Khamsin Virtual Racecar Challenge 2014

[CAEdevice]

Congratulations to Variante but also to BigSpace for the the progress! All the cars are nearer and faster, next race looks more and more interesting. Impressive the general efficency of CSR car.

About CAEdevice: same classification of the previous race, but a bit disappointing this time: I had problems with the flows under and over the diffuser. Hope to solve them for Monza.

PS: My car was at first place in the two fast corners

[RicME85]

Thoroughly disappointed with the result.
Got a concept car built that I will be doing some back to back tests with.

[variante]

Thanks Matteo.
It will be hard to win at Monza, though...

[CAEdevice]

Thanks Matteo.
It will be hard to win at Monza, though...

Yes, it will be difficult because many cars are quite near now, but your car seems to adapt easily to fast tracks (even if I didn't think that Magny Cours was a low downforce circuit).

[machin]

A very good result for Variante today.. which just goes to show how good a nice "simple" car can be; if you can't fully optimise all the little winglets then you are probably better off without them; concentrate on building a nice smooth tub and getting the basic wings to work well first.

(even if I didn't think that Magny Cours was a low downforce circuit).

Variante's car worked well, but CSR's car with more downforce (and the attendent drag) was very close.... a car having a Cd.A of 1.1 and a -Cl.A of 2.65 would've been quicker than Variante..... easier said than done, I know, but we have seen cars in the Khamsin challenge with aero coefficients not far off those figures in last year's competition...

[machin]

I've (roughly) added the cars from Magny Cours 2014 to the aero coefficients chart... showing that both Variante and CSR have pushed the boundary and we now have a new "Best Limit" curve for low Cd.A values as shown below:-



Looks like I will have to re-draw the Monza chart so that it extends to lower Cd.A values! Well done guys!

[cdsavage]

Well done to Variante. Looks like I didn't take off quite enough downforce and drag.

[astracrazy]

Yep well done to Variante for the win and Big Space for the big improvement. But well done to everyone really as we are starting to settle down with the rules. Everyone was able to run legal and we are having less illegal cars on the first submission.

There seems to be a clear midfield now which is quiet close. I think this will become more defined after race 3

I also found it interesting/fascinating how we had to cars with the exact same time.

Sorry for the much up with the graphs a day early, i didn't know google will update the images which are posted around the internet. Note for the future not to update the day before

I'll be uploading pictures of the cars from race 2 later. These are not renders and i do not have time for that, they are basically screen grabs of the cars in sketchup.

Lastly I haven't had any feedback from the race visual, do you all use if first? Did you like it? Would you want anything changed?

[CAEdevice]

Lastly I haven't had any feedback from the race visual, do you all use if first? Did you like it? Would you want anything changed?

Hi Luke, I appreciate the visual: even if there is a discalimer, it seems that speed values are reasonable true.

[RicME85]

Just had chance to look at the result files in Paraview, seems I have experienced similar issues to Variante with the meshing.
Have extra faces around the bottom of the sidepods and randomly, the central section of my diffuser has extra faces but only on one side making my diffuser not symmetrical.

[julien.decharentenay]

Congratulations to all. Great to have a consistent flow of chat and discussion.

A couple of hiccups on the race 2: RicME85 & Variante have noticed some extra faces that have been added to their car design. I am sorry for this issue.

The simulation process is as follows:
1) The submitted geometry is downloaded and split into their individual components;
2) Each component goes though a mesh cleaning stage consisting mainly of "hole" filling. Holes are defined as edge that are only attached to one face. This process is automated using netfabb, which fills the hole by adding faces... Hence the extra faces.
3) The "cleaned" mesh are used in the CFD analysis.

You can have a look at the cleaned mesh in the result archive - from memory - or look at the meshed geometry.

The best option is to have a fully watertight geometry - but it is unlikely to happen with SketchUp... The next best thing is to get the model through netfabb to check the "hole" and the "hole filling" process. I will try to prepare a few slides (upload them on slideshare) and organise a quick Hangout on Air next Sunday evening http://www.timeanddate.com/worldclock/f ... p1=47&ah=1.

[astracrazy]

I think and it seems plug ins cause the problems.

A solution for both of you could be to cut down the area the you use the plug in? i.e instead of covering one big area, split it into 3-4 sections?

I think what Julien is suggesting would be good if you have these programs so you can check for any problem areas before uploading.

[CAEdevice]

A solution for both of you could be to cut down the area the you use the plug in? i.e instead of covering one big area, split it into 3-4 sections?

Yes, exaclty the solution that worked well in my case.

[machin]

OK, so after the Magny Cours Race it is clear that we've previously been looking at Cd.A figures that are way too high. Variante and CSR have both shown that we should all be looking for Cd.A values in the range 0.8 to 1.2 .

The Blue "Best Limit" curve shows roughly what Cd.A and Cl.A figures you should be aiming for.. but don't let that stop you; as Variante and CSR have shown; the "boundaries" are there to be stretched and pushed back!

[machin]

Sure, but I was looking at the values of Cd.A of Monza 2013, and they are higher than I would expect (also considering a 20% more drag for the diffrent rules).

It is worth remembering that Cd.A is a function of the car design, not a direct function of the track design; i.e. if Cd.A values were higher than expected at Monza last year then there is only one reason for it: the car's that were submitted had high Cd.A values. I think that we are only just seeing that everyone has previously been working with Cd.A values that were too high; as per graph above, we should be aiming to see Cd.A values in the region of 0.8 to 1.2, maybe even lower, whereas previously we were seeing Cd.A values of the submitted cars in the range 1.2 to 1.5.

I think it is good that we are seeing a progress in the design of the cars.