[MVRC] Mantium Virtual Racecar Challenge 2018

Post here information about your own engineering projects, including but not limited to building your own car or designing a virtual car through CAD.
User avatar
CAEdevice
49
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

I think that there should not be an "intro class" but one (or more, I hope) open source car bodies with complete cooling ducts, hx, exhausts and engine inlet ducts.

This car bodies (provided by the top teams) should be completed with wings, winglets and (maybe) front wheels cover and mirrors.

The only limit should be to frozen the provided car body (it would simplify the rulecheck too). These car should be theoretically able to win races (same power).

The best customer team for each chassis can sum up the points with the chassis provider, so that we would have a "constructor championship".

Ps: this is an idea to make the design easier, but I am convinced that newbies should invest more time in learning the basics of CAD modeling.

wpsiatwin
wpsiatwin
2
Joined: 12 Jun 2018, 16:34

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

CAEdevice wrote:
20 Aug 2018, 20:25
I think that there should not be an "intro class" but one (or more, I hope) open source car bodies with complete cooling ducts, hx, exhausts and engine inlet ducts.

This car bodies (provided by the top teams) should be completed with wings, winglets and (maybe) front wheels cover and mirrors.

The only limit should be to frozen the provided car body (it would simplify the rulecheck too). These car should be theoretically able to win races (same power).

The best customer team for each chassis can sum up the points with the chassis provider, so that we would have a "constructor championship".

Ps: this is an idea to make the design easier, but I am convinced that newbies should invest more time in learning the basics of CAD modeling.
How did you improve your CAD skills because one thing I’ve been found is that universities don’t put much emphasis on learning CAD on engineering courses. So for me it’s a lot of trial and error which is a very inefficnet method.

User avatar
CAEdevice
49
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

A15013950 wrote:
20 Aug 2018, 22:31
How did you improve your CAD skills because one thing I’ve been found is that universities don’t put much emphasis on learning CAD on engineering courses. So for me it’s a lot of trial and error which is a very inefficnet method.
Sad but true.

The only thing I modeled when I was at university (2002...) was a coffeepot.

I know, it is difficult to find time and motivation, but now you have Youtube tutorials, Student Editions (SWX, Autodesk, ...), Open source tools (FreeCAD, Blender), peer2peer torrents. But, I repeat, I know it is difficult.

Anyway, I hope my proposal could help for MVRC.

User avatar
machin
162
Joined: 25 Nov 2008, 14:45

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

CAEdevice wrote:
20 Aug 2018, 20:25
I think that there should not be an "intro class" ....
This car bodies (provided by the top teams) should be completed with wings, winglets and (maybe) front wheels cover and mirrors.
That’s basically what you’ve done already, but we don’t have a huge number of people knocking on the door, even though what you’ve provided is a really good basis for a competitive car... I thought there would be more people on this forum who would want to stand up and show you how to improve your own car!!! 😉

Maybe it is just the CAD skills issue that is holding people back...?

...I think I might have done something similar to a coffee pot when I was at Uni too! 😂

Just long hours playing around modelling “stuff” was how I learnt...
COMPETITION CAR ENGINEERING -Home of VIRTUAL STOPWATCH

roon
roon
412
Joined: 17 Dec 2016, 19:04

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

machin wrote:
20 Aug 2018, 23:24
I thought there would be more people on this forum who would want to stand up and show you how to improve your own car!!! 😉
Nah, first you have to start an F1 team, spend millions of dollars developing and building a car, release a few blurry photos of it 'round about February, and then log on to a motorsports forum. Only THEN will you find out just how wrong your car is. 8)

User avatar
variante
138
Joined: 09 Apr 2012, 11:36
Location: Monza

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

This year I invited my teammates from the aerodynamics department (Formula Student, not F1) to join the challenge. Most of them were really interested, but at the end no one participated. They were afraid of time.
Somehow they also knew this challenge could have become an addiction...

I think it was a wrong choice from their side: it may be wiser to sacrifice an exam rather than an aero challenge. That's why I'm twice as good as the best one of them at aero design. That's also why I'll never get a degree, but this is another story... #-o

I like the idea of CAEdevice (about constructor championship). I'm not sure how many would join as constructors, but maybe it's worth a try. Anyway, wouldn't it only work with parametric files (so the customer would have to work on the same software as the construcor) or with modular cars? An LMP car is not very modular, an F1 is closer to the ideal.

