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

[wb92]

New version is up.

Going to test it overnight.
Just one thing: in MVRC template lref = 3.4, however in MVRC_Long lref = 1.0... which one is correct? I would guess 3.4...?

EDiT: version 2003 (Windows) still pops-up "cannot run runFile"

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

Thanks!

[LVDH]

EDiT: version 2003 (Windows) still pops-up "cannot run runFile"

I will have to try again this evening. For this, I might get into touch directly with you guys.

[variante]

Yeah, i confirm.
In case Yinlad message got unnoticed, be aware that you can still run the case, after you have normally built it using MFlow. Then, all you have to do is:
> right click on your car folder
> click "Open in blueCFD-Core terminal "
> write "bash runcase.sh" in that terminal
> press enter
> wait a few hours...

[jjn9128]

Is it an openfoam version issue?


Further to variante if (like me) you don't have the right click into blue core set up, open the blue core terminal then change directory to your case folder. E.g. For me it's:
" cd c://

cd user/jjn/documents/mvrc/r001_v01a"

If you type "dir" it will show you what's in a particular directory.

[yinlad]

EDiT: version 2003 (Windows) still pops-up "cannot run runFile"

I will have to try again this evening. For this, I might get into touch directly with you guys.

I don't know if was specific to one run or not, but I've noticed for my latest test the HX volume flow at the top of the report seems correct, but in the Virtual Stopwatch section at the bottom of the report it's outputting '6' as the flow value. Definitely not double the value at the top fwiw

[Joe7218]

EDiT: version 2003 (Windows) still pops-up "cannot run runFile"

I will have to try again this evening. For this, I might get into touch directly with you guys.

I don't know if was specific to one run or not, but I've noticed for my latest test the HX volume flow at the top of the report seems correct, but in the Virtual Stopwatch section at the bottom of the report it's outputting '6' as the flow value. Definitely not double the value at the top fwiw

Can confirm, my value at the bottom is also '6', meanwhile the value for heat ex flow rate at the top is definitely not '6'

[yinlad]

It's not a huge deal, so long as I'm getting the correct reading at the top, I know to keep it above 1.5. But it may be confusing for new entries

[LVDH]

Just one thing: in MVRC template lref = 3.4, however in MVRC_Long lref = 1.0... which one is correct? I would guess 3.4...?

Overlooked this here. The one is actually correct. I compensate for the correct wheelbase way later in the post-processing.

I am now on this strange windows issue. I hope I can somehow replicate it. It is so frustrating that it affects so many of you, but so far it passes my tests.

[wb92]

Just one thing: in MVRC template lref = 3.4, however in MVRC_Long lref = 1.0... which one is correct? I would guess 3.4...?

Overlooked this here. The one is actually correct. I compensate for the correct wheelbase way later in the post-processing.

I am now on this strange windows issue. I hope I can somehow replicate it. It is so frustrating that it affects so many of you, but so far it passes my tests.

So lref = 1.0 is correct. Okay.
I use 201909c to solve and 2003 to post-pro, so I can still develop the car as usual
I noticed new, different color maps on post-pro, need to get used to them, while for CpT is great, that for Cp needs more contrast, but can live with it

[LVDH]

So lref = 1.0 is correct. Okay.
I use 201909c to solve and 2003 to post-pro, so I can still develop the car as usual

When the first submission gets close, you should take care of verifying if this approach is OK.

I noticed new, different color maps on post-pro, need to get used to them, while for CpT is great, that for Cp needs more contrast, but can live with it

Yes, at some point I set a better looking color map as default, if you like the old one, you can always go back.
The missing SLIC visualizations are what I hate the most about my software, the other thing is that there is not a drop-down menu for things like the color maps. This will be fixed soon, though.
For the surface Cp plots, I think, you are forced to use what I set and I agree, it is a bit ugly. I hope to improve it at some point. While it is ugly, it is at least more useful than the old options.



The new Windows version of MFlow for you guys is up. I can now replicate the issue, I just do not understand why my fix seems to work...

[wb92]

wb92 wrote: ↑
06 Apr 2020, 19:51
So lref = 1.0 is correct. Okay.
I use 201909c to solve and 2003 to post-pro, so I can still develop the car as usual

When the first submission gets close, you should take care of verifying if this approach is OK.

I think that might be not needed as new MFlow runs now. We'll see tomorrow if everthing goes well.

For the surface Cp plots, I think, you are forced to use what I set and I agree, it is a bit ugly. I hope to improve it at some point. While it is ugly, it is at least more useful than the old options.

I meant Cp slices. For the rest it's definitely better.
Meanwhile I disabled some post-pro options (streamlines, isosurfaces, etc.) and that gave huge time reduction to see what's going on with the car. However I increased drastically number of pictures to generate, still seems quicker than last year.

[LVDH]

I meant Cp slices. For the rest it's definitely better.
Meanwhile I disabled some post-pro options (streamlines, isosurfaces, etc.) and that gave huge time reduction to see what's going on with the car. However I increased drastically number of pictures to generate, still seems quicker than last year.

Streamlines and the surface visualizations take the longest. The iso surfaces are pretty fast, deactivating them will not change much. The streamlines are faster, the fewer triangles the monitoring surface contains, but as you have all hopefully learned, you should then not trust the measured value if you do not have enough triangles:
https://mantiumchallenge.com/porous-media-openfoam/

Maybe you can show an image of what you do not like, there is probably a solution to that. And if you have a color transfer function, you want to see in MFlow, just send it.

[wb92]

So MFlow_2003 works like a charm

Easy, it's not drama with colors. It's just that in previous Cp slices there was more contrast between colors, so you could see the areas where for example FW, RW, diffuser works the strongest. It was just bit easier to quickly spot it on, but as mentioned, no drama. Glad that software works without errors and initial phase of WBR20 development can go without delays ^^

[jjn9128]

How do we change what images are output? e.g. I'm getting an "out of space" error because I'm outputting 50 X slices, but all I'm really interested in is CpTot - can I deactivate Ux, Cp...etc? Similarly I'm not interested in the +/- 0.5 on the surface Cp contour.

[yinlad]

How do we change what images are output? e.g. I'm getting an "out of space" error because I'm outputting 50 X slices, but all I'm really interested in is CpTot - can I deactivate Ux, Cp...etc? Similarly I'm not interested in the +/- 0.5 on the surface Cp contour.

I also got the out of space error, but I don't actually think I was out of space, so not sure what was going on there