Data for a VPP?

[Joakim]

Hi, I'm trying to make my own VPP as a hobby. I've developed it for a month now, but there are still a lot of uncertanties.

1. Drag due to seakeeping. I have found some papers on this subject, but none of them gives all the data needed. I haven't found any reference on what kind of sea conditions are asumed for each wind speed in common VPP programs (IMS etc.). Fetch and duration of the wind have huge effects. My VPP gives easily 50 sec/mile Beat VMG differences depending on the assumptions made. Some links on the subject: http://www.rina.org.uk/rfiles/HISWA/2002/17th%20-9-%20Sailing%20yacht%20performance%20in%20calm%20water%20and%20waves.pdf
http://www.seakayak.ws/kayak/kayak.nsf/NavigationList/NT0000C07A
http://www.curtin.edu.au/curtin/centre/cmst/publicat/ches99_paper.pdf

2. What is the best source for sail forces? How is the aerodynamic effect of the mast and hull modelled? Is the vertical wind profile kept constant or does it depend on the wind speed?

3. How important it is to take into account yaw moments? Surely they are not easily calculated for each sail inventory and trimming of the sails.

4. How to guess hull parameters for hulls without all the data? Can LCB and LCF be somehow taken from an IMS certificate? How to approximate Cp, LCB and LCF for on unmeasured old hull?

5. I have seen some conflicting opionins on induced drags of foils. All seems to agree with Cd=CL^2/PI/ARe, but the calculation of ARe can be b^2/A or 2*b^2/A. Which is more accurate? What about rudders very close to water surface? How does IMS actually use the effective draft, which seems to be considerably smaller than by Gerritsma et. al. (http://www.rina.org.uk/rfiles/HISWA/2002/17th%20-9-%20Sailing%20yacht%20performance%20in%20calm%20water%20and%20waves.pdf)

6. How accurate are different VPP programs with respect to each other and real life? In what conditions are the errors largest?

7. Why is it so hard to get information about this? Most measurements are done by academia and most VPP-programs as well. I have tried to contact a few persons at different universities, but only one of them replied. I have collaborated with several universities in different area of fluid mechanics and it has never been this hard to get papers, data, models and programs.

8. Is PCSAIL the only free VPP available? I haven't found any other and it is not good enough for my needs (inaccurate and very slow due to Excel).

9. I would really appreciate a copy of "Predicting the speed of sailing Yachts" from 1993 by van Oossanen.


Joakim

[DaveB]

What have you been using to develop it?

Would you be willing to share it with the group? I'd be interested to see what you've got.

Principles of Yacht Design's got some discussion of them and fundamentals... might be helpful.

Cheers,

Dave

[Joakim]

I have used a lot of sources.

I started from the sail forces by WB-Sails: http://www.wb-sails.fi/news/SailPowerCalc/SailPowerCalc.htm

I converted this java into C and added automatic depowering, reefing, jib size choosing and spinnaker set up. I also modified the downwind coefficient, which were not very accurate. I also took vertical wind profile into account:
http://www.redcoopers.net/VPP.htm

Skin friction is calculated with the standard ITTC formula or based on the roughness of the surface. For Reynolds number I'm know using 0.9*LWL instead of 0.7 often reported.

Lift and induced drag of keel and rudder is calculated based on wing theories developed for aircrafts:
http://www.mh-aerotools.de/airfoils/hdi_aoawing.htm
http://www.grc.nasa.gov/WWW/K-12/airplane/short.html
http://www.vacantisw.com/tutorial.htm

Form factors are still a bit open question. Now I'm using 1.0 for hull and 1.1 for foils.

Wave making drag is based on Delft series, but I developed a simpler model, since I don't have any LPP or knowledge of all the variables needed for all hulls of interest. http://www.rina.org.uk/rfiles/HISWA/2002/17th%20-9-%20Sailing%20yacht%20performance%20in%20calm%20water%20and%20waves.pdf


Propeller drag is based on my own knowledge and some measurements reported: http://www.irisoft-yacht.com/helices/revue-de-presse/.

