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

[Koldskaal]

Quick question. When looking at some old CFD runs, specifically the html results page MFlow generates when post processing. The 'totals' section doesn't actually seem to be the sum of the values below. Is this expected behaviour?

I have noticed this too, I assumed the remainder were porus forces from the radiator.

[Joe7218]

When simulating, should we include all the mandatory parts (suspension, halo etc)? thanks

Yes but only the "external" ones.

I also seem to be getting this error "Warning: Rear wheels seem to be positioned incorrectly. In MVRC they should be in global (3.4,0,0)."

Are you using the stls from the website? If not, then I would recommend doing so They should be in the right position.

Thanks!
Worked it out! turns out i didnt have a "rr_wheels folder
I also get an error when i try and run the case, "Error: cannot run "runfile"!"

[yinlad]

Quick question. When looking at some old CFD runs, specifically the html results page MFlow generates when post processing. The 'totals' section doesn't actually seem to be the sum of the values below. Is this expected behaviour?

I have noticed this too, I assumed the remainder were porus forces from the radiator.

hmm, I seemed to be missing around -0.4 Cl from the total compared to the manually summed figures.

And my Cd figure 'total' is double the summed parts. That, I can put to the porous surfaces, but I don't know that they would be causing 0.4 of lift

[variante]

@Machin: could you please update Virtual Stopwatch with the new tracks?

@Newcomers: don't forget to use Machin's software for each race to understand what direction to give to your project (accent on downforce, efficiency, speed,...).

hmm, I seemed to be missing around -0.4 Cl from the total compared to the manually summed figures.

And my Cd figure 'total' is double the summed parts. That, I can put to the porous surfaces, but I don't know that they would be causing 0.4 of lift

Are the wheels included in the count? Because otherwise that's where most of the missing lift is coming from.
@LVDH: It would be good to have wheels numbers included in the data sheet.

I also get an error when i try and run the case, "Error: cannot run "runfile"!"

Same here. LVDH said it's probably fixed in the new MFlow update, so let's just wait a few hours(?).

[yinlad]

If you're getting run errors, you can open BlueCFD terminal and navigate to the directory that contains run.sh and use the command 'bash run.sh' to get it going.

It does rely on some general terminal use knowledge though

[LVDH]

Hi guys,
I am in my office today, all by myself, and will try to address the issues you have detected. In the past so many people have bugged my about the total not summing up with the individual listed parts, that I have changed that. The reason I never included everything is because it helps people draw wrong conclusions, so now we are open to that now.
I think the bugs you have found, have been fixed in MFlow but have still not been released. I might get that done today.
I will let you know about the new version as soon as it is up.

[Joe7218]

If you're getting run errors, you can open BlueCFD terminal and navigate to the directory that contains run.sh and use the command 'bash run.sh' to get it going.

It does rely on some general terminal use knowledge though

How long should the process take then? mine has been running all day and still hasnt finished haha.

Hi guys,
I am in my office today, all by myself, and will try to address the issues you have detected. In the past so many people have bugged my about the total not summing up with the individual listed parts, that I have changed that. The reason I never included everything is because it helps people draw wrong conclusions, so now we are open to that now.
I think the bugs you have found, have been fixed in MFlow but have still not been released. I might get that done today.
I will let you know about the new version as soon as it is up.

Thanks!

[LVDH]

How long should the process take then? mine has been running all day and still hasnt finished haha.

CFD takes a very long time. For these cars you should expect runtimes around 15h on a decent computer.
If you have a very good computer and you are new to MVRC, there is a funny gotcha: If you PC has enough memory (maybe above 32GB) you can create a super huge mesh, because you car geometry was not closed and you meshed the inside of your car. This will create a huge mesh which will take very long to get simulated. You can check the checkmesh.log file in the case folder. It should tell you that your mesh has around 6e6 cells.
Computers that run out of memory in this scenario, crash during meshing causing a cascade of strange error logs.

[yinlad]

Mine have been solving between 9-12 hours on my PC this week, 16gb ram and 6 core 3.5ghz. Though simplefoam only seems to utilize one core, but spreads memory use across 6 processes.

Edit: Worth noting that this is running the MVRC profile, so for 2500 iterations

[wb92]

Mine have been solving between 9-12 hours on my PC this week, 16gb ram and 6 core 3.5ghz. Though simplefoam only seems to utilize one core, but spreads memory use across 6 processes

What case settings do you run?

Edit: Worth noting that this is running the MVRC profile, so for 2500 iterations

Okay - got my answer.

I'm running on MVRC_Long up to 2000 iterations, on 4 cores i7 7gen, 32gb RAM and it takes around 12h (mesh + solving).
This gives me quite good correlation to the official results, but I guess I might check regular MVRC settings in the software.

Some general hints: be sure that you run openfoam without hyperthreading (HT or SMT), then check if your RAM runs in dual-channel mode

[jjn9128]

After some issues last year I was advised by CAE to run "MVRC" but edit the minimum mesh to 0.6 and run for between 1500-2000 iterations. Should run faster, especially for early development!

[LVDH]

New version is up.

[LVDH]

So here is the list of teams so far:

AGR
Baltic
Elbflorace
Ether Sim Sport
HP-Racing
JJR Racing
Koldskaal
Kostov Racing
MercuryMotorsport
MT Race Engineering
Pure Power Racing
Salamander
SHM Racing
Sracing
TalnoRacing
Team BargeBros
Team Logozzo
TF
Variante
Vision Formula Racing
WBRacing

Looks like I will have to dig out my plans for a qualifying if it continues like this...
But I do not what to complain, I think it is good that the championship is growing and with the rule switch that will hopefully take place mid-season, this should be the best year we had so far.

[CAEdevice]

But I do not what to complain, I think it is good that the championship is growing and with the rule switch that will hopefully take place mid-season, this should be the best year we had so far.

Hi, what do you mean with rule switch?

Is TF back? Good news!

[machin]

@Machin: could you please update Virtual Stopwatch with the new tracks?

Virtual Stopwatch is now uploaded with all new tracks (and still has last year's tracks just for reference).

Follow the usual link:=

http://www.competition-car-engineering. ... Timing.htm

As Variante says; the best thing to do is design your first car, run the CFD on it and then see on Virtual Stopwatch whether more downforce or less drag will give you the better performance... (The lower box underneath the lap time tells you this)



Good luck with your designs!!!