wide light project · 5.1.5 yaw sweeps hull, keel, & rudder ... project prospectus syrf wide...
TRANSCRIPT
WIDE - LIGHT PROJECT Final Report – November 2015
Andy Claughton
Land Rover BAR
Sailing Yacht Research Foundation
Warwick, Rhode Island, USA
ii Wide-Light Project
About the Sailing Yacht Research Foundation
The mission of the Sailing Yacht Research Foundation is to develop and catalog the science underlying sailboat
performance, resulting in more accurate sailboat handicapping formulae for the benefit of all racing sailors. The
Foundation serves an international user base of those focused on sailing performance prediction, including naval
architects and handicap rule makers.
The Foundation supports research efforts using scientific and engineering principles, utilizing the best research
facilities and individuals, and reports all findings in public forums. In addition, the Foundation maintains and
continuously updates a free-access library of past research, data and papers relevant to the science of performance
sailing.
The Foundation is sustained through charitable contributions and grants, and its success will be measured by the
widespread use and impact of its research, ultimately resulting in fairer yacht racing, which in turn will stimulate
more participation.
SYRF Board of Directors
Stan Honey, President
Steve Benjamin, Chairman
Scott Weisman, Treasurer
Dina Kowalyshyn
Jay Hansen
Gary Weisman
For more information, contact:
Sailing Yacht Research Foundation
1643 Warwick Avenue, Box 300
Warwick, Rhode Island 02889 USA
www.sailyachtresearch.org
The Wide-Light Project by Sailing Yacht Research Foundation is licensed under a Creative Commons Attribution
4.0 International License.
iii
Contents
Contents .................................................................................................................................................................... iii
Figures ...................................................................................................................................................................... iv
Tables .........................................................................................................................................................................v
Supporting Documents ...............................................................................................................................................v
Acknowledgments .................................................................................................................................................... vi
Abstract ................................................................................................................................................................... vii
1. Introduction & Rationale ....................................................................................................................................1
2. Methodology ......................................................................................................................................................1
2.1 Process ........................................................................................................................................................2
2.2 Test Matrix .................................................................................................................................................2
2.3 Tank Testing ...............................................................................................................................................3
2.3.1 Tank Model ........................................................................................................................................4
2.4 CFD ............................................................................................................................................................4
3. Experimental Results ..........................................................................................................................................5
3.1 Practicalities ...............................................................................................................................................5
3.2 Sailing Yacht Resistance Breakdown .........................................................................................................6
4. Computational Results .......................................................................................................................................9
4.1 Questionnaire ..............................................................................................................................................9
4.2 Star-CCM+ Results ..................................................................................................................................10
4.3 FINE/Marine Results ................................................................................................................................12
4.4 OpenFOAM Results .................................................................................................................................14
4.5 FlowLogic Results ....................................................................................................................................16
4.6 SHIPFLOW Results .................................................................................................................................18
5. Analysis and Discussion ...................................................................................................................................18
5.1 Results and Comparative Charts...............................................................................................................19
5.1.1 Upright Resistance Trim and Heave – Tests CB-1 & HKR-1 ...........................................................19
5.1.2 LCG Variation – Tests CB-2 & 3, HKR-2 & 3 .................................................................................21
5.1.3 Heeled Resistance Tests ...................................................................................................................23
5.1.4 Heel with Yaw – Tests CB-6 &7, HKR-6 & 7 ..................................................................................25
5.1.5 Yaw Sweeps Hull, Keel, & Rudder – Tests HKR-8, 9, 10, & 11 .....................................................28
5.1.6 Yaw Sweeps Twin Rudder – Tests HKr-1 & 2.................................................................................30
5.2 Lift Curve Slope .......................................................................................................................................30
5.3 Discussion.................................................................................................................................................31
6. Conclusions ......................................................................................................................................................32
7. Future Work .....................................................................................................................................................32
References ................................................................................................................................................................34
iv Wide-Light Project
Figures
Figure 1. Typical drag area curve from an upright resistance test ..............................................................................5
Figure 2. Typical heeled and yawed resistance test results. .......................................................................................6
Figure 3. Resistance breakdown for analysis .............................................................................................................7
Figure 4. Typical variation of R0/RU and Te versus speed for a yacht derived from tank tests. ................................8
Figure 5. Star-CCM+ Resistance Curves .................................................................................................................10
Figure 6. Star-CCM+ Yaw Sweep Tests ..................................................................................................................10
Figure 7. Star-CCM+ Yaw Sweeps at 15° and 25° Heel ..........................................................................................11
Figure 8. Star-CCM+ Yaw Sweeps at 15° and 25° Heel, Twin Rudder ...................................................................11
Figure 9. FINE/Marine Resistance Curves ...............................................................................................................12
Figure 10. FINE/Marine Yaw Sweep Tests ..............................................................................................................12
Figure 11. FINE/Marine Yaw Sweep at 15° and 25° Heel .......................................................................................13
Figure 12. FINE/Marine Yaw Sweeps at 15° and 25° Heel, Twin Rudder ..............................................................13
Figure 13. OpenFOAM Resistance Curves ..............................................................................................................14
Figure 14. OpenFOAM Yaw Sweep Tests ...............................................................................................................14
Figure 15. OpenFOAM Yaw Sweep at 15° and 25° Heel ........................................................................................15
Figure 16. OpenFOAM Yaw Sweep at 15° Heel, Twin Rudder ..............................................................................15
Figure 17. FlowLogic Resistance Curves .................................................................................................................16
Figure 18. FlowLogic Yaw Sweep Tests..................................................................................................................16
Figure 19. FlowLogic Yaw Sweep at 15° and 25° Heel ...........................................................................................17
Figure 20. FlowLogic Yaw Sweep at 15° and 25° Heel, Twin Rudder ....................................................................17
Figure 21. SHIPFLOW Resistance Curves ..............................................................................................................18
Figure 22. Trim and Heave Comparison – Zero Heel ..............................................................................................19
Figure 23. Resistance Comparison – Zero Heel .......................................................................................................20
Figure 24. Upright Resistance Data, Ratio to Tank Test Results .............................................................................21
Figure 25. Canoe Body Only Upright LCG Shift, Ratio to Tank Test Results ........................................................22
Figure 26. Hull, Keel, & Rudder Upright LCG Shift, Ratio to Tank Test Results...................................................22
Figure 27. Heeled Resistance Test Data Ratio to Upright Value .............................................................................24
Figure 28. Canoe Body Only Heeled Resistance Test Data, Ratio to Tank Test Results .........................................25
Figure 29. Hull, Keel & Rudder Heeled Resistance Test Data, Ratio to Tank Test Results ....................................25
Figure 30. Canoe Body Only Yaw Sweep Data at 15° and 25° Heel .......................................................................26
Figure 31. Canoe Body Only Yaw Sweep Data, Ratio to Tank Test Results ...........................................................26
Figure 32 Hull Keel & Rudder Yaw Sweep Data at 15° and 25° Heel .....................................................................27
Figure 33. Hull, Keel & Rudder Yaw Sweep Data, Ratio to Tank Test Results ......................................................27
Figure 34. Hull, Keel & Rudder, Summary of Fitted Lines to Yaw Sweep Data ....................................................28
Figure 35. Heeled Resistance Ratio and Effective Draft Comparison .....................................................................29
Figure 36. Comparison of CFD Results for Twin Rudder Configuration ................................................................30
Figure 37. Lift Slope Comparison ............................................................................................................................31
v
Tables
Table 1. Test Matrix, Speed & LCG Indices ..............................................................................................................3
Table 2. Model Dimensions .......................................................................................................................................4
Table 3. List of CFD Codes ........................................................................................................................................4
Table 4. CFD Stakeholder Questionnaire ...................................................................................................................9
Table 5. Comparison Summary – CFD to Model Test .............................................................................................31
Supporting Documents
The defining documents of the project specification and the results from each CFD program are available online
through SYRF’s Technical Resources Library. These documents may be viewed and downloaded from the
following links:
Planning Documents:
Project Plan SYRF Wide Light Project Plan Issue 2.1.pdf [180 KB]
Invitation to CFD Participants SYRF Wide Light Project Invitation to Participate.pdf [274 KB]
Project Prospectus SYRF Wide Light Project Final Prospectus- Rev A.pdf [400 KB]
Project Run Matrix SYRF_-CFD test Program-Rev A.xlsx [1 MB]
3D Model of Project Hull & Appendages SYRF-TH01-CAD for CFD-02-RhinoV4.3dm [2 MB]
Results:
Compiled Results Spreadsheet SYRF-WL Results.xlsx [1 MB]
CFD Participant Questionnaire SYRF-WL CFD-Questionnaire-Compiled.xlsx [97 KB]
Tank Testing Results & Report SYRF-WL Tank Results.zip [9 MB]
Tank Testing Photos SYRF-WL Tank Photos.zip [326 MB]
Star-CCM+ Results Submission SYRF-WL Star-CCM+ Results.zip [236 KB]
FINE/Marine Results Submission SYRF-WL FINEMarine Results.zip [99 MB]
SHIPFLOW Results Submission SYRF-WL SHIPFLOW Results.zip [56 MB]
OpenFOAM Results Submission SYRF-WL OpenFOAM Results.zip [158 MB]
FlowLogic Results Submission SYRF-WL FlowLogic Results.zip [1 MB]
vi Wide-Light Project
Acknowledgments
This project was funded by the Sailing Yacht Research Foundation.
