F1technical.net

I thought I'd make a new thread where we can discuss any potential rule changes for next years MVRC challenge, as a few ideas have been posted in the challenge thread but its hard to keep track of them. I personally have been thinking of a few ideas and it'd be interesting to hear what everything else has in their head too.

1. Reducing the cooling demand
I'm sure everyone involved is aware that the cooling demand is very high. I'd propose either reducing it by about a third to bring it more inline with what F1 teams use, either through reducing the mass flow requirement or increasing radiator porosity in CFD. I'd also propose the following rule changes.

6. Heat Exchangers

6.1 There must be up-to two [old: one] heat exchanger on each side of the car, meeting the following rules:

This is to allow some centre-line cooling as f1 teams do.

6.1.2 The planar shapes may be rotated around the Y-axis only.[

I propose deleting this to allow more freedom

6.1.3 Each shape must be at least 180,000mm2 in area.

And deleting this too, as its currently redundant - I think everyone runs radiators much bigger than this. If a design can meet the cooling demand its fine, if not then there's a power penalty.

2. Resurfacing some parts like the suspension - I'd be happy to volunteer to do this

3. Possibly updating the tyre geometry/model. One thing I noticed, compared to CFD images from when I worked in F1, is the front wheel wake is very biased towards the bottom, i.e. there's almost no upper wheel wake. This means that on my car for example, I have a downforce-generating front wheel vane which is completely opposite to what is seen on F1 cars. I think if we updated the tyre geometry we could get some more realistic wake patterns.

4. Changing the engine inlet/misc systems geometry - I'll come back to this later as this post is already getting a bit long


5. F1 style rear/front wing endplates - I don't really have any idea how to write rules to do this

6. Realistic noses - possibly a minimum area rule?

Is there anything you think I've missed?

I agree with pretty much everything.

LegendaryM wrote: ↑

22 Sep 2022, 15:35

2. Resurfacing some parts like the suspension - I'd be happy to volunteer to do this

2.1 being able to change suspension's angle of attack. Either by rotation of the whole assembly around the wheels' center line, or by being able to choose from a couple of suspension models (identical to each other, aside from the incidence of the elements). F1 regulations allow from 0° to 10° (nose down) for front susp, and -10° to 10° for rear susp.

LegendaryM wrote: ↑

22 Sep 2022, 15:35

5. F1 style rear/front wing endplates - I don't really have any idea how to write rules to do this

Let's impose a minimum radius (on X planes).
A radius rule might be difficult to check, but if some flexibility is used, i don't see problems. Basically, all those without a thin endplate and with a gentle curvature are good to go, without actually measuring anything.

LegendaryM wrote: ↑

22 Sep 2022, 15:35

6. Realistic noses - possibly a minimum area rule?

That plus a rule to prohibit holes (like: only one continuous line inside the nose-rulebox on Y planes).
Or a template to fin inside the nose. All nosecones are pretty much in the same position, so it's would be easy to make a common template without interfering with everyone's design.


And if you want further realism, you could also impose the 75mm rule on the sidepods (again, with much flexibility).
But that would obviously decrease design variety.

I think with this competition you want to not go down the F1 rabbit hole of rules upon rules. It makes it harder for you guys then harder for the organisers to scrutinise.

For me the bulkiness of the minimum chassis is just *bleugh* especially how it impinges the front of the tunnels.

variante wrote: ↑

23 Sep 2022, 16:44

LegendaryM wrote: ↑

22 Sep 2022, 15:35

5. F1 style rear/front wing endplates - I don't really have any idea how to write rules to do this

Let's impose a minimum radius (on X planes).
A radius rule might be difficult to check, but if some flexibility is used, i don't see problems. Basically, all those without a thin endplate and with a gentle curvature are good to go, without actually measuring anything.

What could be done is create a rulebox where a shape should fit in. A rulebox that measures from the lowest part of the leading edge radius of the front wing main plane. and then take this box 20mm further outwards and 20mm further upwards wherein the leading edge stays a constant shape.

This allows a significant amount of freedom, while forcing a certain shape to be present.

LegendaryM wrote: ↑

22 Sep 2022, 15:35

6. Realistic noses - possibly a minimum area rule?

That plus a rule to prohibit holes (like: only one continuous line inside the nose-rulebox on Y planes).
Or a template to fin inside the nose. All nosecones are pretty much in the same position, so it's would be easy to make a common template without interfering with everyone's design.


