m/s. 59 Figure 3-37 displays torque curves for the cases with and without shaft fairing for a TSR 2.75 at 1.5 mIs. The fairing reduces the torque peaks, as well as shifts the peaks approximately 12 degrees to the left, or earlier in the rotation. A similar effect was observed at 2 mIs. Friction is the likely cause of the reduced torque peak, while different vortex interactions, as well as the reduced torque peaks resulting in less revolution speed fluctuation, are the most reasonable explanation for the phase shift in torque curve (revolution speed fluctuations discussed below in Section 4.1.3). Figure 3-35: Shaft fairings. 60 0.300 0.250 II / 0200 / \\- Ao 0. i5u 7 IAoA=0(1.5s) __________________________________________I-Awith shaft fairing (1.5 mis) 0100 /1 AoA=0 (2 mis) I— with shaftfairing (2 m/s)• _ 0.050 0.000 I I I 1.00 1.50 2.00 2.50 3.00 3.50 4.00 TSR Figure 3-36: Ck vs. TSR with and without shaft fairing (1.5 and 2 m/s). I—Run225--withFairing I I I ____ Run205.1 — without Fairing __100 E 80 II II z 60 0 40 20 0 50 100 150 200 250 300 350 400 Theta (degrees) Figure 3-37: Torque vs. Revolution Angle with and without shaft fairing at 1.5 mis, TSR=2.75. 61 Tests were also conducted for a single blade with the shaft fairing, as shown in Figure 3-38. Figure 3-39 provides torque vs. revolution angle with and without the shaft fairing. The fairing appears to smooth out the torque curve downstream of the turbine near 2700 as one might expect, though the general effect of the fairing was to reduce the average torque by 4 % (Ck = 0.15 1 without the shaft fairing for a single blade vs. 0.145 for the case with the shaft fairing). This reduction in power is likely caused by additional friction between the fairing and shaft, as well as an increase in frontal area of the shaft increasing the effective blockage of the turbine and causing more flow to pass around. Figure 3-38: Single blade with installed shaft fairing. 62 Figure 3-39: Single Blade Torque vs. Revolution Angle with and without shaft fafring at 1.5 mIs, TSR=2.75. 3.3.9 Summary Considering the data above, it becomes possible to summarize the improvements to be gained from each parameter by comparing to its baseline configuration. NACA 0012 end plates were shown to increase the baseline Ck value by 12.2% and 16.6% at 1.5 mIs and 2 rn/s respectively, though in general the contribution of tip losses to overall device performance will reduce with increasing aspect ratio. Angle of attack provided a notable improvement over the baseline case of 0° (tested with arm profile B), as at 1.5 m/s 3° and 5° increased the Ck value by 21.1% and 14.8% respectively, while at 2 m/s 3° and 5° increased the Ck value by 17.3% and 2 1.6% respectively. Table 3-6 below summarizes the incremental improvements achieved over the 3-armed baseline (profile C) for the following cases: 2 arms at the ends only, cambered blades at 0° and 5°, and shaft fairing application. 2 z 0 0 0 I- Theta (degrees) 63 Table 3-6: Maximum Ck and percent increase over free-stream baseline. Case Maximum Ck % change 3 arms (baseline) 0.272 -- 2 arms 0.303 11.4% Cambered blade J° AoA) 0.285 4.8% Cambered blade (50 A0A) 0.319 17.3% Shaft fairing 0.255 -6.3% Using this data, it is possible to hypothesize the maximum efficiency of a free-stream, 3- bladed rotor. As moving from 3 to 2 arms yielded an increase in Ck of approximately 0.031, it seems reasonable that in the absence of all arms, the Ck may increase by an additional 0.062; however, one must recognize that removing end arms will also allow for tip losses (Ck approximately 0.02 at this aspect ratio). Assuming tip losses may be eliminated by some other hypothetical means, the maximum efficiency of this device would be approximately 0.3 65. The shift to cambered blades at 5° further increased Ck by a value of 0.047, bringing our theoretical maximum Ck value, without arms or tip losses using the cambered blade at 50, to 0.412. Two other major components affecting rotor design and not examined as part of this thesis are solidity and foil shape. Recognizing these, a rotor with Ck of 0.45 in the absence of all losses seems to be a reasonable theoretical maximum after using numerical codes or an extensive experimental program to pin-point optimum solidity and foil shape/angle of attack. 3.4 Ducted Turbine Figure 3-40 provides a dimensioned plan view of the venturi-type ducting installed around the turbine, while Figure 3-41 illustrates the ducting position within the tank. A top and bottom was installed as shown in Figure 3-4 1, and a large Plexiglas window was installed to allow for removal of the turbine while leaving the ducting in place, as well as to facilitate visualization. The duct shape was determined based on previous NRC trials [11] as well as what was suitable for the current experimental setup. Results for the venturi-type ducting (Section 3.4.1) and ducting with flow deflectors (Section 3.4.2) are discussed below. 64 I- 75.1 12.1 40.4 \ / Figure 3-40: Plan view of ducting (inches). 65 Water ‘1 depth = 84” External duct width=64.6” Tank width = 144” 3.4.1 Venturi-type Ducting Figure 3-42 provides Ck vs. TSR for the free-stream and ducted turbine at 1.5 mIs. The ducted turbine greatly enhances power output from the turbine, though Ck is still calculated based on the turbine area, and not the duct frontal area affecting the flow. Secondly, TSR is calculated relative to the free-stream velocity, and not relative to the accelerated velocity through the duct, explaining why the highest Ck value is occurring at a higher TSR value for the ducted case. ernal width4O.4” I... Figure 3-41: Cross-section of towing tank with ducting and turbine. 66 Figure 3-42: Ck vs. TSR for the free-stream and ducted turbine at 1. 5 mIs. It is interesting to compare the power harnessed by the ducted turbine vs. the power that would be extracted by a free-stream turbine of capture area equivalent to the duct (approximately 32.5” x 63.1”) operating at the Ck values obtained in previous tests. This is provided in Figure 3-43 which indicates the ducted turbine captured a peak of 501 W, while a free-stream turbine of equivalent capture area may be expected to harness 560 W, not accounting for Reynolds’ effects. Therefore, the ducted configuration tested was approximately 12% less efficient than an equivalent-sized free-stream turbine, having a peak Ck value based on the capture area of 0.239, vs. 0.272 for the free-stream device. A free-stream device of equivalent size to the ducted device tested may be capable of generating more power due to more flow passing through the device given there is less blockage in the absence of a duct, as well as the increased diameter and blade size of the free-stream device is capable of producing larger torque forces on the shaft. 0.50 0.45 0.40 0.35 0 30 . o 0.25 0.20 0.15 0.10 0.05 0.00 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 TSR 67 600 Figure 3-43: Extracted Power (W) vs. TSR for the tested ducted turbine and a free-stream turbine of equivalent capture area at 1.5 m/s. Considering torque curves for the free-stream vs. the ducted turbine, Figure 3-44 illustrates the torque curves for the free-stream turbine at 1.5 mIs, while Figure 3-45 provides torque curves for the ducted device at 1.5 mIs. The most significant (and surprising) result is the decrease in amplitude of the torque curve for the ducted configuration once a TSR of 2.75 or greater is reached. A similar decrease in torque ripple was observed in the 2 m/s ducted tests beginning at TSR = 2.5. It is convenient to define a torque fluctuation coefficient calculated as follows from values of the torque curve: CTF = Tm — Tmj Equation 6 avg where: = maximum torque Tmjn minimum torque Tavg = average torque 500 400 300 200 100 0 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 TSR 68 CTF facilitates comparison of torque curve fluctuations, which are a key parameter in the mechanical design of the device as reduced fluctuations may greatly enhance both reliability and operation life of the device. z 0 0 0 I— Theta (degrees) Figure 3-44: Torque vs. Revolution Angle for free-stream turbine at 1.5 mIs. 69 Table 3-7 below tabulates torque fluctuation coefficient for both a free-stream and ducted turbine in the runs shown above. Table 3-7: Torque fluctuation coefficient for a free-stream and ducted turbine. CTF TSR Value Free-stream Ducted Percent Change 2.25 6.48 9.54 47.2% 2.5 5.44 5.71 5.0% 2.75 4.24 1.45 -65.8% 3 3.8 1.25 -67.1% 3.5 5.4 1.26 -76.7% This decrease in CTF is primarily due to the duct constraining the flow and not allowing it to expand and slow in way of the downstream blade, thus increasing the available power; altered vortex interactions compared to the free-stream case may also be increasing performance of the downstream blade though flow visualization would be required to be certain. Lastly, tests were also conducted with the ducted configuration and the shaft fairing. As for the free-stream result, a slight decrease in performance was observed for all runs, except for TSR=2.75 at 2 mIs which showed a 6% increase in performance. This E z 0 0 I- Figure 3-45: Torque vs. Revolution Angle for ducted turbine at 1.5 m/s. 70 point is believed to be an outlier, but may warrant future investigation should the device be re-examined. 3.4.2 Ducting with Deflectors In place of testing a large variety of duct shapes which are both expensive and laborious to construct, 4 deflectors were fabricated to be placed at various locations within the duct to adjust the flow. Figure 3-46 below illustrates deflector positioning and size, with additional details in Appendix B. The configurations tested were as follows: • All four deflectors • Blades spinning towards (deflectors 1 & 3) • Blades spinning away (deflectors 2 & 4) • Downstream (deflectors 1 & 2) • Upstream (deflectors 1 & 2 while running in opposite direction; equivalent to 3 & 4 with flow direction as shown on diagram) The rationale behind the use of the deflectors was to reduce the cross-sectional area, and thus increase the speed and available power, in the blade positions where the turbine is generating the most torque (900 and 270°). Additionally, deflectors were offset from the ducting to allow for flow to pass in-between, limiting separation that may occur behind the deflector. This design was developed by Yasser Nabavi and Voytek Klaptocz [281. 71 Figure 3-46: Ducting with deflectors. Figure 3-47 below provides Ck vs. TSR for the various deflector configurations, as well as for the plain venturi-type duct. The configuration without deflectors produced the highest Ck values, and this is likely due to the deflectors reducing the flux through the ducting assembly and thus reducing the available power to be extracted by the rotor. The 4-deflector and upstream deflector designs appear to be the least efficient, likely due to increasing resistance to the flow before the rotor, while the downstream deflectors as well as 1+3 and 2+4 yield similar peak Ck values. Blade /2 rotation / Flow into turbine 72 Figure 3-47: Ck vs. TSR for duct and deflector configurations. The primary significance of the deflector designs is observed when examining the torque curves of the various configurations. Maximum Ck values were observed at TSR values of 2.75 and 3, and Figure 3-48 and Figure 3-49 provides torque curves for the various configurations at TSR = 3. The downstream deflectors (solid dashes) greatly reduce the torque fluctuations observed, believed to be due to higher torques at the downstream blade caused the smaller cross-sectional area and resulting higher flow velocities. Conversely, the deflectors upstream of the turbine appear to cause much greater torque fluctuations due to the increased velocity passing past the blade upstream of the turbine, which is already producing the majority of the torque. 0.500 0.450 0.400 0.350 0.300 0.250 0.200 0.150 0.100 0.050 0.000 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 TSR 73 Figure 3-48: Torque vs. Angle of Revolution for ducted and deflector configurations. Figure 3-49: Polar plot of Torque vs. Angle of Revolution for ducted configurations. E z 0 I- Theta (degrees) 90 — Run204.1 — Free-stream —a—— Run 485— Ducted no deflectors —÷-— Run407.1 —4 deflectors —a--— Run 445— Spinning towards Run 525— Downstream deflectors 180 270 74 Table 3-8 below provides the maximum Ck values and corresponding CTF for the various ducted configurations examined, as well as for the free-stream case. The downstream deflectors offer a 62% reduction in the torque ripple experienced by the shaft over the case without deflectors. This is considered due to the reduced cross-sectional area in way of the deflectors at the downstream positions of the blades, which increases flow velocity and thus lift extracted in this position, resulting in a torque generation more comparable to the 900 position upstream of the shaft. Lastly, in the figure above it should be noted that reduced torque fluctuations resulted in reduced revolution speed fluctuations (Section 4.1 .3)shifting the peaks back closer to their theoretical position near 90°. Table 3-8: Maximum Ck and corresponding CTF for ducted turbine configurations (1.5 mIs). Case - Ck Value % Ck Change CTF % C Change No deflectors 0.473 — 1.25 -- Downstream deflectors 0.442 -6.6% 0.47 -62.4% All four deflectors 0.393 -16.9% 1.4 12.0% Spinning towards deflectors 0.426 -9.9% 1.17 -6.4% Spinning away from deflectors 0.442 -6.6% 1 .23 -1 .6% Upstream deflectors 0.407 -14.0% 2.61 113.6% 3.4.3 Summary As for the free-stream case, it is possible to quantify the effect of the various ducting configurations compared to the baseline free-stream case. Table 3-9 below provides maximum Ck value, Ck percentage increase over the free-stream baseline, and coefficient of torque fluctuation. Table 3-9: Maximum Ck, percent change, and torque fluctuation coefficient Case - Ck Value % Ck Change C % CT Change Free stream (baseline) 0.272 -- 4.24 -- No deflectors 0.473 73.9% 1.25 -70.5% Downstream deflectors 0.442 62.5% 0.47 -88.9% Al four deflectors 0.393 44.5% 1 .4 -67.0% Spinning towards deflectors 0.426 56.6% 1.11 -72.4% Spinning away from deflectors 0.442 62.5% 1.23 -71.0% Upstream deflectors 0.407 49.6% 2.67 -37.0% 75 As expected, ducting around the rotor increases power output; however, the power obtained from the ducting design tested is less than what may be expected from a free- stream turbine of equivalent cross-sectional area. Recognizing this, ducting (especially with modifications such as the downstream deflectors) is demonstrated to greatly reduce torque ripple. Additional potential benefits such as structural support for the bottom bearing and to facilitate mooring render ducting a prospective enhancement to a turbine design requiring a comprehensive cost-benefit analysis. 3.5 Drag Force No previous documentation has been found on the forces parallel to the free stream flow acting on the turbine rotor, and subsequently the shaft bearings. These forces are a combination of drag forces on the shaft and supporting arms, as well as the component of the lift and drag forces on the turbine blades acting parallel to the free-stream flow. For this thesis, the combination of forces parallel to the free-stream flow will be referred to collectively as drag forces. A means of approximating the drag force on the turbine was devised by measuring the force at the top bearing using the force balance, estimating the centre of action of the drag forces, and balancing moments about the bottom (self-aligning) bearing to solve for the magnitude of the drag force. Figure 3-50 below illustrates the location of the assumed and measured forces. Analytical calculations demonstrated that the blades and arms may be expected to account for approximately 83-93% of the forces parallel to the free-stream flow, while the shaft and arms account for the remaining forces. Given the centre of the blades and arms is 21.5” above the bottom bearing and the centre of the shaft is 26” above the bottom bearing, this results in an assumed centre of force about 22” above the bottom bearing to within approximately +1- 15%. The broad range is due to the simplified analytical calculations as well as the dynamic nature of the system, but is 76 sufficient for this preliminary investigation. With the top bearing 68” above the lower bearing and the force balance measuring the load parallel to the free stream on this top bearing, it is possible to use moment calculations and determine that the drag load at the turbine is 68/22 times the in-line load measured at the force balance. Figure 3-50: Figure 3-51 provides drag force of a free-stream turbine vs. TSR for angles of attack at 00 and 30 for TSR values between 1.5 and 3 at velocities between lmJs and 2mJs. Using this measured drag force (D), it is possible to calculate a drag coefficient (Cd) for the turbine as follows: Cd= D ).p.v2.A Equation 7 Drag coefficient vs. TSR for these same trials is provided in Figure 3-52. The data for velocities of 1.5, 1.75, and 2 m/s collapses reasonably close together, while the data for lmIs yields slightly higher drag coefficients. As these drag forces are a combination of Flow Top shaft bearing 68” V I I Centre of Side view providing location of assumed centre of drag force. 