[MVRC] Mantium Virtual Racecar Challenge 2016

[LVDH]

Hi Matteo,

send me what you have.

Some questions arrived regarding the measurement surfaces producing different results. I do not have screenshot available right now but this is what you should watch out for:
The surfaces are not part of the CFD mesh. The CFD results get interpolated onto the triangles of these surfaces. So on them the triangulized is very important. If your surface is only a rectangle then it could be triangulized into only two triangles. This would be the coarsest way possible. Obviously you could have more and more triangles representing this surface. I did some tests to find out when the results somewhat converge. It seems like you should aim for a uniform triangulization of about 10mm. You will see that then even the cooling inlet which on most cars I have received logs for a has recirculation zone get pretty much the same value as the outlet surface.

[CAEdevice]

ok, I am going to send you a case folder in five minutes. I noticed that outlet surfaces disappeared from the cooling flow logs. I launched a new simulation (to avoid sending you geometry affected by errores casued by another misunderating of the rules), but it is still running.

Meanwhile, can you confirm that I have to double the cooling flow of the report before comparing it to the rules requirements? How the engine power reduction will be computed?

[LVDH]

Meanwhile, can you confirm that I have to double the cooling flow of the report before comparing it to the rules requirements?

Yes

How the engine power reduction will be computed?

I still have to discuss this with Chris and Richard.

[CAEdevice]

Ok, thanks.

Some other questions about the next race: how will be the car submitted? By email o by a page on the website? There will be a resubmission procedure (the rules talk about "penalties" for small details)?

[LVDH]

Hi,
I am working on a file upload feature on the MVRC page. However still without much success and I am getting worried. If anyone has a good idea for a working Wordpress plugin, please send me a message.
There will not be any resubmission process anymore, so make sure to follow the rules.

[CAEdevice]

You could use the "file request function" of Dropbox. The problems with Wordpress plug in often depends on the hosting service more than on the plugin itself (but we can use email for the first race).

I completely agree about abolishing the re-submission process, but it would be better a more complete description of the the penalties attribution.

[HP-Racing]

I don't think we can use email for the submission because the ASCII stls are huge. Our car has 260mb in this format.

[CAEdevice]

A question about the stl quality.

Most of CAD export STL using a couple of parameters: distance and angle tolerance. I am using the settings for a "fine" mesh (I can improve it but I would have 100mb to send).

I also have the possibility to set a limit for the form factor of the triangles (no deformed triangles). The weight of the stl increases very much, but do you think it could help the simulation?

[CAEdevice]

I don't think we can use email for the submission because the ASCII stls are huge. Our car has 260mb in this format.

A Dropbox/Drive link should solve the problem (a free account is enough).

[LVDH]

Guys, please read the rule book. Rule K1.6! I will impose a file size limit (if at any point I figure out how).
Also setting the parameters of the triangulation too high does not always help. It can actually deteriorate the meshing process a lot.
So don't go crazy with the stl resolution, just make sure that your car does not look like an F117.

[CAEdevice]

This should work in Wordpress (it avoids the use of the - often limited - hosting space):

https://wordpress.org/plugins/simple-dr ... load-form/

(It has not been updated from 2 years, but there are lots of similar application, also based on GoogleDrive)

[CAEdevice]

About the "closed surfaces check": I receive a "not closed surface" warning when testing the "body" component.

I am 100% sure that the STLs is closed and that the quality is very good: I opened the two files written by the utility and I had a confirmation of that. The second "surface" is not actually present in the original STL, it is "created" by the utility. It is not relevant becasue it is a small triangle (so small to be hardly shown by CAD, reaching the accuracy limit) placed out of the simulation domain (50-100m from the car).

[cdsavage]

I don't think we can use email for the submission because the ASCII stls are huge. Our car has 260mb in this format.

As Andre said, there is a limit for the combined size of the STL files, before compression (80mb for ASCII STL).

I completely agree about abolishing the re-submission process, but it would be better a more complete description of the the penalties attribution.

At the moment the aim is to stay with a system similar to KVRC, but without resubmission. So just warnings for things that need to be fixed for the next round, and penalties of 1% of the lap time for more serious issues. If this becomes a problem, then for later races we could consider adding another category in the middle, maybe 0.25% of the lap time.

[LVDH]

About the "closed surfaces check": I receive a "not closed surface" warning when testing the "body" component.

I am 100% sure that the STLs is closed and that the quality is very good: I opened the two files written by the utility and I had a confirmation of that. The second "surface" is not actually present in the original STL, it is "created" by the utility. It is not relevant becasue it is a small triangle (so small to be hardly shown by CAD, reaching the accuracy limit) placed out of the simulation domain (50-100m from the car).

As your mesh is almost valid you do not have to worry. The problem with your simulations is related to the exhaust outlet surface and the engine inlet surface being on top of other parts. As soon as you fix that you will have perfect simulations.
And as long as your simulation runs perfectly out of the box the MVRC staff do not have to touch your files and you are in control. As soon as you submit files that violate the checks and we have to clean the files for ages we might decide to reject your design.

[CAEdevice]

About the "closed surfaces check": I receive a "not closed surface" warning when testing the "body" component.

I am 100% sure that the STLs is closed and that the quality is very good: I opened the two files written by the utility and I had a confirmation of that. The second "surface" is not actually present in the original STL, it is "created" by the utility. It is not relevant becasue it is a small triangle (so small to be hardly shown by CAD, reaching the accuracy limit) placed out of the simulation domain (50-100m from the car).

As your mesh is almost valid you do not have to worry. The problem with your simulations is related to the exhaust outlet surface and the engine inlet surface being on top of other parts. As soon as you fix that you will have perfect simulations.
And as long as your simulation runs perfectly out of the box the MVRC staff do not have to touch your files and you are in control. As soon as you submit files that violate the checks and we have to clean the files for ages we might decide to reject your design.

ok, just to be sure about the engine inlets/outlets: is this geometry allowable?