Khamsin Virtual Racecar Challenge 2015

[variante]

What is your frontal area guys btw? I wonder if mine is not too high, at 1.75m^2.

Mine should be around 1.64m^2

Chris' "open source" model: 1.66m^2

[CAEdevice]

What is your frontal area guys btw? I wonder if mine is not too high, at 1.75m^2.

Mine should be around 1.64m^2

Chris' "open source" model: 1.66m^2

Your model looks quite efficient

[machin]

The CoP position is not that insignificant as it was said in the thread. I ran some tests with the small app and just changed the CoP from the ideal 1.65m and the lap time difference changed more than I expected!

It is worth bearing in mind that a COP of 2.0m means that you have twice as much rear downforce as front downforce.... that is quite a big difference... and will of course mean the rear of the car has a much greater capacity for cornering than the front. A COP of 3m means that you have no front downforce at all so the front of the car will behave like that of a car having no downforce regardless of whether your overall Cl.A is 1 or 10.

The main point I want to stress is that whilst COP variations on their own make quite a bit of difference, you shouldn't be thinking about COP as an independent variable from the Cl.A and Cd.A values... I'll try to explain below:-

Lets say you design your car with as much downforce as possible at both ends and, run it through CFD and get the following results:
Cl.A = 10, Cd.A = 3.33 (efficiency 3:1), COP = 1.95m that results in a lap time of 39.34 seconds.

The COP of 1.95 means 35% of your downforce is at the front end, and 65% at the rear. Or put another way: the Cl.A of the front half of your car is 3.5 and the Cl.A of the rear half is 6.5.

Now, because you have already designed your car with as much downforce as possible at both ends, in order to get a balance the only option open to you is to reduce rear downforce (you can't simply "move the COP forward" as that would require more front downforce). In order to balance the front of the car you need a COP of 1.65m which means you need 22.2% more downforce than the front. That means your rear Cl.A needs to be 3.5 x 1.222 = 4.28. Lets say you make an adjustment to your model and run another CFD test: you might find you took off too much downforce, so you need to make another adjustment and re-run the CFD and lets say you now have perfectly balanced your car. Your overall Cl.A will have now fallen to 7.78 (3.5+4.28). If we assume that you have the same efficiency of 3:1 after the modifications your overall drag will have fallen to 7.78/3 = 2.59.

if we plug in the numbers Cl.A = 7.78, Cd.A = 2.59 and COP = 1.65m we get a lap time of 39.02 seconds. So the additional 1 (or more) modifications to your model and the associated CFD runs has only gained you 0.32 seconds of lap time....

That is why we are recommending that for the high downforce rounds teams design as much downforce as possible at both ends of the car and not worry too much about the balance: the extra work involved getting the balance correct yields very little gain in lap time; if you have already maximised downforce at both ends.

I hope this little explanation also reassures potential entrants that they do not need to spend a lot of time refining their cars for the first few rounds: it should be possible to get a fairly good result without it.

(Of course if the front of your car looks like it will generate very little front downforce there is no point putting a huge wing on the back, and vice versa.. a little thought is required).

[MadMatt]

Thanks machin for the long explanation, I will play a bit with this if I have time.

[CAEdevice]

Well, the first race will be in about ten days, it seems that about 18 teams have been preregistered. What about e brief presentation on that forum for who never wrote untill now?

[MadMatt]

I agree, guys let's show your progress!

[julien.decharentenay]

Thanks for letting me know that there was two more teams registered. Welcome to AMR Racing and TF. I will have to send them some more details on the submission process...

Things are starting to look up on our side...

[CAEdevice]

Something more than a teaser: MP002 (that would have raced the car if KVRC had been F1 based) and MP003 almost completed.

[AratzH]

Looks really good Matteo!! Nice double front wing? I hope to be able to post some teaser pictures soon...

[CAEdevice]

Looks really good Matteo!! Nice double front wing? I hope to be able to post some teaser pictures soon...

Hi AratzH, I'm glad you will be involved in KVRC also during 2015!
The double front wing is for the high DF races only (and it is not enough... but I'm leaving for some time and I had to complete the car in advance).

[machin]

I must admit... I hadn't looked at the KVRC bodywork rules for this year up until now.. and... I'm really impressed! The internal parts "templates" are really good; even down to the side crash structures; brilliant!

This image is too good not to post (for all those people who have a passing interest, even if they're not enrolled into the championship itself), showing (half) the internal components that the bodywork must "clothe":-



=D> =D>

[machin]

Something more than a teaser: MP002 (that would have raced the car if KVRC had been F1 based) and MP003 almost completed.

https://dl.dropboxusercontent.com/u/522 ... teaser.JPG

Both look awesome Matteo: a huge step on from last year I would say....

[CAEdevice]

Something more than a teaser: MP002 (that would have raced the car if KVRC had been F1 based) and MP003 almost completed.

https://dl.dropboxusercontent.com/u/522 ... teaser.JPG

Both look awesome Matteo: a huge step on from last year I would say....

Well... the strange thing is that I'm happy with the F1 performances and not very much with the LMP1!

[cdsavage]

A few points about the submission process:

-Was anyone planning on submitting using STEP, rather than STL?

-For STL submission, there was the provision to allow submitting a single STL, with each of the groups inside that file with their own 'solid label'. Was anyone planning to use this method? The alternative is to output an STL file for each of the groups (body, front wing, etc), and then submit a single ZIP file with these STLs inside it. We would probably prefer that everyone use the second option, it should make things easier. 7-zip is a good free option for generating ZIPs.

-As specified in the rulebook, the submission must include a text file specifying the offsets of the parts which are allowed to be positioned (footbox template, front suspension templates, side impact templates). This should be included in the ZIP file.

-A reminder about geometry quality, internal faces, intersections etc - while I know it can be a lot of work, ideally there should be no internal faces and no intersections. Before the groups are split off, the geometry should form a single surface with no holes. When the groups are split from the body, it should be possible to join them back together with the body to form the single surface with no intersections, overlapping faces, etc. The meshing process will probably be able to deal with some intersections and overlapping faces as long as the whole car forms a closed volume, but it would be best not to rely on this. Here is a basic example cut in half to show what I mean by no intersections or internal faces - http://www.mediafire.com/download/3c258 ... xample.zip.

-For the rendered images that we'll be generating, we will possibly add decals/logos onto the car - if you have one or two logos that you like us to include then feel free to submit these, but we may end up not using them if it becomes too time-consuming.

[MadMatt]

I do it with separate STL files already, so I am happy with this!