Lift coefficients of the sail

Hello everyone,
I'm working on a project where I need lift coefficients of the sail of a boat. I cant find anywhere some charts where they're given for some models of boats.
Thank for any help you can give me !

ps: I'm not a native english speaker hence my bad phrasing etc. so if I'm not making myself clear enough please tell me

 

I do sympathise - last time I was looking I found it hard to find good aerodynamic data for fabric sails. I ended up with some approximate data (see graphs below) based on combining and averaging several sources of information, unfortunately this was in 2016 and I dont now seem to have the references I used.

Please note the following points:
This data was the best I could find at the time, I am hoping that others here are soon going to offer something better. I am sure some people here have good CFD results.

My project at the time was to compare alternative configurations of lifting hydrofoils, both in 'foil assist' mode and in 'full flying' mode. Hence I was not all that bothered that I didnt have really good rig data, I just wanted some rig data to apply to all the hydrofoil models for purposes of comparison - i.e. I was more interested in the relative merits of different hydrofoil configurations rather than in absolute speed prediction.

I think most of the information came from wind tunnel tests on models rather than CFD, and some of the wind tunnel tests were from the early days of experimental aerodynamics. I recall that some of the wind tunnel tests included drag from rigging, some did not, I think some may even have included drag from hull and superstructure.

I have refered to a generic rig - this means a typical bermudian rig of moderate aspect ratio.

Some of the data sources gave higher Cl values than the curve below - I recall that I ignored really high values since I did wanted a 'safe' performance estimate i.e. not an over-optimistic one.

There is no lift at small AoA, this is because the rig is 'luffing'

 

I don't actually have a copy of Marchaj and as far as I can tell it isn't available anywhere online, but this paper has some data from one of his tests of a Dragon-class rig.
Modelling the aerodynamics of upwind sails,P.S.Jackson

 

This may be of use to you.

 

Yull,
Have you ever considered this as a test? There is no "correct" answer for your question, only an upper limit and a lower limit. Each Naval Architect is expected to have his/her own answer for the un-answerable question you/they were poised. When I left my last position before I retired, everyone wanted a copy of my research and data sheets, these were developed over an entire 35 year carrier, but none were the actual answer. I could bracket the answers, but none I could give you is "correct" answer. The reason you could not find one is BECAUSE THERE IS NONE! You need to develop your own opinion of what is the maximum or minimum and the answer could be based upon on things you cannot express in definitive terms. This is what a Naval Architect is paid for. Just asking for an answer will always be incorrect.

 

Yull,
Funny, I just connected in order to start a thread regarding the correlation between wing section parameters and aerodynamic coefficients. But I need to write correctly the question first.

For your question, you can guess the sail max lift coefficient (2D) with airfoiltools.com & Eppler 376 airfoil section.
It seems to be a good proxy for teardrop mast+ full batten sails like A-Cat or sail like windsurf or Moth foiler with the mast in a sleeve.

Airfoiltools is based on XFOIL. There is also XFLR5 a bit more user friendly but some XFOIL features are not available (at least H parameters and other useful for separation).

I think Cl=1.7 is a quite optimistic maximum to be considered.

For classic rigs with fixed masts it is much more complicated as explained in Ad Hoc's workpaper.
Good research
Erwan

 

SailPowerCalc https://cdn2.hubspot.net/hub/209338/news/SailPowerCalc/SailPowerCalc.htm

 

Yull, the problem with your question is that the answer to the question as posed has no practical value.

The most basic, primative model that has any relevance to boat design or performance modeling has to take into account the sail adjustments that are found on all real sails.

For a Bermuda rig, the basic external parameters are true wind angle and true wind speed (TWA, TWS), and sea state.
The most basic set of sail parameters has to include main sail reefing, mainsail twist control, jib twist control, boom angle, and jib sheeting angle. A realistic hull performance model is usually needed to convert TWA and TWS to apparent wind conditions.

You just can't get anywhere without having a handle on all of these. When a sailor wants to change heading, or the conditions change, they will adjust most or all of the settings mentioned above, and your plot isn't a curve, but a series of surfaces that consider the range of sail settings for each heading and wind speed.

So the starting point is to develop Cl, Cd, vertical center of side force, and vertical center of forward force (yes, they are different) data for all AoA conditions for each reef state, and for the range of sail twist for each reef, and for all combinations of boom and jib sheet positions that make sense. Then you match these to a particular hull, which has righting moment constraints (transverse and longitudinal) and a complicated resistance profile, and you iterate and eventually identify the regions of the above dataset that have application to that hull.

Until you have all that, you really have nothing at all.

One way you get all that is to collect all the performance data you can, preferably from race trackers or long distance cruiser logs, identify the sails in use, and reverse engineer what the sails are doing.

 

Thank you all for your help, I'll try to dig in the informations you sent and find what I need. On the off chance I still have more questions, I'll come back there !

 

Yull,
As mentionned in my former msg I am investigating a similar issue: Spanload distribution on an A-Cat rig in different conditions corresponding to
1-Elliptical spanload distribution
2-Bell-Shaped spanload distribution

It does not require any cfd skill not even VBA , just a basic EXCEL calculation sheet.

For the Elliptical distribution, As the righting moment is constant,
I must find the maximum apparent wind velocity which makes a full use of the righting moment without being overpowered.
For the Bell shaped spanload, and for the same righting moment I must find the minimum apparent wind velocity necessary to use the full righting moment.

The link between true wind and apparent wind would required a polar of the boat, unfortunatly not available most of the time. So I use windward & downwind datas collected from experienced sailors including on forums.

For the true wind gradient I use the famous log formula (I checked it vs Frank Bethwaite plot in his famous book "High Performance Sailing" (Some differences between 0 and 3 feet height, so not important).
From that, I got the apparent wing along the span, I use canonical formulas:
for the elliptic/circular spanload y=[(1-x^2)]^0.5
for the Bell Shaped spanload* y=[(1-x^2)]^1.5
Of course I split the[0-1] spanload domain into 100 lines (0 to 100%) if you prefer

Then( That is where I arrived) I want to add/sub the downwash/upwash angle to the apparent wind angle, according to the derivative formula of the lift equation.(I am messing a bit with the canonical angle, probably Pi*Radiant units, I have to check).

I can calculated the canonical lift and the centre of effort just with Excel sum function, not by primitive/ Integral calculation, (of course there is a little difference 42.165% for the elllipse instead of 42% theorical AFAIR.)

And I also get the implicit lift for each strip of 1% spanload sail area, so I have to find the best couple of (2D lift coefficient , angle of attack) which minimize section drag for this Lift, at least it is my candid view.

And for this last step, I start to investigate PARSEC method, but much too complex for my issues, as I consider changing only the camber instead of 11 design parameters of the PARSEC approach.

The generic "wing/section" can be the Eppler aboved mentionned, in order to mimick more or less an A-Cat or Moth rig.

In these calculations, I suspect a"trick" in calculation of the side & forward forces, as the angle of attack is not constant along the span, but I did not checked yet, and PhilSweet provides you with a more rigorous presentation of this kind of issue.

If your case study is the rig of a Schooner, not sure this candid approach will help much.

Best luck anyway,
Erwan

 

That was extremely usefull for me.
Thanks to science.