77 resistance on the shaft and arms, as well as components of lift and drag on the foil parallel to the flow, Reynolds effects will be present and it is apparent that at the lower Reynolds numbers in the lmIs tests the result is increased relative drag forces on the device. A linear trend line fit through the combined 1.5, 1.75, and 2mIs data points yields an equation with slope of 0.41 and y-intercept = -0.16 (R2 0.91). This enables a rough approximation for the drag coefficient of the tested device at varying TSR values over this range of Reynolds numbers. One must exercise caution if attempting to extrapolate these results directly to other vertical axis turbines of different solidities, or proportionally larger shaft and arm sizes, as all of these will affect the magnitude of the drag forces generated. Figure 3-51: Drag Force vs. TSR for a free-stream turbine at varying velocity. 1800 1600 1400 1200 1 1000 I- ci 600 400 200 0 0 0.5 1 1.5 2 2.5 3 TSR ‘:3— —u— 1 mIs, AoA=0 1 mIs, AOA=3 —- 1.5 mIs, AOA=0 1.5 mis, AoA=3 —+- 1.75 mis, AoA=0 —.-- 1.75 mis, AoA=3 —s-- 2 mis, AoA=0 2 mis, AoA=3 3.5 78 1.4 Figure 3-52: Drag Coefficient vs. TSR with trend line for data at v=1.5, 1.75, 2mIs. As for the torque curves, it is possible to plot drag data as a function of revolution angle. Figure 3-53 provides drag force vs. revolution angle at the TSR values for which optimal power is typically being generated. The most drag is being produced in the vicinity of 900 as one might expect, since this is where peak torque is typically being generated, and a large component of the lift generating this torque is in the free-stream direction, resulting in drag on the device. Of note are the smaller peaks for the TSR=2 and TSR=2.25 cases, which occur at frequencies of approximately 57.8 rad/s and 64.1 rad/s respectively as determined by performing a Fast Fourier Transform (FFT) on the data set within the analysis software. This occurrence is discussed further after examining the single-blade case below. Figure 3-54 provides drag force vs. revolution angle at 2 mIs for TSR2, 2.25, 2.5, 2.75. It is apparent that these high frequency oscillations have disappeared, and clean drag curves are obtained with peaks near the 90° position. 0 0— 4 1.2 y0.41x0.16 R’=O.91 C) —--1m/s,AoA=0 0.8 . . —.— 1 mis, AOA=3 —&- 1.5 mis, AoA=0 6 ________________________ —A— 1.5 mis, AoA=3 -— 1.75 mis, AoA=0 —.-- 1.75 mis, AOA=3 0.4 -8--2mis,A0A=0 2 m/s, AoA=3 0.2 0 I I I I I I 0 0.5 1 1.5 2 2.5 3 3.5 TSR / 79 4 •Thnn uu An . aJd ‘1 k ? : -0 50 100 150 200 250 300 30 41 -“nfl I LIlt) 1000 snn .\ inz a,C, 0 ‘J O) CU a fl\. in’ —e—Run 1043 2.0 TSR —Run 1044 2.25 TSR —h--Run 1045 2.5 TSR —- Run 1046 2.75 TSR lic Revolution Angle (deg) Figure 3-53: Drag Force vs. Revolution Angle at 1.5 mIs, AoA=O. 100 150 200 250 RevolutIon Angle (deg) Figure 3-54: Drag Force vs. Revolution Angle at 2 mIs, AoA=O. 80 Considering the experimental tests with only a single blade attached to the shaft, Figure 3-55 provides drag coefficient vs. TSR for both the single and 3-bladed case at 1.5 m/s with AoA=3. A single-blade device has approximately 2/3 of the drag coefficient of a 3- bladed turbine. Examining single-blade drag vs. revolution angle (Figure 3-56), drag is again being generated in the 90° and 270° regions, as is torque. The high-frequency oscillations, however, are apparent at TSR values of 2, 2.25, and 2.5, and are less apparent at TSR=2.75. Figure 3-55: Drag Coefficient vs. TSR for a single and 3-bladed device at 1.SmIs, AoA=3. 81 iIuu loop nn u: Figure 3-56: Drag Force vs. Revolution Aiigle for a single blade at 2 mIs, AoA=3. Table 3-10 provides expected frequencies based on the turbine revolution speed for both one primary pulse (ie. at 90°) and two primary pulses (ie. at 90° and 270°) per blade per revolution. Approximate observed frequencies obtained from a FFT on the recorded data are also provided. The observed frequencies are where one may expect based on the turbine blade frequency and the two pulses per blade; however, there is the unique frequency at about 57-63 radJs that isn’t easily explicable by blade pulsing, and that disappears at higher drag forces. Additionally, these oscillations appear with both the single-bladed and 3-bladed device at about the same frequency, making it unlikely that this is due to arm forces or flow around the foils separating and then re-attaching. If that was the case, this frequency near 57-63 Hz would appear at much different values when comparing the 3-bladed and single-blade device. Considering this, it is reasonable to conclude that the oscillation was at a natural frequency of the force balance / load cell configuration. At higher velocities and drag forces, this oscillation has disappeared, hinting to the fact that these higher loads were perhaps capable of dampening the motion at the force balance. Lastly, the vortex shedding frequency on the shaft was predicted to z 41 (3 0 U ::: —--Run 1643 2.0 TSR ât’’k 1ERU 2:5 TSR . AM” \j 50 100 150 200 250 300 350 400 Revolution Angle (deg) 82 be approximately 345 radls, while the natural frequency for the shaft was predicted to be 622 radls, both of which are too high to be responsible for the oscillations discussed here. Table 3-10: Expected and observed experimental drag force frequencies. Expected Experimental Frequencies (radls) Radians I sec Blade Frequency J Expected Frequency Expected Frequency Primary Observed(2 pulses per blade) (3 pulses per blade) Fre encles (raWs) 2.0 TSR 6.54 1961 J. 3923 58.84 18.28 39.65 57.93 3 Blades 2.25 TSR 7.21 21.62 43.25 6487 2130 42.60 63.90 2.5 TSR 8.11 24.32 4863 72.95 24.32 48.63 63.84 2.0 TSR 6.91 6.91 13.82 20.73 6.47 12.88 57.99 Single Blade 2.25 TSR 7.64 7.64 15.28 22.92 — 15.77 2.5 TSR 8.37 8.37 16.75 25.12 — 15.58 59.31 3.5.1 Summary The drag force measurements above provide an insightful first look into the magnitude of drag forces that may be expected on a vertical axis hydro turbine. The high frequency oscillations (57-63 Hz) appear to dampen out at higher velocities and drag loads, indicating that they are likely caused by a natural frequency in the flexibility of the load cell / force balance system. Lastly, the equation approximating Cd [Cd = O.41*tsr — 0.161 only accounts for forces parallel to the free-stream flow, and much further work is required to understand the interaction between parasitic drag forces, lift/drag forces acting on the turbine blades, and the net forces observed by the bearings, which are likely to have a variable direction during turbine revolution. 83 4 DISCUSSION Below, measurement errors and repeatability are discussed, followed by a comparison with the numerical predictions and a general discussion on sources of error. 4=1 Measurement Accuracy Typically, when considering measurement accuracy and error, one must consider both systematic error and random error. Random error is the experimental error that occurs given no two runs will yield exactly the same result due to random variation in the experimental setup and surrounding conditions. Systematic error results from an erroneous method that is repeated with each test and consistently provides a similar inaccurate result. Random errors are addressed below in the form of measurement uncertainty, ensemble averaging for obtaining torque curve data points, and run repeatability. Revolution speed variation is a source of systematic error and is examined in Section 4.1.3. 4.1.1 Instrumentation Uncertainty and Data Point Averaging Precision of the recorded values affects measurement accuracy, and this uncertainty is typically specified with the instrumentation component being used. Error is also attributed to the DAQ component reading and amplifying the signal, as well as any other signal conversion devices. Table 4-1 below provides the uncertainty associated with the torque sensor and angular encoder. Table 4-1: Torque sensor and encoder uncertainty (percent of rated output) and absolute error. Item Torque Sensor Encoder Sensor 0.20% 0.10% Digital-Analog Converter -- 0.50% DAQ Card 0.10% 0.04% Sum 0.30% 0.64% Absolute Error (extreme case) 1.5 Nm 2.27 deg 84 These maximum errors due to instrumentation are very small, and given the number of data points recorded and the averaging techniques applied, these uncertainties do not provide a good understanding of the accuracy of each data point. Considering the torque curves, it is more useful to know the standard deviation of the ensemble averaged data used to obtain the plots. Figure 4-1 and Figure 4-2 provide standard deviation of the data points used for obtaining torque curves of the free-stream device at 1.5 and 2 mIs with the gearbox drive-train at TSR=2.5. Representative 95% CI obtained from the standard deviations are also provided for three locations on the first peak and are circled. The magnitude of the standard deviations are similar for each plot, though one difference is that at 2 m/s the torque values do not drop significantly below zero. The play in the coupling in way of torque values about zero and the resulting steep slopes contribute to the fluctuating standard deviations observed. Revolution Angle (Deg) Figure 4-1: Standard Deviation and Torque vs. Revolution Angle for a free-stream device with gearbox drive-train at 1.5 m/s and TSR=2.5 (N 34). 85 Figure 4-3 provides a similar plot for the free-stream device using the chains and sprockets drive-train at 2 mIs with TSR=2.25. The combination of dampening from the chains and sprockets system, as well as lack of play in the coupling, significantly reduces the standard deviation values to be consistently less than 4, though the peak torque values have also been decreased by a factor of approximately 3 from the gear-box drive-train case. Revolution Angle (Deg) Figure 4-2: Standard Deviation and Torque vs. Revolution Angle for a free-stream device with gearbox drive-train at 2 mIs and TSR=25 (N 52). 86 70 __________________________________ —a— FnsemLt Averaged Torque__F 60 120 180 240 300 30 -10 0 Revolution Angle (Deg) Figure 4-3: Standard Deviation and Torque vs. Revolution Angle for a free-stream device with chains/sprockets drive-train at 2 rn/s and TSR=2.25 (N —33). Lastly, for the case with ducting and gearbox drive-train (Figure 4-4), the reduced torque fluctuations also lead to reduced standard deviations. In this case, the standard deviation is consistently less than 3, with peak torque values ranging up to approximately 80 Nm. The standard deviation above is used to create error bars in efficiency plots when comparing with theory (Section 4.2.2). 87 Torq:: : 60 7 z50 : I / :1 40 : 0 60 120 180 240 300 360 Revolution Angle (Deg) Figure 4-4: Standard Deviation and Torque vs. Revolution Angle for a ducted device with gearbox drive-train at 1.5 mIs and TSR=3 (N 45). 4.1.2 Run Repeatability Given the time constraints due to working in the towing tank facility and the number of parameters requiring investigation, it was not possible to conduct a large number of repeated runs for completion of a comprehensive statistical analysis. Table 4-2 below compares Ck values for repeated runs using the gearbox drive-train for both free-stream (arm profiles B and C) and ducted tests. The percent difference between a series of runs completed at a given set of conditions and their respected mean is provided. For free-stream runs, 18 of the 29 repeated runs have a percent difference of less than 1% in magnitude from their respective mean value. 6 of the 29 are between 1-2%, while the remaining 5 values are between 2-4%. This repeatability is acceptable considering carriage speed, torque, and revolution speed are all being recorded and used for the calculation of the Ck value. Examining the ducted device, 75% of the points have a percent differences less than 2%, with the remaining points having differences of 2.74% - 88 4.2%. A larger error for the ducted device is reasonable given the size of the duct being towed through the water resulting in large disturbances in the flow and increased forces, and thus flexing, on the mounting structure. Runs noted as being in the opposite direction were performed towards the wavemaker instead of the dock so as to investigate consistency between directions. This enabled runs with duct deflectors upstream of the turbine to be performed without having to move deflectors from the downstream position. 89 ° H CD n e2 . a ’ a d - CD &Q < CM ‘ a , (F C D a t j M a , 0 C C d t Q CM I (F a, CM g - i D as CD CD o z - . o a CM - CM Q t< Cd ‘ t a d — CD CD C a fr i CD 0 CM B C D • C e* fl ‘ CM a C 0 a ’ CM , - CD C D C D CD CD CM CD C D) CD CD ‘4 - o t fr 4) CD 0 I— ’ - CD CD C d 0 CD CD CD 0’ a a C.) a F) 0’ 0 Cd fri a -1 C fri a t a F) a Cd fri C t a fri C) .t F) -I bi bi bi bi e n bi bi en F’. ’. bi en e n e r’ . p.. ’ r’ .’ F’. ’ F’ .’ r’ .’ F’ ) p.. ’ F’ ) e e bi en e en e n p. ) z o 3 -— r , 0 tW a a . a - n 0 C, ’ C, ’ 0) Cn (,‘ 0) 0 ) Co - ‘ - ‘ - ‘ r) — — — F’ ) C C 0 — — C 2 0 ’ 3 N N N N F’ ) P. ) N N t’. ’ P. ) F’ ) F’ ) P. ) N N N N P. ) F’ ) P. ) P. ) N N N N N r N N N N N r N N N 5 C. ) s i - 4 M ,, P P P P P P P p P P P P P P P P P P p p p p P P P P P P P o o p p p o p p p O p p a a . a a a a a 6) 6) F’ .) F’ ) F’ ) F’ ) F’ ) P. ) F’ ) P. ) p. ) P. ) F’ ) F’ ) P. ) F’ ) F’ ) P. ) F’ ) F’ ) F’ .) . . — — — a c i Co - J 01 C I a C o O l 6 ) 6 ) - J ‘ J “ 4 ‘ 4 - a 4 — — — — C, ’ Co in C I 0 0 0 — — 6) 6) 6) ) 0 0 F’ ) F’ ) 0 0 “ 0 Co 4 6) Co — C i — Co a Co Co Co Co F’ ) - J Co - J ‘ 4 - i a Co . ) 4 a a Co Co Co ‘ C I Co CO P. ) — C, ’ 0 Co 0 ‘ 4 Co ? 0 . b p P P J . C o 0 C I ’ 4 a J C o C o ’ C o C o o F ’ )F ’) .w F ’) C .) a C o L jC o ’ r ’ )o -4 F ’) ’ 0 ’ O iO • 0 a Co 0 — Co ‘ ) a CO Co Co Co Co Co Co ) Co a C. ) Co Co 0 Co in P. ) Co Co P. ) Co Co Co C. ) Co - , x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x ’ F’ ) F’ ) P. ) F’ ) F’ ) e e e en e e,’ in a a a a 3 > 0 , 0 ‘A r a a a 0 D F) C.. 0 C C O 0 t ‘a 51 r a C . Table 4-3: Sample chain/sprockets drive-train repeated run percent variation in Ck Run # Speed (m/s) Nominal TSR Ck % Oifference* 1045 1.5 2.5 0.1284 0.73% 1045b 1.5 2.5 0.1266 -0.73% 1085 2 2.5 0.1367 -1.03% 1085b 2 2.5 0.1395 1.03% * cakulateci as (Run-Mean)/Mean for each condition Just as Ck values should be equal for each run at the same conditions, torque curves should also match over a revolution cycle. Figure 4-5 provides torque vs. revolution angle for repeated runs at 1.5 m/s and TSR2.5, while Figure 4-6 and Figure 4-7 provide Cartesian and polar plots respectively of repeated runs at 2 mIs, TSR2.5. It is evident that the peak locations are very repeatable, providing consistent knowledge on which regions of a revolution are in need of performance enhancement. The polar plot is a nice visualization tool, accentuating that torque is generally created as a blade passes across the flow upstream of the turbine. 91 200 -. f —Run204.1 150 IRun204.3 Lz-Run204.4 — 100. oI— V I ‘f ‘V 1 1 0 4 50 100 150 200 250 300 350 400 Theta (degrees) Figure 4-5: Torque vs. Revolution Angle for repeated runs with gearbox drive-train at 1.5 m/s, TSR=2.5 (arm profile C). 200 150 —.---Run214.1 HRun214.2 Run214.3 __e_Run2144 V [-Run214.5 E100 z 50 0 V “ I I 0 50 100 150 200 250 300 350 400 Theta (degrees) Figure 4-6: Torque vs. Revolution Angle for repeated runs with gearbox drive-train at 2 mIs, TSR=2.5 (arm profile C). 