And if you want further realism, you could also impose the 75mm rule on the sidepods (again, with much flexibility).
But that would obviously decrease design variety.

If you just impose a minimum + maximum length and width that'd make it a whole lot easier. Plus, you could impose a shadow rule, this does not disallow holes, but does make them more complex.

I don't think that any of the noses I have seen are impossible. The extreme ones are impractical in real life due to time and budget constraints. And well, holes are definitely possible, we have seen them since 2008.

Outside of these propositions, I do agree with jjn9128. It can very quicky get out of hand in terms of scrutineering. Not only that, but requiring more complex rules would reduce accessibility to the competition.

wesley123 wrote: ↑

23 Sep 2022, 17:41

variante wrote: ↑

23 Sep 2022, 16:44

LegendaryM wrote: ↑

22 Sep 2022, 15:35

5. F1 style rear/front wing endplates - I don't really have any idea how to write rules to do this

Let's impose a minimum radius (on X planes).
A radius rule might be difficult to check, but if some flexibility is used, i don't see problems. Basically, all those without a thin endplate and with a gentle curvature are good to go, without actually measuring anything.

What could be done is create a rulebox where a shape should fit in. A rulebox that measures from the lowest part of the leading edge radius of the front wing main plane. and then take this box 20mm further outwards and 20mm further upwards wherein the leading edge stays a constant shape.

This allows a significant amount of freedom, while forcing a certain shape to be present.

[/quote]

A radius rule is quite complex, and is still open to loopholes like the Aston Martin rear wing. Its also quite difficult to design a surface to meet that for people who aren't that experienced with CAD.

Maybe a simpler option would be to specify a mandatory part in the rollover region, that we then have to join our front/rear wing onto?

wesley123 wrote: ↑

23 Sep 2022, 17:41

What could be done is create a rulebox where a shape should fit in.

I think that someone would still find a way to put sharp corners in (=endplates, strakes).

wesley123 wrote: ↑

23 Sep 2022, 17:41

I don't think that any of the noses I have seen are impossible. The extreme ones are impractical in real life due to time and budget constraints. And well, holes are definitely possible, we have seen them since 2008.

I agree. But i think the intent of these new MVRC rules was to get as close as possible to current F1 rules.

LegendaryM wrote: ↑

24 Sep 2022, 12:06

A radius rule is quite complex, and is still open to loopholes like the Aston Martin rear wing. Its also quite difficult to design a surface to meet that for people who aren't that experienced with CAD.

Yeah, that's why i talked about a "flexible" enforcement of the rule.
We already have rules that are "flexibly" enforced, like the minimum thickness rule.
As long as someone isn't trying to take advantage of the radius rule, he shouldn't be penalized even if he doesn't really meet the rule.
Some controversies might arise, but such rule (for sidepods) actually existed in the old management era, and all went smoothly.

As for the Aston Martin loophole, a bigger radius might do the job. Need to check though.

LegendaryM wrote: ↑

24 Sep 2022, 12:06

A radius rule is quite complex, and is still open to loopholes like the Aston Martin rear wing. Its also quite difficult to design a surface to meet that for people who aren't that experienced with CAD.

Maybe a simpler option would be to specify a mandatory part in the rollover region, that we then have to join our front/rear wing onto?

The rulebox doesn't need to force a radius per se, but simply a continuation of the leading edge shape it connects to. Let shape and radius be free, as long as it continues the leading edge shape within the rulebox, forcing it to end perpendicular at the top of the box. This doesn't mandate a radius or any complex shape, it just forces the leading edge shape to stay the same within this box.

I'd be all-in for a simplified ruleset where the metrics would be derived as less of a downforce/drag winner, and more of a downforce and wake defined championship. The beauty of the CFD work you folks have put in is that you don't have to prove it on track, and your concepts could actually lead to revolutionary change in motorsport.

It then comes down to the focus of achievement in the championship. If you measured the downforce of the car, and had a multiplier based upon how far the wake extends behind the car (smaller = better), the focus would then become designing a clean car that can race nose-tail without artificial mathematical constraints.

I love following this championship! Please take my words as encouragement and not as criticism.

This is an interesting discussion. Thanks for starting it!