The project team would like to thank Judel/Vrolijk & Co. who were kind enough to make available one of their
towing tank models for the project. The availability of this model was one of the main reasons that this project
could proceed.
The project was generously supported by the Wolfson Unit MTIA and the University of Southampton, along with
the Computational Fluid Dynamics (CFD) stakeholders.
Principal Investigator
Andy Claughton, Land Rover BAR
CFD Participants
Benoit Mallol, Numeca
Jason Ker, Ker Yacht Design
David Egan, Ennova Technologies
Sandy Wright, Wolfson Unit MTIA, University of Southampton
Lars Larson, Chalmers University of Technology, FLOWTECH Int. AB
Michal Orych, Chalmers University of Technology, FLOWTECH Int. AB
Rodrigo Azcueta, Cape Horn Engineering
Matteo Lledri, Cape Horn Engineering
Tank Testing Participants
Martyn Prince, Wolfson Unit MTIA, University of Southampton
Etienne Gauvain, Wolfson Unit MTIA, University of Southampton
Project Advisory Committee
Jim Teeters, SYRF Technical Director
Dina Kowalyshyn, CDI Marine Company
Robert Ciesla, The Boeing Company
Stan Honey, SYRF Advisory Council
Steve Benjamin, SYRF Advisory Council
SYRF Project Management
Myles Cornwell, SYRF Executive Director
McKenzie Wilson, SYRF Research Analyst
vii
Abstract
The modern era of high performance sailing has ushered in a new design paradigm for modern race boats.
Compared to past hull shapes, the current in-favor design is both wide and light – it is in this description that the
Wide-Light’s project’s name originates.
This project provides insight into the accuracy of existing modelling methods in predicting performance of Wide-
Light designs. It is intended for this information to better inform and equip handicapping systems and box rules to
address Wide-Light designs.
Five different CFD stakeholders carried out “blind” CFD analysis on an identical test matrix using different
computational codes and approaches. The same test matrix was run as a tank test for both canoe body only and
appended (heel, keel, rudder) configurations as a control for the hull geometry. The CFD results were compared
with the tank test control results to determine CFD model accuracy. This project illustrated the accuracy of
commercial CFD codes in predicting the forces on a Wide-Light sailing yacht. Additionally, this project provides
a comprehensive set of data against which researchers may develop and validate their own numerical tools.
viii Wide-Light Project
(this page intentionally left blank)
1
WIDE - LIGHT PROJECT
1. Introduction & Rationale
The report that follows describes the Wide-Light Project, a project conceived and sponsored by the Sailing Yacht
Research Foundation (SYRF). Following its 2013 annual meeting, SYRF solicited a program proposal from Andy
Claughton who was then engaged to manage the project.
The “Wide-Light” label is meant to describe the design of the modern high performance sailboats. These so-called
Wide-Light boats present many of the hydro-dynamic effects that are a challenge to predict: semi-planing hull
forms, immersed transom effects, spray creation, keels operating close to the water surface, rudders, dagger-
boards and canting keels that generate vertical force, the list goes on. The complexity of these interacting effects
is challenging for the designer and a minefield for yacht handicappers who are obligated to handicap all boats
equitably, both new and old.
It is therefore the goal of this project to provide data and conclusions of what might be seen as best practice or
state-of-the-art modelling methods so as to better inform and equip handicapping systems and box rules to address
Wide-Light designs. Projects such as the Delft Systematic Yacht Hull Series and the nine model series performed
at the Canadian National Research Council (NRC) have established the current database of hydrodynamics, but
they are no longer representative of today’s racing fleets. This project is the first step toward expanding the public
database to include modern yachts.
The Wide-Light project aligns with the SYRF mission to develop and catalogue the science underlying sailing
performance and handicapping, by:
a) Publishing an assessment of alternative methodologies to specify and analyze sailing yacht hydrodynamic
resistance using computational tools,
b) Making the data available for researchers, and students as an accessible experimental data set for a
contemporary sailing yacht, and
c) Demonstrating how this type of study can be used to inform the handicapping process.
It is expected that the project will also be particularly valuable for academics as it will allow students, researchers
and lecturers to frame new research projects most effectively.
2. Methodology
This project’s methodology draws upon well established procedures in ship hydrodynamics for comparing CFD
data with experimental tank data (Reference 1). Unlike in typical commercial and military studies where there are
2 Wide-Light Project
numerous tight controls on the process, for the Wide-Light project the methods were accelerated to more quickly
generate results.
A “test matrix” was developed to reflect a typical evaluation program for a sailing yacht, including tests of the
bare canoe body and the appended hull over a range of speeds, heel and leeway angles. This matrix was
distributed to the CFD stakeholders who used their usual procedures to generate data for each point in the matrix.
The analysis was performed at model scale to avoid conflating the uncertainties of the scaling procedure with the
CFD comparison. The stakeholders completed a “CFD Questionnaire” documenting the methodologies used to
perform their CFD analysis.
The CFD calculations were delivered to the project leader before the tank testing was performed, therefore
ensuring a blind test of the CFD codes. However, due to misinterpretations of the original test matrix, some CFD
points were re-worked to correct for mistakes. The towing tank test was performed using the same test matrix as
the CFD stakeholders, unfortunately due to limited tank availability two of the twenty planned test sequences
were omitted. It is the intent of SYRF to complete these missing tank runs when the opportunity arises, but the
timing on this is currently unknown.
2.1 Process
The project process was as follows;
1. Procure a contemporary racing yacht tank test model.
2. Prepare a test matrix appropriate to this model that could be used in a physical model test and a CFD
study.
3. Agree terms with a working group of leading CFD practitioners who were prepared to engage in a blind
comparison of data whilst sharing their methodologies.
4. CFD practitioners perform analysis and submit results.
5. Conduct towing tank tests using a state of art facility and methodology.
6. Prepare comparative data of CFD and model tests.
7. Prepare a summary report of the work for publication by SYRF, including not only the technical
conclusions but also a discussion on how the data generated by the project can best serve the research and
educational aims of SYRF.
8. Summarize the way that the methodologies developed can be applied to yacht handicapping.
2.2 Test Matrix
The full test matrix is shown in the Project Prospectus PDF file and is summarized in Table 1. In the test matrix,
there is no mention of “sail trimming moment” or sail plan center of effort height; traditionally, towing tank tests
have been conducted using an assumed sail center of effort height so that an appropriate bow down trimming
moment from the sail thrust can be applied to the model. To take this approach, an estimate of the hull resistance
is required. In this study, each test run had a predetermined longitudinal center of gravity (LCG) to avoid each
contributor applying a slightly different sail trim moment. The specified LCG position for each test broadly
simulates the effect of the sail trim moment based on an assumed hull resistance curve and appropriate sail plan.
3
Table 1. Test Matrix, Speed & LCG Indices
ID Configuration TEST Heel Fn Description
CB-1
Canoe Body Only
Upright Resistance 0 0.10-0.80 Basic resistance test on the unappended hull.
CB-2 LCG Variation 0 0.35 Effect of shifting the Longitudinal Centre of Gravity, i.e. changing fore and aft trim. CB-3 LCG Variation 0 0.5
CB-4 Heel at zero yaw 15 0.25-0.45 Resistance test with the hull heeled.
CB-5 Heel at zero yaw 25 0.25-0.45
CB-6 Heel with yaw 15 0.35 Change of leeway with rudder on centerline, +ve and –ve leeway. CB-7 Heel with yaw 25 0.5
HKR-1
Hull Keel & Single Rudder
Upright Resistance 0 0.10-0.80 Basic resistance test on the unappended hull.
HKR-2 LCG Variation 0 0.35 Effect of shifting the Longitudinal Centre of Gravity, i.e. changing fore and aft trim. HKR-3 LCG Variation 0 0.5
HKR-4 Heel at zero yaw 15 0.25-0.45 Resistance test with the hull heeled.