I have developed a simpler model for seakeeping based on:
http://www.rina.org.uk/rfiles/HISWA/2002/17th%20-9-%20Sailing%20yacht%20performance%20in%20calm%20water%20and%20waves.pdf
http://www.seakayak.ws/kayak/kayak.nsf/NavigationList/NT0000C07A

The righting moment is based on polynominal fit to stability curve (e.g. from an IMS certificate) or on RM2.

I also have a model for aerodynamic drag and sideforce of the hull and rig, which is based on my own ideas.

Drag due to heel is based on: http://www.rina.org.uk/rfiles/HISWA/2002/17th%20-9-%20Sailing%20yacht%20performance%20in%20calm%20water%20and%20waves.pdf
http://www.zwakenberg.de/hulldrag/page5.html

The solver is quite simple. It searches the best sail inventory for every TWA and TWS, taking into account heel and sideforces. It can also find best VMG angles. It consists of several iteration loops (VMG, speed for one TWA, TWS, speed at constant forces, heel vs. sail forces, etc.). It also computes an IMS certificate and comperes it to an existing file.

There is also a module for calculating speeds for motoring in different wind conditions.

The program is now 750 lines of C without comments. It is not ready to be shared and I haven't yet desided will I ever share it. I made the basic VPP in couple of days. The real work just started after that!

I know, that "Principles of Yacht Design" is one good source. I haven't just found one yet. I'm a bit doubtful about it's accuracy. It has been the source for PCSAIL, which I don't find accurate enough.

What I have learned so far, is that predicting Beat VMG is a very delicate prosess. Lots of things have a huge effect on it (aerodynamic drag of hull and rig, seakeeping, propeller drag, wind profile, minor changes in hull drag etc.). During the process I have lost the faith in the accuracy of IMS Beat VMG.

[redcoopers]

Hi,

As mentioned above, I do have a lot of information on making a simple VPP at: http://www.redcoopers.net

However, I also have a spreadsheet which calculates the DELFT resistance and MICHELL resistance of a sailboat. The spreadsheet calculates only the unyawed - upright - wave making resistance of only the canoe body. It works pretty well; a lot could be added to it, but I'm not going to update it. You can find the spreadsheet at: http://redcoopers.servebeer.com

I think I have another spreadsheet which calulcates the entire DELFT wavemaking of a sailboat and it includes all sailing conditions (heel and yaw). I've already combined the hull resistance with aerodynamic forces, so it is a full VPP for excel. I'd have to check around my harddrive for it, but if anyone is intereseted, I'll post it somewhere too.

-Jon

[DaveB]

Geeze, impressive! I'd be very interested in the spreadsheet if you wouldn't mind posting it!

Cheers,

Dave

[Matthew C White]

I am looking for some type of program or spreadsheet to predict the seakeeping of a 60' cruising sailboat. If anyone can give me a lead as to where I can find one it would be greatly appreciated. I am trying to finish a senior design project. Thanks for you time. You can reach me at spidermw81@aol.com.
Thanks,
Matt White

[mpyd]

I would like to have a look to your spreadsheet as well!!! :?:

[DaveB]

I am looking for some type of program or spreadsheet to predict the seakeeping of a 60' cruising sailboat. If anyone can give me a lead as to where I can find one it would be greatly appreciated. I am trying to finish a senior design project. Thanks for you time. You can reach me at spidermw81@aol.com.
Thanks,
Matt White

Hey,

There's a program with maxsurf called seakeeper...

There's also a section in PNA that deals with Lewis sections and helps you estimate RAOs... PNA stands for Principles of Naval Architecture and can likely be had in any library at a naval arch school.

Cheers,

Dave

[Stephen Ditmore]

To Joakin's question 1, see if you can find anything on the MIT Ocean Engineering web site. In the 90s MIT developed a program for calculating resistance in waves that was used by the America^3 America's Cup campaign in both 1992 and 1995 under the direction of MIT Prof. Jerry Milgram.