92 Figure 4-7: Polar plot of Torque vs. Revolution Angle for repeated runs with gearbox drive-train at 2 mIs, TSR=2.5 (arm profile C). Figure 4-8 provides torque curves for the ducted configuration, again highlighting repeatability of the system. It is particularly impressive considering tests were conducted on different days amongst configuration changes. Lastly, Figure 4-9 provides repeated runs with the chains and sprockets drive-train for TSR=2.5 at speeds of 1.5 and 2 mIs. Again, repeatability is reasonable given the flexibility in the chain and sprockets drive train, and flexing of the force balance. 180 270 93 120 Figure 4-8: Torque vs. Revolution Angle for ducted repeated runs with gearbox drive-train at 1.5 mis, TSR=2.75. Figure 4-9: Torque vs. Revolution Angle for repeated runs with chains/sprockets drive-train at 1.5 and 2 mis, TSR=2.5 (arm profile B). 100 80 p z 4, -p 0 0I- I60 1 ‘I40 itI ‘I ‘I 20 Run 484 -j—e--Run 624 I—e--Run 501 [sun 502 0 0 50 100 150 200 250 300 350 400 Revolution Angle E z C. 0 I— 100 150 200 250 Theta (degrees) 94 4.1.3 Revolution Speed Variation When comparing to numerical predictions, it was observed that the peak torque locations were phased to a higher revolution angle than expected (discussed further in Section 4.2.2). Data examination revealed a fluctuating revolution speed due to the torque being generated. This is illustrated in Figure 4-10 for runs at 1.5 m/s at varying TSR values with the chains and sprockets drive-train. The revolution speed (rpm) provided is representative as it is a spline fit through the multiple data points based on very small sampling periods; however, it provides insight into what is occurring. Additionally, as RPM and torque are being plotted vs. revolution angle, any average taken from this plot will be artificially increased compared to the true average over time of the run. When plotted against time, less time is spent at the angles with higher torque generation and revolution speed; however, when plotting against revolution angle equal weighting is given to all points in the revolution, skewing the average. Averages displayed in the legend provide the true average revolution speed when taken over the time duration of the run. It is interesting to note that the peak revolution speed typically occurs earlier in the rotation, or closer to 90° as one may expect. As the motor controller responds to the increasing rpm, it acts as a brake and the torque continues to increase for another 25° or so as the turbine is slowed. This process is repeated for all TSR values. 95 Hypothesizing that the chains and sprockets drive-train, as well as the flexing in the load cells, was adding to the cause of the revolution speed fluctuation, the chain and sprockets drive-train was replaced with a gearbox and the bottom force-balance plate was fixed firmly to the carriage. Figure 4-11 provides the resulting revolution speed and torque values recorded at the same condition as in Figure 4-10 (profile B arms, 1.5 mIs). Much higher peaks were recorded with torque sensor and coupling attached directly in-line with the shaft, and though the revolution signal was much cleaner, fluctuations still occurred on the order of +1- 15-20% of the target value. Given the magnitude of these fluctuations (ie. from -60 Nm to 90 Nm at a frequency of 3 Hz for TSR=2.0), it is not surprising that these fluctuations occurred. Again, the maximum revolution speed peaks appeared closer to 90° where maximum torque was expected, and the subsequent torque peak appeared approximately 25° later. 160 200 Theta (degrees) Figure 4-10: Torque (below) and RPM (above) vs. Revolution Angle for runs with chains/sprockets drive-train at 1.5 mIs. 96 Not surprisingly, the observed reduction in torque ripple when using ducting also corresponded to a reduction in revolution speed fluctuations. Figure 4-12 provides torque curves for the ducted turbine at 1.5 mIs, while Figure 4-13 provides revolution speed vs. revolution angle for the same runs. Worth noting are the way the revolution speed mimics the torque ripple at TSR=l .5, indicating that revolution and torque ripple are closely tied. Secondly, it is interesting to note that the drop in torque ripple at TSR=2.75 greatly reduces the revolution speed fluctuations (ie. from approximately +1- 29% at TSR=2.5 to +1- 8% at TSR=2.75). Importantly, with the reduction in revolution speed fluctuations, the position of the peak also shifts back in revolution angle from 103° to 95°. It has been demonstrated that in the absence of external factors, an increase in TSR value shifts the torque peak to increasing angle of revolution; however, due to the reduction in torque speed fluctuations and revolution speed fluctuations, in this case the torque peak has shifted to the left with the increase in TSR. This is strong evidence that the revolution speed fluctuations are responsible for a phasing of the torque curve when comparing with numerical predictions, with the largest torque fluctuations leading to a peak phase shift of 20-25°. 41) CU 0 CU 41) -o E z 41) D 0 Runll — RPM for Runl 11(62.6 avg) — RunI 1 2-2.25tsr — RPM for Runll2(70.5 avg) — Runl 1 3-2SOtsr RPM for Runhl3(78.O avg) RunI 14-2.75tsr RPM for Runl 14 Figure 4-11: Torque (below) and RPM (above) vs. Revolution Angle for runs with gearbox drive- train at 1.5 mIs. 97 Figure 4-13: RPM vs. Revolution Angle for ducted device at 1.5 mIs. E z 4’, Theta (degrees) Figure 4-12: Torque vs. Revolution Angle for ducted device at 1.5 mIs. I Theta (degrees) 98 Additionally, with the revolution speed fluctuations, torque values observed will be less than the peak torques that would exist in a constant revolution speed system, as some of the torque will have gone into accelerating the turbine revolution speed. 4.2 Comparison with Numerical Predictions Below, an overview of the numerical model used for comparison to theory is provided. This is followed by a comparison of experimental and numerical Ck values and torque curves. 4.2.1 Numerical Model Overview The numerical model used for comparison to experimental results was developed by Nabavi [281 using the commercial RANS code FLUENT. A two-dimensional, incompressible, unsteady solver was used in conjunction with a Spalart-Alimaras turbulence model. An extensive examination into grid density was also conducted, and a fine structured grid around the blades contained within a sliding unstructured ring in way of the turbine blades was used (Figure 4-14). This combination of parameters provided the best compromise between accuracy, computational cost and reliability, though it still took upwards of two weeks to run a ducted turbine simulation. Lastly, domain size was also examined to ensure that the blockage ratio in the 2D simulations (same percent as 3D blockage in the experiments) was consistent with free-stream results. For the free stream device, this corresponded to 8% blockage, and for the ducted device this corresponded to 18%. Extensive discussion on the numerical model is beyond the scope of this thesis, and details may be found in the referenced document [28]. Figure 4-15 provides a sample output from a simulation highlighting velocity contours at a TSR=2 and free-stream velocity of 1 rn/s. 99 JI () S C C C S C -. S 4.2.2 Comparison of Results Firstly, given the 2-dimensional nature of the numerical models, arm effects were not simulated and must be extracted from the numerical results. Figure 4-16 provides the experimental Ck values obtained for tests with arm profile C without blades to examine power absorbed in the bearings and parasitic arm drag, which were subsequently added to the CFD simulation efficiencies for comparison with experimental data. Figure 4-17 and Figure 4-18 provide Ck vs. TSR comparing the numerical and experimental results for the free-stream and ducted device. Ck values from the experiments with only arms have been added to the numerically predicted Ck to facilitate comparison. Error bars shown for the experimental tests are a combination of the maximum 95% CI calculated from the standard deviations from the appropriate representative torque curve Section 4.1.1 plus the potential error due to the 1.5 Nm uncertainty from the torque sensor. It should be noted that this is likely an over-estimate of the error, as the maximum standard deviation for one location on the torque curve was assumed to be applied to the average torque for that condition. Errors on the Fluent prediction are from the 1.5 Nm uncertainty in the torque sensor when adding the experimental negative Ck due to the arms. 101 0.000 . . — ,7 1I .[ I Figure 4-17: Ck vs. TSR for free-stream comparison of experimental and numerical results. 0 C-) 0. 0 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4. -0.010 0020 —y.005x4- 0.0362 + 0.086 - 0.091x + 0.0289 R2=1 - .1.5mfs -0.030 -0.040 • 2 mIs -0.070 y=0.0051x4-0.05123+0.1783-0.2822x÷0.1531 R2 0.9997 -0.080 -0.090 TSR Figure 4-16: Experimental Ck vs. TSR for arm profile C at 1.5 and 2 ni/s. 0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 —.— Experimental 1 .5 rn/s —.—Fluent 1.5 mIs —— Experimental 2 m/s Fluent 2 rn/s 1.5 1.75 2 2.25 2.5 2.75 3 3.25 TSR 102 0.65 0.55 0.45 0.35 ZPementaI 0.25 0.15 0.05 I I I I 0051 5 1.75 2 2.25 2.5 2.75 3 3. TSR Figure 4-18: Ck vs. TSR for ducted comparison of experimental and numerical results at 1.5 mIs. The figures above illustrate reasonable agreement between experimental performance and the numerical simulations. The discrepancies observed are likely due to a combination of both experimental and simulation errors. Experimental errors affecting the accuracy of the results are outlined in Section 4.3 below. Though a detailed discussion on potential sources of error in the simulations is beyond the scope of this report and is discussed in detail by Nabavi [28], factors to consider include: • Turbulence modeling difficulties (including capturing dynamic stall) • 2-dimensional simulations vs. 3-dimensional experiments • Inability to fully correct for lost power due to arms and bearings by subtracting results of tests without blades for Ck comparisons o Flow disturbance created by the arms reducing lift generated by the foils 103 o Upon removing the blades, bolt heads and other attachment components creating drag also get removed o Lost lift on the blades in way of the arm attachments • Trailing edge of blades in experiments was cropped for manufacturing purposes • Truncation and round-off errors during simulation calculations These same factors will also affect torque curve plots. Figure 4-19 and Figure 4-20 compare experimental torque curves (gearbox drive-train with arm profile C at ends only) with Fluent torque curves for the free-stream device. As discussed above, the experimental torque peaks are phased from the theoretical positions due to revolution speed variation. Fluent also predicts shorter, wider peaks, and the lashing in the coupling as the torque transitions through zero is visible in the experimental data. 200 150 1:0 [ExPerlmentl Iso 100 1 200 • 300 350 4 0 -50 1 -100 Revolution Angle Figure 4-19: Torque vs. Revolution Angle comparing free-stream experiments and Fluent at 1.5 mIs and TSR=2. 104 250 Figure 4-20: Torque vs. Revolution Angle comparing free-stream experiments and Fluent at 2 mIs and TSR=2.75 Comparing ducted experimental results to the simulations, Figure 4-21 (v=2 mis, TSR=2) again displays phasing between the expected torque peaks and experimental torque peaks, along with more extreme and narrower peaks. Figure 4-22 compares results for a TSR value of 2.75, which as demonstrated above provides a significant decrease in torque ripple and revolution speed fluctuations for the ducted case. Significantly, this decrease in torque ripple (which is also predicted numerically), and consequently revolution speed fluctuations, aligns the peaks of the two data sets very nicely. In this case, phasing is only approximately of 6° instead of the typical 200250 degrees. This confirms that the torque ripple and corresponding revolution speed fluctuations are the cause of the peak phasing. Also interesting to note is that the predicted and experimental peaks have similar shapes now that the torque curve does not pass through zero. This is indicative that the play in the coupling may be contributing to a backlash effect, leading to recording of higher and narrower peaks than what would be nominally occurring. 200 150 z 100 -p -50 50 0 Revolution Angle 105 400 300 200 100 1 1 ‘Expehmental’ (I) I I I —Fluent = a- 0 0 Ia I— 150 100 100 k h1 200 0 (300 350 40 -200 -300 Revolution Angle Figure 4-21: Torque vs. Revolution Angle for a ducted turbine at 2 mIs and TSR2. 100 •0% 90 80 70 j60 I—.— Experimental I z I—Fluent I50 40 a. a- I \30\j 1 20 / 10 0 I I I I 0 50 100 150 200 250 300 350 400 Revolution Angle Figure 4-22: Torque vs. Revolution Angle for a ducted turbine at 1.5 mIs and TSR=2.75. 106 Lastly, Figure 4-23 compares drag force for the experiments and Fluent at 2mIs and TSR=2.75, with torque being displayed below. Given the number of assumptions in the procedure above for balancing moments to record drag and the assumed accuracy on the order of 20%, the results are in good agreement. The average predicted by Fluent is 1290 N, while the average from the experiments is 1325 N. Two significant factors that will raise both averages in the true application are as follows: • The Fluent simulation does not include shaft drag (predicted to be approximately 155 N) • When drag was being recorded, the turbine used arm configuration B at the quarter-chord positions, and hence more lift will be generated in an optimized design increasing the drag component on the turbine. It is also significant that the drag peak position aligns well with both the theoretical torque peak, as well as the theoretical drag peak. This is correct given that at an angular position, drag reading will be independent of the torque reading, which is directly affected by the motor control and phased due to the fluctuating revolution speed. 2500 > 0 ____________________________________________________ z 0 I ‘I7 .2 500 z j 0 / 50 100 150 200 250 300 350 4 0 -500 Revolution Angle Figure 4-23: Drag Force and Torque vs. Revolution Angle for free-stream Fluent and experiments at 2 mIs and TSR=2.75. 107 4.3 Sources of Error In addition to the revolution speed variation that appears to artificially phase the peak in the torque curves up to approximately 25°, additional sources of error include: • Backlash in the flexible spider coupling used with the gearbox drive-train is potentially affecting the results in two ways: o When torque is transitioning through zero a “bucketing” is observed in the torque curve, which should have a much rounder profile. o When the play in the coupling re-engages, there is likely a “slamming” effect that creates a narrower torque curve than what would actually occur, with a larger maximum height. • Considering runs with the chains and sprockets drive train, inertial effects of the sprockets on the lay-shaft torque sensor appear to be dampening out the maximum and minimum torque values. • Free-surface interactions are an additional potential source of error. With the given time constraints and associated difficulty in producing a structure rigid enough to tow through the tank, the turbine was placed at a depth believed to be deep enough yet still facilitating the structural setup required. After a few seconds of spinning the turbine in a stationary position, disturbance was observed at the free surface; however, with the moving device the accumulation of vortices, and thus large interactions, would be minimized. Waves created by the shaft and surrounding frame as the device was being dragged may also potentially affect the results, though these small variations in pressure are expected to cause error of magnitude well below (if any at all) others identified in the system. Lastly, the comparative nature of these tests examines each parameter under equivalent conditions to observe its effect. • Blockage of the tank must be considered when extrapolating the results to true free-stream conditions. This would most likely occur with the ducted device, given the 18% blockage of the cross-sectional area of the tank. 2-dimensional simulations revealed blockage should not have a large effect on the result [28]; however, predicted decreases in Ck of 18% and 11% for blockages of 17% and 7.5% respectively when moving to a free-stream condition were predicted by 108 Bahaj et al. [29] using actuator disk theory for a horizontal device without ducting. • The angular encoder seemed to wander about 10 or 2° after each run. This is believed to be due to skipping of increments, or truncation error upon digital- analog signal conversion, but was easily managed by resetting the angular position and encoder before each run. • The method of assuming a centre of force and balancing moments for drag force estimation could likely lead to errors on the order of +1- 20%. Additionally, initial readings on the load cells were tared out before each run; however, settling after the previous run led to variation in the initial readings, and if the force balance system settled in an odd manner this may also introduce error to the measurement. Given the large number of unknowns, one must consider a possible error as large as 25%, though 10% is likely more reasonable. 