I think, at the end, that the key for time saving is to make indeed the car modular (changing one piece is easy and doesn't fack up everything else). As for CAD complexity, it would decrease with a modular car. And, generally, it would decrease with "smarter" and simpler rules for bodywork. Also, a more strict set of rule would reduce R&D time (and, however, make the challenge less funny and interesting).

etsmc
etsmc
7
Joined: 04 Apr 2012, 13:20

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

Are we getting ahead of ourselves thinking of next year? I don't think so.
From the challenge website, it does say '2019 will see F1 style cars again' so now would be a great time to start planning.
For this year's challenge, it is underway and I feel it should be left alone, no changes to rules or software. If there is any software development then we could all test it to help out but understand it is for next year.

From last year I have to say the intro class was great, the car provided was excellent it was simple enough to design a front wing, rear wing and diffuser slap it on the car and run a simulation. Don't know if the car was available as a parametric model but it did not need to be and this is how a constructors option should be.

For a 2019 F1 style Constructors series.
- Constructor supplies the car bodywork excluding Floor, wings, bargeboards, nosecone, and diffuser.
- Team designs Floor, wings, bargeboards, nosecone, and diffuser

Rules would only need to be written for these areas so should make the rules simpler, much like we have now a simple flat floor, dual elements rear wing.
Design of intakes and cooling ducts has already been taken care of.
The development of these cars I see following how Variante's 'F1 2017 Aerodynamics and Car Development' thread has progressed.

The cars should be designed in such a way that if team A designs a Floor, wings, bargeboards, nosecone, or diffuser it should be able to fit any teams car (whether it works on that car or not who knows) Constructors, if they wish, could release parts for their cars.

If we are prepared and have a rulebook in place ahead of time I think it would be a success.

User avatar
CAEdevice
49
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

variante wrote:
21 Aug 2018, 00:16
This year I invited my teammates from the aerodynamics department (Formula Student, not F1) to join the challenge. Most of them were really interested, but at the end no one participated. They were afraid of time.
Somehow they also knew this challenge could have become an addiction...

I think it was a wrong choice from their side: it may be wiser to sacrifice an exam rather than an aero challenge. That's why I'm twice as good as the best one of them at aero design. That's also why I'll never get a degree, but this is another story... #-o

I like the idea of CAEdevice (about constructor championship). I'm not sure how many would join as constructors, but maybe it's worth a try. Anyway, wouldn't it only work with parametric files (so the customer would have to work on the same software as the construcor) or with modular cars? An LMP car is not very modular, an F1 is closer to the ideal.

I think, at the end, that the key for time saving is to make indeed the car modular (changing one piece is easy and doesn't fack up everything else). As for CAD complexity, it would decrease with a modular car. And, generally, it would decrease with "smarter" and simpler rules for bodywork. Also, a more strict set of rule would reduce R&D time (and, however, make the challenge less funny and interesting).
When you talked about "time" you make me think that we are in the same situation of real F1: just substitute "time" with "money".

A "top" team can invest more time than a small team, motivation can change the results but it is not always enough. No matter how many rules and limits FIA or LDVH will add: all the available resources (money for F1, time for MVRC) will be spent in the "free" areas. If we would be allowd o develop only rearview mirrors, we wuold spend our resources in ot. The laptime gap between the teams would be super small, but the classification would remain about the same.

Also consider that it took two years for me to obtain any result and that I am doig about the same numer of simulations per year.

The actual problem is to find a balance between "professionals" (engineers, engineering services providers, aerodynamics students, CAD designers...) and "amateurs" (motorsport fans), but (as in "physical" motorsport there is not a definitive solution).

The open source car bodies with cooling/engine inlets/exausts alredy done is the best solution in my opinion.

I can improve the intro car, but I have to say that the small time between the rulebook release and the first race reduced the possibility to do a good job. I will do better in the next season.

Meanwhile, consider that the 2018 intro car has only 20% less downforce than my official car at the beginning of the developement (2 weeks before the frst race)...

User avatar
schwepes
0
Joined: 01 Apr 2013, 10:01

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

etsmc wrote:
21 Aug 2018, 05:31
Are we getting ahead of ourselves thinking of next year? I don't think so.
From the challenge website, it does say '2019 will see F1 style cars again' so now would be a great time to start planning.
[...]
If we are prepared and have a rulebook in place ahead of time I think it would be a success.
Agree. Based on this year's experience I think that there should be a rulebook (or at least some main principles) at least 3-4 months before the 1st round. Especially if there will be so major change in the rules (WEC->F1).

btw. can't wait to see results of this year's first round. Looking at car images I have found some compliance issues, so I expect some penalties. I hope that compliance issues will be put to public discussion.

User avatar
CAEdevice
49
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

I can't wait to see the numbers too. I am wondering how much the new rules affected efficiency (in my opinion about 20%).

User avatar
schwepes
0
Joined: 01 Apr 2013, 10:01

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

Matteo, your car looks really great. Almost every detail, every surface, looks like comprehensively refined.