HKR-5 Heel at zero yaw 25 0.25-0.45
HKR-6 Heel with yaw 15 0.35 Change of leeway with rudder on centerline, +ve and –ve leeway. HKR-7 Heel with yaw 25 0.5
HKR-8 Yaw Sweep 15 0.35
Leeway Sweeps and Rudder Variations at fixed speed and heel angle.
HKR-9 Yaw Sweep 15 0.5
HKR-10 Yaw Sweep 25 0.35
HKR-11 Yaw Sweep 25 0.5
HKr-1 Hull Keel & Twin Rudders
Yaw Sweep 15 0.35 Leeway Sweeps and Rudder Variations at fixed speed and heel angle. HKr-2 Yaw Sweep 25 0.5
2.3 Tank Testing
The tank testing was carried out by the Wolfson Unit for Marine Technology and Industrial Aerodynamics
(WUMTIA) in the QinetiQ #2 towing tank at the Haslar Technology Park, Gosport England.
Standard WUMTIA test and analysis procedures were performed. The report describing the tests is presented in
the PDF Portfolio.
LCG Index
LCG m
LCG_001 -2.488
LCG_002 -2.476
LCG_003 -2.469
LCG_004 -2.459
LCG_005 -2.447
LCG_006 -2.431
LCG_007 -2.403
LCG_008 -2.367
LCG_009 -2.330
LCG_010 -2.294
LCG_011 -2.262
LCG_012 -2.235
LCG_013 -2.206
LCG_014 -2.172
LCG_015 -2.127
LCG_016 -2.784
LCG_017 -2.634
LCG_018 -2.484
LCG_019 -2.334
LCG_020 -2.184
Speed Index
Fn Vs m/s
V_001 0.1 0.681
V_002 0.15 1.022
V_003 0.2 1.363
V_004 0.25 1.703
V_005 0.3 2.044
V_006 0.35 2.385
V_007 0.4 2.725
V_008 0.45 3.066
V_009 0.5 3.407
V_010 0.55 3.747
V_011 0.6 4.088
V_012 0.65 4.429
V_013 0.7 4.769
V_014 0.75 5.110
V_015 0.8 5.451
4 Wide-Light Project
2.3.1 Tank Model
A previously tested model canoe body was donated to the project. To protect the intellectual property of the
designer and the owner, the hull was modified to lines supplied by SYRF, re-faired, painted and marked up. The
model was re-commissioned for testing and fitted with a previously used keel fin, bulb and movable single rudder.
The principal dimensions for this model, designated as model number M1108 by the Wolfson Unit, are presented
below.
Table 2. Model Dimensions
2.4 CFD
An invitation was extended to practitioners who were active in the racing yacht field. Those who felt able to
support the project were allowed to participate. In total, five individuals were designated as CFD stakeholders.
Fortunately, this list represents a large segment of the CFD community, from commercial and open source RANS
codes to computationally less intensive panel codes.
Table 3. List of CFD Codes
Software Contributor Affiliation Type
FINETM/Marine Benoit Mallol, Jason Ker
Numeca, Ker Yacht Design
RANS
FlowLogic David Egan Panel Code
OpenFOAM Sandy Wright Wolfson Unit MTIA, University of Southampton
Open Source RANS
SHIPFLOW Lars Larsson, Michal Orych
Chalmers University of Technology, FLOWTECH Int. AB
Combination Panel Code & RANS
Star-CCM+ Rodrigo Azcueta, Matteo Lledri
Cape Horn Engineering RANS
M1108 Metric Imperial
Overall length 4.88 m 16.0 ft
Design waterline length 4.60 m 15.1 ft
Displacement (appended) 215 kg 474 lbs
Displacement (canoe body) 197 kg 434 lbs
Maximum beam 1.28 m 4.2 ft
Draft to datum 1.15 m 3.8 ft
5
3. Experimental Results
The experimental results and procedure are presented in Report No. 2546D from the Wolfson Unit MTIA. The
report, spreadsheets, and tank photos are available for download from the SYRF Technical Resources Library.1
3.1 Practicalities
The test matrix for the project was developed with the computational (CFD) work in mind. In comparison to the
computational process, the experimental (tank test) process is more complex and requires additional processing
steps to allow for the comparison between the experimental and computational data. The rationale behind the
processing of the experimental data follows.
A typical upright resistance curve, expressed as a drag area (AD = drag/q), is shown in Figure 1. The first few runs
of the upright resistance curve are used to align the model to produce a minimum amount of sideforce and yaw
moment. It is impossible to set the model up so accurately that it runs with no sideforce or yaw moment, but the
data is corrected for the presence of these forces and moments.
Figure 1. Typical drag area curve from an upright resistance test
For the heeled and yawed tests, a series of runs over a range of yaw angles, are carried out at each speed and heel
angle combination. These tests are conducted on both port and starboard tacks.2 The yaw angles should be chosen
so that the sailing sideforce for the heel angle is spanned. Using a predetermined leeway angle sequence that
remains the same for all speeds and heel angles will produce a substantial number of test points that are far
1 Link to results and photos: SYRF-WL Tank Results.zip [9 MB], SYRF-WL Tank Photos.zip [326 MB] 2 The Wolfson Unit is unique in that the experimental set up allows testing on both tacks, giving the ability to determine the
forces from a mean value from both tacks. This approach delivers a more reliable test data set.
0.2
0.3
0.4
0.5
0.6
0.7
4 6 8 10 12 14 16 18
Speed kts
Dra
g A
rea
Upright
Resistance
6 Wide-Light Project
removed from the ‘sailing’ condition. Also, for the heeled tests, it is convenient to determine the down thrust of
the sails (SF tanφ) based on estimated sailing sideforce for the yacht. This may be applied regardless of the
leeway angle set as it is not only easier from a practical view point but also avoids fluctuations in displacement
affecting the determination and interpretation of induced drag.
Typical test data is shown in Figure 2. There is usually some difference in the sideforce for a given leeway on the
two tacks and at higher speeds and heel angles a difference of drag tack to tack at the same sideforce, not the same
leeway, of 1–2% is acceptable. Greater differences than this indicate a misalignment of the center planes between
the hull, keel and rudder.
Figure 2. Typical heeled and yawed resistance test results.
Typically 3-4 tests per tack are done at each speed. It should be noted that although it is tempting, from an
analysis perspective, to make one of these a zero yaw point, there is often some non-linearity close to zero
sideforce that makes it more useful to begin the tests at some low leeway angle. This process is repeated for all
the speed/heel angle combinations in the test matrix.
As part of the tank analysis, the data from both tacks was analyzed to determine the mean line through the data
from both tacks and to average out the sideforce differences at the same nominal leeway on each tack.3
3.2 Sailing Yacht Resistance Breakdown
The total hydrodynamic drag of the yacht is assumed to be the sum of the following components:
𝑅𝑇𝑂𝑇 = 𝑅𝑈 + 𝑅0 + 𝑅𝐼
Upright Resistance (RU) comprising:
Wave resistance (RW)
Appendage viscous drag (RVapp)
Canoe body viscous drag (RVcb)
3 This data is presented in Table 4 of the towing tank report.
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0 200 400 600 800 1000 1200
Sideforce2
Dra
g A
rea 11 knots
Leew
ay
2 d
eg
10 knots
9 knots
Leew
ay 3
deg
Filled Symbols = Port tack
Open Symbols = Stbd. tack
7
R0 (R zero) the increase in drag above RU at zero sideforce from the fitted line to the test data,
RI the Induced drag.
Because this study is not extrapolating the data to full scale it is not necessary to consider the de-construction of
the resistance into viscous and gravitational components.
The combination of the aforementioned drag components is shown graphically in Figure 3. At a speed VS and
heel angle φ the resistance value can be determined by the intersection of the resistance against sideforce squared
line with the equilibrium sailing sideforce line (shown as a dotted vertical line in Figure 3). The requirement of a
typical Velocity Prediction Program (VPP) hydrodynamic force model is to determine these three resistance
components — this approach must be adopted to assist with the comparison of the experimental and computed
data.
