[MVRC] Mantium Virtual Racecar Challenge 2020 (Grand Prix Cars)

[yinlad]

on a side note I've run a test case with just a rear wing to find it's absolute performance in clean air... safe to say I think I have some air flow issues on the whole car

What direction is the flow moving? You might have to nose it up a bit to align it to air box wake. Alternatively a downwash wing just behind the cockpit should help push that dirty air under the rear wing.

after the changes I made for this test it's mostly cooling exit and front tyre wake coming inboard and stalling the wing. Also having never postprocessed in paraview until now, I feel my eyes have been opened significantly....

[jjn9128]

Yeah front tyres suck. If only they could be covered...

[Joe7218]

after the changes I made for this test it's mostly cooling exit and front tyre wake coming inboard and stalling the wing. Also having never postprocessed in paraview until now, I feel my eyes have been opened significantly....

Theoretically, how would one post process in paraview? purely theoretical question

[yinlad]

after the changes I made for this test it's mostly cooling exit and front tyre wake coming inboard and stalling the wing. Also having never postprocessed in paraview until now, I feel my eyes have been opened significantly....

Theoretically, how would one post process in paraview? purely theoretical question

After you run a sim you can find various .vtk files in the 'postProcessing/sampleDict' and 'postProcessing/Sets' folders, you can open these in paraview, which is obviously a lot more useful than just relying on generated pics like I was doing haha. It's worth loading the SurfaceLIC plugin when looking for U_nearwalls. Googling SurfaceLIC should get you some results on how to load that in and select it for a data set

[LVDH]

after the changes I made for this test it's mostly cooling exit and front tyre wake coming inboard and stalling the wing. Also having never postprocessed in paraview until now, I feel my eyes have been opened significantly....

Theoretically, how would one post process in paraview? purely theoretical question

I highly recommend to go through appropriate tutorials first. Either from the Paraview homepage directly, although these are very general, or by going through the Paraview section in the OpernFOAM documentation. If you just jump right into it with our "advanced" simulations, you will just be going through a bunch of avoidable rookie mistakes.

[Thomas2019]

For those with too much time on their hands, there is an ongoing effort with getting the VTK pipeline directly into Blender.
https://bvtknodes.readthedocs.io/en/lat ... Nodes.html
https://blenderartists.org/t/bvtknodes- ... /1161079/3
I played around with it last year, but found it not superior to my usual postprocessing workflow, yet(!). It has made some progress since then, though. And more people try to use it and provide feedback.

Thomas

[AratzH]

You might find this useful. I have a Paraview scene template I use to post process all cases. If you go to File/Load State and choose this file it will give you the option to choose the case location. Once you select everything and wait for a while you will have the whole case loaded with cut planes, surface, VTKs, some streamlines, etc.

[yinlad]

seeing some weird behaviour on my inlet surface, anyone seen this happen before?

[LVDH]

For those with too much time on their hands, there is an ongoing effort with getting the VTK pipeline directly into Blender.
https://bvtknodes.readthedocs.io/en/lat ... Nodes.html
https://blenderartists.org/t/bvtknodes- ... /1161079/3
I played around with it last year, but found it not superior to my usual postprocessing workflow, yet(!). It has made some progress since then, though. And more people try to use it and provide feedback.

Thomas

That sounds pretty cool. I am working on something similar. I will have a look at this. Maybe one day, we can have a cool approach for very cool OpenFOAM post-processing.

seeing some weird behaviour on my inlet surface, anyone seen this happen before?

make sure the engine inlet surface does not just touch the inlet duct but actaully intersects it, so you are sure it is sealed properly.


I just had the honor of checking half of your cars. I found very few violations (Chris found more , better eyes???). I have to say I like the cars very much. After seeing them more in detail, I have to say JJR will be high on top and I like the MT Race Enginneering and the MaccaRacing cars very much.

CFD is pretty much done, so I guess we will post the scrutineering results soon and the results should be out this Sunday, which is great.

[MaccaRacing]

I just had the honor of checking half of your cars. I found very few violations (Chris found more , better eyes???). I have to say I like the cars very much. After seeing them more in detail, I have to say JJR will be high on top and I like the MT Race Enginneering and the MaccaRacing cars very much.

what do you like about my car?... well, I am honoured to read this

[LVDH]

Not sure if I just like the overall lines on the car, the diffuser strakes or if it is just the color scheme. I think the best colors are on WBRacing though and that will probably also be a pretty strong car. In the end it does not matter, the numbers will tell which car is best.

[yinlad]

make sure the engine inlet surface does not just touch the inlet duct but actaully intersects it, so you are sure it is sealed properly.

thanks will check that out. It's an exact face at the moment so will try scaling it up a bit to intersect. I've actually managed to balance the car a little better with my new parts now so round 2 should see 90% of my car change from Monaco...

[LVDH]

make sure the engine inlet surface does not just touch the inlet duct but actaully intersects it, so you are sure it is sealed properly.

thanks will check that out. It's an exact face at the moment so will try scaling it up a bit to intersect. I've actually managed to balance the car a little better with my new parts now so round 2 should see 90% of my car change from Monaco...

Well it is only "exact" when all nodes of the surface are "exactly" on a node of the inlet duct. That is something I would not count on. And I do CAD cleanup on all of your cars and on most them, I do not take any chances and scale the surface up a little bit.

[wb92]

Not sure if I just like the overall lines on the car, the diffuser strakes or if it is just the color scheme. I think the best colors are on WBRacing though and that will probably also be a pretty strong car. In the end it does not matter, the numbers will tell which car is best.

Haha, looks like with the colors the only positive thing so far for the new season! I was wondering if placing color details in such way would work, apparently it did, so thanks for that

#redisfaster

by the way - could it really be that super strong outwash concepts are washed out (pun not intended) due to the narrow mesh refinement outside of the car?

[AratzH]

I actually did a mesh sensitivity study regarding precisely the effect of the coarse region in out wash. I probably don't have the most aggressive execution, but it definitively affects things.
This two pictures are simulations of the same geometry at Z=-0.1. The first one is the MVRC template. The second one is the settings we use at work, slightly different mesher, but same solver settings.






As you can see the out wash is a bit different, the one from the front wing seems stronger in MVRC. The main difference is the tire wake and the out wash generated by the barge boards.
Nothing to be worried about really, since we all play by the same rules, but interesting as a theoretical debate.