It will be also interesting to see what will be the progress in terms of performance/efficiency when we will see some "borrowings" between cars during the season. Some interesting concepts and solutions were put into the competition, though some of them I think are not fully compliant with the rulebook.

User avatar
CAEdevice
49
Joined: 09 Jan 2014, 15:33
Location: Erba, Italy

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

schwepes wrote:
21 Aug 2018, 10:45
Matteo, your car looks really great. Almost every detail, every surface, looks like comprehensively refined.

It will be also interesting to see what will be the progress in terms of performance/efficiency when we will see some "borrowings" between cars during the season. Some interesting concepts and solutions were put into the competition, though some of them I think are not fully compliant with the rulebook.
I don't think that my car performances can grow more than 5% during the "efficiency" races.
The "void" volume behind and especially in front the rear wheels (despite a small volume, that is the reason because many partecipants forgot to check it, I suppose) deeply affected the diffuser flow, in an area that was very efficient during 2017.

User avatar
Team_Bart
2
Joined: 10 Jul 2018, 19:26

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

Hi guys, since I'm new here, probably my opinion/experiences in the first weeks can help designing the (introclass) rules.
First, why did I join the full entry and not the introclass? The reason is quite simple. I think for a beginner it is more interesting/attractive to think about a global car design than directly start working on optimizing some parts. I think (and I thought when I joined this competition) that designing a diffuser or wing directly costs a lot of time and a lot simulations with very small changes to the car. Actually you directly start working on details in that case. From a learning perspective, for a beginner it is probably (in my case) more interesting to see the difference between for example a high and low nose cone than optimizing my wing with some additional small flaps. I think this becomes interesting when you have more experience in aerodynamics. For example, I tried the front radiators in my car, because I think it has some benefits, but am interested why nobody is using them in real live. The best way to find out why is try it yourself :).

The second topic is time. Indeed, as discussed above, this challenge is quite time consuming. Luckily I was not totally new in SolidWorks, so that helps me a lot. Anyway some beginner parts would be nice, but I will come back on this later. I think when there is a bit more time between the rule release and the first race this time is worth it, since the needed CAD time becomes a lot less when you have your basis car. At this moment my biggest problem is finding time for simulations since I have 1 computer/laptop which I used for more than only designing racing cars (unfortunately :) ). I don't know if it is difficult, but a great extension to MantiumFlow would be a possibility to pause the simulation. For example make it possible to run an hour in the morning, do something else and continue in the afternoon. Anyway, if you do not have a good computer, it becomes a problem to run a lot of simulations and with that very difficult to optimize details (like a good diffuser). I think a lot of amateurs will have this problem and prevent them from participating the competition. It might be a good idea to consider this by designing the new rules.

My idea for an introclass is the following:
Probably for next year it becomes a bit difficult, but I think for the intro class, it is a great idea to make a database with parts from the designs from previous years. Especially with F1 style cars we can split the design into specific parts like frontwing, engine cover, floor, etc. New teams then can use a lot of elements which in the basics works well. The goal then becomes to connect those parts to have a proper working car. An introclass team then can use for example a frontwing of team A and an engine cover from team B and designing the rearwing by himself. To prevent copying a car you can require that a team can only use an X number of parts from the same team. In this way new teams have a lot freedom in designing their own car but the needed time and skills are much less. Probably the only drawback is that teams have to submit their cars in parts and that all parts becomes open source. The last we can counter by only making available the parts from previous years.

User avatar
schwepes
0
Joined: 01 Apr 2013, 10:01

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post


CAEdevice wrote: (...) and especially in front the rear wheels (despite a small volume, that is the reason because many partecipants forgot to check it, I suppose)
Hmmm.. did I miss something

User avatar
LVDH
46
Joined: 31 Mar 2015, 14:23

Re: [MVRC] Mantium Virtual Racecar Challenge 2018

Post

Hi guys,
I read the critisism and suggestions. I agree very much have had pretty much all of these ideas. The problem is more my professional and private workload. MVRC is a lot of work, which makes it even harder. I am working on making this better all the time. Some stuff happens behind the scenes. I am for example working on automatic rule checking. This will remove the largest chunk of work and help a lot. One thing will very soon turn into reality. The compliance checks will be made public. This will probably happen this week.
Another thing is turning into reality right now. As I said, that bad weather made it possible to have some flow visualizations:
Image
Better quality can be found here:
https://mantiumchallenge.com/interestin ... mans-2018/

For whoever is interested: These are actually Lambda2 iso surfaces at Lambda2=35000. If you like, I can add these to the automatic post-processing. I can also release images that show much more of the cars. Variante has pretty interesting vortices along his car.