4.4 Sample Application From the findings above, it is possible to develop a sample device for the purpose of replacing diesel generators used to power remote communities. Using dwelling and power usage statistics from the B.C. Hydro Remote Community Electrification Program [30], a device capable of producing 257 000 kWh per year was targeted. At 15 000 kWh per year estimated usage per dwelling, this is sufficient for approximately 17 homes. Multiples of these units (ie. for 34, 51, and 68 homes) are consistent with the larger communities targeted for power generation by B.C. Hydro. A power coefficient of 0.45 was assumed using a ducted device with deflectors and is suitable for the purpose of this exercise. It is likely that a higher value may be achieved through further optimization of the duct and foil, though transmission losses must also be considered. Tidal data for Quatsino Narrows in Northern Vancouver Island was used to assess extractable power from the current. This is considered to be a moderate-high resource, and tidal data is provided in Figure 4-24. Power generation was assumed to begin at a current velocity of 1.5 mIs (minimal extraction is available below this speed), 109 and to cut off at current speeds greater than 3.84 m/s due to structural and cavitation limitations. Generator selection has not been performed as part of this application exercise. Figure 4-24: Tidal current data. The resulting rotor required was a 3.375m x 3.375m device assuming an aspect ratio of one, which is suitable for the forces anticipated. Figure 4-25 illustrates the resulting power output from the device as a function of current speed. Torque is also provided, and the dashed lines show maximum and minimum values due to torque fluctuations. Interestingly, maximum and minimum values are also provided for a free-stream device producing the same amount of power should ducting with deflectors not have been used. It is apparent that the resulting stress on the structure due to the large fluctuations would present a significant reliability obstacle. Lastly, Figure 4-26 provides a sketch of a representative configuration for the device to be moored offshore or in a river near the community. The nominal rating of the device at 2.5 m/s (a typical current speed for rating hydro current turbines) is 41 kW. 8 6 4 0 a 4, > -2 -4 .6 -8 Days 110 Figure 4-25: Power and torque output. Figure 4-26: Representative device configuration. 160000 140000 120000 100000 80000 0 0 60000 40000 I. -20000 -40000 Current Speed (mis) 111 5 CONCLUSIONS AND RECOMMENDATIONS 5.1 Conclusions The research presented above details one of the few available experimental data sets and all associated setup information suitable for the validation of both a free-stream and ducted vertical axis hydro current turbine model. Building upon past NRC research identifying near optimum TSR and solidity ratios, an experimental turbine model (and all associated testing equipment and instrumentation) was built, commissioned, and tested in the UBC campus towing tank. In addition to obtaining repeatable experimental data for use in validating numerical codes, a parametric study was performed yielding baseline data on the effect of a number of parameters. For a free-stream device with span/diameter = 0.75, end plates were shown to increase the baseline Ck value by 16.6% through the reduction of tip losses for the tested aspect ratio. Additionally, changes in angles of attack between 30 and 5° were shown to increase the Ck value by over 21%. Further testing of a 3-armed and 2-armed model allowed for the quantification of arm effects, as well as demonstrated an increase in Ck of 0.047 by applying cambered blades at 50 This yielded a theoretical maximum performance without tip losses of Ck0.412. Accounting for further possible optimization of solidity, airfoil shape, and angle of attack, a theoretical maximum of Ck=0.45 in the absence of parasitic and tip losses is reasonable. Application of a venturi-style duct increased power output by the rotor to a Ck value of 0.473 compared to 0.272 for the free-stream case; however, the power produced was 12% less than what may be expected from a free-stream rotor of cross-sectional area equivalent to the duct capture area. Significantly, the duct provided a decrease in peak torque values, as well as in torque fluctuation coefficient from 4.24 to 1.25, over the free stream case which is very important for cyclic loading considerations. Subsequent duct configuration changes, as provided in Table 5-1 below, led to an additional reduction in torque fluctuation coefficient. The optimal reduction was provided with two downstream deflectors, providing a Ck = 0.442 and a CTF = 0.47. 112 Table 5-1: Maximum Ck, percent change, and torque fluctuation coefficient.. Case Ck Value %Ck Change CTI Free stream (basehne) 0.272 -- 4.24 No deflectors 0.473 73.9% 1.25 Downstream detiectors 0.442 62.5% 0.47 All four detlectors 0.393 44.5% 1.4 Spinning towards deflectors 0.426 56.6% 1.17 Spinning away from deflectors 0.442 62.5% 1.23 Upstream deflectors 0.407 49.6% 2.67 A preliminary investigation into drag force on the turbine was also conducted, and an approximation for the drag coefficient (accounting only for forces parallel to the flow) was found to be [Cd = 0.41*tsr_ 0.16]. A primary source of error was the fluctuating revolution speed of the device caused by the large torque fluctuations involved; however, understanding of this error (up to 25° with the largest torque fluctuations down to only a few degrees for minimal fluctuations) renders the data presented suitable for validation of numerical models. Such a comparison was provided for both a free-stream and ducted numerical simulation created using a commercial RANS solver, and optimal correlation was obtained for the ducted comparison when reduced torque and revolution speed fluctuations were observed in the experimental results. Lastly, a sample case study was presented for a ducted 3 .375m diameter by 3.375m span rotor operating in Quatsino Narrows on Vancouver Island capable of powering approximately 17 homes. 5.2 Recommendations for Future Work Recommendations for tests conducted with the same or a similar setup are as follows: Application of a flywheel between the torque sensor and drive-train as a means to better regulate revolution speed control. Applying a flywheel connected by a shaft out of the top of the gearbox would allow for a variable revolution speed control by when adjusting the added weight, while still registering true torque values observed in the shaft. 113 • Replacement of the flexible coupling with a universal joint without backlash, or an alternative coupling. • Use of a flume tank of suitable size and speed instead of a towing tank, as it would serve as a more reasonable facility for such turbine tests: o Allow for a more rigid, fixed structure o Permit longer run durations o Decrease testing time by not having to return to starting position o Simplify installation and removal of turbine In addition to recommendations for improving the experimental setup used above, general understanding of the model testing of vertical axis hyciro turbines may be greatly improved through the following: • A study investigating how free-surface effects affect turbine performance. To do this, however, a deeper tank may be required so as to ensure interactions with the bottom of the tank are not a factor. • An examination into performance differences (if any) between operation in a flume tank vs. a towing tank, potentially due to differing pressure field development upstream of the turbine • A detailed investigation quantifying blockage effects on vertical axis turbine performance. This may be most effectively performed in a flume tank by reducing cross-sectional area through the addition of a series of false bottoms and walls. Alternatively, tests may also be conducted in tanks of varying dimensions. Key factors suitable for experimental investigation and providing additional understanding of turbine operation and quantification of loading design requirements include: • The complex interactions between blade lift and drag, parasitic drag forces, and drag on the shaft should be investigated to resolve net force fluctuations and directions on the bearings. Given the difficulty in simulating blade arms due to the computational cost of a 3D model, this research is likely best suited to an 114 experimental study instrumented for measuring bearing forces in multiple directions. • Detailed force data on an individual blade of a multi-blade device would be valuable for numerical model validation. This may include the use of strain gauges at the connection point between the arm and blade to resolve radial and tangential forces acting on the blade. The key challenge of such a study would be to get the low-signal strength data recorded using the underwater strain gauges synchronized with the revolution angle and transmitted to the stationary computer for analysis. Numerical models are an invaluable tool for optimization studies pertaining to the duct shape, foil shape, and solidity ratio, as well as for understanding cavitation inception. Such numerical optimization should be ongoing, with the current limiting factor being high computational costs coupled with the high monetary costs to meet them. Lastly, considering the device and its path towards commercial application, a number of factors require close examination and an exhaustive list is beyond the scope of this thesis; however, of primary significance from a hydrodynamics and mechanical engineering perspective are the requirement for: • A detailed cost-benefit analysis assessing the use of ducting • A mooring investigation to best understand how to overcome the fluctuating loads and how to best assure device stability • Antifouling considerations to minimize performance reduction due to marine growth • A detailed examination of cavitation avoidance/management caused by the pressure fluctuations on the blades 115 REFERENCES [1] Intergovernmental Panel on Climate Change Fourth Assessment Report, “Climate Change 2007: Synthesis Report Summary for Policy Makers.” Nov 2007. <http://www.ipcc.ch/>. [2] Rifkin, Jeremy. The Hydrogen Economy. Penguin Group Inc., New York, 2002. [3] Cornett, A. “Inventory of Canada’s Marine Renewable Energy Resources.” National Research Council of Canada CHC-TR-041, April 2006. [4] “Evaluation of Nova Energy Ltd’s Hydro Turbine.” H.N. Halvorson Consultants, Victoria, Dec 1994. [5] Siamack, Shojai and Bernard S. Katz. The Oil Market in the ]980s: A Decade of Decline. Praeger/Greenwood, 1992. [6] “Green Energy Study for British Columbia Phase 2: Mainland — Tidal Current Energy.” Triton Consultants Ltd., Vancouver, October 2002. [7] Courtesy of Blue Energy Canada Inc., Dec 2007. <www.bluenergy.com> [8] Templin, R.J. “Aerodynamic Performance Theory for the NRC Vertical-Axis Wind Turbine.” National Research Council of Canada Laboratory Report LTR LA-160, June 1974. [9] White, Frank M. Fluid Mechanics. Fifth Edition, McGraw-Hill, New York, 2003. [10] Davis, Barry V. “Water Turbine Model Trials.” Nova Energy Limited for NRC Hydraulics Laboratory NEL-002, March 1980. [11] Davis, B.V., D.H. Swan, and K.A. Jeffers “Ultra Low Head Hydroelectric Power Generation Using Ducted Vertical Axis Water Turbines.” Nova Energy Limited for NRC Hydraulics Laboratory NEL-02 1, March 1981. [12] Davis, B.V., D.H. Swan, and K.A. Jeffers. “Ultra Low Head Hydroelectric Power Generation Using Ducted Vertical Axis Water Turbines.” Nova Energy Limited for NRC Hydraulics Laboratory NEL-022, October 1983. [13] Davis, B.V., J.R. Farrell, D.H. Swan, and K.A. Jeffers. “Research and Development of a 50kW to 100kW Vertical Axis Hydro Turbine for a Restricted Flow Installation.” Nova Energy Limited for NRC Hydraulics Laboratory NEL 038, March 1984. 116 [14] Davis, B.V., D.H. Swan, and K.A. Jeffers. “The Ducted Vertical Axis Hydro Turbine for Large Scale Tidal Energy Applications.” Nova Energy Limited for H. A. Simmons NEL-070, March 1984. [15] Davis, B.V. and D.H. Swan. “Commissioning and Testing of a 100kW Vertical Axis Hydraulic Turbine.” Nova Energy Limited for NRC Hydraulics Laboratory NEL-081, December 1985. [16] Gorlov, A. M., “Tidal Energy.” Encyclopedia of Ocean Sciences, Academic Press, London, pp. 2955-2960, 2001. [17] Coiro, D.P., A. De Marco, F. Nicolosi, S. Melone, F. Montella. “Dynamic Behaviour of the Patented Kobold Tidal Current Turbine: Numerical and Experimental Aspects.” Acta Polytechnica Vol. 45 No. 3, pp. 77-84, 2005. [18] Guido, C., S. Francesco, L. Greco, A. Moroso, H. Eriksson. “An Experimental Investigation and a Theoretical and Computational Methodology to Study an Innovative Technology for Marine Current Exploitation: the Kobold Turbine.” Bolletino della Communita Scientifica in Australasia, December 2006. [19] “Cycloidal Tidal Power Generation — Phase 2.” QinetiQ Ltd. for the Department of Trade and Industry Contract T/06/00229/00/REP/2, 2004. [20] Ponta, F. and P. Jacovkis. “Marine-Current Power Generation by Diffuser- Augmented Floating Hydro-Turbines.” Renewable Energy, Elsevier, 2007, in press. [21] “Variable Pitch Foil Vertical Axis Tidal Turbine.” Edinburgh Designs Ltd. for the Department of Trade and Industry Contract T/06/00234/00/REP/2, March 2006. [22] Shiono, M, K. Suzuki, S. Kiho. “Output Characteristics of Darrieus Water Turbine with Helical Blades for Tidal Current Generation.” Proc. of the Twelfth International Offshore and Polar Engineering Conference, Kitakyushu, May 2002. [23] Taylor, Julian. “Crossflow Turbine Developments & Testing for Ultra-Low Head Hydro Applications.” Highquest Engineering Inc. DSS Contract File No.: 51SZ.23216-6-6156, Vancouver, Aug 1987. [24] Alidaadi, M. (private communication), 2007. [25] McCroskey, W.J. “The Phenomenon of Dynamic Stall.” NASA Technical Memorandum B 1264, Moffett Field, Ca, March 1981. 117 [26] Riley, Donald R. “Wind-Tunnel Investigation and Analysis of the Effects of End Plates of the Aerodynamic Characteristics of an Unswept Wing.” National Advisory Committee for Aeronautics Technical Note 2440, Aug 1951. [27] Klaptocz, V. (private communication), 2007. [28] Nabavi, Y. “Numerical Study of the Duct Shape Effect on the Perfonnance of a Ducted Vertical Axis Tidal Turbine.” MASc thesis, University of British Columbia, Vancouver, B.C., Canada, 2007. [29] Bahaj, A.S., A.F. Molland, J.R. Chaplin, and W.M.J. Batten. “Power and thrust measurements of marine current turbines under various hydrodynamic flow conditions in a cavitation tunnel and a towing tank.” Renewable Energy, Volume 32 Issue 3 pp. 407-426, March 2007. [30] B.C. Hydro (private communication), January 2008. 118 APPENDIX A: Design Calculations 119 SCALING AND CONSTANT DEFINITION The term prototype refers to the kill-scale unit. while model refers to the model being tested in the tank. Variable Notation Value Unit Conwnenta Scale Factor SF 22.214 <— Scale factorto make Urn = 4 with Dp = 66.66 a 16.664 Prototype Diameter Up 20.32 m (To make Urn = 3.5, SF = 19.045) 66.66 ft (To make Urn = 3, SF =22.214) Prototype Radius Rp 10.16 m (to chord line if symmetrical, else pivot point( 33.33 ft Prom Airfoil Chord Length Lcp 1.52 m S ft Number of Bodes Nb 3 Blade ReightDlameter ltD 0.75 (Based on value from NEL-002 p. 20) Prototype Span Length Lap 15.24 m 50.0 ft Prototype Cuinent Speed 6 knots Scp 3.09 m!aec Prototype Water DersJry Rhop 1025 kg& Prototype Water ‘[racostly Viacp 1 .OOE-03 kgl(m’a) Model Diameter Urn 0.91 m Use Toos -> Goal Seek.... to sets rnodei diameter by changing SF or Up 3.001 ft Model Radius Rm 0A6 m 1.50 ft Model Chord Length Lcm 0.069 m 2.701 in Model Span Length Lent 0.656 m 2.25 ft Model Water Density Rhom 8 kg!m3 Model Water Vacoeity Viscrn 1 .OOE-03 kg/trn*a) Model Turtine Area Am 0.63 m Model Current Speed Model Cisrent Speed Scm = Scp/agrtlSF) nVaec Scan = 2.00 nvseo =Scp/SORT(SF) 6.56 Nseo Solidity Ratio Solidity Rato SR = NbLOIR SR = 0.45 Tip Speed Ratio lip Speed Ratio TSR = R*cdSc where a = angus frequency TipSpeedRatio TSR= 2.25 ‘N:’kairLp:o3tdadeana lfeantnocebiadea.tl-ennuszciianoe. 