[MVRC] Mantium Virtual Racecar Challenge 2024 (Grand Prix Cars)

[yinlad]

In my opinion the only way to solve the ambiguity of the underfloor strakes would be to introduce the minimum radius rule

Perhaps, but I don't think we should be making a big rule change 2/3rds of the way thru the season. I would vote to just leave as is for this season with the following 2 clarifications:

1) No more than 4 strakes per side in the high-res folder.
2) You cannot use Graph's "infinite strake" loophole. (Each strake must have only 1 closed section in any X-plane?)

Does the z extrusion rule for fences not make Graphs proposed loophole illegal anyway?

For me the intent of the rule is clear. But I can accept that anything that's a part of the floor, in vehicle_body and within the floor volume is fair game if it adheres to the floor rules wrt closed sections in X and y.

It's a cheeky interpretation but should be limited by the bounds of the floor volume and the lower mesh fidelity.

[G-raph]

Does the z extrusion rule for fences not make Graphs proposed loophole illegal anyway?

I don't think so. Extruded in Z just means that each fence needs to be perfectly vertical (and 10mm thick) in front view across every X-plane.

But nothing says the height can't vary. So if what is shown below with the red arrows is allowed, then a more complex one like the blue cutout will also be legal. If you then hide the remaining top section inside the floor you have an infinite amount of fences.



Obviously, having said that, we should agree not to use that loophole. I don't believe anyone is using it anyway.

Minimum radius rules would stop fences appearing elsewhere on the floor (like in the F1 regulations) but my concern is it is always a difficult thing to check. And if it only relies on a visual check, how can you garantee someone is not using a slightly smaller radius to exploit the rules?

[G-raph]

Sorry, double post!

It's the bottom that lost suction, even though all vortices behave as expected.
Here's the little stall that conditioned floor performance and costed me like 0.8s in laptime (if that's even the cause...)
https://i.imgur.com/rYi9zl7.png

Thank you for posting that picture.

I know it is annoying when you get unexpected results, especially after running the long template and using an already proven rear wing. You have been very unlucky.

However, with the results being what they are, this looks like quite a significant stall to me. Because it is on the central part of the rear wing mainplane (as opposed to a small flap trailing edge separation) I would expect it to cost quite a lot of performance on the floor. I would even go as far as saying that in a way you are "lucky" to only have lost 0.28Cl.

[CAEdevice]

Sorry, double post!

It's the bottom that lost suction, even though all vortices behave as expected.
Here's the little stall that conditioned floor performance and costed me like 0.8s in laptime (if that's even the cause...)
https://i.imgur.com/rYi9zl7.png

Thank you for posting that picture.

I know it is annoying when you get unexpected results, especially after running the long template and using an already proven rear wing. You have been very unlucky.

However, with the results being what they are, this looks like quite a significant stall to me. Because it is on the central part of the rear wing mainplane (as opposed to a small flap trailing edge separation) I would expect it to cost quite a lot of performance on the floor. I would even go as far as saying that in a way you are "lucky" to only have lost 0.28Cl.

Also the beam wing is working differently (and worse), despite there is no separation. This might affetct the floor more than the upper wing partial stall.

[LVDH]

OK, so as time has passed, I simply have to accept my sructineering results. The full results are posted here:
https://mantiumchallenge.com/results-mv ... tone-2024/
Only two teams have some small minor violations this time.
I think, we should come up with some wording to get rid of all the additional strakes on the floor. Also, I think the argument, lets not change at this point in the season is a bit flawed as now other teams have to catch up designing strakes like crazy. I always understand that you guys want rules that resemble F1 as closely as possible and S. Perez was pretty nice this year and showed us his floor pretty often and it is not covered with large strakes /fences.


Also CFD images are out.

https://mantiumchallenge.com/cfd-images ... mvrc-2024/
I do not understand why the slices are not in the same position for everyone. I have to investigate this.
But you might remember that we did some modifications to the CFD setup before the season as you were concered about tire wake structures. How do you see them now?

And regarding Variantes correlation issues:
As I understand, the car was tested with the full run CFD setup, then the rear wing was changed and basically an untested configuration was submitted. This did not perform as expected. Just looking at the rear wing stall in the center area is a bit limited as we now know that also the diffuser stalled. These things are all linked together. I assume the diffuser was right at it's limit. Now having a rear wing also at it's limit affected both, with the results, we now see. This happens, usually adding untested parts right before the submission, bites people in the rear area of their body. I think we have many stories like this.

[yinlad]

Something I've noticed is that Variante appears to have leaky mesh on his front wheel deflectors, based on their hollow appearance in these Z slices anyway.