Figure 3. Resistance breakdown for analysis
The first step is to fit a cubic spline curve to the upright resistance results using a least squares fit. This allows the
upright resistance to be determined at any speed. Looking at the heeled and yawed results, unless stall is
occurring, a straight line can usually be fitted to the drag versus SF2 data points at each speed and heel angle. The
slope of the line is determined by the induced drag characteristics of the keel and rudder combined with the
wavemaking effects. The slope of the line may be expressed as an effective draft (Te) derived from the formula:
𝑇𝑒 = √1
(𝑑𝑅
𝑑𝑆𝐹2) 𝜌𝜋𝑉2𝑐𝑜𝑠2𝜙
dR/dSF2 = slope of resistance versus sideforce2 line
ρ = density
V = velocity
ϕ = heel angle
Dra
g
R U
V S Sailing SF2
R0
R I
Resistance vs. SF 2 line. Speed = V S
Heel =
R TOT
Dra
g
8 Wide-Light Project
The intercept of the straight line with the zero sideforce axis (R0) determines the drag due to heel, and may be
expressed as a ratio to the upright resistance RU. Thus, for each tested speed and heel angle, the hydrodynamic
behavior can be expressed as an effective draft (Te) and a heel drag ratio (RH/RU), where RH = RU + R0.
Figure 4 illustrates this approach and shows the results from tests HKR 8-11 expressed as effective draft and heel
drag ratio.
Figure 4. Typical variation of R0/RU and Te versus speed for a yacht derived from tank tests.
These plots show typical behavior — heel drag increases both with heel angle and speed while effective draft
reduces with increasing speed and heel angle as the keel root comes closer to the water surface.
2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.00.90
0.95
1.00
1.05
2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.00.0
0.2
0.4
0.6
0.8
1.0
Solid Line 15 deg heelDashed Line 25 Deg
SPEED m/s
HE
EL
- D
RA
G R
atio
SPEED m/s
EF
FE
CT
IVE
DR
AF
T m
9
Presenting the data in this form gives a much clearer insight into the quality of the results than simply comparing
on a point by point basis. It allows us to see if the computational results actually match the trends of the physical
test results, even if the “absolute” values are somewhat different.
4. Computational Results
Each participant submitted computational results for his specific CFD program. In addition to submitting results
via spreadsheet, each participant provided a brief summary of his methodology and results. The submitted files
from each participant are available via the SYRF Research Library links included in the section footers. The CFD
results are presented in the graphs that follow.
4.1 Questionnaire
Each CFD stakeholder was asked to complete a questionnaire detailing the software and methods used to generate
the results.4
Table 4. CFD Stakeholder Questionnaire
CFD Questionnaire
CFD package?
Mesh Generation Software?
Solver?
Post process and visualization software?
Towing point?
Free surface tracking/capturing method?
Turbulence model?
Numerical Ventilation/Streaking
Ventilation observed?
Correction/Type?
Mesh type?
Computational Domain Size?
Local refinement?
Speed treatment?
Cell resolutions (in mm)?
Mesh sizes (in million cells)?
Time to prepare meshing?
Time to mesh each case?
CPU requirements?
Computer?
MeshSize/meshing/Solving/cores?
Cost effectiveness?
4 Link to compiled questionnaire: SYRF-WL CFD-Questionnaire-Compiled.xlsx [97 KB]
10 Wide-Light Project
0
50
100
150
200
250
300
350
0.0 1.0 2.0 3.0 4.0 5.0 6.0
Dra
g (N
)
Speed (m/s)
Star-CCM+ CB-1, 4 & 5, HKR-1, 4 & 5
CB1 Heel 0° Yaw 0° HKR1 Heel 0° Yaw 0°CB4 Heel 15° Yaw 0° HKR4 Heel 15° Yaw 0°CB5 Heel 25° Yaw 0° HKR5 Heel 25° Yaw 0°
20
40
60
80
100
120
140
160
-2.8 -2.7 -2.6 -2.5 -2.4 -2.3 -2.2 -2.1
Dra
g (N
)
LCG (m)
Star-CCM+ CB-2 & 3, HKR-2 & 3
CB2 Fn 0.35 Heel 0° Yaw 0° CB3 Fn 0.5 Heel 0° Yaw 0°
HKR2 Fn 0.35 Heel 0° Yaw 0° HKR3 Fn 0.5 Heel 0° Yaw 0°
20
40
60
80
100
120
140
160
180
-2 -1 0 1 2 3
Dra
g (N
)
Yaw (deg)
Star-CCM+ CB-6 & 7, HKR-6 & 7
CB6 Fn 0.35 Heel 15° CB7 Fn 0.5 Heel 25°
HKR6 Fn 0.35 Heel 15° HKR7 Fn 0.5 Heel 25°
-300
-200
-100
0
100
200
300
400
-2 -1 0 1 2 3
Sid
efo
rce
(N
)
Yaw (deg)
Star-CCM+ CB-6 & 7, HKR-6 & 7
CB6 Fn 0.35 Heel 15° CB7 Fn 0.5 Heel 25°
HKR6 Fn 0.35 Heel 15° HKR7 Fn 0.5 Heel 25°
4.2 Star-CCM+ Results
The results and all associated files which were submitted can be downloaded from the SYRF Technical Resources
Library.5
Figure 5. Star-CCM+ Resistance Curves
Figure 6. Star-CCM+ Yaw Sweep Tests
5 Link to results: SYRF-WL Star-CCM+ Results.zip [236 KB]
11
40
60
80
100
120
140
160
180
0 50000 100000 150000 200000 250000 300000 350000 400000
Dra
g (N
)
Sideforce2 (N2)
Star-CCM+ HKR-8 & HKR-9
HKR8 Fn 0.35 Heel 15° Yaw var Rudder 2
HKR9 Fn 0.5 Heel 15° Yaw var Rudder 2
ZeroPoint = 146.40Slope = 6.51E-05
ZeroPoint = 55.27Slope = 11.86E-05
40
60
80
100
120
140
160
180
0 50000 100000 150000 200000 250000 300000
Dra
g (N
)
Sideforce2 (N2)
Star-CCM+ HKR-10 & HKR-11
HKR10 Fn 0.35 Heel 25° Yaw var Rudder 3
HKR11 Fn 0.5 Heel 25° Yaw var Rudder 3
ZeroPoint = 55.12Slope = 19.10E-05
ZeroPoint = 147.61Slope = 11.60E-05
40
60
80
100
120
140
160
180
0 50000 100000 150000 200000 250000 300000
Dra
g (N
)
Sideforce2 (N2)
Star-CCM+ HKr-1 & HKr-2
HKr-1 Fn 0.35 Heel 15° Yaw var Twin Rudder 2
HKr-2 Fn 0.5 Heel 25° Yaw var Twin Rudder 3
ZeroPoint = 56.21Slope = 12.46E-05
ZeroPoint = 148.05Slope = 10.89E-05
Figure 7. Star-CCM+ Yaw Sweeps at 15° and 25° Heel
Figure 8. Star-CCM+ Yaw Sweeps at 15° and 25° Heel, Twin Rudder
12 Wide-Light Project
20
40
60
80
100
120
140
160
-2.8 -2.7 -2.6 -2.5 -2.4 -2.3 -2.2 -2.1
Dra
g (N
)
LCG (m)
FINE/Marine CB-2 & 3, HKR-2 & 3
CB2 Fn 0.35 Heel 0° Yaw 0° CB3 Fn 0.5 Heel 0° Yaw 0°
HKR2 Fn 0.35 Heel 0° Yaw 0° HKR3 Fn 0.5 Heel 0° Yaw 0°
20
40
60
80
100
120
140
160
180
-2 -1 0 1 2 3
Dra
g (N
)
Yaw (deg)
FINE/Marine CB-6 & 7, HKR-6 & 7
CB6 Fn 0.35 Heel 15° CB7 Fn 0.5 Heel 25°
HKR6 Fn 0.35 Heel 15° HKR7 Fn 0.5 Heel 25°
-300
-200
-100
0
100
200
300
400
-2 -1 0 1 2 3
Sid
efo
rce
(N
)
Yaw (deg)
FINE/Marine CB-6 & 7, HKR-6 & 7
CB6 Fn 0.35 Heel 15° CB7 Fn 0.5 Heel 25°
HKR6 Fn 0.35 Heel 15° HKR7 Fn 0.5 Heel 25°
0
50
100
150
200
250
300
350
0.0 1.0 2.0 3.0 4.0 5.0 6.0
Dra
g (N
)
Speed (m/s)
FINE/Marine CB-1, 4 & 5, HKR-1, 4, & 5
CB1 Heel 0° Yaw 0° HKR1 Heel 0° Yaw 0°CB4 Heel 15° Yaw 0° HKR4 Heel 15° Yaw 0°CB5 Heel 25° Yaw 0° HKR5 Heel 25° Yaw 0°
4.3 FINE/Marine Results
