[MVRC] Mantium Virtual Racecar Challenge 2022 (Grand Prix Cars)

[yinlad]

One issue I have been having is that on some runs I only get vtk output for total(p) (no U output or static(p). I don't think it's anything to do with mflow, but if anyone else has run into this and what might be causing it I'd appreciate any help

[LVDH]

One issue I have been having is that on some runs I only get vtk output for total(p) (no U output or static(p). I don't think it's anything to do with mflow, but if anyone else has run into this and what might be causing it I'd appreciate any help

CFD simulations can crash. MFlow helps to setup rather stable simulations but things can happen, esp. if you have bad geometries for some reason. You should check the solver.log and make sure the simulation ran to the end. It can fail for reasons of numerical stability, or your computer screwed up (out of memory). It is a bit more unlikely but it can also run out of memory when creating the surfaces used in post-processing.

[spacehead3]

One issue I have been having is that on some runs I only get vtk output for total(p) (no U output or static(p). I don't think it's anything to do with mflow, but if anyone else has run into this and what might be causing it I'd appreciate any help

CFD simulations can crash. MFlow helps to setup rather stable simulations but things can happen, esp. if you have bad geometries for some reason. You should check the solver.log and make sure the simulation ran to the end. It can fail for reasons of numerical stability, or your computer screwed up (out of memory). It is a bit more unlikely but it can also run out of memory when creating the surfaces used in post-processing.

Yeah, usually a geometry issue when that happens to me. Sometimes I can fix it by just increasing the resolution of the STL output for the part in question.

[yinlad]

yeah I found it to be my rear wing causing the issue. I think the end plates were a bit thin in places

[samuel_l25]

[/quote]

Im having a similar (or same) problem as Samuel.
When I try to run the case in MFlow 2207e I get the message:
"MFlow.py:882: DeprecationWarning: QProcess.startDetached(const QString & command) is deprecated"
Maybe you have already solved the problem and can tell me what you did? So I can see whether its for me the same issue.
Thanks in advance!
[/quote]

You can build the case in the mantium software and then run it through the blueCFD core terminal.

i.e. open the directory and type in the terminal 'runCase.sh'

[yinlad]

Done a bit more testing and I'm very confused now. If I run the wheels and mandatory parts separately, I get all samples in post. Same for my cars files. But once I combine them and do a run, the post processing stuff that generates the VTKs crashes and fails.

Not expecting miracles but if anyone has any idea on things to try I'm all ears. I'm going to try a run with mflow 'd' version as a hail mary in case some config change in 'e' is interfering with openfoam some how but it's a wild stab in the dark

[Alonso Fan]

Unfortunately I'm saddened to announce that SHR will not make the first race of the season. The team is hard at work preparing an entry for race 2 onwards with a fast development rate expected with the new rules.

[RicME85]

This might be seen as a daft question...but here goes.
MAND_Safetycell_MVRC_V07 - Is it expected to enclose it or can we create our own bodywork that attaches directly to it? I know it cannot be modified but say, adding surfaces that connect to it are fine yes? So for example I can create bodywork that shapes the cockpit surround better, I dont have to say offset the whole geometry by say 10mm to create the shapes I want?

I was hoping to enter the first race but it isnt looking likely, even with the delay.

[LVDH]

Done a bit more testing and I'm very confused now. If I run the wheels and mandatory parts separately, I get all samples in post. Same for my cars files. But once I combine them and do a run, the post processing stuff that generates the VTKs crashes and fails.

Not expecting miracles but if anyone has any idea on things to try I'm all ears. I'm going to try a run with mflow 'd' version as a hail mary in case some config change in 'e' is interfering with openfoam some how but it's a wild stab in the dark

When you load a case which contains all files, do you see everything in MFlow? You might face an issue once mentioned here, that the files might not be in the same scale (some km long, others only mm?).

This might be seen as a daft question...but here goes.
MAND_Safetycell_MVRC_V07 - Is it expected to enclose it or can we create our own bodywork that attaches directly to it? I know it cannot be modified but say, adding surfaces that connect to it are fine yes? So for example I can create bodywork that shapes the cockpit surround better, I dont have to say offset the whole geometry by say 10mm to create the shapes I want?

I was hoping to enter the first race but it isnt looking likely, even with the delay.

All mandatory parts will get inserted into the simulation, so if you would not model this part, but everything behind and in front of it, you might get a valid car. I would recommend creating some nice looking surfaces which are slightly offset of the safety cell though.

[yinlad]

Done a bit more testing and I'm very confused now. If I run the wheels and mandatory parts separately, I get all samples in post. Same for my cars files. But once I combine them and do a run, the post processing stuff that generates the VTKs crashes and fails.

Not expecting miracles but if anyone has any idea on things to try I'm all ears. I'm going to try a run with mflow 'd' version as a hail mary in case some config change in 'e' is interfering with openfoam some how but it's a wild stab in the dark

When you load a case which contains all files, do you see everything in MFlow? You might face an issue once mentioned here, that the files might not be in the same scale (some km long, others only mm?).

Yeah everything is there. As I say I've run 2 separate cases (mandatory parts and my parts) that both generate the full VTK suite, but when combined break for some reason.

It outputs the all numbers correctly, but something about visualisation seems to be broken.

My plan is to start with the mandatory files and add my bodywork in piece by piece until it stops generating everything. From there I should hopefully be able to fix it

[TalnoRacing]

Finally got a car together and submitted an entry. Could not get CFD done, so I am hoping I didn't get it too wrong.

[jjn9128]

Finally got a car together and submitted an entry. Could not get CFD done, so I am hoping I didn't get it too wrong.

https://i.ibb.co/qgVDd9V/TR22.jpg

Its's very wrong and you should be ashamed of yourself

[LVDH]

Well, it is a very unique nose. I might have to create a rule about minimal surface in an X-cut for safety reasons.
And thanks to that new Aston Martin rear wing, I now understand that we should not have rear wing end plates above the main elements of the rear wing. No idea how to deal with that though.
You could have tried to integrate the air box a bit better...
It is very cool to start seeing the new cars. Lets hope for a very cool season, even if the start is more than rough.

[variante]

Finally got a car together and submitted an entry. Could not get CFD done, so I am hoping I didn't get it too wrong.

https://i.ibb.co/qgVDd9V/TR22.jpg

Looking sexy! And that's my favorite sidepod configuration.

But, since you haven't run CFDs, i'm concerned about your:
- Cooling. Your cooling exhaust might be way too small. I had big issues myself with that.
- Balance. That's a very small rear wing compared to the front wing...
- Wheel deflectors effectiveness. They created more harm than good in my simulations (7.5% downforce drop). I think they need a lot of optimization to work (or a lot of luck getting them right the first time).

[CAEdevice]

Well, it is a very unique nose. I might have to create a rule about minimal surface in an X-cut for safety reasons.
And thanks to that new Aston Martin rear wing, I now understand that we should not have rear wing end plates above the main elements of the rear wing. No idea how to deal with that though.
You could have tried to integrate the air box a bit better...
It is very cool to start seeing the new cars. Lets hope for a very cool season, even if the start is more than rough.

Maybe you could require second order continuity and blended surfaces, with a minimums radius and not intersecting surfaces.

Talno's car has a lot of interesting details, but I am afraid about cooling.