Panel/v-lattice code

[national]

I am after some open source/free software to compare to a Finite volume code of flow past yacht sails.
Are there any about? I would ideally like a vortex lattice code suitable for modelling sail flow.

thanks in advance.

[nico]

the only one i can think of is Tornado (http://www.ave.kth.se/divisions/aero/software/tornado/), which runs under matlab. You'll be able to customize it enough so that you can sail geometries.

[national]

Thanks Nico,

What's your research in? I am also doing CFD PhD in the Uk.
National

[nico]

I am developing a finite volume solver for seakeeping/manoeuvring cases. Mainly looking at better schemes for the advection of density. (VOF,..). What about you?

[nico]

Send me a PM or an email at nico.rousselon AT strath . ac . uk. I have a vortex lattice code that could be of interest. (see VLM http://www.deltavoiles.com/actualite_1.htm)

[Leo Lazauskas]

Originally Posted by nico

Send me a PM or an email at nico.rousselon AT strath . ac . uk. I have a vortex lattice code that could be of interest. (see VLM http://www.deltavoiles.com/actualite_1.htm)

Have you tried to run these codes on very simple geometries?
For example, in inviscid flow there is an exact solution
for the lift of a circular wing.

There are some benchmark solutions at:
http://www.cyberiad.net/wing.htm

Flat circular disks:
http://www.cyberiad.net/lsie_circle_planar.htm

Circular planform with parabolic camber:
http://www.cyberiad.net/lsie_circle_parabolic.htm

I would be very interested to see what VLM gives
for the same cases.

All the best with your studies! If you are working with Sandy Day, please send him my regards.
Leo.

[nico]

Thanks for the links. Unfortunately, i haven't been able (yet hopefully) to do a proper verification/validation of the code. I don't like it, but i haven't had time to do it properly. But results on sail sets compare well to wind tunnel data, and results for rectangular plates are similar to other tools.
I dont think circular wings at the best geometries to compare to, mainly because sail and circle are quite different, and conclusions might not be applicable to both. But if I have a bit of time, i ll give a shot, and post some results.

I am not working with Sandy, but his office is not too far.

There is also a free vlm (http://web.mit.edu/drela/Public/web/avl/) developed by Mark Drela (Xfoil). (it can also model sources and double lines, -> influence of non lifting bodies)

[Tim B]

I am going to be solving a very similar problem before Christmas (which it seems is arriving at a rate of knots). In that I need a pressure distribution (we'll ignore skin friction drag for a moment) to give an estimate of forces and moments of a multi-element rig. My plan was to adapt the method given in "Low-Speed Aerodynamics - From Wing Theory to Panel Methods" by Katz & Plotkin.

Do keep us posted as you progress national, where are you studying?

Cheers,

Tim B.

[national]

I am studying at Nottingham, Doing aero-elastic analysis of sail rigs using a commercial FVM RANS solver coupled to bespoke sail structural package. It's going ok just want to show that the hard work with RANS is worth it over the panel and other methods. Because the meshing is a bit of a hassel especialy when you add masts and the like.

I am yet to have a play with the other codes. will keep you posted. Struglling with mast effects at the moments!
National

[Leo Lazauskas]

Originally Posted by Tim B

I am going to be solving a very similar problem before Christmas (which it seems is arriving at a rate of knots). In that I need a pressure distribution (we'll ignore skin friction drag for a moment) to give an estimate of forces and moments of a multi-element rig. My plan was to adapt the method given in "Low-Speed Aerodynamics - From Wing Theory to Panel Methods" by Katz & Plotkin.

Katz and Plotkin's book is an excellent introduction to the field, but their codes are not always very very efficient. I'm not sure if they have improved much since the first edition, but their numerical methods for the induced drag were almost useless. They converge very slowly (if at all) to the correct solution for rectangular wings and fail on wings with curved planforms such as sails and keels.

That's why I suggest using a circular planform as a test case because there is an "exact" solution for the lift.

Vortex Lattice Methods will not give accurate pressure distributions for lifting surfaces with curved leading edges and/or trailing edges. The pressure distribution near the wing tips (where all the action happens!) are very unreliable.

Good luck!
Leo.

[national]

If this is so then why do so so so many people use them for pressure analysis in FSI simulations?!
You can pretty much count on a hand how many people have done FSI with CFD well!

Anyone know of any decent papers highlighting problems assosiated with these types of codes? or can I just quote the forum!

[Tim B]

I'm aware of the limitations, but at the moment, there is no way to run a full Navier Stokes 3D CFD case for sails AND hull. (Well, there is, I just don't have time, or enough computers to do it) So I've had to go to the next method down, as it were.

Oh well,

Tim B.

[Leo Lazauskas]

Originally Posted by national

If this is so then why do so so so many people use them for pressure analysis in FSI simulations?!
You can pretty much count on a hand how many people have done FSI with CFD well!

Anyone know of any decent papers highlighting problems assosiated with these types of codes? or can I just quote the forum!

VLM is poppular because it is fairly easy to implement and they display remarkable accuracy for rectangular wings and some wings with straight edges. No doubt other people have used them blindly expecting similar accuracy for wings with curved leading or trailing edges.

David Standingford's PhD thesis contains several plots comparing VLM and a method of E.O. Tuck's that is much less sensitive to the location of panels and control points. Also shown are the lift, induced drag and leading edge suction as functions of aspect ratio for delta wings as well as rectangular and elliptical wings. IIRC, Katz and Plotkin presented very few (if any) induced drag calculations for wings with curved planforms.

See:

Standingford, D.W.F., Optimal Lifting Surfaces (including Endplates, Ground Effect and Thickness), PhD Thesis, Dept. Applied Mathematics, The University of Adelaide, July 1997.