The comments I am about to make are based on my own view of what the MVRC challenge should be. I understand that the other competitors and / or the organisers may have a completely different view. I'm ok with that and happy to go with whatever is decided.

I think the challenge is about :
- Rewarding creativity, understanding of CFD and Aero, and CAD skills of the participants.
- Keeping it relatively close to modern F1 cars, but sufficently relaxed to allow creative and original designs.
- Keeping it simple for the organisers, both in terms of enforcing rules and providing CFD results in a short time frame.

With that in mind, here is my opinion on what has already been proposed.


1. Reducing the cooling demand

I fully agree with LegendaryM proposal (massflow reduction, removal of the number, orientation and area limits of the radiators), on the basis that our cars are clearly overcooled compared to current F1 cars. What I found so far is that, despite a wide range of bodywork designs being possible within the rules, they don't make much of a difference as everything is dominated by the large amount of cooling flow hitting pretty much the entire rear-end of the car.

But I understand that finding cooling is part of the challenge, so I would suggest to reduce the massflow requirement from 3 to 2.5. This should make our cars comparable to what F1 cars run at the most demanding tracks (Mexico / Singapore...).


2. Suspension

I think we all agree that both front and rear suspensions would benefit from being re-surfaced. I don't have time to volunteer for that unfortunately, but I would support anyone who would do it.

However I think it should stay a spec part. Having the ability to design or rotate each supension member could be quite daunting for new entrants and unnecessarily complicated for the organisers (as the suspension is supposed to move with car rake angle).


3. Possibly updating the tyre geometry/model.

I understand where LegendaryM is coming from, and agree that the upper wake is too small compared to the lower wake. However I don't believe this has much to do with the tyre geometry itself.
The lower wake may too big because the tyre is dropped quite low under the road, which makes a massive contact patch.
The upper wake is definitely too small because the flow separation point is way too late. This is not because of tyre shape, but because of boundary layer CFD settings. The simplest fix in my experience is to either apply surface roughness or a different type of mesh on the tyre.


4. Changing the engine inlet/misc systems geometry

Yes, it would be nice to have new surfaces if someone is willing to help. But I don't think it is critical.


5. F1 style rear/front wing endplates

I don't think you realise that the current F1 endplates are the result of extremely complicated rules, which would be a nightmare for the organisers to check and enforce. A simple radius check would not work as you need to have exclusions where the wing elements join the vertical endplate.

Only PurePowerRacing has a front wing endplate that appears F1 legal (but I'm sure it actually is not), and we know he has very strong CAD skills. JJR is close but with too tight radius. All the others are miles off the F1 design.
But I don't think it is a problem, as it keeps things simple and allows anyone to be creative, and we can't say that any of our front wing look completely unrealistic.
This is because the regulation boxes do a good job to enforce generic, F1-like wing shapes.

I am totally against the idea of "flexible rules" based on the vague criteria that the geometry "looks like f1". Because where do we stop? It is a massive can of worms. The current rules are very clear, black and white, and I think that is very good for everyone involved.

So if anything, you could make the boxes around the front wing and rear wing endplates even tighter if you really want. But I don't think any action is actually necessary.


6. Realistic noses

To me this is similar to the endplates, I'm not sure there is a need to change anything. You could add a crash structure that we would need to clear, but please don't make it massive.

Zynerji wrote: ↑

24 Sep 2022, 15:43

I'd be all-in for a simplified ruleset where the metrics would be derived as less of a downforce/drag winner, and more of a downforce and wake defined championship. The beauty of the CFD work you folks have put in is that you don't have to prove it on track, and your concepts could actually lead to revolutionary change in motorsport.

It then comes down to the focus of achievement in the championship. If you measured the downforce of the car, and had a multiplier based upon how far the wake extends behind the car (smaller = better), the focus would then become designing a clean car that can race nose-tail without artificial mathematical constraints.

I love following this championship! Please take my words as encouragement and not as criticism.

I love your ambition! Sadly we have to be realistic here, our CFD simulations are at least 20 times coarser than a standard F1 CFD simulation. I don't think we would see any realistic or meaningful wake difference between cars with these settings.

I agree with jjn and graph, I don't think we should aim for replicating f1 rules. Inspiration, sure, but with some simplifications and adjustments to make more room for more creative solutions.

