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

[spacehead3]

The official laptime simulator shows 1920mm as a default: should we consider it the "official" COG?

The optimum CoP is ~1.92m or 46.6% front balance, yes.

[CAEdevice]

The official laptime simulator shows 1920mm as a default: should we consider it the "official" COG?

The optimum CoP is ~1.92m or 46.6% front balance, yes.

Thanks!

[LVDH]

Hi, I am not sure about this rule:

11.3 The rear wing must be attached with at least three sections with one or two such sections inside of the volume RV_RR_WING_PYLON_V01 attaching to the gearbox.

Whats does it mean "at least three sections"?

Three or more...

I have fixed the issue on the support forum. Somehow I cannot find out how to make sure you have to tell my which exact OS you are using, so please make sure to tell in your messages. But support forum working or not, I am trying to replicate issues on Windows. In any case, Windows seems to be able to run Linux quite well now. Maybe someone knows if it makes sense to jam MantiumFlow into a OF Docker Image. This would allow us to use much more recent versions of OF!

[yinlad]

Hi, I am not sure about this rule:

11.3 The rear wing must be attached with at least three sections with one or two such sections inside of the volume RV_RR_WING_PYLON_V01 attaching to the gearbox.

Whats does it mean "at least three sections"?

Three or more...

I have fixed the issue on the support forum. Somehow I cannot find out how to make sure you have to tell my which exact OS you are using, so please make sure to tell in your messages. But support forum working or not, I am trying to replicate issues on Windows. In any case, Windows seems to be able to run Linux quite well now. Maybe someone knows if it makes sense to jam MantiumFlow into a OF Docker Image. This would allow us to use much more recent versions of OF!

Submitted a ticket on the support website

I'm not sure of the mflow process involving OF but after a brief glance at the OF docker docs it shouldn't be too much effort to build a container on top of the OF one that calls it and outputs to a directory (in theory)

[RicME85]

Have been considering the Docker route so would happily test it if you decide to go that route LVDH.
I have an Unraid server so have been playing around with Docker containers a little bit.

[LVDH]

I actually use Docker sometimes to run Python software. Never a GUI so far, though. Looking through the internet, there are various pages that claim that GUIs either work or they do not. I have not had the time to test it yet, though. Maybe someone knows an even better way to do it. The goal is to have OF and MFlow as conveniently usable to Windows users as possible. OF is a C++ software that needs to get compiled and MFlow is Python software. Running Python is only somewhat convenient on Windows. I am spending some time looking into WSL, maybe with that things are easy enough.

[variante]

Suggestions (in order of importance, more or less):

1 - Close the wheels completely (eliminating the rims, to minimize computational effort).
2 - Further taper the airbox (it still sticks out of a reasonably sized bodywork. Is that airbox volume even needed?).
3 - Lighter STL files (maybe it's only me, but i have troubles opening files larger than 10MB).
4 - Suspensions tweaks: give angle to front wishbones (to accommodate wing upwash and avoid stalling them), keep rear wishbones some 25mm off the diffuser volume (they currently intersect), use sharp trailing edges (never had meshing problems).
5 - Proper chassis sculpting to improve radiator design freedom (and the fuel tank should be integrated in the chassis).
6 - Would be nice to have proper wider engine exhaust manifolds (for the sake of realism).

[LVDH]

Suggestions (in order of importance, more or less):

1 - Close the wheels completely (eliminating the rims, to minimize computational effort).

I can have a look at this.

2 - Further taper the airbox (it still sticks out of a reasonably sized bodywork. Is that airbox volume even needed?).

I think it is needed as it helps keeping some other rules simpler.

3 - Lighter STL files (maybe it's only me, but i have troubles opening files larger than 10MB).

I agree on this. Matteo was nice enough to create these parts. However his parts always translate into huge stl files. I am not sure why that is. My plan is to recreate these parts in FreeCAD. This will have to wait until Monday though. My plan is to replicate his geometry as closely as possible. The wheels will stay as they are, so they will remain to be a bit large in file size.

4 - Suspensions tweaks: give angle to front wishbones (to accommodate wing upwash and avoid stalling them), keep rear wishbones some 25mm off the diffuser volume (they currently intersect), use sharp trailing edges (never had meshing problems).

I can have a look at this, but you should not get your hopes up. One reason being is that we have to settle at some kind of geometry at this point. We are getting close to our first race.

5 - Proper chassis sculpting to improve radiator design freedom (and the fuel tank should be integrated in the chassis).

You have to elaborate on this.

6 - Would be nice to have proper wider engine exhaust manifolds (for the sake of realism).

Yes and no, this is way down on my list.

You forgot the most important point. I have to supply you guys with a working version of MFlow. At least I am on it. Something must have changed in Windows and I have to find out. The VM on which I create the executable does not update itself but your machines obviously do. As they probably do it when you have no time for it, consider something like Linux...

