Khamsin Virtual Racecar challenge 2013 (CFD model racing)

[N12ck]

Once a car is submitted if there are any legality issues a time penalty is added, you can't change or modify a car after the round has started,

[cdsavage]

I meant changes for the next round.

[N12ck]

Results, Feedback, Time Penalties, Championship standings will all be published tomorrow.

Thanks
Nick

[N12ck]

French Gp results are up http://www.khamsinvirtualracecarchallen ... -french-gp

[RicME85]

need some formatting to make that page more readable.
really surprised at the result, good for my team to get points on the board but unless something major happens over the next races that its, no chance of a championship for us, just wish the car we was meant to run made it to the grid.
obviously a lot of work went into the csr car, not only does it look good but it works, big surprise to see variante so low down.

[variante]

hey Nick, thanks for your work, but i fear it's not finished yet: those data look wrong...not in the right order maybe.

..to be more precise: looks like my car achieved the fastest lap (without the time penalty), however it's only got L/D ratio of 1; and a downforce level which is the half of my previous car (the current car is nothing else than an evolution of the previous, so...)

[N12ck]

emails getting sent within the next hour, explaining penalties etc, vtk links etc, I will double check the CFD data tomorrow, but this is the data we scrambled in excel, so it may have got mixed up, I'll check and check with Julien

[machin]

hey Nick, thanks for your work, but i fear it's not finished yet: those data look wrong...not in the right order maybe.

..to be more precise: looks like my car achieved the fastest lap (without the time penalty), however it's only got L/D ratio of 1; and a downforce level which is the half of my previous car (the current car is nothing else than an evolution of the previous, so...)

I think you are right; I've just plugged a few of the CFD results from the website into a copy of the Virtual Stopwatch program and got different results to those shown, so I think the numbers are jumbled somehow...

...bear with us on this one, I'll contact Nick and Julien and help them get this sorted.

[N12ck]

hey Nick, thanks for your work, but i fear it's not finished yet: those data look wrong...not in the right order maybe.

..to be more precise: looks like my car achieved the fastest lap (without the time penalty), however it's only got L/D ratio of 1; and a downforce level which is the half of my previous car (the current car is nothing else than an evolution of the previous, so...)

I think you are right; I've just plugged a few of the CFD results from the website into a copy of the Virtual Stopwatch program and got different results to those shown, so I think the numbers are jumbled somehow...

...bear with us on this one, I'll contact Nick and Julien and help them get this sorted.

worked it out what had went wrong, in one or 2 the wrong results were added into the table, but this has been corrected as I hunted out the correct results

just to show how much data there is:

[machin]

Good work Nick!

Some impressive CFD figures there too... the spread of lap times is fairly small; well done to all involved.

On the whole it appears that teams managed to get a better balance to their cars than they did for the test... although the majority still have bias to the rear; so there's definitely some room for a bit quicker laps in all the cars.

I'm looking forward to seeing images of all the cars; certainly the winning car looks to have gone for a simple (and yet still efficient) route so that approach clearly works well...

Bad luck to Variante.... hopefully something that can be sorted for next time...? It is interesting to note that whilst the downforce coefficient for the JJR car is better in the straight ahead conditions, Variante has the better coefficients (less drag, more downforce and better balance) in the yawed condition, and a faster lap time... (ignoring the time penalty)....

[variante]

Bad luck to Variante.... hopefully something that can be sorted for next time...?

Actually i'm quite happy with the result: fastest lap by 0,5s and a lot of room to increase the DF on the front axis (thus the balance). My only concern is that JJR and CSR managed to get a bit more DF on the rear; on the other hand, i haven't started using CFD yet.
I've still got no idea why my car has been declared illegal, but Nick is going to send us all the info. Not something major, i'm sure.

BTW, Julien asked for something like this, right?

[N12ck]

Hi Variante,

The reason for the illegality was the tub, it had a radius spanning the side of the cockpit, it should be straight, I am just finding each competitors email addresses now.

Thanks
Nick

[etsmc]

Well i am happy with a 3rd. looks like there was some improvement over my test car.. producing more down force and less drag but the balance seems to have gone towards the rear.
Here is the car that was submitted (although missing the beam wing)
will be interested in the paraview files to see where improvements could be made.


congrats to other teams and also the guys for running the tests. =D>

[AratzH]

Really nice results. I'm happy to see that the competition is pretty tight. On the other hand it was a shame that our car got a penalty. Im looking forward to see what was it.
It is also curious that having better Cl/Cd ratio than Mercury our time (without penalty) was 0.7s slower! I think the forward biased DF has not been to good. It seems that the 2.2m for CoP was the sweet spot. The good thing is that there is lots of room for improvement, specially in the rear wing and diffuser (wich had half the DF of the front wing )

Now some questions:

1. How are Cl and Cd calculated? Which are the reference areas for them?
2. What is the minimum mesh elemt size? This is critical for designing the gap betwen the planes of the wings. If the smaller element is bigger than the gap, the the plane is considered as a single piece, loosing the benefit of having multiple elements.

Now some pictures of the car:

[machin]

It is also curious that having better Cl/Cd ratio than Mercury our time (without penalty) was 0.7s slower! I think the forward biased DF has not been to good. It seems that the 2.2m for CoP was the sweet spot.

For the best lap time you want to get your downforce split in similar split as the static weight disctribution. This is mandated by the FIA rules as 45:54 Front to rear. With a 3.4m wheelbase that's a Centre of Pressure (COP) of 1.84m.... so most of the team's didn't have enough front downforce to balance the rear... your design was one of the exceptions with lots of front end and little at the rear...

Having said that I think everyone did a good job getting fairly good balance; remember the real teams will CFD test their cars over and over again to get the ideal balance....

EDIT: Reference areas for Cl and Cd are the frontal area of the car, but take note that results on the website are displayed as Cl.A and Cd.A; so they include the frontal area.

EDIT 2:

A little bit of number analysis on the results to show the front to rear downforce split and from that the Cl.A-Front and Cl.A-Rear (only for the 100mph straight ahead case)