The results and all associated files which were submitted can be downloaded from the SYRF Technical Resources
Library.6
Figure 9. FINE/Marine Resistance Curves
Figure 10. FINE/Marine Yaw Sweep Tests
6 Link to results: SYRF-WL FINEMarine Results.zip [99 MB]
13
40
60
80
100
120
140
160
180
0 50000 100000 150000 200000 250000 300000 350000 400000
Dra
g (N
)
Sideforce2 (N2)
FINE/Marine HKR-8 & HKR-9
HKR8 Fn 0.35 Heel 15° Yaw var Rudder 2
HKR9 Fn 0.5 Heel 15° Yaw var Rudder 2
ZeroPoint = 53.47Slope = 11.04E-05
ZeroPoint = 144.10Slope = 6.86E-05
40
60
80
100
120
140
160
180
0 50000 100000 150000 200000 250000 300000
Dra
g (N
)
Sideforce2 (N2)
FINE/Marine HKR-10 & HKR-11
HKR10 Fn 0.35 Heel 25° Yaw var Rudder 3
HKR11 Fn 0.5 Heel 25° Yaw var Rudder 3
ZeroPoint = 145.28Slope = 11.99E-05
ZeroPoint = 52.40Slope = 18.23E-05
40
60
80
100
120
140
160
180
0 50000 100000 150000 200000 250000 300000
Dra
g (N
)
Sideforce2 (N2)
FINE/Marine HKr-1 & HKr-2
HKr-1 Fn 0.35 Heel 15° Yaw var Twin Rudder 2
HKr-2 Fn 0.5 Heel 25° Yaw var Twin Rudder 3
ZeroPoint = 54.05Slope = 12.18E-05
ZeroPoint = 146.12Slope = 10.81E-05
Figure 11. FINE/Marine Yaw Sweep at 15° and 25° Heel
Figure 12. FINE/Marine Yaw Sweeps at 15° and 25° Heel, Twin Rudder
14 Wide-Light Project
0
50
100
150
200
250
300
350
400
450
0.0 1.0 2.0 3.0 4.0 5.0 6.0
Dra
g (N
)
Speed (m/s)
OpenFOAM CB-1, 4 & 5, HKR-1, 4 & 5
CB1 Heel 0° Yaw 0° HKR1 Heel 0° Yaw 0°CB4 Heel 15° Yaw 0° HRK4 Heel 15° Yaw 0°CB5 Heel 25° Yaw 0° HKR5 Heel 25° Yaw 0°
20
40
60
80
100
120
140
160
180
200
-2.8 -2.7 -2.6 -2.5 -2.4 -2.3 -2.2 -2.1
Dra
g (N
)
LCG (m)
OpenFOAM CB-2 & 3, HKR-2 & 3
CB2 Fn 0.35 Heel 0° Yaw 0° CB3 Fn 0.5 Heel 0° Yaw 0°
HKR2 Fn 0.35 Heel 0° Yaw 0° HKR3 Fn 0.5 Heel 0° Yaw 0°
20
40
60
80
100
120
140
160
180
200
-2 -1 0 1 2 3
Dra
g (N
)
Yaw (deg)
OpenFOAM CB-6 & 7, HKR-6 & 7
CB6 Fn 0.35 Heel 15° CB7 Fn 0.5 Heel 25°
HKR6 Fn 0.35 Heel 15° HKR7 Fn 0.5 Heel 25°
-300
-200
-100
0
100
200
300
400
-2 -1 0 1 2 3
Sid
efo
rce
(N
)
Yaw (deg)
OpenFOAM CB-6 & 7, HKR-6 & 7
CB6 Fn 0.35 Heel 15° CB7 Fn 0.5 Heel 25°
HKR6 Fn 0.35 Heel 15° HKR7 Fn 0.5 Heel 25°
4.4 OpenFOAM Results
The results and all associated files which were submitted can be downloaded from the SYRF Technical Resources
Library.7
Figure 13. OpenFOAM Resistance Curves
Figure 14. OpenFOAM Yaw Sweep Tests
7 Link to results: SYRF-WL OpenFOAM Results.zip [158 MB]
15
40
60
80
100
120
140
160
0 50000 100000 150000 200000
Dra
g (N
)
Sideforce2 (N2)
OpenFOAM HKR-10 & HKR-11 (no data)
HKR10 Fn 0.35 Heel 25° Yaw var Rudder 3
ZeroPoint = 76.29Slope = 4.17E-05
40
60
80
100
120
140
160
180
0 50000 100000 150000 200000 250000 300000
Dra
g (N
)
Sideforce2 (N2)
OpenFOAM HKr-1 & HKr-2 (no data)
HKr-1 Fn 0.35 Heel 15° Yaw var Twin Rudder 2
ZeroPoint = 77.50Slope = 23.64E-05
50
100
150
200
250
0 50000 100000 150000 200000 250000 300000 350000
Dra
g (N
)
Sideforce2 (N2)
OpenFOAM HKR-8 & HKR-9
HKR8 Fn 0.35 Heel 15° Yaw var Rudder 2
HKR9 Fn 0.5 Heel 15° Yaw var Rudder 2
ZeroPoint = 184.94Slope = 12.17E-05
ZeroPoint = 74.66Slope = 24.98E-05
Figure 15. OpenFOAM Yaw Sweep at 15° and 25° Heel
Figure 16. OpenFOAM Yaw Sweep at 15° Heel, Twin Rudder
16 Wide-Light Project
0
50
100
150
200
250
300
350
0.0 1.0 2.0 3.0 4.0 5.0 6.0
Dra
g (N
)
Speed (m/s)
FlowLogic CB-1, 4 & 5, HKR-1, 4 & 5
CB1 Heel 0° Yaw 0° HKR1 Heel 0° Yaw 0°CB4 Heel 15° Yaw 0° HKR4 Heel 15° Yaw 0°CB5 Heel 25° Yaw 0° HKR5 Heel 25° Yaw 0°
20
40
60
80
100
120
140
160
-2.8 -2.7 -2.6 -2.5 -2.4 -2.3 -2.2 -2.1
Dra
g (N
)
LCG (m)
FlowLogic CB-2 & 3, HKR-2 & 3
CB2 Fn 0.35 Heel 0° Yaw 0° CB3 Fn 0.5 Heel 0° Yaw 0°
HKR2 Fn 0.35 Heel 0° Yaw 0° HKR3 Fn 0.5 Heel 0° Yaw 0°
20
40
60
80
100
120
140
160
180
-2 -1 0 1 2 3
Dra
g (N
)
Yaw (deg)
FlowLogic CB-6 & 7, HKR-6 & 7
CB6 Fn 0.35 Heel 15° CB7 Fn 0.5 Heel 25°
HKR6 Fn 0.35 Heel 15° HKR7 Fn 0.5 Heel 25°
-300
-200
-100
0
100
200
300
400
-2 -1 0 1 2 3
Sid
efo
rce
(N
)
Yaw (deg)
FlowLogic CB-6 & 7, HKR-6 & 7
CB6 Fn 0.35 Heel 15° CB7 Fn 0.5 Heel 25°
HKR6 Fn 0.35 Heel 15° HKR7 Fn 0.5 Heel 25°
4.5 FlowLogic Results
The results and all associated files which were submitted can be downloaded from the SYRF Technical Resources
Library.8
Figure 17. FlowLogic Resistance Curves
Figure 18. FlowLogic Yaw Sweep Tests
8 Link to results: SYRF-WL FlowLogic Results.zip [1 MB]
17
40
60
80
100
120
140
160
180
0 75000 150000 225000 300000 375000 450000
Dra
g (N
)
Sideforce2 (N2)
FlowLogic HKR-8 & HKR-9
HKR8 Fn 0.35 Heel 15° Yaw var Rudder 2
HKR9 Fn 0.5 Heel 15° Yaw var Rudder 2
ZeroPoint = 43.91Slope = 14.30E-05
ZeroPoint = 124.89Slope = 7.82E-05
40
60
80
100
120
140
160
180
0 50000 100000 150000 200000 250000 300000
Dra
g (N
)
Sideforce2 (N2)
FlowLogic HKR-10 & HKR-11
HKR10 Fn 0.35 Heel 25° Yaw var Rudder 3
HKR11 Fn 0.5 Heel 25° Yaw var Rudder 3
ZeroPoint = 125.32Slope = 11.22E-05
ZeroPoint = 43.27Slope = 20.50E-05
40
60
80
100
120
140
160
180
0 50000 100000 150000 200000 250000 300000 350000
Dra
g (N
)
Sideforce2 (N2)
FlowLogic HKr-1 & HKr-2
HKr-1 Fn 0.35 Heel 15° Yaw var Twin Rudder 2
HKr-2 Fn 0.5 Heel 25° Yaw var Twin Rudder 3
ZeroPoint = 44.46Slope = 13.23E-05
ZeroPoint = 126.10Slope = 11.01E-05
Figure 19. FlowLogic Yaw Sweep at 15° and 25° Heel
Figure 20. FlowLogic Yaw Sweep at 15° and 25° Heel, Twin Rudder
18 Wide-Light Project
0
50
100
150
200
0.0 1.0 2.0 3.0 4.0 5.0 6.0
Dra
g (N
)
Speed (m/s)
SHIPFLOW CB-1
CB1-Pot & BL method-Trim v1 CB1-Pot & BL Method-Trim v2
CB1-RANS-Trim v1 CB1-RANS-Trim v2
34
36
38
40
42
44
46
-2.8 -2.7 -2.6 -2.5 -2.4 -2.3 -2.2 -2.1
Dra
g (N
)
LCG (m)
SHIPFLOW CB-2 & 3
CB2 Fn 0.35-Pot & BL mthd-Trim v1CB2 Fn 0.35-Pot & BL mthd-Trim v2CB2 Fn 0.35-RANS-Trim v1CB2 Fn 0.35-RANS-Trim v2
4.6 SHIPFLOW Results
The results and all associated files which were submitted can be downloaded from the SYRF Technical Resources
Library.9
Figure 21. SHIPFLOW Resistance Curves
5. Analysis and Discussion
Section 3 describes an approach to viewing the hydrodynamic behavior of a yacht hull that can be applied to any
sailing vessel. Applied to Wide-Light designs, three fundamental features are apparent: the upright resistance
increases with speed, the appendages represent a significant proportion of the total resistance, and the effective
draft of the hull and keel reduce as speed and heel angle increase. The model tank test results can be compared
with the CFD results with regards to the two following questions:
a) Do the computational results capture the general behavior?
