[MVRC] Potential rule changes for 2023

[Koldskaal]

The mesh resolution will always be a compromise. MVRC deliberately leans towards coarse meshes to make simulations on small inexpensive PCs possible. We use wall functions on the entire car for this reason. So you can't really trust the boundarylayer behavior too much in these simulations.

I would too suggest that you look at the "near-wall velocity" instead.

[beschadigunc]

The mesh resolution will always be a compromise. MVRC deliberately leans towards coarse meshes to make simulations on small inexpensive PCs possible. We use wall functions on the entire car for this reason. So you can't really trust the boundarylayer behavior too much in these simulations.

I would too suggest that you look at the "near-wall velocity" instead.

Good point and fair enough but the force difference (between my simulation in starccm and this) is massive between the wings and the wake behind is also effected by this. I believe it is not totally mesh realted instead we can fix this via some turbulance model tweaks

[beschadigunc]

few examples of the same wing but one using my STARCCM + settings with mesh coarseness similar to 0.5

[G-raph]

This might sound a bit direct and harsh, but I don't have time for a long explanation, and please remember I'm only trying to help.

First of all, have you put your front wing into the high_res folder? Your surface mesh looks very coarse indeed.

Anyway... the problem is your cascade design, particularly the first element.

- The leading edge is too sharp and too nose down, which makes the boundary layer very thick immediately. Yes, you got away with it with your finer StarCCM mesh and better boundary layer modelling, but it is quite obvious from your LIC plot that your element is misaligned.
I would make it much thicker and rounder, and more nose up. You can get away with a sharp leading edge, as we only run at a fixed ride height, but in that case you have to make sure it is properly aligned.

- The lower surface at the trailing edge is too aggressive. It is pointing directly at the second element, creating a reflex in the camber line. So the flow has no choice but to separate from that element, which is made even worse by the poor boundary layer from the leading edge problem. This is quite obvious even in your StarCCM picture.

[beschadigunc]

This might sound a bit direct and harsh, but I don't have time for a long explanation, and please remember I'm only trying to help.

First of all, have you put your front wing into the high_res folder? Your surface mesh looks very coarse indeed.

Anyway... the problem is your cascade design, particularly the first element.

- The leading edge is too sharp and too nose down, which makes the boundary layer very thick immediately. Yes, you got away with it with your finer StarCCM mesh and better boundary layer modelling, but it is quite obvious from your LIC plot that your element is misaligned.
I would make it much thicker and rounder, and more nose up. You can get away with a sharp leading edge, as we only run at a fixed ride height, but in that case you have to make sure it is properly aligned.

- The lower surface at the trailing edge is too aggressive. It is pointing directly at the second element, creating a reflex in the camber line. So the flow has no choice but to separate from that element, which is made even worse by the poor boundary layer from the leading edge problem. This is quite obvious even in your StarCCM picture.

Thanks for the tips I will implement the design changes to compare the results , yes it was in the high res folder and I use the regular settings 0.5 base size

[Ft5fTL]

Back to the topic, i think we can use the engine air intake section for cooling aswell. Something like 2m3/s for sidepods and around 1m3/s for air intake (dunno if its possible tho needs testing) sounds balanced for me.

My biggest vocal point since 2019 was the bricklike engine model. Koldskaal did a splendid job to create one on the main thread. I had some spare time and did some modifications to that and created new suspensions, wheels, rims, exhaust and a gearbox-crash structure. If MVRC staff wanna take a look to them im more than happy to send them


Apart from those topics, im happy of the rules as it is. Just get rid of the Misc_systems reg box please.

[CAEdevice]

Good job Ft5fTL.

I would only refine some details:
1) More rounded for minimum chassis shape
2) No winglets on the rear wheels hub (replaced by a ruled volume for custom winglets). Yes, I know I modeled that winglet, it was a failed attempt to design something working for all the cars
3) Front wheels brake inlet should be larger than rear one

[jjn9128]

Those wishbones are so thin, sharp and nosed up you're all gonna have a lot of spearation issues. I'd reduce chord a bit and thicken up the LE and mid chord - F1 can nose down the front wishbones up to 10deg.

[Ft5fTL]

Good job Ft5fTL.

I would only refine some details:
1) More rounded for minimum chassis shape
2) No winglets on the rear wheels hub (replaced by a ruled volume for custom winglets). Yes, I know I modeled that winglet, it was a failed attempt to design something working for all the cars
3) Front wheels brake inlet should be larger than rear one

I did not touch to the safetycell. Just there for purely cosmetics. Same for the winglets and the brake inlets. Changing the brake inlets should be easy enough tho. I had similar to real life front brows and lower wings for the front wheel hubs on my preseason car last year. (viewtopic.php?p=1048842#p1048842) I can dig them up if you guys want.

Those wishbones are so thin, sharp and nosed up you're all gonna have a lot of spearation issues. I'd reduce chord a bit and thicken up the LE and mid chord - F1 can nose down the front wishbones up to 10deg.

Should be easy enough.

[jjn9128]

for something like this competition where the wishbones are fixed you want them as compliant as possible for all FW spanwise loadings.

[beschadigunc]

for something like this competition where the wishbones are fixed you want them as compliant as possible for all FW spanwise loadings.

Oh I wish , it almost always separates the current form ( unless you are running AMR 23 style center offloaded wing

[Schifty]

I might think to join the 2023 season, but just wanna know, what you suggest as software to build the car?

[Koldskaal]

I might think to join the 2023 season, but just wanna know, what you suggest as software to build the car?

I think you can use pretty much any 3D modelling software. Some notable free ones are: Blender, FreeCAD, Fusion360 (hobby-version). All of them have pros and cons. Go for whatever you are most experienced with, and or interested in. I use SolidWorks

[Schifty]

Thanks just wanted to know, its asked to be CFD compatible

[beschadigunc]

I would say we use the tire model from ft5ftl going forwards comparing that with images looks much better than the current version