Quite surprised at the comparison in wake management, even if only in this narrow view of the wider picture

[variante]

And regarding Variantes correlation issues: As I understand, the car was tested with the full run CFD setup, then the rear wing was changed and basically an untested configuration was submitted. This did not perform as expected. Just looking at the rear wing stall in the center area is a bit limited as we now know that also the diffuser stalled. These things are all linked together. I assume the diffuser was right at it's limit. Now having a rear wing also at it's limit affected both, with the results, we now see. This happens, usually adding untested parts right before the submission, bites people in the rear area of their body. I think we have many stories like this.

The diffuser did not stall. Only the rear wing, in a that tiny region. But that was enough to substantially decrease aiflow extraction from the floor, I suppose.

I can't test every modification in MVRC_long mode, because it takes me 60 hours. The wing was the same used successfully in Monaco anyway, so it didn't seem critical.
So, I couldn't realistically detect nor predict such problem, but it still facked up my race, and the championship.

[spacehead3]

I'll repeat that I've had great correlation the past 2 years with "MVRC" (not long) template, and takes about 24 hours to run on my old machine.

[variante]

I'll repeat that I've had great correlation the past 2 years with "MVRC" (not long) template, and takes about 24 hours to run on my old machine.

I also enjoyed manageable correlation by using the fast option, until Silverstone. Before that, I had never experienced anything critical in long mode that didn't at least give hints in fast mode.
So, I don't think the medium mode would offer any guarantees against the long mode either.

[CAEdevice]

I'll repeat that I've had great correlation the past 2 years with "MVRC" (not long) template, and takes about 24 hours to run on my old machine.

Same for me, standard template gives reliable results.

Ok for me to change the rules to avoid floor small strakes outside the fences volume but allowing them inside the lateral side box, but, to be ready with a new floor for Monza I need to know the new rules preferably before the end of the first week of august.

In order to have a better compliance with the FIA F1 rules, as for the floor winglets, especially for Monza, I'd suggest to include a rule to make blended rear/front wing 2021 mandatory.

[yinlad]

In order to have a better compliance with the FIA F1 rules, as for the floor winglets, especially for Monza, I'd suggest to include a rule to make blended rear/front wing 2021 mandatory.

Now we're talking

Edit: to actually take this a bit seriously, I'm all for both the wing and strake amendments but agree whatever we decide should be decided quickly to give everyone time to comply. Happy to follow whatever decision Andre makes

[spacehead3]

So, I don't think the medium mode would offer any guarantees against the long mode either.

Respectfully, that logic doesn't really make sense. It's a better mesh, it very much should give better results. If you haven't experienced that then perhaps you have just been lucky so far and / or had a very stable flow field. The reason that I stopped using "fast" was that I was seeing opposite trends to "mvrc" and "long".

Obviously it's up to each person to decide if they want to trade speed or accuracy...You're clearly beating me so I guess your method works overall! Just trying to help!

[variante]

So, I don't think the medium mode would offer any guarantees against the long mode either.

Respectfully, that logic doesn't really make sense. It's a better mesh, it very much should give better results. If you haven't experienced that then perhaps you have just been lucky so far and / or had a very stable flow field. The reason that I stopped using "fast" was that I was seeing opposite trends to "mvrc" and "long".

Obviously it's up to each person to decide if they want to trade speed or accuracy...You're clearly beating me so I guess your method works overall! Just trying to help!

I meant to say that a better mesh doesn't imply perfect reliability, which is what we'd need to avoid correlation issues. It only decreases the chances of them occurring, at the cost of increased computational time.
So, as much as I appreciate your opinion, I can't help but underline that you're still exposed to problems like the one I experienced.

[CAEdevice]

The differences between "standard" template and "long" is small, the difference between "fast" and "standard" is huge.

In my opinion "fast" template is unuseful: I prefer to use customs settings based on "standard" (iterations reduced to 2000 and base mesh dimension increased by 5%). I get the results in 9 hours with a 10 years old workstation built with second hand components.

[G-raph]

Also CFD images are out.

I do not understand why the slices are not in the same position for everyone. I have to investigate this.
But you might remember that we did some modifications to the CFD setup before the season as you were concered about tire wake structures. How do you see them now?

Thank you for the CFD pictures. To make it fair, here is my Z-slice from Race 03 :




I think the tyre wake structures are more realistic with the new settings. The main issue I found is that the front wake is a bit more different than it used to be between the Fast and Standard/Long settings, which adds to the correlation challenge.