b) Do the absolute values of the data points agree?
Although CFD results are acknowledged to differ point by point with the experimental data, their relative
differences can still be used as a reliable comparator. This approach can still be adopted for this study even though
there is only a data set for one hull.
9 Link to results: SYRF-WL SHIPFLOW Results.zip [56 MB]
19
-0.5
0.0
0.5
1.0
1.5
2.0
0.5 1.5 2.5 3.5 4.5 5.5
Trim
(d
eg)
Speed (m/s)
CB-1
-0.04
-0.03
-0.02
-0.01
0.00
0.01
0.02
0.5 1.5 2.5 3.5 4.5 5.5
He
ave
(m
)
Speed (m/s)
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
-0.5
0.0
0.5
1.0
1.5
2.0
0.5 1.5 2.5 3.5 4.5 5.5
Trim
(d
eg)
Speed (m/s)
HKR-1
-0.04
-0.03
-0.02
-0.01
0.00
0.01
0.02
0.5 1.5 2.5 3.5 4.5 5.5
He
ave
(m
)
Speed (m/s)
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
5.1 Results and Comparative Charts
The experimental and computational results are available in a combined spreadsheet from the SYRF Technical
Resources Library.10
5.1.1 Upright Resistance Trim and Heave – Tests CB-1 & HKR-1
The trim and heave curves for the tank test and CFD calculations are summarized in Figure 22 for the bare canoe
body and the appended model.
Figure 22. Trim and Heave Comparison – Zero Heel
10 Link to Results Spreadsheet: SYRF-WL Results.xlsx [1 MB]
20 Wide-Light Project
0
50
100
150
200
250
300
350
400
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Dra
g (N
)
Speed (m/s)
Resistance Curve CB-1 & HKR-1
Tank CB1 Tank HKR1Star-CCM+ CB1 Star-CCM+ HKR1OpenFOAM CB1 OpenFOAM HKR1FlowLogic CB1 FlowLogic HKR1FINE/Marine CB1 FINE/Marine HKR1
0.010
0.015
0.020
0.025
0.030
0.035
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Dra
g A
rea
(m2 )
Speed (m/s)
Speed vs. Drag Area CB-1 & HKR-1
Tank CB1 Tank HKR1Star-CCM+ CB1 Star-CCM+ HKR1OpenFOAM CB1 OpenFOAM HKR1FlowLogic CB1 FlowLogic HKR1FINE/Marine CB1 FINE/Marine HKR1
The Upright Resistance curves from the tank results and the computed results are shown in Figure 23. The left
hand plot is the force comparison, and the right hand plot is presented as a drag area, where:
𝐷𝑟𝑎𝑔 𝐴𝑟𝑒𝑎 =𝑅𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒
12⁄ 𝜌𝑉2
Resistance is in N,
ρ = water density (1000 kg/m3)
V = speed (m/s)
In the Drag Area plot, it is important to note the suppressed zero — the resistance at low speed is due to surface
friction resistance, and the increase in drag above that is due to residuary (wave making) resistance. For the canoe
body, only the friction and residuary resistance are of similar magnitude. For the appended model, the friction
resistance is nearly always greater than the residuary resistance. The friction resistance can be calculated from
published data using the wetted area and waterline length, while only the residuary resistance requires model tests
or CFD to predict.
Figure 23. Resistance Comparison – Zero Heel
A “ratio plot” approach is used to more clearly evaluate the magnitude of the differences between the results. The
results are expressed as the ratio of
𝐷𝑟𝑎𝑔 𝐶𝑜𝑚𝑝𝑎𝑟𝑎𝑡𝑜𝑟
𝐷𝑟𝑎𝑔 𝐵𝑎𝑠𝑒𝑙𝑖𝑛𝑒
21
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
0.5 1.5 2.5 3.5 4.5 5.5
Dra
g R
atio
(-)
Speed (m/s)
CB-1 Upright Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
0.5 1.5 2.5 3.5 4.5 5.5D
rag
Rat
io (
-)Speed (m/s)
HKR-1 Upright Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
for each test point. Where tank test data is available, the tank drag is used as the baseline and therefore the tank
test drag ratio is constant at 1.00. The results for the upright resistance tests are shown in Figure 24.
Figure 24. Upright Resistance Data, Ratio to Tank Test Results
In summary, Figure 22 shows that in broad terms the CFD prediction of the body motions agree well with the
towing tank values. Figure 24 for the CB-1 (canoe body only) case shows that most CFD points, for resistance
prediction, lie within 7-8 % of the tank value. However, Figure 24 also reveals that the agreement for OpenFOAM
and FlowLogic is less in the HKR-1 (appended) configuration. The lack of agreement for OpenFOAM is largely
because the computational scheme inherent in OpenFOAM meant that the computational time needed to
accurately capture the drag of the keel and rudder was far beyond that which could be devoted to this project. As
such, a simplified mesh was employed that reduced the accuracy of the appendage resistance predictions, but
which did not affect the prediction of heeled resistance ratio and effective draft.
5.1.2 LCG Variation – Tests CB-2 & 3, HKR-2 & 3
The ratio plots for the LCG shift tests on the hull alone (CB-2 & 3) are shown in Figure 25.
22 Wide-Light Project
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
-2.8 -2.7 -2.6 -2.5 -2.4 -2.3 -2.2 -2.1
Dra
g R
atio
(-)
LCG (m)
CB-2 Fn 0.35 Heel 0° Yaw 0° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
-2.8 -2.7 -2.6 -2.5 -2.4 -2.3 -2.2 -2.1
Dra
g R
atio
(-)
LCG (m)
CB-3 Fn 0.5 Heel 0° Yaw 0° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
-2.8 -2.7 -2.6 -2.5 -2.4 -2.3 -2.2 -2.1
Dra
g R
atio
(-)
LCG (m)
HKR-2 Fn 0.35 Heel 0° Yaw 0° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
-2.8 -2.7 -2.6 -2.5 -2.4 -2.3 -2.2 -2.1
Dra
g R
atio
(-)
LCG (m)
HKR-3 Fn 0.5 Heel 0° Yaw 0° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
Figure 25. Canoe Body Only Upright LCG Shift, Ratio to Tank Test Results
The ratio plots for the LCG shift tests on the appended hull (HKR-2 & 3) are shown in Figure 26.
Figure 26. Hull, Keel, & Rudder Upright LCG Shift, Ratio to Tank Test Results
23
Figure 3 from the tank report below shows typical behavior for a change of LCG – there is a discernible resistance
minimum for each speed at a given LCG position.
The CFD results mirror the tank behavior, although the rate of change of resistance moving away from the
minimum varies slightly as shown by the ratio plots, Figure 25 for the hull only results and Figure 26 for the hull
keel and rudder results.