I think one thing we should remember is, right now, Andre is the only person who checks the cars. And of course more rules (or more complicated rules) means more time spend checking the cars. I think Andre already has a paraview script setup to assist in the scrutineering process. But imo we should aim for 100% automated scrutineering even if it means simplifying the rules. I already sent Andre a python script that can check certain "kinds" of rules. I'll see if I can share my script with the rest of you.

My current script can only handle things like: "is this reference surface visible from these angles" and "is this body inside this reference volume". It could be an interesting challenge to expand it to include the section rules for the front wing, and minimum radius rules.

I had a go at trying to program something for minimum radius rules. In theory you could use VTK to create x-slices of the relevant stls and then find the circumscribed circle to sample windows of 3 points. like this:

(Thick black lines are supposed to be x-slices of the stl)
This method requires a certain degree of refinement for the stls. But at least it is something concrete that is not reliant on Andres subjective judgement.

I agree with most of your suggestions, expecially with G-Raph comment about mesh and surface roughness of the wheels.

I add a point we talked about in the other thread.

I suggest to not double cl.A coefficient when simulating the laptime. It is not professional and I think that with car development and rule fine tuning we can reach decent laptimes without tricks... and, after all, none of us is (I suppose) Adrian Newey.

Other ideas:

1) To measure the pressure resultant on the brake ducts, that should be higher than a defined value. It would add realism without any complicated rule (we already do this for engine inlet).

2) To pin a post with a list of all our suggestions so that we can easily read them in the future, without scrolling the thread.

3) To increase a bit the clearance between internal minimum volume (e.g. safety cell) and external maximum volume (e.g. car body). 10 mm would be enough to make cad modeling (e.g. blended surfaces) easier. Alternatively: to round some edges of the internal minimum volumes.

4) Re-work non only the dimensions, but also the alignments of internal mechanical parts, filling some internal space that produces unuseful cells during meshing

CAEdevice wrote: ↑

27 Sep 2022, 21:39

Other ideas:

1) To measure the pressure resultant on the brake ducts, that should be higher than a defined value. It would add realism without any complicated rule (we already do this for engine inlet).

2) To pin a post with a list of all our suggestions so that we can easily read them in the future, without scrolling the thread.

3) To increase a bit the clearance between internal minimum volume (e.g. safety cell) and external maximum volume (e.g. car body). 10 mm would be enough to make cad modeling (e.g. blended surfaces) easier. Alternatively: to round some edges of the internal minimum volumes.

4) Re-work non only the dimensions, but also the alignments of internal mechanical parts, filling some internal space that produces unuseful cells during meshing

1) yes this is a good idea
3) I agree with as well, I think the easiest thing to do would be to resurface the chassis and gearbox so they are tangent continuous and generally a bit nicer. I think LVDH said they had done some work improving the CAD so maybe this has already been adressed.

2) Yes I will try to pin some suggestions to the top. That kind of puts me in control, maybe we should have a vote or something but I think the organisers and people who have been involved for many years should have the biggest say.

It might be a daft idea, but I'd like to contribute regarding the endplates. As said before in this thread, replicating the rules of F1 is to complicated, but inspiration is good. And I think we should strive to make appearances as close the real life cars.

How about a rule stating that all parts connecting the wings with the pylons and endplates must be of a perfect circular curve between two different angles?

For example:

The angles are set with 0/360 at the upper end of the Z-axis.

1. All bodywork connecting the rear wing to the pylons must form a perfect circular shape between 340 degrees and 190 degrees when intersected in the X plane. The circular shape must have a minimum radius of 50 mm.

UlleGulle wrote: ↑

03 Oct 2022, 15:49

It might be a daft idea, but I'd like to contribute regarding the endplates. As said before in this thread, replicating the rules of F1 is to complicated, but inspiration is good. And I think we should strive to make appearances as close the real life cars.

How about a rule stating that all parts connecting the wings with the pylons and endplates must be of a perfect circular curve between two different angles?

For example:

The angles are set with 0/360 at the upper end of the Z-axis.

1. All bodywork connecting the rear wing to the pylons must form a perfect circular shape between 340 degrees and 190 degrees when intersected in the X plane. The circular shape must have a minimum radius of 50 mm.

... or that all the elements connecting airfoils (or "wing profiles") to the endplates must be a single blend of one or more sections