CFD - Computational Fluid Dynamics, Motorsport, Formula 1

[Websta]

Thes pictures are a great find. We can see the floor suction peaks at the front and on the kink line, and the low pressure footprint of the bargeboard vortex protruding from the front suction peak.

There are a lot of other interesting details, such as the shape of the tyre contact patch.

Should help a lot in discussing underbody aero

Some good points, Shelly.

[Jersey Tom]

I wouldn't read too much into that tire footprint shape. May just as well be an intersection of the cambered tire and a flat plane.

[shelly]

I agree with you JT - the almond shape seems to suggest exactly that. But it is nice to see that in cfd a team uses that as contact patch representation, which is the most logical choice when you do not have more detailed infos.
Problem is that contact patch shape has a strong influence on aero.

[astracrazy]

if anyone can help me out here viewtopic.php?f=3&t=13385 would be a great help! Thanks guys

[george1981]

Hello, this might not be the best thread to ask the question but I thought it was a good place to start.
I'm trying to model a large component in CFD and one possibility that crossed my mind was to model it at 1/10 scale. I thought this might be a fairly standard approach and there might be some reasonable scaling rules I could apply. However I haven't seen it done any where and can't find much if anything on the subject. Do people make scale models in CFD? I know it's common in windtunnel testing.
Would anybody be able to point me in the right direction?
Cheers
George

[superdread]

Hello, this might not be the best thread to ask the question but I thought it was a good place to start.
I'm trying to model a large component in CFD and one possibility that crossed my mind was to model it at 1/10 scale. I thought this might be a fairly standard approach and there might be some reasonable scaling rules I could apply. However I haven't seen it done any where and can't find much if anything on the subject. Do people make scale models in CFD? I know it's common in windtunnel testing.
Would anybody be able to point me in the right direction?
Cheers
George

It doesn't matter how large your real component is. CFD is all in a computer and they certainly don't care how big the numbers you give them are. Also it compromises the result because there are things (e.g. the speed of sound) that don't scale with your model.

As CFD is a discretized method (i.e. you take the normally continuous space and chop it into pieces) the spatial density of your data points determines how good your result will be. But as they cost computing power you have to balance quality of result against computing time.

[godlameroso]

Of course you can key certain optimizations into your calculations that not only yield better correlations but also require less computing power. Of course those are trade secrets...

[Richard]

Does anyone have experience of XFlow? http://www.xflowcfd.com/

We're currently using ANSYS CFX with Harpoon for mesh generation. We're finding xflow is delivering output in something like 25% of the analysis time required for CFX, presumably due to LES (see Cam's post on previous page) .

We had a couple of PhD chaps in our offices a couple of years ago who were saying that LES would revolutionise flow simulation, but we've been stunned when we try in a finished product.

We need to dig a bit deeper into the quality of that output and do some benchmarking, so any feedback from others would be really useful.

We've also heard it is used in F1, so that makes this is on topic

[flyboy2160]

Does anyone have experience of XFlow?...

... and if you do, how much did it cost? I can't find any pricing on any XFlow company or distributor site. A couple of CFD forum posters mention very high prices, but don't mention if that is a single seat or site license.

(When dealing with a previously very expensive CAD software from ca ompany who originally also didn't list prices anywhere, I was told by a sales rep that the pricing was dependent on how much the traffic would bear. They would originally quote a very high seat price. If a large company didn't object, they got their price. If you whined and told them to take a hike, they would lower the price. This situation was ameliorated by market competition.)

Thanks.

[N12ck]

does anyone have any previous with Star CCM, I have used openfoam before, but has anyone got any tips with Star CCM or little tricks with it?

Thanks
Nick

[cdsavage]

does anyone have any previous with Star CCM, I have used openfoam before, but has anyone got any tips with Star CCM or little tricks with it?

Thanks
Nick

Have a look around in here, the videos cover mainly advanced stuff but a number of them have demonstrations of the software which might help you pick up some of the basics.

http://www.cd-adapco.com/downloads/webinar_recordings/

[astracrazy]

what is the best way to simulate underbody aero with solidworks? just create a false floor and put the floor/diffuser just above it?

[stez90]

not sure what solidworks can do, but you need a moving floor and if you are testing ONLY the underbody you will need a "standard" body over it to ensure that the upper surface of your model doesn't alter the result.

[astracrazy]

so you can't just test the floor and diffuser, you'd need some sort of body on top?

[stez90]

yes, otherwise a different shape of the upper surface will alter the result.. you need a simple streamlined body, nothing difficult.. check the thread "diffuser profile - convex or concave", there are great examples..

viewtopic.php?f=6&t=10943