3 Eased on qn. p. 21 of NEL-002. Sets oplara TSR according to aodduty from NEL002 RPM and Tip Speed Prototype RPM RPMp = TSRtScp,1Rp2pi)60 rev!min RPMp = 6.53 rev!min Omegap = 0.68 red/a Prototype Tip Speed TSp = RPMpt2’PlQ’RpitSO tWa TSp = 6.94 nfl Model RPM RPkti = TSRS2*pii))60 rev/mm RPMm = 93.97 revimin Omega_m= 9.84 rad/a Model Tip Speed TSm = RPMm’2PlO’Rm/60 TSm = 4.50 nfl 120 Reynolcfs Number Estimation Point in rotation Pr = 0 deg Vt’here 0 deg is ckectly into the cuirent (rotating cow). Prototype Re: Rep = Rliopwencp’Lqwiscp Rep = 1.567E÷07 Where: Vencp a Prototyçe encounter velocity Vencp = TSp + Scpcos(Pr) ni/s Vencp= 10.03 Wa Model Re: Ran, = RhonrVencn’LcmMscm Rem a 4.460E÷0S Where: Vencm — Model encounter velocity Vencm = TSm ÷ Scncos(Pr) iNs Vencrn= 6.50 iNs Stagnation Pressures (to aid with calculating required P range for transducers) Prototype Stagnation ft Pstsgp = 1l2opVencp2 Re Pstagp= 5.16E÷04 Pa 7.6 psi Model Stagnation P: Patagro - 112*Rhon?VencnV2 Pstsgm = 2.IIE÷04 Pa 3.1 psi 121 . - 0 a t . . 0 — r - . r P o P — — — — o c o . 0 0 0 - a - — cc cc cc c a j.j — 0 — 3 a S ‘ 1 0 r m r 0 0 5, z 0 5, z C) r m 0 5 a a a a a a:= a a t a a U I g * — p p i13 ta a 3 a S a . 5 a a 0 a 0 0 I 3 a iS a C C 3 a In L’ 3 (‘ a MODEL DRAG FORCE ESTIMATION Model Chord Lrr Lcn 0066 m Max ntrn Mode Arfoi Width Wan, 210% perert of thorO (p.72 NEL-002 for NACA 52.41-021 Acojat Max Model Mrfoil W& kaam O.044 m Model Spar Length Lam OS.tf 0 rn Carnage Sceet 5cr. cc other 2000 rn’s Crag Coefficiem ororat shaft Cdsnat From te Wi,re 40! for Re> 10000 Cdshaft = 12 Assure LID vey large) Drag on cenrat shalt =.SRhornscnArCsrtLaraftCceraft Lahaft 132n Dsht C.24826rn 1_9 in Drag on oensat shalt 152.7 N Crag on Mounbng Ans Cdshaft = 1.2 rag on amt =0fRhorSert2’Camm’Larrn’Cdshaft Larrs= C0Q57n, Darn = 0.0254 n I in Cranonairre: 07.1 N Tha Drag #of tens CemtalSlsat 1 1527 N Mounting errs: 6 742 N six rms 1’3oorfl2igodraginfreestericondthon Net ES cireoton force (fror ‘o velocity arc MA) -1515.’ ITotauFvrce: 2112 N I 47&2tf Alternate Dresawe-based scenato for deterninino &ao 126 kg 2 n*-earl 020c3873fG8 m Turbine Area = 0.60 Fressr.re = rr,crngtHead G6 ?a Force P-essu-eWea Fvrce 1252 N flit tf Current Speed: pressure is als V2’rho’W2) 123 Angular Natural Frequency of Loaded Beams Reference: Saths. Peter. Wnd Fcrces in Erneerrg. Perganxri Press, Oxford. 1072. = coefildemfron reference ad1e 6.23 p. if?) = Young’s MoöJus (Pa) YM 6.E+10 E= ln= Length of bean, rn) ln 1.t2m lassurnedSiset) n Mass perunit engthkg/m) Arex Densty &89E+10 Pa Mr Aiwniium (matweixcom) I.93E+11 Pa Mr 304 Stainless Steel 0.C4SD en 0.00508 en = Area’dersity 1.88 kgfm For a fixed-free candlever For a fixed-hinged cantilever For a hlnged.hinged cantilever Angu ar Natural Freq.ency, o = .,“scrifl1’I:M’ n; Shaft atr d ametr Shaft thicW’ess: cd_shaft 1.9 in LShai 0.2 in I = Area rrcment ci tiertia of beam x-secocr ‘nj I = piC864:od_shaft’4- (co_shaft -2’t_shaft:rM: I = 1.63E.07 m4 ps:’:cd_shaft’2- (od_shaft -2’t_shaftr2) 14 0.000588 122 rn2 2700 nocies c (ra&s) Seq it-t: 3.52 221.4 35.2 2 — 22.4 1409.0 224.3 3— 61.7 3881.1 617.7 4 — 121 76112 1211.4 s 200 12580.5 2002.3 4 nodes 2 15.4 a. 5 .7 Seq (1*) 1043 3145.1 178 1542 6541.9 500.5 2:2 11 VdO.( 10412 I (lOWS 1151.L? y’z3-I 4 nodes a. Seq (It) 1 — 9.87 620.8 98.8 2 30.5 2484.7 3954 3 88.9 5692.0 890.0 4 158 9038.6 1581.8 5 247 15536.9 2472$ 124 Slit Exoon Due to blade pulsinw ItissafetassirethatthewahwJ eencapulseatarateoftherwinterciblaces’therpm, asanwtg a pdse occurs as a blade passes a specific port in the rotation. Pulse frequescy = model rpii • ntrtbw of blades = RPf*n ‘Nb = 281.9 puls.esimin = 4.70 It Due to vorWx shecking on central shalt Ratio of staface 0 free stream velocty. a4tha ha = od_shatt2One_m f mha 0.119 TIn ratio is so icw that the retadot for a stationay cylindera be deemed ok licm’Swt’c hat /Vscm 98327 tg(Res_m; = 4.96 For use in table raenced below Sts_m = 1.3 {appr) Table p. 140 ‘Wet Forces in Enneedng by Saotrs. Peter. Vol.3 1972 4 nodes tee ilti 1 35.2 2 224.3 4 nooes tea Ha) 1 1542 2 500.6 4 nodes tsq lal 1 98.8 2 396.4 Slit Strength and De*ecfion od,aft 0.0426 m L5h 0S0508 m inc Lsthofteam{m 1.52 m Shaft Re)nolcfs flutter Res_n, = Resjn = Shaft Sirotlral sitter Excitation Frequeroy = San’Sts_m/ oti shaft = 74.8 It kiLn blade pulstig 4.70 Hz kline t ntex fldderg 74.6 It Ecitation Freouenoes: Panl Frnencies for flfli.LIw Natal Frequencies for a 4Jjjpqjjj. Nafl Frequencies for a hinged-hinged oflever is? it 0.2 a Area torn of nets of beam xsecscn 1.63E.07m4 Drag Fata 21112 N (from Drag Etnacsnwcataheetcle-caeescenaio) 125 With Bottani Locational Beaiing: iSchigley p. 071) CAsiaice between bangs U’S!: 75 in t005 n Cstarce from tçcerbearriglc force (AC): 55 hi 1.270 ii A — Fa Distance from krceie lower beaing (CS’. 25 hi 0.835 ‘i SunFx0; Fa+FbF Sun Ma = 0; F’AC = Fb’AB Fb= F’ACIAB 1408 N ICFa F-Fe F Fa 704.0 N ShearFrjits: FrcniA->C: 704.0 ___ Frcn,C->B -1407.9 Moment AtM 0 At C: B4.O5 Nm AtS: 0 Stress: 4f’yil 0.00413 m 894.0 1 1.83E-O7mM Slress 132E+08 Pa Stress = 132 MPa Tensile Yield Stress is: 215 MPa FoS = 1.62 Yield Stress for Aluminum 276 ?4a Yield Stress for Stainless SteeL 215 bPs Dëeodons: E= 1.93E÷11 Pa 6.SSE+10 PaforAlLmilnum I = 1.63E47 mM E 1.93E+1t Pafor3o4Stalnless Steel Ft 2ttE.03N Ce’eotionfrona->a (x hioreasflg ten B-> Candn’usibetween C and xrncter 0.554 m The = F’AC’x O’E’ ‘AE:’(t2 + ADA2 - ABA2) Ybc -0.0071 m Ce’aclion toniC -, A: (xnustbebetween 0.825 and 1.905 n) x incer 0.79 m Yes = F’BC’(I-x) I I6’E9’AS)’ (t2 + BC’2 -2’ABx) Yca 4.0063 m 126 Cantilevered Desiwr (Schigley p 6tZ4: Cistance between berrgs (AS): 16 in 0.4572 ni siance fran lower beating to force (BCX 37 in 0.9306 in Sun rxO: Fa+FFb Sun * = 0; Fs3?2 = FEC F; = P’BCAB Fa 4341 N 0771bf A Fb Fs4F Fb 6453 N 1452l Shear Force: PcrnA->B: 4341 N Pb FrcrnB->C: -2112 N Fran C -> E’rc 0.0 Mcment AIB: lists • CfttC: 0.0 F Sflsv 0.02413m 1984.8 Mi 1= 1.63E-C7ni’4 Stress = 2.04E+08 Pa Stress = 294 liPs Tensite Yield Stress is: 215 liPs FoS = 0.73 Yield Stress for Alugnaim: 276 ½)a ;mnseb.ocrn) Yield Stress for Stainless SteeL 215 MPa MacDeL =FBC)’2U6t9 •:BC-31n) Schçeyp.9 E ¶cCE+11 Pa 6.89E+lOPaforAhminsn 1 1.83E-C7rtt4 6= l.93E÷11 Pafor30dStainlessSteel SC = 00Q98 m 2111.0 N hr = 1.52 miflthofbesnibnsbotanbeaing)’:letcniBbowid( Max Deft.: -0.038 in Shaft Critical Speed Distance beffisen berrgs (AS): f&8 in __________ 1.748n A — Ciexcefrcnixcerbearrgiowl 1ACi 425 in 1.022 in Distance fran iii to cC (CD) 13.5 in O.343n wit c Disance fran w2 to lower beai(ng (D0(: 15Sf’ in 0.362 in w2 Critical Speed (no adthtional weilt) _______ Bongai = (pil)lerqht2 sz(E’T?n) Iengtft 1.743w 1.035+11 Pa = 1.63E-0 in4 m 1.Btce2Ooe3 ngal = 420.02 radls oniegal = 4011 ‘pm 127 0 C C -l n - j C) • 3 0 1 “ !. h H i - 4 0 * J 0 O , , W I 4 — O - l t * W - & — I 0 • G4 - . 0 0 * ‘ ) - - “ a : . - “ 4 ’ 4 - . - 4 0 - — . a 1 - : 1 1 . [1 . • . ‘ p p . ‘ . . . . i: 9 1 ii : 00 AIRFOIL STRENGTH CALCULATIONS Mc0 8ar L-;th: 0335 rn 131311 332 b 351 0 3trjt xt 36r iJt rtI: Cs rlSaJteS .ft :-‘so 114 93415 srt5 FDL: Cisance ho— srrol ercs 10 25% r0Itt 01 5(101 53511 Idax Wons.nt for ;upport a lJ4 of the GØan 1ron the ends: 293 Nm 9ee l: CMI rre4Isls’j eIoeI 511515 r 14: 6—37 5 Satettj Faotor: 9.92 Uel5ct Of .rtha: Uax’Y 0118995.: 312.1 ‘15 C t’; on 8is ste s8eofra— #oI ysfol fld ACA 348.13 11 1322 1011s -.4 1.ZIE B 213 5t 0 3Cg 51101 sIance A)0153 Fol r 381111817348 14.1Y2i MCrett N.’n1 0% 0333 003 2302 2% 0.314 35.37 2.131 4% 0227 2.723 6% 0.341 7313 1325 5% 0 333 ‘ 03.47 2.338 ‘0% 0.359 ‘21.33 £422 -- A 1l:0 J: V/. ...-_,I : :.: /1 \ : \ ‘ 4a e *2% 595% ‘1C.l - — —. — r.r ‘10% 0.252 ‘5-3.21 ‘4% 0 385 14.87 5.333 ‘3% 0.112 :10.94 11.373 1% 0.323 237.31 14.351 20% 0.13’ 25233 13.397 22% 0.131 25C 33 21.595 34% 0.133 3-1631 35.355 21% 0.171, -325.62 29.261 23% 0.17S -315.81 23.543 24% 0.132 -242.35 21.836 30% 0225 -44.53 15.297 32% 0.222 -237.11 14.331 34% 0:33 -213.58 113’S 35% 0.247 -134.37 3.363 33% 1231 -153.21 5.9’ 1 40% 0.274 -131.35 4.522 42% 0.265 -135.47 2-354 44% 0.322 -73.12 1.525 46% 0.315 -3174 3.723 46% 0.323 -23.1’ 3.131 50% 0.343 0103 .3332 52% OW’ 2337 3.131 54% 0.371 5274 2.723 55% 0.334 7312 1.523 55% 0.335 03.47 2.335 30% 0312 131.34 4.332 52% 0323 13621 3.311 54% 0435 33.57 8.533 31% 0353 210.35 11.375 54% 0433 237.31 14.331 70% 0483 25334 72% 0.43’ 2!C.35 21 .335 74% 0.333 31531 23.343 75% 0.514 -329.53 25.251 79% 0.321 -31331 24.345 73% 0.335 •293.05 21.333 50% 0343 -243.33 11.33’ 52% 0.552 -237.31 14.351 64% 0.373 -213.93 11.376 33% C’.353 -133.37 3.363 95% 0334 -194.21 5.311 50% 0.517 -131.34 2322 92% 0.531 -133.47 2.334 54% 0343 -79.12 1.323 94% 01.333 -32.74 2.723 93% 0.372 -23.37 3.131 ‘00% 0.536 0.00 ‘3.333 129 End4apaflerl Faa: Max Mcrentlor e44 at06orted ze!m - L4IWLSm,2I.sfl - Lsm0.) Max Mcre or cr4 onm — 114 Nn (lax flt4tll c?Is rr’tlalfl tefl) Max SIrfS% r tar 1.224—35 3 taf4 nolan 225 Cicrçsfla[(%j DflrcoPIan9Fo.IIcrl Mcmoi-t;Nri 0% 0.333 0.033 __________________________________________________ 2% 0.314 443 4% o.rr 17:54 4% 0,240 24433 9% 0.345 31213 0.282 47.753 14% 0.295 54.432 2 L// 10% 0.359 40’344 22% 0.151 77.430 ‘EE . Tt I 34% 0.223 101 449 Cnt 34% 0247 C4 132 34% 0.240 ‘04.433 4C% 0.274 108.323 32% 0.225 110.151 44% 022 111.417 45% 0.318 111321 44% 0.329 111594 50% 0.333 111345 0.357 111194 54% 0.373 112321 54% 0.354 111417 54% 0.395 110.151 50% 0313 C1.523 52% 0325 106433 54% 0439 04.152 55% 0353 101459 58% 0455 44494 70% 0453 54495 72% 0494 51.159 74% 0.538 85493 74% 0.91’ 94.794 75% 0.321 81477 75% 0.535 7741 80% 0.349 71449 82% 0.532 54.732 84% 0,575 50.773 85% 0.593 54442 88% 0.524 47.753 80% 0.817 40599 52% 0.530 31253 54% 0.445 24.433 54% 0,448 17.154 54% or: 4s53 ‘00% 0388 0.032 130 VERTICAL LOWER BEARING SUPPORT CALCULATIONS Dragtoroecnrnode. 2112W 262 Sf E”sjre to ciarge wee1 on aag ‘:r ;treadnhee: rot ctezkir; :r’eert saee4s Njntr of struts: 2 Force pernin.s dueto motel drag: 156.D N Strtt argon: 77 In 1.956 m crorn bottom of suppcctng beam on sib-carrIage to plate with bearirgi Dstartce from bottom of st-carnage to water 19 In SLomerged length of tearng support ann: 56 In 1.473 rn ire. urtoe’water want Stppcc arm tinrerslons: Lergoli — 4 In iparailei to noel rec1anguIartube’ 0.1:16 In wIdth — 2 In iperper4lcular to howl 0.0606 In Tnicsness — 0.1575 In 0.0046 In irerta— 1.75458E-06 n’4 Speed: 2 rn/s Reyndd Number 2.016—0! Drag cc’ellclent otsupport arm: 0.2 fora 2:1 eitpse ir tirtuien: tow rflrerce V4r:e p. 4.63 9etererlce Area: —urnoe’water span length parallel to foe 0.150 n”2 Drag ne to a sngle tearmg support ann: 60 N Totat trag force on a bearIng support arm: II1C N Dl&ance to cewe otforce: 50.5 ir rouT appronlmallon: 1.25 m Maxlntrr Morrent —Fccoe Distance to centre of’crce 1431 Mn Maxlmur t—Myil &14EO7 MLrrinurn Yle.d Stress: 2.765—0! Pa FoS = Yrnax — —F’dlstance to ftrct2 rn; dls1ance to force - 3.IengthI :reference Sct-ley p. 960’; Ymax= 4.011601 Iii 5— 6.60’t-€1D PacrAlLmnun 131 APPENDIX B: Component Drawings 132 Fi gu re B -I : ¼ sp an ro to r as se m bl y. QT Y: 1 iJ A it f l1 E I lA it U T F O U bA rb RE V IS IO NS :1 0 8 A l) Ex is tin j h o Ie s— 11 4- 20 UN C - 28 /3 Pl ac es / r i z z z z z - z z z z z z r z z z z z z z z [— —— a-- - - 11 4- 2O U N C -2 B A 3 Pl ac es 1? 4- 2O U N C- 2B 8 3 Pl ac es N 1 0 0 I.2 47 — / E ys ti ii hc 4e s— / 3. 43 8 - — - 01 .8 75 / / , 4_ - - - - SE CT IO N A -A Ty pi c& o f a ll th re e )o ca tio ns 1. 47 9 A dd a dd iti on al 9 th re ad ed ho le s to e x is tin g s ha lt M AT ER IA L: St ai nl es s St ee I 44 0C tn ,th uu .L A TI rW . . W U bL tl iG ld . b r 11 1 A a tw r ,l IC , l it li L t P D W I. O i. A 11 L (A A U N L A tA L T d* N l’ tO L tA w .iH A V P *) A Ui Il U ti ti li U T C T iU .W lt li U tl U li ul iu ii A U b Jl Il k u tT ID U t ii Il il li u li l T ur bi ne Sh af t r il t U tt i bA T U U . U rC UC r ib ‘ • D U tD t S A .E ;R DU .l C Ji W l’ 4t i” li R M eb SU D L bA T K lit PU X T. U C D C . ‘ Tl U PU tU tC U L rn U , A a rD A U U ED ID U T n il U tli U lit L U li i lt i, il i UB C C H t.l C tl ) T i c’ t 00 2- 01 2- RO l Fi gu re B -2 : M ai n sh af t. c T 00 2- 01 3- RO O RE V IS IO N S: J A - 2. 40 6 TH RU 4 pl ac es T . 97 5 11 4- 20 UN C - 2 B .7 5 4 pl ac es A M AT ER IA L: A lu rn in ur n- 60 61 SE CT IO N A -A QT Y: 1 0 . o r tH It h U L U .0 H !T V I fl J tO L tt II U . L • 1 I Tu rb in e T op Be ar in g Su pp or t b It A S L I S fl A II P ti IZ O V O IK * .V 1 W 3 .t 1W 1 1 1 5 1 1 5 5 5 5 1 1 1 1 1 1 5 1 UB C Fi gu re B -3 : U pp er be ar in g m o u n t. R EV IS IO N S: • c 40 6 TH RU $ p la ce s TH RU 8 0 SE CT IO N A- A QT Y: I A I .r 4 r tk I) P t IN L t O aI n 1 tt 4 bI It A 1 . . t PI A M I’ t L Z V In & I* C IA . U ’ h k IJ 1 ex ,C ft 4 I4 tt N Z F N tt 4 J 4 A L JH fA c; t5 , L ,t te ,h W W ri 00 2- 1) 14 -R OD 0 R 2 .4 7 5 A M AT ER IA L: A Iu ni in ur n- D 61 T ur bi ne B ot to m B ea ri ng S up po rt , r A e G 8 4 A fl e fl IW U T V IZ pq C ’d cc n r c . In . r D d , a n D e s r d rI o .a a ht ,a y ra n d u A A V U t # t. It c qt C K tt )$ Y UB O C A re a 3 O ,, a C A It [a lt Fi gu re B- 4: Lo w er be ar in g m o u n t. It h P th L )L )J .M I1 I A fl .I tL Jf iU . L IT [L M L C t, 4 rt k lo p : N it lU C II i t — l t . A tf tI tA il A $ o t4 w e (Z 8 ta il ! A I.J !* A L U I’ II U L II ;U L JN C O U IU tY ( U I- N :I IL ,’ 4 .L . : lp I, :U N lu te N : I, W S U t1 RE VI SIO NS :I ft5 ” TH K , rØ .4 Q 6T H RU 8 5 c i / / / ‘ 8 p’ ac es U N C -2 B C ) C ) C) C C ) ø C ) 03 50 0 TH RU 0n 17 5 — - — 6. 00 M AT ER IA L: A lu m in um -6 06 1 QT Y: I Tu rb in e Bo tto m Pl at e [.W A tt4 tS , I :. lt UB C “ ‘ it 00 2- 01 6- R U U Fi gu re B- 5: Lo w er be ar in g su pp or t p la te . RE VI SI ON S :1 I Ic 4 tN 4 4 :. t * h tL t, It . 01 2 t’ tH tt hI tU I. .A W i L I* W .. L tU M . OO V 11 1 2. IL N T tI 1 Z 10 1S t T H L tI ’t tI tY . 1 2 L 4 l. 2 Il H tA tA ;L St O O P tO Ia .5 V at tM W m oL U t’ Il O fl tt t0 0 0 tA U tt N RE CT AN GU LA R TU BE 4x 2x 01 87 5 TH K El ’) 5- , ED El ’) ED El ’) ED . 75 3. 25 Th is pa rt fo rm ed th e tw o la rg e v er tic al su pp of ls po sit io nt hg th e low er be ar in g pl at e. Th e e n ds w e re fa ire d w ith a pi pe th at w a s sp lit a n d gl ue d in pl ac e to c re a te a ro u n de d pm fll e. El QT Y: 2 0O M AT ER IA L: A lu ni in um -6 06 1 Tu rb in e R ec ta ng ul ar Su pp or t lI t , , . 4 f l 1 S S tS tt 5 It s D C NO T SC H_ E PR t,4 ER AW IN O - R o tS sa en st L S tT U N F l. 0 0 5 1 4 1 1 fl RI P1 SI 4O L I4 FI M ’)1 OI IP O t 11 0 lO IN S Il l t a t M UO N P fi C tv C D lt N . IS ti O d 4 4 n 5 1 5 Id Y ll U N S IN 0 0 0 T 11 1, 1. 10 UB C LI Sl E ‘ IC R O E N SH EE T 1 D0 ‘1 00 2- O’ l7 RO O Fi gu re B- 6: V er tic al su pp or ts fo rl ow er be ar in g su pp or t p la te . R EV IS IO N S: i . o c . 2 5 0 — 8 0 N - I I I — 12 50 2 .0 0 0 — I T C lt JC 4’ lt C A ll c e n r tM lI o Is lI L t I U I £ r • fl tA A A . I 11 41 11 L .Z L t.A V t In A fl * a A U I u s C ,I A W It N hN Ø lI IJ I A . IJ IF A C IS IA 1 * 5 3 IN C le A . z 0. 40 6 T H R J 4 Pl ac es “ V 8 Q C ) 0 C ) LI ) CA h . 0 0 0 — I . 2 5 0 — 1 . 2 0 — — — — - — - — — - - - - — — — - — - - - - — - — M AT ER IA L: A lu nh in um -6 06 1 QT Y: 4 7. 11 C R *l M U td tA t M Q K ZI W O flJ le 75 13 Iu r1 n e A ng le Su pp or t UB C IA M C 1 C F 1 00 2- O’ 19 -R OO Fi gu re B -7 : A ng le br ac ke tc o n n e c tin g v e rt ic al su pp or ts to bo tto m pl at e. RE V IS IO N S: I I . 00 0 I ‘ 19 QT Y3 fl L *J L A L 5 I t. L T ñ IH 5V PL A t 5 5 1 4 5 5 5 5 ?L L A JI TF A CS S 4) 55 54 11 55 1. 54 - 0. 26 6 TH RU \ / 5 3 1 X 82 ,O 0 Pl ac es FR OM O TH ER SI DE E e 4 L ) A ,y )C 4 ç\ô ç SE C TI O N A -A M AT ER IA L: A lu m in un i-6 06 1 T IZ SM A SL FT L 00 2- 02 8- RO D ri s 5. 54 5. 14 m m .p Lo nt Ts an jrS 5 4 . 0 0 . 0 1 50 5_ F FR O M OA AW I4 IG U 5S W I5 PU T 14 5 . 1. 1S t 24 01 54 14 1 ‘ O T t R H EE T 1 I Str ut C ol la r T op Fi gu re B -8 : To p o fs ha ft c o lla r fo r m o u n tin g a rm s to m a in sh af t. I. tN It h o at A H il V I M IM tt IJ 5 U — I I I • :I . a :N T tK IO Ø tI H I. t L W C a: . a : rI C A Z s * tA x a . 