[LVDH]

Hi, I am not sure about this rule:

11.3 The rear wing must be attached with at least three sections with one or two such sections inside of the volume RV_RR_WING_PYLON_V01 attaching to the gearbox.

Whats does it mean "at least three sections"?

Three or more...

I have fixed the issue on the support forum. Somehow I cannot find out how to make sure you have to tell my which exact OS you are using, so please make sure to tell in your messages. But support forum working or not, I am trying to replicate issues on Windows. In any case, Windows seems to be able to run Linux quite well now. Maybe someone knows if it makes sense to jam MantiumFlow into a OF Docker Image. This would allow us to use much more recent versions of OF!

Submitted a ticket on the support website

Thanks, it seems there is even more broken with the support sites. I am one it and will try to answer as soon as I have a better one than I do right now.

[spacehead3]

4 - Suspensions tweaks: give angle to front wishbones (to accommodate wing upwash and avoid stalling them), keep rear wishbones some 25mm off the diffuser volume (they currently intersect), use sharp trailing edges (never had meshing problems).

What do you guys think about allowing the suspension to be rotated some amount (maybe +/- 10 degrees?) around the respective wheel center? And then we could just include the rotated parts with our submissions.

[CAEdevice]

4 - Suspensions tweaks: give angle to front wishbones (to accommodate wing upwash and avoid stalling them), keep rear wishbones some 25mm off the diffuser volume (they currently intersect), use sharp trailing edges (never had meshing problems).

What do you guys think about allowing the suspension to be rotated some amount (maybe +/- 10 degrees?) around the respective wheel center? And then we could just include the rotated parts with our submissions.

I think this is a very very good idea (provided we define a single front wheel axis: the wheels have camber so we have two, so each suspension would rotate around its own wheel). Maybe 10° is too much (considering the rod attachments), but 5° would be OK.

On the contrary I would not change the section of the arms/rids for two reasons: 1) the "fat" section is an help to match the flow without separation 2) The present section orientation (almost neutral) is a compromise (as everything for the common parts).

[variante]

5 - Proper chassis sculpting to improve radiator design freedom (and the fuel tank should be integrated in the chassis).

You have to elaborate on this.

I meant this:

But, it's really not important now.

4 - Suspensions tweaks: give angle to front wishbones (to accommodate wing upwash and avoid stalling them), keep rear wishbones some 25mm off the diffuser volume (they currently intersect), use sharp trailing edges (never had meshing problems).

What do you guys think about allowing the suspension to be rotated some amount (maybe +/- 10 degrees?) around the respective wheel center? And then we could just include the rotated parts with our submissions.

Yeah, sound good.
Suspension guys would go nuts about that, but who cares

[LVDH]

4 - Suspensions tweaks: give angle to front wishbones (to accommodate wing upwash and avoid stalling them), keep rear wishbones some 25mm off the diffuser volume (they currently intersect), use sharp trailing edges (never had meshing problems).

What do you guys think about allowing the suspension to be rotated some amount (maybe +/- 10 degrees?) around the respective wheel center? And then we could just include the rotated parts with our submissions.

Yeah, sound good.
Suspension guys would go nuts about that, but who cares

The guys checking the rules would go nuts as well.
Anyway, maybe I am missing something here and it is indeed a good idea. Please keep it as a suggestion for next season as I see no way of having enough time to implement this aspect.
In a bit, I can have a look at what is happening to MFlow in Windows and best would be I can figure it out today as it will otherwise have to wait until Monday. I am also looking into how WSL works. This might make things significantly easier for everyone.

[Koldskaal]

What do you guys think about allowing the suspension to be rotated some amount (maybe +/- 10 degrees?) around the respective wheel center? And then we could just include the rotated parts with our submissions.

Yeah, sound good.
Suspension guys would go nuts about that, but who cares

The guys checking the rules would go nuts as well.
Anyway, maybe I am missing something here and it is indeed a good idea. Please keep it as a suggestion for next season as I see no way of having enough time to implement this aspect.

I think a better idea would be to rotate the sections individually. This is pretty easy to do with the builtin tools in OF if the upright and wishbones are in seperate stls:



All the arms can be made as y-extrusions from the origin that are rotated/translated into place at the beginning of runCase.sh:


Maybe something that can be implemented for next season?

[variante]

I think a better idea would be to rotate the sections individually.

Yes, although slightly more complicated, this would be closer to ideal.
Since we're on it, one observation: if you give pitch to each suspension member first and position them later, the more obliquely positioned elements will end up having smaller pitch than you originally intended. For example, the 10° pitch will end up being 7.1° relative to the car's direction, once the suspension element has been repositioned to (let's say) 45°.
Not even a problem, really, but worth mentioning for awareness.