5.1.3 Heeled Resistance Tests
The results of the heeled resistance tests are best expressed as a ratio of 𝑅𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑎𝑡 𝐻𝑒𝑒𝑙
𝑈𝑝𝑟𝑖𝑔ℎ𝑡 𝑅𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒 at each test speed. This
“Heel drag ratio” for the tests is summarised in Figure 27. For example, for the canoe body tests at 15 degrees
heel the Heel drag ratio for the Star-CCM+ result at 1.7 m/s is 0.947; in other words, when heeled, the drag of the
hull is 5.3% less than with the hull upright.
24 Wide-Light Project
0.85
0.90
0.95
1.00
1.05
1.5 2.0 2.5 3.0 3.5
He
el-
Dra
g R
atio
(-)
Speed (m/s)
15° Heel, HKR-4
0.85
0.90
0.95
1.00
1.05
1.5 2.0 2.5 3.0 3.5
He
el-
Dra
g R
atio
(-)
Speed (m/s)
25° Heel, HKR-5
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
0.85
0.90
0.95
1.00
1.05
1.5 2.0 2.5 3.0 3.5
He
el-
Dra
g R
atio
(-)
Speed (m/s)
15° Heel, CB-4
0.85
0.90
0.95
1.00
1.05
1.5 2.0 2.5 3.0 3.5
He
el-
Dra
g R
atio
(-)
Speed (m/s)
25° Heel, CB-5
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
Figure 27. Heeled Resistance Test Data Ratio to Upright Value
The data from Figure 27 is re-cast as a ratio plot in Figure 28 for the canoe body tests, and Figure 29 for the hull
keel & rudder tests.
Although it is not a condition that the yacht can ever sail in, the resistance of the yacht when heeled at zero yaw
angle provides a reliable indicator of the hull’s characteristics. With this understanding, Figure 27, which only
illustrates CFD results for the canoe body, shows consistent hull behavior, with the heeled hull consistently
having less resistance than the upright hull because the wetted area reduces as heel angle increases. A similar
behavior is shown for the appended hull, but the resistance reduction is somewhat reduced because the keel and
rudder wetted surface do not change with heel. The towing tank results also show a trend of heeled resistance
increasing as speed increases, and this is generally captured by the CFD results. Again, the ratio plots show that
the trends are captured, but the absolute values differ to the same degree as the upright resistance results.
25
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2
Dra
g R
atio
(-)
Speed (m/s)
CB-4 Heel 15° Yaw 0° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2
Dra
g R
atio
(-)
Speed (m/s)
CB-5 Heel 25° Yaw 0° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2
Dra
g R
atio
(-)
Speed (m/s)
HKR-4 Heel 15° Yaw 0° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2
Dra
g R
atio
(-)
Speed (m/s)
HKR-5 Heel 25° Yaw 0° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
Figure 28. Canoe Body Only Heeled Resistance Test Data, Ratio to Tank Test Results
Figure 29. Hull, Keel & Rudder Heeled Resistance Test Data, Ratio to Tank Test Results
5.1.4 Heel with Yaw – Tests CB-6 &7, HKR-6 & 7
The results of the yaw sweeps on the bare canoe (CB-6 & CB-7) are presented in Figure 30 as plots of Drag vs.
yaw angle and Sideforce vs. yaw angle. Because the model is heeled these do not show a discernible minimum at
zero yaw. In fact the drag minimum is at a negative yaw angle because this aligns the centerline of the heeled
waterplane more nearly with the hulls direction of travel through the water.
26 Wide-Light Project
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
-2 -1 0 1 2 3
Dra
g R
atio
(-)
Yaw (deg)
CB-6 Fn 0.35 Heel 15° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
-2 -1 0 1 2 3
Dra
g R
atio
(-)
Yaw (deg)
CB-7 Fn 0.5 Heel 25° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
20
40
60
80
100
120
140
-2.0 -1.0 0.0 1.0 2.0 3.0
Dra
g (N
)
Yaw (deg)
CB-6 Fn 0.35 Heel 15°, CB-7 Fn 0.5 Heel 25°
15° - Tank 25° - Tank15° - Star-CCM+ 25° - Star-CCM+15° - OpenFOAM 25° - OpenFOAM15° - FlowLogic 25° - FlowLogic15° - FINE/Marine 25° - FINE/Marine
-10
0
10
20
30
40
50
60
70
80
90
-2.0 -1.0 0.0 1.0 2.0 3.0Si
de
forc
e (
N)
Yaw (deg)
CB-6 Fn 0.35 Heel 15°, CB-7 Fn 0.5 Heel 25°
15° - Tank 25° - Tank15° - Star-CCM+ 25° - Star-CCM+15° - OpenFOAM 25° - OpenFOAM15° - FlowLogic 25° - FlowLogic15° - FINE/Marine 25° - FINE/Marine
The yaw sweep results are presented as ratio plots to the tank data in Figure 31.
Figure 30. Canoe Body Only Yaw Sweep Data at 15° and 25° Heel
Figure 31. Canoe Body Only Yaw Sweep Data, Ratio to Tank Test Results
The results of the yaw sweeps with the keel and rudder (at zero rudder angle) fitted (HKR-6 & HKR-7) are
presented in Figure 32 as plots of Drag vs. yaw angle and Sideforce vs. yaw angle. Here the drag minimum is
27
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
-2 -1 0 1 2 3
Dra
g R
atio
(-)
Yaw (deg)
HKR-6 Fn 0.35 Heel 15° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
-2 -1 0 1 2 3
Dra
g R
atio
(-)
Yaw (deg)
HKR-7 Fn 0.5 Heel 25° Ratio
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
20
40
60
80
100
120
140
160
180
200
-2.0 -1.0 0.0 1.0 2.0 3.0
Dra
g (N
)
Yaw (deg)
HKR-6 Fn 0.35 Heel 15°, HKR-7 Fn 0.5 Heel 25°
15° - Tank 25° - Tank15° - Star-CCM+ 25° - Star-CCM+15° - OpenFOAM 25° - OpenFOAM15° - FlowLogic 25° - FlowLogic15° - FINE/Marine 25° - FINE/Marine
-300
-200
-100
0
100
200
300
400
-2.0 -1.0 0.0 1.0 2.0 3.0
Sid
efo
rce
(N
)
Yaw (deg)
HKR-6 Fn 0.35 Heel 15°, HKR-7 Fn 0.5 Heel 25°
15° - Tank 25° - Tank15° - Star-CCM+ 25° - Star-CCM+15° - OpenFOAM 25° - OpenFOAM15° - FlowLogic 25° - FlowLogic15° - FINE/Marine 25° - FINE/Marine
closer to zero yaw because this corresponds to the minimum Sideforce from the appendages. The yaw sweep
results are presented as ratio plots to the tank data in Figure 33.
Figure 32 Hull Keel & Rudder Yaw Sweep Data at 15° and 25° Heel
Figure 33. Hull, Keel & Rudder Yaw Sweep Data, Ratio to Tank Test Results
28 Wide-Light Project
0
50
100
150
200
250
0 50000 100000 150000 200000 250000 300000
Dra
g (N
)
Sideforce2 (N2)
HKR-8 Fn 0.35 & HKR-9 Fn 0.5 Heel 15°
Tank HKR8 Tank HKR9Star-CCM+ HKR8 Star-CCM+ HKR9OpenFOAM HKR8 OpenFOAM HKR9FlowLogic HKR8 FlowLogic HKR9FINE/Marine HKR8 FINE/Marine HKR9
0
50
100
150
200
250
0 50000 100000 150000 200000 250000 300000
Dra
g (N
)
Sideforce2 (N2)
HKR-10 Fn 0.35 & HKR-11 Fn 0.5 Heel 25°
Tank HKR10 Tank HKR11Star-CCM+ HKR10 Star-CCM+ HKR11OpenFOAM HKR10 OpenFOAM HKR11FlowLogic HKR10 FlowLogic HKR11FINE/Marine HKR10 FINE/Marine HKR11
Just as with the previous figures, the ratio plots in Figures 31 and 33 show that the CFD results capture the trends
shown by the tank test results, with similar divergence of absolute results that are seen consistently throughout the
tests.
5.1.5 Yaw Sweeps Hull, Keel, & Rudder – Tests HKR-8, 9, 10, & 11
As shown in Figure 2 of Section 3, the results of the yaw sweep tests can be captured in a single straight line on a
resistance vs Sideforce2 plot. The yaw sweep data are presented in Figure 34.
Figure 34. Hull, Keel & Rudder, Summary of Fitted Lines to Yaw Sweep Data
As described in Section 3.2, the character of the Resistance vs Sideforce2 plots can be expressed as a heeled drag
ratio RH/RU and an effective draft, expressed in meters. Figure 35 shows the data expressed in this way.