4A NP t: :a aW ,r hA .,. W 1A 0! ftA C A tS tW C A C tJ A RE VI SI ON S: r . 2 6 6 V . 80 0 / _ J 2. 37 5 W . 37 5 / 12 0 A pa rt C E 0 I ½ I 0 ‘F0 a n n 21 9 SE CT IO N A- A QT Y 3 M AT ER IA L: A lu rn in um -6 06 1 St ru t C ol la r B ot to m A S A a ,. G 4 . L .J C A I , r a . T A .. m . I. C I f l . a n s s o a w . I s O A tD u T — 1 tt tQ Q tI qv c . 1 U U 2- U )4 ’< UU Fi gu re B -9 : B ot to m o f s ha ft c o lla r fo r m o u n tin g a rm s to m a in sh af t. QT Y: 3 = K v F . N > FR O M PR O V ID ED M A TE RI A L C 1 ! 00 2- 04 1 - Ra G RE V IS IO NS :I BO TH EN D S ;- 1. 00 26 .2 50 — : L’ J M A TE RA L St ai nl es s St ee L 44 0C U .3 f5 sp ar P C 9C C S K * W If l# a * ta (r .. 7 A T. I Fi gu re B- 1O : 3/ 8” sp ar fo rf oi l a ss em bl y. RE V IS IO N S: i 1. 00 = = - = - - = = - = = - - — F = = - . — - i- .- t. = = — ‘ — — — — — — QT Y: 3 FR O M PR O V ID ED M A TE RI A L 0. 25 0 Sp ar A - r . . f t c s n .f ,, r p I o ,E .r .r u r E ft DD N OT SD A ..E aR D U 4 L .t ,4 rt I’ , . - . N t ft ]l U U t h ft tI .A 4 ftf tft I- t. L tz V. I 4 ft .A L o . V .fl t_ K N L, ft I lA Te ,: U lh tt L V V ei l a P N C A u IA L . TL TM IA & .tS T A C IC IN W I u n Sd 1E 1 O. - L _Q P2 IQ Q - Fi gu re B -i l: 1/ 4” sp ar fo r f oi la ss e m bl y. BO TH EN D S M A F t I 4C i. Ar m pr of ile A Fi gu re B -1 3: A rm pr of ile s A a n d B w ith fa iri ng s. a 2. 28 H Ar m pr of ile B 0.3 8 — EE EE -- - H 1. 00 H Fi gu re B -1 2: ¼ sp an bl ad e a ss e m bl y as fo r a rm pr of ile s A a n d B. RE V IS ON S: I U II tA k. L S fl M I t L t, f lT ,, A * D . R P H Il U M A L . S .I M I, _ tS IU B & A A W A L IN St ru t M ou nt in g Bl oc k TO , . , a n g - . . r w A tO A TO A W I M A X H A O T . . c q c c A r IA y f l . A GA ID JD AT A4 Th 1 C .I t. .A tI Z IA / . 43 8 L - • 37 5 — . 0 0 0 1 0U 2- 03 D -R U U /, 2 5 0 ’T H R U . 00 0 4, g 0 LI) C ) . 57 7 • . 50 0 • . 42 3 . 43 .8 I I II II I II I I I I I li ii I I I i i II II II II II II 11 11 1 I I 0T h’ :6 M A TR LA L Al un iin um -6 06 1 U ’ Fi gu re B- 14 : A rm bl oc k u se d fo r a rm pr of ile s A a n d B. — A hi nu n — ho le s th ro ug h u n L es s o th er w is e s pe ci fi ed — 30 c tI o n s qu be d t 21 25 In c h de ep — 6 s e c ll o n s re qu h’ ec l a ± 0. 37 5 In ch es de ep In w .y of ’ r n s — 6 e n ds r ’ e q re d c 21 25 in ch es de ep w iih c o c in er b o re s 0. 31 3 de ep . 34 4 C ou nt er su nk to 0, 31 3 de ep Fi gu re B -1 5: 63 4- 02 1 fo il fa br ic at io n c o m po ne nt s. i5 9 1* 21 D RI LL O R 10 — 22 U N (2 PL A CE S) — s tr u t a tt ac hM en t to p p la te — 6 s e c ti on s re qu ir ed a t 0. 18 75 In ch es de ep a s ia 1. 62 7 Th .6 23 — 1. 42 5 03 63 - F , I I I I 0. 30 2 1. 12 5 09 38 1 0, 68 8 1. 37 5 I Is - r - _ - 1 ’ 0.49 0 t Ø O .2 5 1 (2 PL A CE S) RO 25 0 (T YP ) (C OU NT ER SU Nk TO 0. 0€ 23 OC EP ) (2 P1 _A CE S) 0 0 - 3 ’ #2 9 tR IL L PU ? 8— 32 lI t — s- Le iit a tt ac hM en t bo tio r’ i pl at e — 6 s e c ti on s re qu Ir ed a t 0. 18 75 in ch es de ep Fi gu re B -1 6: 63 4- 02 1 fo il fa br ic at io n c la m pi ng c o m po ne nt s fo r ¼ sp an c o n fig ur at io n. RE V IS IO N S: c .2 66 TH RU N / 05 31 X 82 0n 0j 2 ) * h C 1 .t : li lt li t tr w til lS lo U t li n t P u s 1 0 1 . X r U t 2 U U tM lS .S ll h L A Il It il C t V . ll !A i’ P lt lL Il l, 51 21 2U 1 ti ll C U O U T tN U t ti li lC U tI T U I* O IW C L U T S QT Y: 6 0. 35 9 X 82 .O M AT ER IA L: A lu m th ur n- 60 61 en d pl at e 2 TO 11 21 11 1 24 1( 11 44 11 1 P 2 1 fl It 1 4 0 4 4 2 2 1 UB C li tr e 14 2. _t Il S SH E ( 1 U U 3- en tW la te 2 00 Fi gu re B- 17 : C ir cu la r e n d pl at es . RE V lS O N S: 0T h’ 6 C R $d *i S. M E I I .t If R C fX tC O C ,! E L IM I. o 4 N U D W S * L O C F T M C DD P .T S D A I F C M C A W IN ’ / M A TE RI A L: 4J um in um -6 O6 1 • L L t * I tI lC )f t H U t U t U sf aJ cA .a , U U fl It U ? , “ 4O A . UM C O A W ItN En d Pl at e 1 Fi gu re B- 18 : N A CA 00 12 pr of ile en d pl at es . UB C C H t. K tt ) OH U A t , U 1 .t U tH L A H T tU U _t .H Z I U U 3- en d i- la te - - - / Fi gu re B -1 9: R ot or a ss e m bl y w ith a rm pr of ile C. Fi gu re B -2 0: B la de as se m bl y w ith a rm pr of ile C. RE VI SI ON S: l 0. 25 0 TH RU 2 PU ce s tH It h L L JJ .. A H If lb If la .L E U K II . Ib ’ fl - 43 1 8 8 — 0 8 / & fl N M AT ER IA L: A lu ni in ur n- 70 75 QT Y: 9 St rL it M ou nh ng B lo ck T n S m .a iM ,. L fi t P lo ta l. r u r t r . , DO N D SO A_ E RO FA CR AW I4 G A * % G c . LR D V .a A E flK LP T QM PR ( M G ’C Io rf l a th rf ld I1 U D tD T aS tt, i US C ‘4 L Il t L ,t tI tl K tt ll V , k e L ,. ’t , fl I, ;& W U R 4 jl P1 4. 41 fi tI t L ”L ,X tP C t 1 ‘ I QO 3-O O1 -RO Fi gu re B -2 1: B lo ck fo r a tt ac hi ng pr of ile C a rm s to sh af tc o lla rs . RE V IS IO N S: I SE CT IO N A- A 1. 05 0 92 5 . 42 5 . 00 0 // // // // // /4 4 ’/ // // // // .2 - (.3 3) M at er ia lP ro vi de d (N AC A 00 12 Ex tru sio n) M AT ER IA L: A lu ni in ur n- 60 61 QT Y: 3 To p A rm N A t JC ,4 L 4 tS N UB C )I A V tt ) 1 O F I . ‘ I 00 3- 00 3- R0 2 Fi gu re B -2 2: Pr of ile C u pp er a rm . RE V IS IO N S: QT Y: 3 bA te I C 1 00 3- 00 6- RO O A øi 5O ( . 3 3 ) J I SE C TI O N A -A . 00 M at er ia l P ro ic 1e d (N AC A 00 12 Ex tru sio ni M AT ER IA L: Al um in un i-6 06 1 C en tr e A rm T .I A C R * A .M !f l, tt * O tt a T ? a r . D t r 1 5 & .E C IJ C R A W I4 G N A bI A A U k_ ,tA I4 ttN UB C Fi gu re B -2 3: Pr of ile C c e n tr al a rm . RE V IS IO N S :1 A L L tZ fl tt JW iI Jb t I W * P (3 S I IE * A 2 iIS 4t IC A .L 5 tk P ’ t O 5 fl If lA P P C S II ” fl 4 2 L O N O IA W tn , P ta ,l U .M J I lI * :: tp U tS c I 2 if lt 1 • 05 • 00 A ‘ 0 ‘ < ‘ // // // // // /1 _ z SE C TI O N A -A M at er ia l Pr ov id ed (N AC A 00 12 E: tru si on ) M AT ER IA L: QT Y: 3 Bo tto m A nn UB C 0 1 i 00 3- 00 4- RO D U i U I Fi gu re B -2 4: Pr of ile C lo w er a rm . 3 tc b m o u n tin g pi ec es in w o y o F c e n tr 1 or ’m o f n o n — c c ri be re d fo i’ s hc pe ’ II 0,7 50 0 .5 o I Fi gu re B -2 5: R ep la ce m en t 6 34 -0 21 c e n tr al fo il pi ec es fo r u se w ith a rm pr of ile C. G cr t 2 5’ de ep I ’ 2. 57 2 ‘ — 1. 62 5 0. 67 7 0. 67 3 V 0, 28 19 0. 56 79 -s O .O 45 4 \_ _t — L 2’ o 6. E 56 00. 37 G 3 m o u n tI ng pi ec es In w ay o f c e n tr a l a rm ’ (M Fi gu re B- 26 : 63 4- 42 1 c a m be re d bl ad e fo il c o m po ne nt s. 8ci 00 C CC’, C C.) C Ic C -1 -< APPENDIX C: INSTRUMENTATION AND DAQ COMPONENTS Instrumentation: • 2 of PT-Global SG-PT4000-500 lb s-type load cells www.sensor-technik.co.ukldatasheets/pt4000. pdf • Futek Torque Sensor, 0 - 369 ft lb, 0.2% accuracy, aluminum, 2mV/Voutput, 7” length (TRS300) http://www.futek. com/product.aspx?stock=FSHO I 992&acc2=acc • Accu-Coder 776-B-S-2048-R-PP-E-P-A-N 1-7/8” through-bore encoder (2048 increments per revolution) http://www.encoder.com/model776.html • Extech 0 - 18 Volts DC, 3 Amps, 2 digital/four digit display power supply • BK Precision triple output 12V, 5V, and 0-30Volts DC, 5 Amp, 2 digital/three digit display power supply • BK Precision 0-18 Volts DC, 5 Amp programmable power supply with Labview RS232 • U.S. Digital encoder digital-analog converter (used with encoder) www.usdigital.com/products/edac/ Drive-train: • 3HP microMAX motor 182TCZ TEFC from Marathon Electric with Parker SSD AC 690+ vector drive controller and braking resistor kit (may be used for both driving and braking turbine) (7/8” shaft; 230V, 4.6A, 5400 max. safe rpm) www.marathoneIectric.com/motors/docs/manuals/SB548. pdf www.ssddrives.com/usa/Resources/PDFs/Catalog/690%2OSeries%2OAC %2ODrives.pdf • CONEX gearbox B091020.LAARJ, TEXTRON fluid and power. Ratio 20:1, SHC 634 lubrication, helicoidal gear geometry (used with gearbox coafiguration) www.akrongear.com/documents/catalogs/textron/Series%20B%2023293- 0503. pdf 159 Data Acquisition Hardware: • 1 cDAQ-9172 8-slot USB Chassis with rail mounting kit http:Ilsine. ni .com/nips/cds/view/p/Iang/fr/nid/202545 • 1 NI 9205 32-Channel +1- 10V 250 ks/s 16-bit analog input module used with encoder and carriage speed http://sine.ni.com/nips/cds/view/p/Iang/fr/nid/202571 • 1 NI 9237 4-Ch 50 ks/s per channel 24-bit analog input module used with torque sensor http://sine. ni com/nips/cds/view/p/Iang/fr/nid/202632 160 APPENDIXD: RUN LOG 161 O t t , a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a p a a a in in in in in in in in in in N ) N ) N ) N ) a a a a a a a a a a a a a a o 0 0 0 0 0 0 0 0 0 a in a a o ia CO m in a in a o io Co - I G% in a in i- o o j o in a U ) a a 0 La CO - a ” in a in o C 1 b 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 C a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a M a a !P. r a a r N )N )N )t ’3 k o ó b b à o b o o o o o o o o o o o à b — ,. 0 L 0 it 0 0 Li i Li i Li i Li i Li i Li i Li i Li i Li i Li i Li i Li i Li i Li i Li i Li i Li i Li i Li i Li i 0 0 0 0 0 0 0 0 0 0 0 C 0 0 0 0 0 0 0 0 UI ‘ 4 a a a a a in in in u N) in N ) N ) N ) N ) N ) N ) N ) N ) a a a a a a in in in in N) N ) N ) N ) N ) N ) N ) N ) N ) N ) a a a a a a . ,. q u i i jL ii k a jL a h o in O O h ii L ii W O a U iL b f l i t a a 0 0 i n m 0 0 i n i n 0 0 W 0 0 O 4 f l e L t u i U i = i n i n i n w w C O i n a a i n i n i n i n i n i n i n i n a a i n i n W W N ) N • 0 i n W a 0 O 0 i n i n 0 i n C O a 0 O C O t h i n t b 1 M M N N ) O 0 J W i n L t W O L t in a - i a N ) a N ) in a a 0 in a a o in N ) LD a Li i a 0 0’ , Li i in a a a a v i - 4 N ) in N ) a in - i a ’ a a a in i W a a L D L O C O C O C O C O % J % 1 & a ,n r n L ii L ii a f l W W a a U J U 3 W W M N ) . 1 w D a o i n v i a a a a 4 - 4 a a a a - i - J o c a a a i e o o v i o o a a C o C O i n i n C O c o N ) N ) .J 1 a 0 a a a o w in in in in ‘ . i - i L9 L9 P 9 W N ) N ) in U I C O LD L9 U ) N ) N ) a a w La N ) a o o V i Li i CO CO 4 “ a a a 4 in ‘ j p J j p p p 8 8 6 8 6 8 p p p p p p p p p 6 6 6 6 6 6 6 p p p p p p p p 8 p 6 6 8 8 - a m a in in in u i a in N ) a a a c o - i a a CO CO a a it a in a a U I a 0 o N ) in “ J in S Li i “ - J N ) V i . Qo N ) Cf l N ) 4 N ) N ) “ a’ . a in CO lb a a ’ a ’ CO 0’ . 0 0 a a in - J % I w w %1 Lb t% ) a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a h a a a a to to tO to to to to to to to to to to to to to a a a a a a a a a a a a a a a a to a a to a to a to a to a to C to a a to tp M M M M M M M M M M a a a a t- ’ a to w to M M M M M M M a a a a a Ib a to to V it o V it o M M M M M M a M M M a a b b a a a b b b a b b b M a a a to to a a a Q v iM a to C W a a to to a a to C m a a a v it o o a 4 a o o w a w to a a to c o o a ‘ a to to to to 9i 9.’ 9.’ N N N 9.’ N N %J to ?° p p !-‘ !-‘ N N 9J 9J a ø p i a , A J M 9 1 tf l, % J %4 to D N N N N b lb M M a a a t h M a a a t h b t o a w a a u , • w t h a t n o t o t o t o w w w w t n a a a a to to cc ’ to tu M a a a to v i %I v i a N to to v i - j o M to a a v i to a M M N O O i a ‘ . J o i a i v i o a a a a o i L n c o t o t o t o a a w t o t n v iv it n % ,J a b p p p p p p p p p p p p p 6 6 6 b p p p p p p p p p p p p b 6 b 6 b b b b p p p p p p p p p p p w w O c ’ M t o O M M a a c O t 0 a O t o M v i a O O a m i v i t o t o M M % js to w to o io o ,v it o to to a v ia M a a a 0 ON t 0 Lb W L M US Lb LU U S L U L U U S U S U èL U LU L U L U W V S L U W W W W M M M M M M M M M M M M M M M M M M n h N tU tU N ’ ia Ia ia 0 0 0 0 0 0 0 0 0 0 0 0 i a n i a n ia ia ia I i 0 0 0 0 0 0 0 0 0 0 ‘ a c t Lb Lb a US N ’ Ia C Lb CO 01 U ’ a U ) N ’ Ia 0 Lb CO - j %4 01 01 U ’ 0 U ) N ’ Ia 0 Lb CO %I 01 U I 0 U ) N ’ Ia 0 Lb CO 01 U ’ 0 U ) N ’ Ia 0 Lb CO %J 01 o t o o o t t t o o o o o o o o o o o o t o o i a i a i a i a n I a I a I a I a I a I a ia i a o o c o p o p p p p p p o o o o o o c o ia ia ia ia U ’ U ’ 0 U ’ 0 0 O 0 0 o 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Lb 0 Lb 0 0 0 kS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 U i U ’ U’ U i U i U i U i U I U i LI ! U i U ’ U i LI ! LI ! U i U i U i LI ! LI ! 10 Lb lb ID Ia Ia Ia Ia Ia US US LU LU N ’ N ’ N ’ N ’ N ’ N ’ N ’ N ’ N ’ N ’ N ’ N ’ Ia Ia Ia Ia Ia p- I N ’ N ’ LU US N’ N ’ N ’ N ’ N ’ N ’ N ’ N ’ N ’ N ’ Ia Ia Ia Ia Ia Ia Ia Ia N ’ N ’ v J w U ’ U ’ U ’ a a a 4 n w w w w N ’ N ’ w w U ’ a w w w w w w M N ’ I a I a n C o C O LI ! Lb I t o Ia Lb Lb 1. 10 1 M M M C I 0 0 0 1 0 1 %J %4 I d LU US L bL b 0 0 . LU Lb Lb Oh % J Lb Lb 0 0 0 I- h O 0 M M U i L fl th b h a LI ! M O CO 0 N ’ CC 0 Ia o a Lb N ’ a a U ’ CO C l ‘ - 5 ‘ 4 CC C l LU U ’ CF i ‘ 4 N ’ N ’ 0 0 Ia CO US US Lb 0 CC 00 US N ’ Lb 0 0 N ’ LU U ’ a U ’ N ’ Ia C l ‘ 4 U I CC 0 1 * 2 C c ‘ 4 L b L b o 0 0 1 C l o o w o a a w w w w N ’ N ’ ’ 4 j N ’ N ’ w w o o 0 1 N ’ 4 t o C o a a o o U ’’ 4 ’4 N ’ M I a L b L D O W W L b N ’ I a L b a U ’ U ’ L b L b I a I a 0 0 N ’ N ’ L b L D W W L b L b O O L b L b J C O L b L b U ’ U ’ % 4 ,j U iU ’ U ’N ’0 O I a J Lb 4 a p p d o . j U f r p g i s j j f l 4 6 6 6 6 6 6 6 P 6 2 6 2 P p p p p b 6 6 6 6 6 6 6 6 6 p p 6 6 6 6 p p p p p p p p 6 6 6 6 6 6 p p p p g S S e 8 S S 2 2 t t S C ’ U I a ’ v i b a a a a a a a a a a a w v i w w w w v i w w v i w w w w w w v i w w w w w w w w w v i w o M M o o o o o o o o o o a a a a a a a a a a w w u i w w I M w w w w n a M M o n I- o O v iO O - . i c m v i a v i r i- o c m v i a v i a v i i. a e - o v ic m -i cm y a w i n c v i O O c m v i a v iM o v i ic m v i o o o a a a o o o o o o a v iv iv iv iv iv iv iv iv iv iv iv iv iw v iv iv iv iv iv iv ii a v ia 0 0 v i 0 v i 0 0 0 0 0 0 0 0 0 0 0 0 W v i v i v i v i ID v i v i ID v i v i v i v i v i v i v i 10 v i v i v i 0 10 0 v i C v i C v i 0 v i 0 v i 0 v i I-k M M l- l- N ) v i I- I- N ) N ) I- I- I- ’ I- )-k I-k N ) 0 N) N ) N ) N ) N) N ) N ) N ) N ) N ) N ) I- ’ I- I-k Ia I- I- v i v i v i v i N ) N ) N) N ) N ) N ) N ) N ) N ) N ) I- ki b b ki h to a a a Ln tn L.a L.a a a k3 to Li i to to to ci ci a i*i a a c i ci Lii Lii ki k3 Lii Lii ci ci to to ci Lj ci ci b a t V i I - v i e o I n v i v i t c m v i v i v i v i a a o o o o o o a a v i v i a c m o 4 a c o y N ) 0 1 0 1 W v iO l 0 1 M M If l V i W V .’ W y M M N ) v i W O O CO ‘ . 4 J Q 1 O lt O v i v i W O O C O ’- J O 1 4 C l 01 N N P P o w w w ° P N ’ . i 2 2 aJ U I V i I-k ia v i U I 1-4 ia a a CO CO N ) N ) - . a i a — j — j — i % l cm cm v i v i a a . ‘ - J v i 0 jb 1-4 v i v i v i v i - J N ) CO 1 0 0 . v i CO v i CO CO N ) N ) v i v i a v i v i a v i 4 I-h N ) i V. ’ a a ‘ . i % I y. ’ a w O l v i v i N ) v i v i v i V i v i v i I-k 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 p p 6 6 6 6 6 6 6 p p p p p p p p p p b v iW v iC O J v if h M v i c m C O c m - .a c m N ) 0 W 0 C O W C O v ia 0 J a N ) 0 W c m M N ) .a c m c m v io f lN ) 0 v i0 v iW N ) l- v i v iM N ) J 0 0 v iv i ½ ,jC I Jo Q I v i v i ‘% 1 v i 0 c m v i I 0 ‘ J v i % JN ) O W C O W M O O 0 % ja I-k i- ’v i W y CO ‘ I !: o i m m 01 0’ . a’ . a’ . a’ . o’ . 0’ . 0’ . a’ . 0’ . 0’ . i 0’ . 0’ . v i v i v i v i v i v i u ’. v i v i v i v i v i v i v i v i v i v i v i v i v i v i v i v i v i U , U , V i U I U I 0 0 0 0 0 0 0 0 0 0 . 6 . 8 .8 . 8 N - 0 0 0 0 0 0 0 0 0 q ’ i n ’ 8 8 8 8 8 g g 8 8 8 8 8 8 8 8 8 8 8 8 g g 8 8 8 8 8 8 8 8 8 8 8 8 8 N N . 6 v i D . 6 v i N M 0 0 CD CD . 8 . 6 . i N Ln v i - - . a - - . a v i bi i. v i ‘ vi - . a L v i . n ‘ - . . v i Lii 3 0 0 . 8 . 8 N v i v i ‘ - 1 “ 4 a ’. D i 0’ . D i v i v i v i v i . 6 . 6 . 6 - 6 v i LU v i J N N “ - I ‘ ‘ 1 D i D i U ’ v i - . J - . I V i V i . 8 . 8 LU v i U I V i V i U’ ) - 8 - 6 . 8 - 8 v i LU v i v i N W D , ‘ . ‘ - J - J - - 4 IJ N 0’ . Q I- I- - J . D p p a ’. v i 2 2 N i- i N CO C O .6 . 8 4 - . 4 W ‘ .0 - 1 . J N a . a’ . a’ . ci’ . 0 0 ci’ . i.- 0 m a’ . o ‘ . a’ . — J U i 0 iD iD LU iD - Ca 8 Vi 6- i- I- I- 0’ . 0’ . 0 0 0 0 4’ . . 6 N N “ 4 “ .1 CD CO . E’. . 8 iD iD 0 . 8 CO CO N i “ .1 4.0 0 LU . 8 . 6 v i i. V i a’ . v i 0’ . Lb ‘ . 4 Vi - I LO CO 1 0’ . v i I- . 1). % I . l N v i I. 0 U i CO frb 0 0 . 8 N fr 0 0 0 D i V i 8 . 8 % I ‘ . 4 ’. 4 0’ . 0’ . V i v i U I U I. 8 . 8 . 8 . 6 LU . 6 LU W N N . 4 . 4 0’ . 0’ . V i U i CO CO U’ . U i . 6 . 8 . 8 . 8 0’ . 0’ . . 8 .8 LU W LU U i N 0 0 1 0 0 C D C D W W L b 0 N N .8 .8 .1 % J L b L b 0 0 0 0 C D O D L U L U C a C a N N - i 0 iD v i . 8 Lb Lb ‘ .9 ‘ .0 V i U i CD CO - . I “ 4 v i ‘ .9 . C - 8 01 N N L V - 8 U i D i D i ‘ . 0 ‘ . 0 N N N i) ‘ . 9 iD ‘ . 4 “ 4 iD iD 00 CD . 8 N 0 0 O 2 0 0 9 Q 8 O 0 0 0 0 0 0 O O 0 0 O Q 0 0 O 0 0 0 0 Q Q Q Q O 0 C 0 CO 0 . J CD ‘ I- ’ U ’ 0 N N N N N N - “ , I N I- I-’ . 