Jim Teeters of Teeters Yacht Development (or something like that) in Newport, RI is one of the authors of the IMS VPP. Look for papers by him published by SNAME.

If you want to know about the history of early VPPs look for the name Hugo Meyers (now deceased, formerly with IBM but also a yacht designer) and the Pratt Project under MIT Prof. Justin Kerwin.

[sorenfdk]

Hi,

I think his name is spelled Hugo Myers (without an "e").

PCSail may not be the most accurate, but then again: What can one expect from a spreadsheet VPP? I think it is pretty OK for preliminary work, and You can't beat the price...

Does anybody know if there exists a version of PCSail that uses SI-units instead of feet etc.?

Happy easter to You all!

[tspeer]

You can calculate the added resistance RAO's using SMP, available from NTIS (http://www.ntis.gov). This should correspond to the linear theory in the Gerritsma paper.

Unfortunately, the best known VPP's have been developed for rating purposes, not accurate performance prediction. The data are often skewed when entered into the VPP because the VPP is known to under- or over-predict the performance for certain cases. And authorities, such as US Sailing - whose technical program is paid for in part by our dues - does not often make the data they collect public because they don't want designers to use the data to improve boat performance! They are afraid that the designers will exploit weaknesses in the prediction and make the ratings less accurate.

For a good example of this process, and information on effective span, see http://www.rina.org.uk/rfiles/IJSCT/Discuss/teeters.pdf.

But getting good aero- and hydrodynamic coefficients for a specific design, and determining the small differences due to design alternatives, is hard. That's precisely why design teams pay big bucks for tank and wind tunnel tests. CFD is coming along, but both the cost and the expertise are beyond the typical small craft designer.

Past CSYS papers are a good source of info: http://wseweb.ew.usna.edu/nahl/csys/default.htm

Here's a commercially available VPP:
http://www.wumtia.soton.ac.uk/brochures/VPP.html


See also:
http://www.rina.org.uk/rfiles/IJSCT/Discuss/hansen.pdf
http://www.curtin.edu.au/curtin/centre/cmst/publicat/rinamodernyacht98.pdf
http://www.curtin.edu.au/curtin/centre/cmst/publicat/chesapeake97.pdf
http://www.curtin.edu.au/curtin/centre/cmst/publicat/sailingscience99.pdf
http://www.rina.org.uk/rfiles/IJSCT/Discuss/keuning.pdf
http://www.nap.edu/books/NI000359/html/
http://www.curtin.edu.au/curtin/centre/amecrc/papers/prc98/
http://sic.epfl.ch/SA/publications/SCR98/scr10-page24.html
http://jeanphilippe.boin.free.fr/isope2002.pdf
http://jeanphilippe.boin.free.fr/newzel2002.pdf
http://www.gidb.itu.edu.tr/staff/%7Einsel/Publications/Cesme.PDF
http://www.ussailing.org/imsbakup/spinnakertests.htm

[Stephen Ditmore]

Thanks for the above contribution, Tom. I hope you don't mind if I add http://www.schwenn.com/ to your list of commercially available programs worth checking into. Note also: I found a number of other interesting sites when I did a google search on "Velocity VPP", including http://www.sailrater.com/

[Joakim]

Thanks a lot for this information. I've already found many of the papers listed, but I've to look closer to some of them. The best paper so far has been:

http://www.rina.org.uk/rfiles/HISWA/1998/15th%20-6-%20Approximation%20of%20the%20hydrodynamic%20forces%20on%20a%20saili.pdf

But there are some errors and unknown things in this paper.

1. How is LCB defined in equation (10) (% , m, LCB (m) /LWL) ???

2. In equation (16) the third term is somehow wrong. I have seen another version ( A2 * (Tc + Zcbk) / Vk^(1/3) ) with the same coefficients. The latter gives more reasonable results.

3. In Fig. 16 I can reproduce all other resistances expect the residual resistance of keel, which is about 800 N at Fn=0.5, but with equation 16 I get something like 400 N.

