# Windsurfing Sail and mast design

Discussion in 'Hydrodynamics and Aerodynamics' started by matt_bob, Apr 13, 2015.

1. Joined: Apr 2015
Posts: 9
Likes: 0, Points: 1, Legacy Rep: 10
Location: Pretoria, South Africa

### matt_bobJunior Member

Good Day

My final year project for engineering undergrad requires that I design a windsurfing sail and mast arrangement for an attempt at the world speed record.

I have made good progress but I am now stuck with calculation of lift and drag coefficients and forces. I am using XFLR5 as it is the only program I have found that allows me to make my own aerofoil shapes easily. I am open to other suggestions, also tried Airfoiltools[dot]com, and Ansys Fluent will require alot of work for each aerofoil and I am not even sure if it can work out the coefficients.

I have a few camber ratios that I am considering for the design based on graphs from Principles of Yacht Design Second Ed. But I am struggling to make them exactly in XFLR.

Is the only way of doing this by manipulating the spline until u get the right camber ratio?

And because the mast is going to affect the flow, will i get more accurate results if I add that in the aerofoil shape? [see attachment for example]

Any help would be greatly appreciated.

Regards
Matt

#### Attached Files:

• ###### Capture.PNG
File size:
6.3 KB
Views:
597
2. Joined: May 2008
Posts: 2,514
Likes: 297, Points: 83, Legacy Rep: 1082
Location: Beaufort, SC and H'ville, NC

### philSweetSenior Member

Does your project emphasize the aerodynamic design of the sail, or do you just need numbers to complete a decent enough force balance so you can design the structures?

3. Joined: Oct 2005
Posts: 441
Likes: 25, Points: 28, Legacy Rep: 97
Location: France

### ErwanSenior Member

You need to start with the max righting moment, the true wind speed, and a proxy of the windsurf polar curve.

You'll get the apparent wind speed, and with other datas including area & aspect ratio, you'lll get a proxy of operating Reynolds number for XFLR5, also
use Ncrit:1 in the toolbox and try to minimize drag for operating AoA you'll probably get more power than necessary.
Crew aero-drag is probably a good share of the global drag-pie.
You can optimize sailing area & aspect ratio for each true wind speed and one righting moment, or conversly.
For a candid like me, there is to chew around this kind of project.

Best wishes

EK

4. Joined: Apr 2015
Posts: 9
Likes: 0, Points: 1, Legacy Rep: 10
Location: Pretoria, South Africa

### matt_bobJunior Member

Shot for the quick replies guys.

@phil
Yeah, my supervisor said here is the true wind speed and I want a total sail area of this. Design a sail that can go faster than the world record. So I need to take into account everything, twist, camber, material, mast bend etc etc.

@Erwan
What I have done so far was assumed a board speed that was faster than that of the world record and calculated the respective true winds speed per height from water according to the boundary layer. From that calculated the apparent wind speed and angle with height from the waters surface, and then the Reynolds number from that apparent wind speed. From that got the range of the Reynolds number and used that range for the XFLR tests.

What is Ncrit? I left it at 9, which was the default but I set the transition zone at 15% for the top section and 10% for the bottom, as the sail experiences turbulent flow over majority of the surface area and if I recall correctly in PYD by Larrson et al there is a seperation zone that happens approximately at those distances aft. Or am I mistaken with that?

My drag currently is maxing at 0.2 and my lift at 2.14 in one case.

Regards
Matt

5. Joined: Dec 2004
Posts: 1,709
Likes: 82, Points: 48, Legacy Rep: 467
Location: Sydney Australia

### CT 249Senior Member

Out of interest, how are you going to calculate twist and how deep is your head going to be?

6. Joined: Apr 2015
Posts: 9
Likes: 0, Points: 1, Legacy Rep: 10
Location: Pretoria, South Africa

### matt_bobJunior Member

I have no idea how I am going to calculate twist for the sail yet. As for the head, I was looking at that yesterday after finishing with the cambers and I think that I am going to make it 0.6m initially and then refine after I have worked out the forces.

Still trying to figure out the dimensions for the sail though. Working with a 4600mm mast and the area has to be 6m^2, area was specified in the project requirements. I am going to check out some of the sail programs like sail7 and see if they have a feature that I can use to tackle this aspect.