0 0 0 CD CD N N N t.4 I CO ‘‘ .1 i.b a’ . a’ . . 8 . 6 . 6 . 0. V i v i N N 0 N V iS P O N 0 1 V i v i 4 V i .8 4 D i C O D i ’ .D .4 W 0 ’ .f l L U 0 v i f r G ’ .D i 4 .. 4 L U L U i 0 0’ . N V i iD CD - 6 N “ - I ‘ . ‘ . 0 . LU v i CO ‘ . Lv - J LU CO LU J - . 4 . 8 N V i u Q LU Ca Q’ . N 01 v i v i u i D i N “ J 0 0’ . 0’ . U i — 1 m m m m m m t h o , o , c a , W N ° W U) O U I Ø W C D % W C W W M C t t n v t U ) U ) U ) U ) U) U ) U ) U ) U ) U) U ) U ) U ) U ) U ) U) U ) U ) U ) U ) U ) U ) U ) U ) U ) U) U ) U ) U ) U ) U ) U ) U ) U ) U ) U ) U) U ) U ) U) U ) U ) U ) U ) U ) U ) U) U ) U ) o O M M N M M M M M M M M w p w p w U ’ w w w p w w L 9 p t p w a y 9 a L n a U ’ 4 a b i o o o j . . J L n u , U 1 U 1 0 0 0 L t U ) U iU ) U U ,i b C U ’ U) - J %J Jt . i- r w c o U )a 4 0 0 W y O Q U I U 0 Ø 4w a U) C a a a a o c U U ,C C 0 U )b 0’ U ’ U ’ U ’ U ’ & P J C % J a w k o a U ’ a a a L n w w w U ’ U ’ t a U ) !“ !‘- ‘ t ! ° N’ N’ 9’ 9’ 9’ 9’ ‘ 9’ N’ N’ 9’ 9’ !“ “ i O 9 9 9, N’ N’ 9’ 9 RI a % I — J a c c %J % i V i v i v i a M i-’ . — J — i — i J o e i v i v i a a — i — i v i - & I- h v i v i v i v i — i — i v i v i v iM Q % Q v i a ’ c c i3 iv i v i v i a o a ’ v i a o M a %J U I I M M P U Ja ’ U’ a w o o v i L O W U’ v ia v i M W v i M M ‘ i 4a W L M M I- A v iv iw v i U ’ a I )M M I- I. A W c C v i’ 4 i ’ c c v iv i% d c c k f l I I b b p p p p p p p p p p p p b p p p p p p p p p p p p p p p p b p b b b b p p p p p p p p p p 6 p b b cc 0 a ’ a v i v i v i o c c v i c o v i 0 0 , n 0 • t t i ’ t t t t t t t ’ o t t t t t t t t . t t t t t t t t t t t t t t t t t t t f l O Q Q f lQ f lQ b n Q n pa pa pa 0 0 0 0 0 0 0 0 0 0 P4 P4 P4 P4 P4 P4 P4 p4 pa pa pa pa pa pa pa pa ia pa pa pa pa pa pa pa pa pa pa pa pa pa pa pa pa pa pa pa pa pa Li i a u l Li i v i Li i Li i Li i Li i U iu i a ” ti i a a a a a a a a it it LD L L 40 it Lu L O W 0 0 0 0 0 0 0 0 0 0 0 ö pa pa pa v i v i P4 P4 P4 pa pa pa pa p4 P4 P4 P4 pa pa pa pa p P4 p4 P4 P4 P4 pa pa pa pa Lu Lu Lu Lu P4 P4 P4 p4 pa pa pa pa Lu Lu P4 P4 P4 P4 pa pa %4 P4 P4 0 ‘ - 4 ‘- 4 P4 P4 Lii - J P4 a Lii ki ‘ - - J P4 LU ib it ‘ J “ 4 P4 P4 “ 4 44 P4 P4 Li i Li i 0 0 “ 4 ‘ - 4 P4 P4 ‘ - 4 ‘ - 4 P4 P4 P4 ‘ - J ‘ ‘ i a ‘ - 4 C i L u a C i v i v i v i v i o L u o a o o v i o v i v i C i v i L b 4 L , 1 a C i a a a ‘ 41 a a v i v i P 4P 4 P4 pa pa Co Co U 1 Q Lb Lb - - U lv i tO L D Co O ’4 - U i U lL u Lu Ci C iU 1 Lii L il a Lu Lu J ii 91 3 0 Lu Li i Li i Li i a i 40 r v i N N 9) P pa pa a a N N N N N to 9 9J 9’ 9 P P t’ hh P P P - 4 Lu Li i 0 tO 0 0 0 pa pa P4 pa Ci o a n Lb j %J ‘ .4 Lb it pa P4 ‘ .4 Lb ‘ .4 a P4 Lu Ci Ci 0 pa C i - J Lu Lu ‘ .4 ‘ .1 Lu Lu c c Ci P4 Pa pa a v i Lu Ci pa v i Lu it Ci - to v i a v i , , , ‘ . cc v i Lu v i to Ci w pa . I CC pa 0 v i a co P4 v i e i v i cc a pa Ci a c c v i - Lu ha pa pa - 4 co Lu v i pa Ci P4 Ci Ci a it a Co P4 Ci Co pa a a Ci - J , , , ‘ . pa Lu c c v i v i Ci P4 P4 Lb 0 Co 0 pa p-L P4 Lb pa pa 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 pa pa pa pa o o o o o o o o o a a a a o pa p a o o o o o o o pa pa pa pa pa pa 0 0 0 0 0 0 P pa pa i- ,’ a 0 0 0 w pa pa - i cc P4 P4 ‘ - 1 - 4 Lu a pa P4 ‘ - P Co v i v i u a u . a pa pa pa ‘ 0 iD v i v i Lu ‘ .u pa pa it L P4 P4 0 0 C i Ci Lu Lu pa pa - - J pa pa ‘ .1 Ci Lu Lu pa a a a v ’ i p a p a C iU ’ cc P4 C O pa L o w “ 4 — 1 0 0 a cc to - - — I - J o p a w a ci i ‘ - 4 iQ to p a 0 0 pa !‘ - w a v i P4 pa P4 a - J “ 4 4 ‘ .1 0 pa v i Lu 0 Co ‘ - 4 0 Co Lu Co Ci ‘ - 4 Co pa a Lu Li i a pa P4 P4 40 Ci Ci Ci pa ‘ - 4 pa Co pa 0 ‘ - 4 40 Co v i Lu Li i 0 Co pa pa a Co pa a ‘ - 4 Co 0 0 - a IL S 1 U IU IU IU I t t t t t t t t t t t t t t t t t t f i f i n n n n n f l Q O n n n f if i f in i i i i i i i i i I M M M M N M M M M 4 J - . j ’ , j UI U I U i U i U I U I UI U I U I U I U I U I a V i a in tn a in a a a M M M M J M a a M io J a w % .4 c D tD a a U I W U U I U I W W W a th L D U I L n O U I w w r f l a c c Q n P ta M P’ ° f• ‘ in a - J . a W U I 0O Q % J V i w a a o a t n % J e I W W a L D t f l 3 w c o w M U I w W v f l O J 4 J I 3 I J f l2 2 sn !- h’ b b 6 b 6 b 6 6 6 6 b 6 b 6 ,6 6 c o o b in ia iu o b b o O o w w a o i n m w w w w w u I w U J w a a Q o ,% r n w p 3 o w w o a o a w a o a a w c t , . , J a November 2006 runs Aim Profile ‘B”, chains aid sprockets dth’e-troia Target Coaditiora — — 4dievtd Carditons Or 0ta — nm no ii (mis) o 158 v (mis) ca TSR Avg Torque otorqueM Pawn Cli Initial load Mg 4mg Cd (nd/s) - (rad/s) — f?fraj {radiaec) (W) (t4) (N) - 1001 1.00 3.28 1.50 1.00 323 1.47 0.29 5.82 1.66 -0.0053 -6.95 25743 1002 lix) 3.331.? 1.00 51.1174 -016 636 -5.39 0.0157 -6.25 257.070.320 1003 lAX) 437 100 1.00 4.36 199 -1.63 735 -12.77 0.0605 -3.56 262.95 0.139 1004 lAX) 432 2.2 1.00 418 225 -2.91 8.78 -23.58 0.0613 -5.06 289.75 032 1005 1.00 5.47 2.50 1.00 5.84 249 -3.22 9.80 -3137 0.1003 -5.96 351.44 1. 1006 1.00 6.01 235 1.00 539 2.74 -2.16 10.71 -23.25 0.0739 -5.81 36432 1164 :i: lix) 636 !.00 1.00 5.51 297 -031 1172 -3.68 0.0117 -6.60 402.41 1.284 1010 1.50 4.20 1.25 1.50 4.11 125 0.21 7.40 2.11 -0.0020 - 1.50 4321.50 1.50 417148 -2.55 8.76 -12.58 0.0212 311 247.lsa 1042 150 5.74 1.7 154) 5.66 3 -457 10.19 46.63 0.0441 -146 351.02 0.498 1043 150 5.55 2.00 1.50 6.54 139 -7.43 11.77 -8738 0.0833 -4.71 459.28 0.65 1044 130 73322 1.50 7212.20 -913 12.97 -13434 0.1113 -0.74 461050.65 1045 1541 5.20230 1.50 1.11 254 -9.29 14.60 -13559 0.1214 -613 607.63036 11 1.50 9.0223 1.50 334273 -630 16.09 -10137 0.0960 -5.05 699220.99 1047 1.50 9343.00 1.50 1.14289 -238 16.44 -39.11 0.0371 -4.56 789.271. 1048 154) 10.55 32 1.50 10.24 3.13 1.56 18.44 28.68 -0.0272 1049 1.50 1141 330 150 1132 3.51 6.75 20.73 139.93 -0.1327 1081 1.75 5.74130 1.75 5301.49 -533 10.27 -54.76 0.0527 -710 467.72 0.487 :i: 1.75 53013 1.75 6.55174 -7.62 1197 -91.18 0.0545 -5.91 489.9803 1063 1.75 1552.00 125 739199 -1255 13.67 -17153 0.1025 -631 596360. 1064 1.73 5.6122 1.75 1332.23 -1435 1536 -220.41 0.1311 -7.77 690.20031 1065 1.75 9.57 2.50 1.75 9.47 248 -1324 17.05 -224.07 0.1341 -627 76536 0.797 1066 1.75 10.53 2.7 1.75 10.43 273 -935 18.77 -175.55 0.1051 -6.89 896.18 0334 1067 1.75 1143 3.00 125 1138 298 -4.27 20.48 -87.47 0.0524 -612 1035.48 107 2 50 7 DO 2 50 DO 2 50 1 7 I DO 2 1 50 7 DO — 1280 1.50 410 12 1.50 4.09 125 0.48 7.35 333 -0.0035 - 1241 1.50 432 1.50 1.50 431 150 -1.34 834 -1.136 0.0112 -5.63 403.01 0. 7 1242 150 5.74 1.7 1.50 5.63 172 -5.25 1014 -5317 0.0503 -7.00 38353 0.54 :i.: 1.50 536 100 130 632 139 -7.96 1173 -93.44 0.0884 -7.05 425.77 0. 1244 130 731 22 150 7.07 215 -1133 12.73 -149.42 01414 -732 486.53 0190 1245 1541 5.20 250 1.50 1.11 247 -1123 14.60 -164.05 0.1553 -7.03 560.49 0394 1245 130 9.02 23 1.50 134 273 -8.67 16.10 -139.52 0.1321 -6.48 636.23 0.902 1247 1.50 934 100 130 9.63 294 -5.01 17.34 -8634 0.0823 -734 740.01 1069 1245 1.50 10.65 52 1.50 10.45 3.19 4.93 16.80 -17.43 0.0165 — 1249 150 1143 3.50 1.50 1126 144 3.77 20.27 7638 4.0724 1080 14111b 2*182b 1083 1084 1085 1086 1087 1200 12.01 1202 1203 1204 1205 1206 1207 2.00 2.00 2.00 2.00 230 2.011 - 2.00 2.00 1.00 tAXI tAXI 1.00 1.00 1.00 1.00 1.00 5.47 6-55 736 £75 9.34 1034 12.03 13.12 2.73 323 333 437 432 5.47 531 6.55 1. I. z; 2. I 2. 2. 3. 1. 1. 1. I 2. I 2. 3.i 2.00 2.00 1.99 1.99 1.99 139 1.99 1.99 1.00 lix) 1.00 130 1.00 1.00 1.00 130 5.45 632 7.66 1.76 9.75 1034 1191 12.82 233 326 3.71 4.36 5.03 525 6.12 6.51 125 149 175 201 223 245 2.75 294 124 1.49 159 199 229 240 279 297 -1.97 -7.61 -1120 -1736 -1933 -17.36 -13.16 -1.57 1.04 0.54 4.47 -1.99 -332 4.22 -2.87 -1.38 9.81 11.73 13.79 15.77 17.54 19.32 21.44 23.08 4.91 5.88 6.68 7.85 9.05 945 1101 11.72 -19.31 -89.21 -153.09 -283.24 -346.03 -33911 -282.19 -3615 520 3.18 -3.13 -15.60 -31.83 -39.91 -3156 -16.23 0.0077 00358 0t6 0.1142 0.1391 0.1367 0.1136 0.0145 4.0162 4.0101 0.0099 0M95 0.1010 0.1266 0.1001 0.0515 4.99 -19.14 -17.45 -12.34 -15.19 434 -22.54 -22.06 -7.02 -6.80 -9.26 -7.74 -1.06 -7.94 -7.67 68.59 690.26 758.00 808.61 928.89 931.18 1324.84 1558.44 193.71 218.13 267.76 260.17 302.28 329.52 349.61 1-i 1. 0.614 01.94 0.I5 0.8Y 0.96 1M1 tILl! 170 November 2006 runs Arm Profile ‘ chains and sprocket EWe-train. Tirget Canditio’ AdiicMtd Cacidilicns Drag Data ninno is(rnhj ca TSR v(rnh) c 73W AvgTarqwe otarrzeM Pcwr Ck irñtiaIIcad Avg dra4 cii 12160 1.75 4.78 1.2 1.73 410 1.25 -0.10 8.64 -0.14 03905 — :i: 1.73 1741.50 1.75 5191.49 -2.88 1023 -29.47 0.0176 -14.86 378.160.602 1262 135 6301. 1.75 6.65 1.74 -8.13 1197 -9736 0.0582 -1641 597370.623 1.75 2.00 199 22.50 -14.93 639.71 0.666 ThE 1.75 5-6122 1.75 8542.23 -16.93 15.37 -260.19 01556 -16.95 7572.20.789 iii? 1.75 230 2.48 28.80 -15.40 807.67 0141 ilil 1.75 3 23 1.75 1138 2.59 -7.32 20.48 -150.00 0.0698 -15.12 930.94 0.969 12167 1.73 3.00 2.98 13.16 -17.30 1073.23 1.118 1280 200 5.47 12 2.00 5.45 125 -2.14 9.80 -2100 03984 — 1281 239 8.55130 2.00 5.52149 -7.61 11.74 -89.37 0.0358 -7.24 591.670.472 flfl 2.00 7.65 1. 2.00 7.66 1.75 -12.24 1178 -168.62 0.0676 -12.14 685.35 0.346 1283 2.00 5-73 2.00 2.00 8.19 1.88 -19.51 2473 -287.42 01154 -1103 765.73 0.611 1284 2.00 9.84 2.2 1.99 9.83 2.23 -2234 17.59 -405.90 0.1631 43.06 89737 .716 1285 2.00 10..94 2.50 1.99 10.64 2.44 -1836 19.15 -355.02 0.1459 -13.57 1004.41 0.849 1286 239 120323 1.99 11.232.5.8 -1721 20.21 -347.82 0.1399 -13.65 1127.640199 2.00 13.12 3.00 1.99 1337 rr -6.14 24.43 -149.93 0.0606 -1339 1405.58 1121 1288 239’ 8.56 . ‘AcA £JAI 6.49 2.49 -5-88 1169 -80.37 0.0322 1340 1.541 4.10 12 1.50 4.10 125 1.10 7.38 6.79 4.0085 1341 1.541 432 130 1.50 4.97 152 4.66 8.95 -5.93 0.0056 1342 1.50 5.74 I. 1.50 531 174 -1.52 10.29 -15.59 0.0147 1343 1.50 6.56 2.00 1.50 5-31 192 -5.15 1135 -58.52 0.0558 13.44 1.50’ 738 22 1.50 7.15 2.18 -9.60 12.87 -123.52 01169 1345 1.54) 8.20 250 1.50 5-10 2.47 -10-66 2458 -155.51 0.1472 1346 1.54) 9.02 2. 1.50 9.17 2.80 -9.10 fl51 -150.19 0.1423 -ii:y 1.54) 9.84 3.00 1.50 9.78 2.99 -163 17.51 -99.08 0.0939 1348 1.50 10i65 3.2 1.50 10.60 123 -1.89 19.07 -36.05 0.0342 1349 1.50 11.48 330 1.50 1144 149 2.61 20.59 53.57 4.0509 1380 2.00 547 1.2 2.00 5.45 125 -0.54 9.81 -5.30 01)021 1381 2.00’ 6.56 130 2.00 653 1.50 -2.97 1176 -3433 0.0140 1382 2.00 7.66 17 2.00 710 139 -7.80 14.03 -1.09.46 0.0639 1383 2.00 5-73 2. 2.00 5-74 2.00 -15.02 15.73 -23623 0.0949 1384 239’ 9.84 22 2.00 934 223 -2172 17.54 -380.97 01530 13.85 239 10.94 250 1.99 1039 2.47 -2165 19.42 -420.36 01689 1326 239 1203 2. 1.99 11.89 273 -1723 21.40 -558.75 0.1483 1387 2.00 13.12 3.00 139 13.30 105 -9.43 23.94 -223.80 0.0908 0.. 1400 1.00 273 12 1.00 2.67 In 1.37 4.81 639 4.0209 iIii 139 3.28 150 1.00 109 14.1 0.46 5.56 2.58 4.0082 1402 139 3.13 1 1.00 177 172 0.40 6.79 219 4.0085 Thi 1.00 437 2.00 1.00 4.39 200 4.63 7.89 -4.98 0.0158 1404 1.00 432 22 1.00 4.90 224 -1.20 8.82 -9.68 0.0307 1405 139 5.47 250 1.00 544 248 -0.34 9.79 -3.32 0.0105 1406 1.00 6.01 2.7 1.00 6.01 274 032 1012 332 4.0112 1407 1.00 6.56 3.00 1.00 653 298 1.75 1176 2035 4.0652. 1408 1.00 7.21 32 1.00 7.11 125 338 12.80 50.87 -0.1615 1409 1.00 7.65 3.50 1.00 7.66 150 628 1178 86.50 4.2746 1460 1.75 4.78 1.2 1.75 4.66 1.22 015 8.40 129 43908 1461 1.75 5.74 1.50 1.75 3.68 1.49 -3.07 10.23 -31.40 0.0188 1462 1.75 6.70 1.7 1.75 6.62 173 -3.60 11.91 -4293 0.0257 1463 1.75 7.66 2.00 1.75 7.56 1.98 -750 1161 -102.07 0.0610 1464 1.75 8.51 22 1.75 730 201 -7.85 1185 -108.68 0.0650 14Mb 1.75 8.61 2.23 1.75 8.44 221 -7.75 15.19 -117.71 0.0708 1.75 937 230 1.75 930 259 4.29 1712 -11209 0.0571 1466 1.75 10.53 2.7 1.75 1031 270 -336 18.56 -66.00 0.0395 171 August/September 2007 Tests — Free Stream, Gearbox Drive-train r Target Ccndhons Achieved ccndi:ons Measured Data ninno v w TSR v RPM w TSR AvgTorque Power Ck (TWSI (rad/s) I) ciern 1W) Exp I . 11 1.50 5.74 1.7! 1.50 €470 5.73 [75 5.55 3t55 0.032 Vt nyc 12 1.50 7.38 22! 1.50 7040 7.57 225 15.37 113.31 0107 13 1.50 SiC 2.53 1.50 77.90 8.15 242 1617 131.91 0125 Nov20O6aems(ProfileB) 14 1.50 922 23! 1.50 5560 815 273 12.17 102.09 0.10! IAoA=0 15 1.50 7.26 21! 1.50 7030 7.35 224 15.50 114.64 0.109 34-021 blades 1! 1.50 8.20 252 1.50 79.30 822 253 1531 127.14 0.122 17 1.50 9.22 2.7! 1.50 5560 8.95 273 11.56 106.31 0102 30 2.00 9.54 2.2! 2CC 9250 9.55 227 33.95 322.59 0.131 3 2.00 9.24 2.2! iCC 9250 9.71 222 35.10 34193 0.135 32 2.00 924 2.2! 2CC 9180 931 222 3428 331.22 0132 33 2.00 9.24 22! iCC 9170 9.72 222 3317 329.50 0 131 34 2.00 1034 2.52 2CC 12192 10.77 24! 32.45 349.49 0.139 35 2.00 12.23 22! 2CC 113.22 11.2! 271 24.32 236.15 0.11! Exp End Plates NACA 0012 40 1.50 4.10 1.2! 1.50 39.10 429 [2! -1.28 -5.21 -0.02!41 1.50 4.92 1.52 1.50 47CC 4.92 1.52 5.11 2!.14 00242 Nov2006atms (Profile B) 42 1.50 5.74 135 1.50 64.70 513 1.7! 63! 35.24 0034 AoA=0 43 1.50 6.66 2.02 1.50 5150 6.54 [33 9.72 63.59 0062 63.4-021 blades 44 1.50 7.38 2.2! 1.50 7040 7.3’ 225 1634 124.82 0.115 45 1.50 8.20 2.50 1.50 77.60 8.14 245 18.15 14710 0.140 46 1.50 9.22 2.7! 1.50 25.70 8.2’ 273 142! 125.03 0.119 47 1.50 9_4 302 1.50 93.30 9.77 235 757 73.92 0072 43 1.50 10.56 32! 1.50 101.22 10.59 3.23 0.29 3.27 0.003 go 2.00 5.47 12! 2.50 52.10 5.4! [2! 2.42 13.20 000! Cl 2.00 6.!€. 1.5: ICC €2.70 6.!€ [50 15.61 101.79 CiG4 @2 2.00 7.56 17! 2CC 7180 7.52 1.74 19.12 145.59 005! g3 2.00 8.75 202 200 6150 8.54 12.7 2537 222.50 0.08! @4 2.00 954 22! 2CC 9150 9.5! 221 38.55 375.13 0152 @5 2.00 1034 2.50 2CC 10252 10’S 24! 38.33 39190 0155 CC 2.00 1223 2.75 iCC 1132D II.!! 271 27.23 32223 0.129 Exp End Plater Circular 80 1.50 4.10 12! 1.50 3903 425 [24 -uS -4.82 -GODS - 81 1.50 432 1.53 1.50 47,CC 4.22 [50 5.11 2514 00243 Nov2006ms (Profile B) 82 1.50 5.74 1.7! 1.50 €4.80 5.71 1.74 6.2! 3!.59 0034 AoAO 83 1.50 636 202 1.50 5260 6.!! 200 9.02 0.05€ 63.4-021 blades 84 1.50 735 22! 1.50 7030 735 224 16.14 112.76 0112 85 1.50 820 2.52 tSO 77.90 8.1! 242 17.79 145.25 2.13’ ec 1.50 9.22 2.7! 1.50 2560 8.9€ 273 13.29 119t7 0.113 87 1.50 9.24 302 1.50 93.30 9.77 235 5.4’ 53.42 0.052 8e 1.50 10.55 325 1.50 101.12 1O.!8 3.23 -3.22 -3170 -0031 100 2.00 5.47 125 2CC €2.10 5.45 1.2! 4.42 24.10 0.010 101 2.00 6.56 1.50 2CC 52.53 6.55 1.50 1135 72.37 0.031 102 2.00 7.56 17! 2CC 7250 7.52 1.74 152.2 115.97 0.055 103 2.00 8.75 2.02 2CC 62.70 8.5€ [38 22.76 197.01 0.079 10.1 2.00 9.54 225 2CC 92.50 9.55 221 33.53 325.59 0130 106 2.00 10.94 2.50 iCC 122.70 10.75 240 3374 362.56 0.14! 106 2.00 12.23 2.7! 2CC 112.9: 11.82 270 25.24 29526 0.119 110 2.20 5.47 12! 2CC €210 5.45 [2! 1.12 6.11 0.002 030 !0PI10 Repeatwloendptaes 111 1.50 636 2.02 1.50 5260 63! 230 6.37 54.84 0.052 172 1.50 7.38 22! 1.50 70.50 7.35 225 152€ 113.34 0.107 113 1.50 8.20 2.52 1.50 75CC 8.15 242 16.68 13536 0.128 114 1.50 9.22 2.7! 1.50 55.50 8.92 274 12.33 11173 0.10€ 0.20 #DW’0 Repeatwlo end plates 115 2.00 835 2.00 200 52.70 8.5€ 138 23.15 250.38 0.050 110 2.00 9.54 22! 2CC 9290 9.72 222 33.94 330.02 0132 117 2.00 10.94 2.52 2CC 10232 10.75 24! 32.13 345.37 0.136 118 2.00 12.03 2.7! 2CC 11310 11.54 271 24.33 265.01 0.115 172 August/September 2007 Tests — Free Stream Ininnol v I w ITSRI v IRPMI w ITSRIAvciTornuelPowerI Ck 4 £ 121 6.50 4.92 1.50 1.50 2710 423 LEG -340 -15.75 -0215 WEx15” NovemberArm On 120 1.50 4.10 128 1.50 3910 4.09 1.25 -250 -1146 -0211 (BLADES REMOVED) 122 1.50 574 1.75 1.50 84.70 5.73 1.75 -4.16 -2152 -0.223 I Nov2006arrns(ProfiIeB) 123 1.50 5.56 2.00 1.50 62.50 5.55 2C%Z -4.54 -30.40 -0229 AoA=0 124 1.50 7,36 225 1.50 70.50 7.35 225 -5.52 -41.47 -0:3; 634-021 blades 125 1.50 5.20 2.50 1.50 75.40 5.21 2.50 -5.40 -52.52 -0.253 129 1.50 902 2.75 1.50 86.10 921 2.75 -719 -55.50 -0 253 127 1.50 9.84 320 1.50 93.90 913 3.02 -5.53 -54.52 -0.283 : 128 1.50 10.55 325 1.50 121.70 1054 124 -9.78 -104.10 -0395 125 1.50 1145. 3.50 1.50 18940 1145 3.42 -10.98 -125.73 -0119 200 #DPtO! 140 2.00 5.47 125 2.00 82.20 5.45 1.25 -4.25 -23.22 -0.209 141 2.00 5.36 1.50 2.00 52.50 5.54 LEG -5.15 -3319 -0.214 142 2.00 7.65 1.75 2.00 7330 7.54 1.75 -5.41 -4818 -0223 143 2.00 5.75 2.20 2.00 83.50 3.72 tOO -7.55 -56.70 -0.227 144 2.00 934 225 2.00 93.50 912 224 -915 -9013 -0.235 145 2.00 1094 2.50 2.00 124.20 1059 242 -10.5.3 -117.59 -0.247 149 2.00 12.33 2.78 2.00 114.50 1135. 