Planing hull modelling

Has anyone modelled planing hulls with CFD to any sucess?
I am guessing it is pretty damm difficult with free surfaces and mesh deformation and boat that has to rotate and lift etc. anyone managed it?

 

I think it could be done with OpenFOAM, but it would need an adjustment to allow heaving and trimming. It's not impossible, but usually Savitsky is quicker and easier.

Simulating yacht hulls is much easier (if you ignore heave and pitch) and often gives better insight into what is going on.

Tim B.

 

Sorry to appear naive but what is Savitsky?
I am CFD engineer and am new to the marin e thing. Is it a VPP progam or panel method?

Also why do you say modelling yachthulls is easier? Do you mean just modelling the free surface?

 

Savitsky is essentially a momentum method. It is reasonably well known in the marine world, the paper is available from SNAME in the US.

Yacht hulls don't tend to heave and trim as much as power-boats, so the effects can be ignored (at Fn<0.3 roughly). That makes the solution somewhat easier.

Where in the UK are you?

Tim B.

 

The Savitsky method is a semi-empirical formula for calculating/estimating planing craft resistance. It's pretty simple to plug in to a spreadsheet, so is quick and easy - nothing like CFD! There are a myriad of problems doing CFD on a planing hull - the extra hassle (and expense) has to yield significantly better results than traditional methods (like Savitsky) to warrant it's use. Currently this is hard to justify, as the dynamic aspects of planing are difficult to replicate, as are loads of other bits that you are most likely aware of. Validation is also tricky. If you can crack it, and find a cost effective way of doing it, then CFD could become more mainstream. Best of luck!

 

Well OPEN-Foam is free, open-source software... and you can have an 18 node cluster for as little as £6600. Even if you decided to go to CFD in a major way, you can still get more information than a tank-test will give you for less money than it would cost to run the tank. Naturally, bigger clusters will let you run cases faster.

The problem with validation is that most tank test results are, shall we say, open to interpretation. Consequently, you can only ever say that the result is similar. Whether you trust the CFD or the tank test is up to you as a designer.

The advantage of CFD is not really in getting a quick cut-and-shut answer. It is about finding potential problems with the design (separation across the bow at high leeway angles for example) which would not have been obvious in the towing-tank.

National, I am sure, is well aware of this.

Tim B.

 

CFD is definitely a Good Thing, and has a great future - but it is not a Holy Grail. Tank testing will be around for a while yet, and the results will be trusted more. Sure TT results require interpretation, but so does CFD output.
Tank testing is done at the more detailed level of analysis, Savitsky is used at the very earliest stages. CFD currently sits in the middle somewhere. There is no great advantage to using CFD at the start of the design process, becasue Savitsky is adequate and much cheaper. As designs evolve CFD is very useful and lets you see the effect of design changes. But unvalidated CFD is, well... invalid.
One of the biggest problems with CFD is that anyone with a computer can have a go. I'm not casting aspertions on National, who I am sure is highly skilled, but the fact remains that the the quality of an individual CFD'er is unknown - and anyone can make a pretty picture - especially using free software. Tank tests are only carried out by a handful of experienced teams (due to the limited availability of tanks), so there is more 'trust' in the answers they provide.

Back on topic, modelling a planing craft accurately would be a complex problem because there are so many (constantly changing) parameters to account for. That leads to a lot fo assumptions piling up on each other. Definitely worth a crack at though.

 

I am based in Nottingham. I am currently doing a Phd with the CFD group at Nottingham Uni but am going for a work placement in a few months time. They asked about the feasibility of modelling planing vessels. I thought this would probably be a little bit too advanced but thought some of you lot might have had a bash already.

Anyone know of any decent papers or any of the more advanced applications?

 

It is a very tough problem!

Here are a few references to papers by Tuck. The last is available as a pdf.

Keep in mind that calculating the pressure distribution over some simple flat wings is a tough problem. (Compare results from CFD or Vortex Lattice Methods to the exact solution for a wing of circular planform to see how difficult it can be!).

With planing you need to add a free surface and many other complications. Some researchers have found integral representations of the problem, but they are impossibly difficult to evaluate.

A promising start on slender planing surfaces was made by Casling a few years ago. See the last paper cited here.

http://internal.maths.adelaide.edu.au/people/etuck/index.html

41. TUCK, E.O. "Low-aspect-ratio flat-ship theory",
J. Hydronautics, 9 (1975) 3-12.

68. TUCK, E.O. "Linearized planing-surface theory with surface
tension. Part I : Smooth detachment", Journal of the Australian
Mathematical Society, Series B, 22 (1982) 241-258.

69. TUCK, E.O. "Linearized planing-surface theory with surface
tension. Part II : Detachment with discontinuous slope", Journal
of the Australian Mathematical Society, Series B, 23 (1982) 259-277.

100. TUCK, E.O. and DIXON, A. "Surf skimmer planing
hydrodynamics", Journal of Fluid Mechanics, 205 (1989) 581-592.

180. TUCK, E.O. " Slender planing surfaces ",
for J.N. Newman 70th birthday volume, to appear in
Journal of Engineering Mathematics, 2007.

 

The only paper, using finite volume methods, i can think of was presented at Marine Cfd in 2005 at Southampton:
VOF-Dynamic Mesh Simulations of Unsteady Ship Hydrodynamics
M Visone, Blue Group, Italy
C Falletta P.L. Ausonio, Ship-Yacht Designers & Consultants, Italy
P Bertetti and R Gandolfi, AZIMUT, Italy
D Paterna and R Savino DISIS Univ. of Naples "Federico II", Italy
In this paper, turbulent flow simulations around the hull of a motor yacht,
at steady speed through calm water, have been performed to compute
the hull incidence in static and dynamic conditions. The presence of the
water-air interface has been taken into account with a Volume-Of-Fluid
(VOF) technique that allows the tracking of the wave form at each time
during the numerical simulation. Validation of the computational
simulation has been carried out, comparing the result of different model
tests. The lift, drag and pitching moment have been computed in the final
equilibrium positions. In particular, the contribution from wave drag has
been calculated separately. The results have been compared with the
experimental data, showing a good agreement.

There are using a sliding interface, and an unstructured grid using tetrahedras, so not perfect for free surface computations.

This website (www.azcueta.net) shows some interesting results. He uses a single grid strategy (moving not deforming mesh).

Another way is to use overlapping grids, this is being developed here http://www.iihr.uiowa.edu/~shiphydro/projects.htm.

So, this is still a good research topic, and not yet ready for efficient commercial use.