[MVRC] Mantium Virtual Racecar Challenge 2017

[LVDH]

You do not have to rerun the whole case. Just download MantiumFlow once more, start the GUI, open the case and then hit the post-processing button. If it still fails I will have to see why.

[Alonso Fan]

You do not have to rerun the whole case. Just download MantiumFlow once more, start the GUI, open the case and then hit the post-processing button. If it still fails I will have to see why.

I tried redownloading and post processing again to no avail

[LVDH]

OK, thanks for the files. It appears that the Windows OF version creates different log files than the Linux version. The names for the force coefficients and the bin files are for example mixed up.
I can fix this. It used to be that every different version creates different log files, but now even in between the OS the logs have to change. Why? Who knows. Give me a day.

[Alonso Fan]

OK, thanks for the files. It appears that the Windows OF version creates different log files than the Linux version. The names for the force coefficients and the bin files are for example mixed up.
I can fix this. It used to be that every different version creates different log files, but now even in between the OS the logs have to change. Why? Who knows. Give me a day.

how are the actual results when compared to your simulations? are they similar?

[RicME85]

Have set up the latest version of MFlow along with blueCFD and everything is running beautifully.
One thing I noticed when I ran the previous OF setup with the VM was the RAM usage was very low. I have 20gb in my machine and assigned 18gb to the VM (was running on Win7 and had got it running on next to no RAM) but it never used more than 6gb. Currently at time 194 and Im getting around 95.gb to 10.5gb of usage. I am also able to use my PC whilst it is running which is handy. Oh and it works great on Windows 10 with the latest updates unlike the previous OF solution.
Is MFlow supposed to start the simulation? When I hit start it just flashed up a CMD window for the faintest of moments before disappearing, I had to run 'blueCFD-Core 2016 terminal' then navigate to the case and bash runCase.sh to get it started.

[CAEdevice]

My feedback about the new solver: it requires less RAM, but it is a bit slower. I think that most of the partecipants will find it faster because it can run in 16gb environemet without using all the RAM. This did not happen to me with OF1606+ and Vbox because I use a 128gb workstation. It looks very stable: good job LDVH, thanks!

[Alonso Fan]

Is MFlow supposed to start the simulation? When I hit start it just flashed up a CMD window for the faintest of moments before disappearing, I had to run 'blueCFD-Core 2016 terminal' then navigate to the case and bash runCase.sh to get it started.

yes I had to do that to start the case too

[LVDH]

Did you guys apply the patch as described? MantiumFlow should start the simulation, but OF needs the patch for that. The fix of my software is uploading right now, so it should be available in a few minutes.

[Alonso Fan]

Did you guys apply the patch as described? MantiumFlow should start the simulation, but OF needs the patch for that. The fix of my software is uploading right now, so it should be available in a few minutes.

I did yes

[CAEdevice]

Did you guys apply the patch as described? MantiumFlow should start the simulation, but OF needs the patch for that. The fix of my software is uploading right now, so it should be available in a few minutes.

No, I still use the command window not the GUI (but I copied the patch files)

[LVDH]

Did you maybe change the location of the blueCFD installation? If yes you should reconfigure MantiumFlow (either the default_settings files or do it the GUI).

[RicME85]

applied the patch and the directory was correct in MFlow

[RicME85]

Chris - re legality check. In mine you mentioned the diffuser fences should be in a separate file, the diffuser was generated from the online geometry tool hence the parts being in one file. I didn't want to separate them and ruin the geometry. If they really have e to be separated then that's something that could do with being changed on the tool if possible (Andre?)

[rjsa]

I didn't get an email

[CAEdevice]

I compared OF1606+ and OF4.1 results: the difference is very very small (less than 2%)