2.74 -12.59 -150.58 -0.262 Exp - 108 1.50 4.10 125 1.50 39.30 4.05 1.24 -2.57 -10.49 -0.213 A Single Blade, B Arms 1’1 1.50 4.92 1.50 1.50 47.30 422 LEG -0.59 -3.39 -0.2035 ov2006arms (Profile B) 102 1.50 5.74 1.75 1.50 5450 5.71 1.74 2.85 14.74 2014 103 1.50 6.35 2.20 1.50 6250 6.55 2.02 405 25.73 2.025 4-021 blades 194 1.50 7.35 2.25 1.50 7040 7.37 225 6.25 35 75 2 037 105 1.50 5.20 2.50 1.50 75.20 5 15 242 6.85 69.95 2 056 100 1.50 902 2.78 1.50 8550 595 2.74 2.32 83.52 2279 107 1.50 9.84 320 1.50 93.50 2.79 298 585 85.90 2 052 108 1.50 1055 3.25 1.50 121.50 1052 3.24 7.64 81 15 2 077 172 1.50 S.20 2.50 1.50 75.20 5.15 2.42 6.47 62.33 2 056 200 #DPfl! 100 2.00 5.47 1.28 2.00 52.20 5.45 1.25 -2.16 -‘1.50 -0.205 iti 2.00 5.85 1.50 2.00 52.50 5.85 150 235 2.15 2 001 182 2.00 7.65 1.78 2.00 72.90 7.63 1.74 585 42.35 2 017 183 2.00 5.75 220 2.00 8310 5.70 1.92 792 5285 2024 184 2.00 9.84 2.25 2.00 93.40 2.75 223 13.12 125.25 2 051 185 2.00 1014 2.50 2.00 123.70 10.55 245 15.32 17714 071 las 2.00 1225 2.75 2.00 i’SdO 11.57 2.71 1525 19229 2 077 NewArms: 3,OAoA= 0 7 Arns (profile C) oA = 0 1-021 blades Ifree-stream InmnoI v w ITSR v IRPMI w TSRAvgTorquePower Ck 201 1.50 4.92 1.50 1.50 2510 4.91 LEG 15.68 75.97 2.073 202 6.50 5.86 220 1.50 62.40 6.53 1.92 21.55 14051 2 133 203 1.50 7.35 2.25 1.50 7050 7.3€ 2.24 30.60 225.16 2213 284 1.50 5.20 2.50 1.50 7520 5.15 2.42 3375 275.78 3251 209 1.50 9.84 320 1.50 93.40 9.75 205 25.82 275.51 2 253 287 1.50 11.48 3.50 1.50 136.70 11.35 3.47 1720 195.59 3.158 293.1 6.50 7.35 2-25 1.50 72.20 7.55 230 29.19 220.59 3206 294.1 1.50 5.20 2.50 1.50 7540 521 250 32.72 263.50 2.254 293.2 1.50 735 2.25 1.50 7020 7.35 234 30.17 22118 2.209 294.2 1.50 6.20 2.50 1.50 7730 LIE 2.42 33.01 269.1$ 2254 2921 1.50 5.35 2.20 1.50 52.40 5.53 1.92 2129 139.25 3131 294.3 1.50 5.20 2.50 1.50 75.50 523 251 34.35 282.57 9.257 298.1 1.50 202 2.78 1.50 85.50 5.95 273 32.14 28712 2.272 293.3 1.50 7.35 225 1.50 7050 7.39 225 29.47 217.77 3206 294.4 6.50 520 2.50 1.50 77.90 5.15 242 32.75 25727 2253 2.00 #D1V•0 173 ist /September 2007 Tests — Free Stream 211 2.00 656 1.50 2.00 62.50 8.54 1.50 2818 184.21 0.073 212 2.00 6.75 200 2.00 82.70 6.66 1.98 44.9 367.67 0.155 2121 2.00 6.75 2.00 2.00 82.80 6.67 1.98 46.15 399.95 0.159 213 2.00 84 2.25 2.00 92.70 9.70 222 66.51 54.2.29 0.219 214 2.00 10:34 2.50 2CC 102.92 10.77 2.46 64.51 604.79 0.277 215 2.00 12.03 2.75 2CC 11100 11.72 2.88 59J4 700.31 0279 218 2.00 1112 3CC’ 2CC 0.00 0.50 0.00 0.000 217 2.00 15.31 3.50 2CC 0.00 0.00 0.00 0.000 213.1 200 9.84 2CC 92CC 9.89 222 56.26 535.59 0214 214.1 2.00 10.34 2.50 2.00 10320 10.80 247 8199 891.19 0276 213.2 200 9.64 2.25 2.00 92.90 9.72 222 56.06 545.30 0217 214.2 2.00 10.94 2.50 2.00 103.20 10.80 247 84.24 693.89 0277 2143 200 10:38 2.50 2.00 10310 10.80 247 63.25 663.20 0272 2121 200 8.75 200 2.00 82.80 6.67 1.98 48.15 399.95 0.159 2144 2.00 10.38 2.50 2.00 ICC.10 10.79 2.47 8417 692.47 0278 215.1 2.00 12.03 2.75 200 112.00 11.82 270 61.49 728.62 0.290 213.3 2.00 9.84 2.25 2.00 92.80 9.71 222 56.09 544.81 0.217 214.5 2.00 10.38 2.50 2CC 103.00 10.78 246 6464 585 0276 215.5 200 1203 2.75 2CC 112.00 11.72 288 5934 700.31 0279 Shaft Fairing 2007 Arms (profile C) L0A =0 634-021 blades tree-stream Irunnol v I w ITSRI v IRPMI w ITSRlAvaTorciuelpowerl Ck 221 1..50 4.92 1.50 1.50 47.10 4.93 1.50 11.44 56.40 0.053 222 ISO 5.56 2CC 1.50 62.50 8.54 1.99 19,11 125.01 0.118 223 1.50 7.38 225 1.50 70.30 7.36 224 28.55 210.07 0.199 224 1.50 8.20 250 1.50 77.90 8.15 249 31.08 253.41 0239 225 1.50 9.02 2.75 1.50 85.80 8.98 274 30CC 289.41 0.255 228 1.50 9.84 ICC 1.50 93.30 9.77 298 24.8.5 24187 0.229 227 1.50 11.48 3.50 1.50 106.70 11.38 3.47 14.92 189.75 0.180 2241 1.50 8.20 2.50 10 77.90 8.15 249 30.41 247.95 0234 0.00 401 V0! 231 2.00 5.56 1.50 2.00 62.50 8.54 15) 28.19 171.33 0.088 232 2.00 815 2CC 200 83.20 8.71 1.99 41.79 383.92 0.145 233 200 9.84 225 2CC 92.70 9.70 222 5t94 503.95 0201 234 2.00 10.34 2.50 2CC 103.20 10.80 247 63.81 689.25 0275 235 200 12.03 2.75 ICC 112.60 11.81 270 57.85 683.35 0.272 txp NewArms: 2 ONLY 2 arms only 2007 Arms (profile C) AoA = 0 634-021 blades tree-stream Irunnol v w ITSRI v RPMI w I TSR AvgTorquePowerj Ck 241 1.50 4.92 1.50 1.50 46.10 4.83 1.47 15.78 78.94 0.072 242 1.50 6.56 2CC 1.50 6200 6.49 1.98 21.35 142.44 0.135 243 150 7.38 2.25 1.50 70.30 7.36 224 2993 220.23 0.208 244 1.50 8.20 150 1.50 77.80 8.14 248 33.27 270.92 0256 245 150 9.02 275 1.50 85.40 8.94 272 35.25 320.45 ‘1303 248 1.50 9.84 3.CC 1.50 9320 9.75 297 3291 321.03 0303 247 1.50 11.48 150 1.50 106.50 11.36 3.46 26.53 30128 0285 243.1 tso 7.38 225 1.50 70.20 7.35 224 31.63 232.40 0220 244.1 ‘150 8.20 250 1.50 77.80 8.14 248 34.04 277.19 0282 0.00 $01VOl 251 200 8.56 1.50 2.00 62.30 8.52 1.49 3129 294.89 0.062 252 200 8.75 200 2.00 82.60 8.85 1.98 47.98 414.84 0.185 253 200 9.84 225 2.00 9290 9.72 222 59.27 578.31 0230 254 2S0 10.38 150 2CC 104.00 10.89 249 67.82 73824 0294 255 2.00 12.03 2.75 2CC 113.30 11.86 271 68.11 783.98 0.313 253.1 200 9.84 225 200 9290 9.72 222 6327 61521 0245 2541 200 10.94 2.50 2.00 103.00 10.87 249 69.30 753.63 0.300 174 August/September 2007 Tests — Free Stream Cantered Blade: MA =0 2007 Arms (profile C) AoA =0 34-421 blades free-stream runnol v w ITSRI v RPM w TSP AvgTorquePower Ck 281 1.50 4.92 1.50 1.50 47.20 4.94 1.51 15.03 74.25 0.070 282 1.50 6.56 2.00 1.50 62.50 6.54 1.99 23.36 156.87 0.148 283 1.50 7.38 2.25 1.50 70.50 7.39 2.25 33.00 243.85 0.230 284 1.50 8.20 250 1.50 78.00 8.16 249 34.52 262.64 0267 285 1.50 9.02 2.75 1.50 85.9C 8.99 2.74 33.53 302.00 0265 286 1.50 9.84 2.00 1.50 93.40 9.78 2.98 291.5 291.81 0.278 287 1.50 11.48 3.52 1.50 !C€52 11.37 3.46 191 225.74 0213 283.1 iso 7.38 225 1.50 71.00 743 2.27 3201 245.31 0232 294.1 1.50 6.20 2.50 1.50 77.9C 6.15 2.49 34.72 263.09 0268 0.00 UDIV,’O! 291 2.00 5.56 1.50 2CC 62.50 5.54 1.52 31.50 20e06 0.062 292 2.00 6.75 2.00 2CC 82.80 8.87 1.98 5233 453.51 0.181 293 2.00 9.84 2.25 2CC 9210 9.54 220 81.13 569.85 0235 294 2.00 1o34 2.50 2.CC 103.00 10.78 246 88.59 717.86 0268 295 2.00 12.03. 2.75 2CC 113.10 11.84 271 83.18 747.86 0.298 293.1 2.00 9.84 2.25 2.CC 92.70 9.70 222 60.88 568.75 0235 294.1 2.00 10.34 250 2CC ‘02.60 10.78 246 85.25 702.18 0260Irunnoi V j w ITSRI V RpM JJSF?J vg iorque vower i UK Cambered Bladr MA — & 301 1.50 4.92 1.50 1.50 47.10 4.93 1.52 10.12 4929 0.047‘ — 302 150 6.56 2.00 1.50 8240 6.53 1.99 24.53 11021 0.151 )7Arnis (profile C) 303 1.50 7.38 225 1.50 70.20 7.35 224 31.57 231.95 0219 oA= 6 304 1.50 8.20 2.50 1.50 77.90 8.15 249 3817 318.93 0299 1-421 blades 305 1.50 9.02 2.75 1.50 85.60 8.96 273 37.84 337.23 0.319 free-stream 306 1.50 9.84 3CC 1.50 93.30 9.77 298 34.00 332.02 0.314 307 1.50 11.48 2.50 1.50 108.50 11.3’S 3.46 2213 258.13 0244 324.1 1.50 6.20 2.50 1.50 77.80 8.14 248 37.33 309.36 0.292 0.00 *0IV0! 311 2.00 6.56 1.50 2CC 60.10 6.29 1.44 28.50 17926 0.071 312 2.00 6.75 2CC 2.CC 84.40 6.83 202 51CC 450.53 0.160 313 200 9.84 225 2.CC 9290 9.72 222 81.54 539.39 0239 314 2.00 10.34 250 2.CC C3.10 10.79 247 891! 751.93 0.300 315 2.00 1203 275 2.00 “210 11.79 269 82.57 611.67 0.324 314.1 2.00 10.34 2.50 2CC 103.20 10.80 247 70.22 758.49 0.302Inwno v I w ITSRI v I RPM w I TSR I AvgTorquelPowerl Ck 3 New Arms Only No blades ice-stream 341 1.50 4.92 1.50 1.50 47.00 4.92 1.52 -3.40 -18.73 -0.016 342 tso 6.56 200 1.50 62.50 6.54 1.99 4.22 -27.61 -0.026 343 1.50 7.38 225 1.50 70.40 7.37 225 4.39 -32.35 -0.031 344 1.50 6.20 2.50 1.50 78.20 8.18 249 -5.01 -4’ .01 -0.039 345 1.50 9.02 2.75 1.50 85.90 6.99 274 -5.55 -49.90 -0.047 348 1.50 9.84 3.00 1.50 92.90 9.83 3.00 -8.19 -10.84 -0.057 347 1.50 11.48 320 1.50 10920 11.43 3.48 -7.38 -84.35 -0.080 0.00 #011/,0! 351 2.00 6.66 1.50 2CC 82.80 6.66 1.50 4.22 -27.85 -0.011 352 2.00 8.76 2.00 2CC 83.20 8.71 1.99 -5.47 47.63 -0.019 353 200 9.84 225 2CC 93.80 9.82 224 -5.49 -63.72 -0.025 354 2.00 10.94 250 2CC 104.20 10.91 249 -7.51 -61.91 -0.033 355 2.00 1203 275 2CC 114.20 11.35 273 4.31 -93.33 -0.040 175 August/September 2007 Tests — Free Stream Irunnol v w ITSRI v I RPM! w TSR IAvgTorquelPowerl Ck S. ‘ lade 361 1.50 4.02 1.50 1.50 47.00 4.02 1.50 2.86 14.01 0.013iflQ 5 382 1.50 6.56 2.00 1.50 58.30 6.10 1.66 8.56 62.23 0.049 arms profile C 363 1.50 7.38 2.26 1.50 73.50 7.69 2.34 11.97 91.32 0.986 1634-021 blade 364 1.50 0.20 2.50 1.50 78.10 8.17 2.49 16.38 125.72 0.119 lee-stream 385 1.50 9.02 2.76 1.50 85.00 8.09 2.74 17.82 180.22 0.151 306 1.50 9.84 3.00 1.50 93.50 9.79 2.98 19.17 187.50 0.177 367 1.50 11.48 3.50 1.50 10883 11.39 3.47 18.09 206-DO 0.195 368 1.50 7.38 2.25 1.50 70.43 7.37 2.25 11.02 87.13 0.382 369 1.50 820 2.50 1.50 78.20 8.18 2.49 15.01 122.86 0.116 0.00 #DIV:C! 371 2.00 6.56 1.50 2.00 62.50 6.54 1.50 745 4814 0.019 372 2.00 8.76 2.30 2.00 83.00 0.89 1.99 15.82 135.78 0.064 373 2.00 9.84 2.25 2.30 03.43 9.78 2.23 24.12 235.70 0.394 374 2.00 1094 2.50 2.20 103.60 1Q54 2.48 27.50 298.20 0.119 375 2.00 12.03 2.76 2.00 113.20 11.65 2.71 29.63 351.37 0.140 376 2.00 10.94 2.50 2.00 103.60 10.63 2.48 27.06 302.80 0.121 InnoI v w ITSRI v RPM I w TSR fAvgTorquePower Ck Exp F One Blade 381 1.50 4.02 1.50 1.50 47.00 4.02 1.50 1.37 6.74 0.006airing. 382 1.50 6.55 2.00 1.50 52.50 6.54 1.99 8.31 52.40 0.060 New Arms:3 383 1.50 7.38 225 1.50 70.40 7.37 2.25 11.27 83.04 0.07-9 A0A=0 384 1.50 820 2.50 1.50 7820 8.18 2.49 14.60 119.50 0.113 634-021 blades 365 1.50 9.02 2.75 1.50 65.00 0.99 2.74 17.05 15320 0.145 free-stream 386 1.50 9.84 2.30 1.50 03.50 9.80 2.99 10.00 185.16 0.175 387 1.50 11.48 2.50 1.50 108.70 11.38 3.47 18.43 209.58 0.198 384.1 1.50 820 2.50 1.50 78.10 8.17 2.49 14.29 116.81 0.113 0.00 #0MG! 391 2.00 6.56 1.50 2.00 62.60 5.55 1.60 3.30 21.52 0.009 392 2.00 8.75 2.00 2.3C 82.00 8.80 1.98 14.41 125.03 0.060 393 2.00 9.94 2.25 2.30 03.42 9.78 2.23 22.83 232.96 0.093 394 2.00 10.9k 2.50 2.30 103.63 10.64 2.48 2727 235.70 0.119 386 2.00 12.03 2.75 2.30 11323 11.88 2.71 29.72 352.44 0.141 176 177 August/September 2007 Tests — Ducted, Gearbox drive-train Tpet Achieved conditions Measured Data nirno v [TSR v RPM w TSR Torque Power Ck (mis) I mls) (rads) (Nm) tW Ducts+4Bumns 400 1.60 ‘.25 1.53 4.35 1.24 8.10 1712 2.C’0 401 1.60 .W t.50 .91 1.53 449 42.77 2.040 i PrafileCarms 402 1.60 ‘.7! 1.53 on 1.74 13.91 70.02 2.C!AoAO 403 1.60 2.CC 1.52 6.24• 1.22 23.90 S-an 244634-021 blades 404 1.60 2.22 1.53 7.20 2.24 2&90 .5449 2.72 400 1.60 2.20 1.52 9.11 2.43 33.3€ 23202 3.254 400 1.60 2.72 1.52 9.92 2.72 4’.72 42025 2.443 407 1.60 3cC 1.52 931 227 42.75 41525 2.253 40* tio 3-2! 1.13 13.05 12* 34.55 30045 2.341 403 1.60 230 1.50 11.31 2.49 2e.34 252.25 2.270 410 1.60 3.? 133 12.0- 3.63 :8.34 22713 2.2-9 431.1 1.60 *0 *40 4.93 453 932 42.53 2.C41 4n 1.60 230 1.53 6.54 1.32 22.20 4159 2.32 4311 1.60 2.20 1.53 9.14 2.45 31.9C 29274 3.242 437.1 1.60 3CC 1.53 3.7C 2.25 43.30 42047 2.297 436.1 1.60 3.90 133 11.26 2.40 :6.14 294.73 2.290 420 2.00 2.32 0.0-3 CCC 2.2CC 421 2.00 ‘30 2.33 0.03 CCC 2.CCC 422 2.00 1.79 2.32 72.; 7.SC 1.74 44.73 34CC? 3.3€ 420 2.00 1CC 2.33 23.3 9.72 1.29 53.25 431 33 2.73 4 2.00 2.22 2.32 5C. 947 2.3 53.32 470.14 0.’90 425 2.00 190 233 109.7 II.’? 2.55 31.92 91205 2.304 420 2.00 172 2.33 112.7 11.62 2.73 3’.22 553.53 2.250 429.1 2.00 2.72 2.33 12233 il.9C 2.73 51.20 50413 2.290 Exp Duct 2bum 440 1.60 .20 1.53 47.33 4.92 1.53 v.28 2425 2.C23ps 441 1.60 2.00 1.52 62.53 9.54 1.22 2942 90C1 2.016 ProflleCanns 442 1.60 2.2.5 1.53 7043 7.37 2.25 I.26 3.3C45 2.2-9 AoAO 440 1.60 2.20 1.53 77.52 9.12 2.40 L4.90 20422 2.344 634-021 blades 444 1.60 2.71 1.53 22.33 9.91 2.72 51.37 42227 2.434 442 1.60 2CC 1.52 51.22 936 2.27 45.32 421.12 2.425 Bump din. opposite 443 1.60 2.25 1.52 ‘CC.53 12.53 121 39.93 405131 3.397 turbine spinningtowards 1.60 1.20 1.53 3*43 11.40 2.42 29.74 24007 3.322 443-1 1.60 2.90 1.53 77.53 9.11 2.4-0 46.30 270.73 2.350 Duct No Bumps 490 2.00 .72 2.33 72.93 7.52 474 49.36 37032 2.’6C 431 zoo 230 2.33 23.32 9.95 1.29 57.33 49332 2.55 402 00 2.22 2.33 90.12 9.22 2.13 53.3 54171 2.224 zoo 2.90 2.33 12.22 13.74 2.45 39.SC 50103 2.391 zoo 2.72 2.32 12.52 11.73 2.73 93.36 ICC.74 3.433 ‘rofile C arms AoAO 1-021 blades 480 1.60 .20 1.0 37.1 4.91 1.5-3 5.73 42.03 2.242 431 1.60 ICC 1.53 02.53 9.54 1.22 jOtS 37.37 3.97 432 1.60 2.2.2 1.53 70.12 7.24 124 34.51 253.33 2.230 483 1.60 2.20 1.52 72.32 7.95 142 47.12 27463 2.264 484 1.60 2.72 1.53 5143 9.63 1C3 54.7’ 443.39 2.441 433 1.60 2.OC 1.53 02.33 932 2.25 51.51 6CC 21 2.473 1.60 1.22 1.53 CC.52 l3.62 12$ 439? 401.72 2.420 437 1.60 3.50 1.53 -Cc522 11.33 145 1&.97 390.23 9.374 400.1 1.60 1.20 1.53 47.33 4.92 1.53 13.35 51.12 2.242 480 zoo -.72 2.32 72.53 7.05 1.74 56.33 42703 2.71 481 2.00 2CC 2.33 23.32 9.73 1.29 34.97 50325 2225 402 2.00 2.22 233 1343 9.47 110 723€ 609.21 2.227 433 2.00 230 233 102.52 12.73 2.45 37.20 1041.13 2.4’! 484 zoo 2.72 2.33 l243 11.17 2.62 96.97 114133 2.452 zoo 172 2.33 112.53 11.76 2.62 59.95 1177.23 9.465 Exp Duct 2 Bwiips Upstream 22 ProlIle C arms AoA=O 634-021 blades Duct 2bwnps Profile C arms 0 4-021 blades ‘umps e diagonally app. turbine rotating away 604 2.00 2.50 2,33 102.53 1172 24.4 93.44 95424 9.267 604 2.00 2.75 2.33 112.53 11.72. 2.53 93.59 l0l37 0.425 544 130 .5C’ CCC’ aD 541 1.60 2.C 1.53 0253 654 2.33 31.12 20401 3.52 542 1.60 2.25 1.33 70.23 .36 2.24 3.4.4.46 25122 0.235 540 1.60 2.50 133 77.33 B.3 2.4! 46.442 377.43 9.367 444 1.60 2.74 1.53 0113 9.9 2.72 L3 41459 1252 645 160 300 1.33 02.73 97 2.34 44.395 42043 0.407 640 1.60 330 133 C 11.24 3.45 27.734 31434 2.257 .547 1.60 2.75 1.53 43.43 L9 2.72 4.!.51 433.23 0.404 604 2.00 2.50 233 102.33 1177 2.45 92.1! 55113 0.255 544 2.00 2.75 2.33 112.73 11.00 2.73 9197 1072.44 2.424 600 1.50 5C 133 47.9 4.94 1.01 12. 5541 2.667 641 130 2CC 1.53 42.33 655 2.34 27.92 -a: 03 0.73 602 1.80 2.2.5 133 70.43 7.37 2.25 31.14 224.57 0.2’7 604 1.60 2.54 1.53 77.53 3.12 2.47 .44.6! 362.13 2.243 570 572 ‘TO (14 1.60 2.75 1.33 24.33 936 2.73 51.95 442.14 0.440 1.80 3.00 1.53 02.53 9.65 2.35 493 454 03 3.442 1.60 2.50 133 1te73 11.36 3.47 31.154 25453 2.335 140 .75 133 4543 934 2.73 11.3! 45539 9.434 2.00 1.75 2.33 7233 7.62 .7S 64.052 40027 2.00 2.25 2.33 3743 9.15 2.03 ,53•95 652.01 2.222 2.00 2.30 2.33 ‘02.33 1:74 2.45 93.2! 67013 3.39.7 2.00 2.75 233 11253 II.?! 2.73 94.104 1110.33 3.443 August/September 2007 Tests — Ducted run no v TSR v RPM w TSR Torque Power Ck s_Direction 601 1.60 2.75 13 55.33 9.93 2.72 34.66 457.72 0.461 wvemaker 502 1.50 2.75 133 55.33 933 2.72 53.96 452.19 0.455 deck 606 1.60 3.00’ 1.53 22.33 9.2 2.34 49.32 454.15 0.455 wavemaker 504 1.80 3CC 133 52.03 9.72 2.35 47.9! 454.79 0.440 dock SIO 2.00 2.75 2.33 i2.53 11.72 2.73 193.25 15233 0.471 wavemaker 611 2.00 175 2.33 11243 11.77 2.53 99.95 116234 3.444 Duct: 2 Bumps Downstream 153 6253 654 02 21.56 .404 ProfileCarms 622 1.60 2.25 1.33 70.43 7.37 2.25 29.157 215.25 2.263 oA=O 620 1.80 230 133 75.33 9.17 2.40 41765 33159 0.35 634-021 blades 624 1.80 2.75 1.53 55.33 3.92 ‘.72 31.976 46233 3.435 624 140 3.00 1.53 22.33 9.72 237 40.12 465.13 0.442 624 1.60 3.30 1.43 0543 11.35 3.4! :9.393 24033 3.322 rur towards dock 537 1.60 171 133 53.23 3.33 2.72 5132 44735 0.442 ..... towards wavemaker 604 604 644 s,4.f 178 August/September 2007 Tests —Ducted run no v TSR v RPM w TSR Toraue Power Ck 620 1.50 1.50 621 1.50 2.00 622 1.50 2.25 623 1.50 2.50 624 1.50 2.75 625 1.50 3.00 632 2.00 2.25 633 2.00 2.50 634 2.00 2.75 652 2.00 2.25 653 2.00 2.50 654 2.00 2.75 0.00 #0/V/OF 0.00 #0MG! 1.50 62.60 6.56 200 28.3 165.52 0.175 1.50 70.20 7.35 2.24 34.845 256.16 0.242 1.50 77.60 8.13 2.48 44.77 363.81 0.344 1.50 85.40 834 2.73 54.33 485.88 0.459 1.50 92.90 9.73 2.97 50.65 492.75 0.466 0.00 #DNiOI 0.00 #DIVIO! 1.50 85.30 8.93 2.72 54.32 465.22 0.459 #0/ViOl 0.00 #DIV/0! 1.50 62.50 6.54 1.99 29.695 194.35 0.184 1.50 70.40 7.37 2.25 33.9 249.92 0.236 1.50 77.60 8.13 248 47.862 368.94 0.368 1.50 85.30 8.93 272 53.92 481.65 0.455 1.50 92.90 9.73 2.97 50.776 493.97 0467 2.00 91.90 9.62 2.20 60.86 778.18 0.310 2.00 02.60 10.74 2.46 94.58 1016.19 0.405 2.00 112.60 11.79 2.70 97.82 1153.44 0.460 #0/V/OF 0.00 #0MG! 1.50 62.50 6.54 1.99 26.96 176.58 0.167 1.50 70.20 734 224 32.725 240.23 0.227 1.50 77.70 8.14 2.48 43.967 357.75 0.338 1.50 85.50 8.95 2.73 53.186 476.20 0450 6.50 93.00 9.74 2.97 48.05 467.96 0.442 0.00 1.50 85.40 8.94 2.73 52.752 471.76 0.446 2.00 90 9.42 2.15 66.472 626.48 0.250 2.00 102.2 10.70 245 91.06 974.56 0.389 2.00 112.5 11.79 270 104.24 1229.14 0.490 Exp •25 1’ Duct: Barge S C arms =0 4-021 blades 600 601 602 603 604 605 606 604.1 1.50 1.50 1.50 2.00 1.50 2.25 1.50 250 1.50 2.75 1.50 3.00 1.50 3.50 1.50 2.75 Exp Duct no bumps repeat 26 Profile C armsAoA =0 634-021 blades Exp 27 Duct wi shaft fairing 1 Profile C arms AoA=0 L021 blades 640 641 642 643 644 645 646 644.1 1.50 1.50 1.50 2.00 1.50 2.25 1.50 2.50 1.50 2.75 1.50 3.00 1.50 3.50 1.50 2.75 179

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0068009/manifest

Comment

Related Items