Regards
Matt

7. Joined: May 2004
Posts: 5,372
Likes: 255, Points: 93, Legacy Rep: 3380
Location: Italy (Garda Lake) and Croatia (Istria)

### daiquiriEngineering and Design

You need to set your design goals and specifications. Once you have set them up, it will be much easier for you to gather your thoughts and decide on the path to follow. "An attempt at the world speed record" is not a sufficiently detailed set of goals to meet. You will need at least the maximum speed, the wind speed, weight of the windsurfer, size limits, materials to be used for the mast and sail and their mechanical properties, and things like that.

So, what are your design goals and specifications?

8. Joined: Oct 2005
Posts: 441
Likes: 25, Points: 28, Legacy Rep: 97
Location: France

### ErwanSenior Member

As far as I understood, you re sailing close to the water surface, with strong wind, the wind boundary layer might be more turbulent than in a Wind tunnel where Ncrit=9, that is why it's a conservative assumption to use Ncrit=1. That's all I ve understood

Regards

E

9. Joined: Oct 2005
Posts: 441
Likes: 25, Points: 28, Legacy Rep: 97
Location: France

### ErwanSenior Member

You mention interesting workpaper :" PhD from Larson and al" separation ?

Please, Would you have the exact title of this workpaper in ordervto Googlize it

EK

10. Joined: May 2004
Posts: 5,372
Likes: 255, Points: 93, Legacy Rep: 3380
Location: Italy (Garda Lake) and Croatia (Istria)

### daiquiriEngineering and Design

Ncrit is a correlation factor which transforms the value of the turbulence level "t" of the fluid, into a form used by the e^N criterion for boundary layer transition. It is calculated by the formula:
Ncrit = −8.43 − 2.4*ln(t)​
t is the turbulence level (or intensity):
t = v'/Vinf​
v' is the root mean square turbulence velocity and
Vinf is the free-stream velocity.

Once the Ncrit has been assigned, the program calculates the local value of the N (via numerical integration of a rather complex function of various boundary layer parameters) over the foil surface and compares it to the Ncrit value. For N<Ncrit is the flow is modeled as laminar, while for N>Ncrit it is modeled as turbulent.
In reality the transition occurs over a finite distance, which is defined by further two factors, N1 (beginning of the transition zone) and N2 (end of the transition zone), each one calculated via other logarithmic formulas. Ncrit is placed in between these two. But these details are IMO beyond the scope of this explanation.

When it comes to the values of Ncrit to use for practical calculations, Ncrit=9 corresponds to a case of an object with a smooth surface moving through the still air, which is not your case.
Ncrit=3-4 would be more suitable for a moderately-rough airfoil moving through air with low level of turbulence, but that is not your case again.
The windsurf sail is not a smooth surface and the wind it moves through is always somewhat turbulent, so Ncrit=1-2 would be the most appropriate in your case, imo.

Cheers

11. Joined: Feb 2007
Posts: 543
Likes: 165, Points: 53, Legacy Rep: 160
Location: California, USA

### Doug HalseySenior Member

Be careful. The professor who assigned this project might be trying to trick you. Maybe the correct answer is that no windsurfing sail has any realistic chance of setting the world speed record. At least not a conventional sail like the one you seem to be trying to optimize. Maybe the professor is trying to see if you can come up with a revolutionary new idea, instead of getting bogged down in the details of a futile task?

12. Joined: Oct 2005
Posts: 441
Likes: 25, Points: 28, Legacy Rep: 97
Location: France

### ErwanSenior Member

Thanks Daiquiri for providing the academic perspective.

I agree with Doug, try to think out of the box within your rule-box.

Start with the balance of forces and moments, the clearer the picture of drags, the better the solution you will work out.

Best regards

EK

13. Joined: Oct 2005
Posts: 441
Likes: 25, Points: 28, Legacy Rep: 97
Location: France

### ErwanSenior Member

With 50 knts apparent wind, the drag of the dude hooked at the wishbone would be around 150 Newtons. (0.5 sqm body area, Cd=0.8)

14. Joined: Nov 2014
Posts: 3
Likes: 0, Points: 0, Legacy Rep: 10
Location: alberta

### barney831New Member

Kenney, B.C., 2001.
Hyperwind Sailing.
Catalyst, J. Am. Yacht Res. Soc., 6, 24-26.

15. Joined: Dec 2012
Posts: 1,270
Likes: 25, Points: 0, Legacy Rep: 271
Location: Hampshire UK

### SukiSoloSenior Member

Note that in the full size foiling Moth tank tests the drag of the helm was 40% of the total drag. The tester debated whether to use a dummy helm and were pleased they did - for this very reason. Lots of their other data was valid but it helped build up an overall picture. A windsurfer may generate more drag tan a relatively static flat Moth helm.

Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.