4. In equation (24) and also in some other equations there are keel parameters, which are not well defined for bulb keels. How to set TR for example? TR is not defined at all in this paper, but it seems to be tip chord / root chord.

5. What kind of interpolation would be best for VPP usage? Most equations only give values for discrete Fn or Heel values.

Joakim

[tspeer]

1. I believe LCB in the paper is %LWL, + aft of midships.

2. I can't help you with the discrepancies in eq. 16. The term in the paper is the cube of your expression - maybe there was a transcription error somewhere.

4. The common aeronautical practice is to extend the leading and trailing edges to the end and measure the (virtual) chord there, even though it's buried in the body. I believe you're right, TR is the taper ratio.

I think the best interpolation would depend on the particular term. For example, in Fig 18, the induced drag varies with sideforce^2. The ideal would be to find some relationship that reduces the shape of the curve to a straight line, then interpolate based on those parameters. You might try Fn^2, for example.

Probably the most common approach is to simply interpolate linearly between the points given. If there are enough break-points, this is will be within the accuracy of the fit to the data.

[Joakim]

1. Do you mean, that LCB in this equation would be typically be from 0 to +8% or from -8 to 0%? In all other equations LCBfpp is used which is meters from the front edge of LWL and the actual term used is LCBfpp/LWL, which makes sence. It is very annoing not to define all variables exactly!

4. If I have a keel with a 0.85 m root chord, a 0.65 m chord just above bulb and a 1.1 m long bulb, then the tip chord is 0.5X. Is that correct?

I think there is not enough data for linear interpolation. E.g. residual drag of hull is given at Fn from 0.1 to 0.6 at 0.05 intervals, which means that drag can change very much between the intervals. E.g. one yacht has residual drag of 185 N at Fn=0.35 and 686 N at Fn=0.4. At Fn=0.375 linear interpolation would give 436 N and my current interpolation (6th order polynomial fit at two parts) gives 367 N.

[nico]

In winvpp, i think the residual drag is interpolated with a cubic polynomial curve. I agree that a linear interpolation is not correct. Particularly for Fn around 0.4 where the values change a lot.
LCBfpp is the distance in meters from the forward end of LWL
LCB is as presented in table 2, and should be negative a LCB of -3.28 means a LCBfpp of LWL/2+3.28/100*LWL.
What programming language are u using?

[tspeer]

Oops. I may have the sign wrong. As a practical matter I don't think you'll find any hulls that actually have the LCB ahead of midships.

I'd say the taper ratio of the keel is 0.76, or a little less when you extrapolate the blade to the bottom of the bulb. The bulb is ignored for the purpose of defining the reference area.

[Joakim]

I'm using C language.

I had already assumed that LCB would be the % values in table 2, but using a % value in residual drag equation does not seem to fit the logic used in other equations, which only uses non-dimensional parameter derived from actual measures (e.g. LCBfpp/LWL, VOLc^1/3 / LWL) and all the ratios (Cm, Cp) are in fractional from not in % (e.g. Cp=0.54 not 54%).

I have also checked the excel sheet that Nico has sent here and in it LCB is given in %, but with a wrong sign, but it is only used to calculate LCBfpp. Even more confusing....

Joakim

[nico]

in my spreadsheet LCB is positive but i have changed the sign in the equation that computes the residual drag -*- = +

[nico]

I think that the negative LCB in % is the way delft have described LCB from the start (for the last 25 years). With the LCBfpp/LWL they are probably moving to a form that is used most often used (ie. LCB=0.56).
% is just an easier way to write fractionnal number

[MikeJohns]

Joakim

Did you try http://www.vacantisw.com for his foil lift/drag prediction software?

[Guest]

http://www.wb-sails.fi/news/SailPowerCalc/SailPowerCalc.htm

Aerodynamic calculations in this site seem a bit over-optimistic. Does anyone know where I can find a table of cl cd coeficients for a simple sail? Thanks in advance...