Khamsin Virtual Racecar Challenge 2015

[machin]

Cool, welcome along to the challenge! Chris (CDSavage) is the main man to ask any questions you might have about the championship, but as you can see from this discussion it is a friendly bunch of people taking part all willing to offer advice and guidance.

[LuPi]

Great,

but it's 50 pages long,

[RicME85]

First half is just discussion about the formation of the rulebook really

[julien.decharentenay]

@LuPi Thanks a lot for registering. It is indeed a long thread (and there is also the 2014 and 2013 threads)... Designs for the next race (Monaco) are to be submitted by 14th May. To get started Chris has prepared a number of documents:

• Rule book http://www.khamsinvirtualracecarchallen ... ects=0&d=1: which describes the rules that the design has to comply with as well as how to prepare the geometry for submission. The file also contains CAD information that goes along with the rules to assist with understanding how to comply;
• Additional parts (wheel and suspensions I believe) http://www.khamsinvirtualracecarchallen ... ects=0&d=1 and http://www.khamsinvirtualracecarchallen ... ects=0&d=1. They should help you getting started if you need them.

To assist with testing your design, you can download the CFD "framework" used from http://www.khamsinvirtualracecarchallen ... -click-cfd and test options in term of lap-time performance using http://www.competition-car-engineering. ... KVtest.htm.

I hope this help... But I haven't got through the participating side (only organising), so I would not have the beginning of the start of an idea on where to start in practice. Jacopo (team Variante) wrote the following article last year on how he is using SketchUp for his design http://www.sketchup-ur-space.com/2014/j ... tchup.html.

[julien.decharentenay]

And OCCFD v1.1.004 is finally out. Please download from http://www.khamsinvirtualracecarchallen ... edirects=0 and let me know what does not work...

If you have convergence issues (ie OCCFD reports divergence - but it won't change if something does not work), you can try More Options > KVRC Very Stable (and let me know if it solves the divergence issue). Hopefully you won't have to resort to it...

[RicME85]

Interesting, I hadn't seen that article from Variante, I have always wondered how he makes such good models in Sketchup.
Before KVRC when the laughable Formula Sketch was running and I first got in to Sketchup modelling I used the same ribbed approach to car design after breaking down Codemasters F1 models.
Now I am more skilled and knowing Variantes approach I am definitely going back to it but using the 2d plane aspect to (like those who build 3d models from blueprints)

[CAEdevice]

And OCCFD v1.1.004 is finally out. Please download from http://www.khamsinvirtualracecarchallen ... edirects=0 and let me know what does not work...

If you have convergence issues (ie OCCFD reports divergence - but it won't change if something does not work), you can try More Options > KVRC Very Stable (and let me know if it solves the divergence issue). Hopefully you won't have to resort to it...

Thank you very much!I've just installed OCCFD 1.004 and I'm running the first test (re-running the geometry processed during the first race).

EDIT: The analysis is running, the mesher seems a bit slower, but much more robust and stable, thank you!
The model that were not working have been meshed without any isses in "standard mode".
In the afternoon a comparision between the result produced by OCCFD 1.003 and OCCFD 1.004 (same model, the one that run Race1).

EDIT: The analysis has been completed. Just a minor isuue (frontal area computation) but depending on my installation.
The result of OCCFD 1.003 and 1.004 are comparable (the difference in DF and DRAG are less than 2% when simulatin the car I used during 1st race, so every detail that has been simulated with the previous versions will have the same behaviour).
The mesher is much more robust now.

[julien.decharentenay]

Just following up on CAEdevice post above (and email), I updated OCCFD to patch 005. The only difference is to allow for OCCFD to recognise ImageMaick 6.9.1 (in addition to the previous version of ImageMagick)... If you really want to update OCCFD, you can download it from http://www.khamsinvirtualracecarchallen ... edirects=0

[CAEdevice]

Thank you Julien, now everything works perfectly!

[julien.decharentenay]

Thank you Julien, not everything works perfectly!

Did you mean "not" everything works perfectly???

[CAEdevice]

Thank you Julien, not everything works perfectly!

Did you mean "not" everything works perfectly???

Sorry! My fault! "now" not "not"

[julien.decharentenay]

Thank you Julien, not everything works perfectly!

Did you mean "not" everything works perfectly???

Sorry! My fault! "now" not "not"

Glad to hear. Thanks for testing.

[CAEdevice]

Hi Julien, should we expect any difference between the results of the standard and "very stable" solution?

[julien.decharentenay]

Hi Julien, should we expect any difference between the results of the standard and "very stable" solution?

Expect some difference between the standard and the "very stable" version. I can't really advise on the effect and magnitude of these difference - the major difference is that the standard version uses a Gauss scheme for gradient calculations, while the 'very stable' one uses a least square with a limiter. I would assume that the 'very stable' is more dissipative, but I can't really say how this would affect drag and downforce...

On another note: I received a request from Rajib Dey, editor from Sketchup-ur-space, for article or tutorial. Would anyone be willing to spend some time on this?

[CAEdevice]

I don't use SketchUp this year, but if you think it can be interesting I can produce a clip with Spaceclaim + OCCFD (also useful to explain the right format for the submission).