29
0.90
0.92
0.94
0.96
0.98
1.00
1.02
1.04
2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0
He
el-
Dra
g R
atio
(-)
Speed (m/s)
HKR-8 & 9 Heel 15°
0.0
0.2
0.4
0.6
0.8
1.0
2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0
Effe
ctiv
e D
raft
(m
)
Speed (m/s)
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
0.90
0.92
0.94
0.96
0.98
1.00
1.02
1.04
2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0
He
el-
Dra
g R
atio
(-)
Speed (m/s)
HKR-10 & 11 Heel 25°
0.0
0.2
0.4
0.6
0.8
1.0
2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0
Effe
ctiv
e D
raft
(m
)
Speed (m/s)
Tank Star-CCM+ OpenFOAM
FlowLogic FINE/Marine
Figure 35. Heeled Resistance Ratio and Effective Draft Comparison
In broad terms, the Star-CCM+ and FINE/Marine codes are closest to the towing tank results in both absolute
terms and capturing the slope of the Drag vs. Sideforce2 line. The FlowLogic results capture the slope, but the
OpenFOAM results are at some variance to the experimental result for the reasons discussed in 5.1.1.
While no tank tests were carried out on the twin rudder configuration, a similar summary of the Resistance vs.
Sideforce2 results for the twin rudder configuration are shown in Figure 35.
30 Wide-Light Project
0
50
100
150
200
250
0 50000 100000 150000 200000 250000 300000
Dra
g (N
)
Sideforce2 (N2)
HKr-1 Fn 0.35 Heel 15°, HKr-2 Fn 0.5 Heel 25°
Star-CCM+ HKr-1 Star-CCM+ HKr-2
OpenFOAM HKr-1 OpenFOAM HKr-2
FlowLogic HKr-1 FlowLogic HKr-2
FINE/Marine HKr-1 FINE/Marine HKr-2
5.1.6 Yaw Sweeps Twin Rudder – Tests HKr-1 & 2
Figure 36 shows the Resistance Sideforce2 plots for the twin rudder results derived from the CFD calculations. No
tank tests were made with the model in this configuration.
Figure 36. Comparison of CFD Results for Twin Rudder Configuration
5.2 Lift Curve Slope
This report has focused on the comparison of resistance and side force with speed and heel angle, as these are the
main drivers of performance. However, the data was also analyzed to determine a simple lift slope from the HKR-
8, 9, 10, & 11 tests. These results are presented in the Table below.
The value was determined from the slope of the line 𝑆𝐹/𝑉2 versus yaw angle where Sideforce (SF) is in N and
Speed (V) in m/s.
The results are presented in Figure 37 and reflect a satisfactory degree of consistency – the trend of lift slope
reducing as heel angle increases is captured by CFD. However, these results reveal that the reduction of lift slope
as speed increases is not captured.
31
Figure 37. Lift Slope Comparison
5.3 Discussion
The fundamental aim of this project was to explore how computational methods might be employed to inform
force models used in VPP based handicap rules. The fundamental components of a sailing yacht’s resistance are
described in Section 3.2.
𝑅𝑇𝑂𝑇 = 𝑅𝑈 + 𝑅0 + 𝑅𝐼
Upright Resistance (RU) comprising:
Wave resistance (RW)
Appendage viscous drag (RVapp)
Canoe body viscous drag (RVcb)
R0 (R zero) the increase in drag above RU at zero sideforce from the fitted line to the test data, and,
RI the Induced drag.
The table below describes how the computational results compared with the physical model test.
Table 5. Comparison Summary – CFD to Model Test
Component
Upright Resistance Figure 24 shows that the computational methods are tolerably capable
of predicting the upright resistance of both the bare canoe body and the
appended hull. Generally there is an offset of a few % across the speed
range, and the CFD generally under predicts the resistance. As
discussed in Section 5.1.1, for these wide, light hull forms nearly 50%
of the resistance is skin friction drag that can be calculated simply from
published data and hull wetted surface area.
The computational methods are also sensitive to the effects of shifting
the center of gravity fore and aft.
Heeled Resistance Figure 29 shows that the advanced RANS codes (Star-CCM+ &
FINE/Marine) are able to accurately predict the effects of heel on
resistance.
Fn 0.35, Heel 15° Fn 0.5, Heel 15° Fn 0.35, Heel 25° Fn 0.5, Heel 25°
Tank 13.4 12.3 9.5 8.8
Star-CCM+ 10.7 11.5 8.2 9.1
OpenFOAM 9.8 10.8 8.0
FlowLogic 11.5 12.5 9.2 10.3
FINE/Marine 10.5 11.2 8.1 8.9
0.0
5.0
10.0
15.0
32 Wide-Light Project
Resistance due to yaw angle Figures 31 and 33 show that the computational methods are all able to
predict the effect on sideforce of changing yaw angle.
Heel Drag Ratio The upper plots in Figure 35 show that the computational methods are
all able to predict the increase of drag at zero sideforce as speed
increases, although of course the “offset” found in the upright
resistance curves is carried through into the results.
Induced Resistance The lower plots in Figure 35 show that the tank test derived effective
draft reduces as speed increases, this trend is not well captured by the
computational codes. For handicapping work it is crucial that the force
models capture this effect because it is where designers can seek to find
a performance advantage by optimizing the hull and keel volume
distributions.
6. Conclusions
This first phase of the Wide-Light project brought together several respected CFD technicians with experience in
evaluating the hydrodynamics of sailing yachts. They were able to collaborate on setting up the tank test program
for an existing model of a boat fitting the parameters of Wide and Light design. Using the geometry of the boat
and the design of the tank test program they performed pre-test evaluations of that model. CFD results were
compared with the tank test results.
The study described in this report is the most comprehensive ever undertaken for publication in the public
domain, and has met the goals that were set. The following broad conclusions may be drawn.
A body of physical test data relating to a defined geometry has been published and is available for
validation of other data.
Commercial CFD codes may be used to confidently predict the variation of the forces on a sailing yacht
hull as speed heel and leeway change. These studies do not need prohibitively large mesh density to
achieve valid results.
Less computationally heavy codes, e.g. FlowLogic can produce data to capture the typical global behavior
of a sailing yacht hull.
Yacht Handicapping organizations can rely on correctly configured CFD studies to generate data for a
wide variety of yacht hull shapes. (Reference 2)
7. Future Work
A second phase of the Wide-Light project is planned to build upon the lessons from Phase 1. Specifically, Phase 2
is intended to take one or two of the promising CFD programs of Phase 1, provide their contributors with the
geometry of a small fleet of designs and let them evaluate those designs and the combinations of
speed/heel/leeway that are of interest to handicap rule-makers. The fleet of boats included in Phase 2 will
represent realistic variations from a baseline design in the critical parameters that drive performance:
displacement and beam for a fixed length are an obvious choice of variations.
33
The proposed Phase 2 program will build upon the success of an earlier study conducted in the 1990s: the nine
model tank test program conducted at the National Research Council (NRC) in Canada. The NRC's program has
been immensely useful in providing the differences in hydrodynamics for a range of beam and displacement
variations that bracketed the fleet of boats racing worldwide in the 1990s. Since the NRC study, there has been
rapid growth in the number of new Wide-Light designs with significantly wider beam and lighter displacement
than the designs included in the NRC’s nine-model series. With these new Wide-Light designs represented at
every major big boat regatta, it is paramount that handicap rules treat them fairly. To equitably rate them against
the existing fleet of heavier and narrower designs, it is imperative that the developers of those rules have a
database of hydrodynamics of these Wide-Light designs.
To build the most accurate and informative database requires understanding the nuances of existing CFD
programs. Wide-Light Phase 2 intends to continue refining our understanding of CFD programs so as to better
inform handicappers in fairly rating these new and popular designs. As Phase 1 has demonstrated the value of
CFD analysis in predicting differences in hydrodynamics between designs, and because a multiple model tank
series is prohibitively expensive, Phase 2 will not use tank testing. Instead, Phase 2 will focus solely on the use of
the most accurate CFD programs to develop a database of the hydrodynamics of Wide-Light designs. Upon
completion, Phase 2 will produce the first publicly available database of Wide-Light hydrodynamics.
34 Wide-Light Project
References
1. Lars Larsson, Frederick Stern, Michael Visonneau, “Numerical Ship Hydrodynamics, an assessment of
the Gothenburg 2010 Workshop.” ISBN 978‐94‐007‐7188‐8
2. ORC Technical Committee, ORCi Documentation.
<http://www.orc.org/rules/ORC%20VPP%20Documentation%202015.pdf>.
3. Keuning, J.A., Sonnenberg, U.B., Developments in the Velocity Prediction based on the Delft Systematic.
Yacht Hull Series, The Modern Yacht Conference March 1998.