[KVRC] Khamsin Virtual Racecar Challenge 2016

[CAEdevice]

...hopefully the whole set of data: DF, drag, CoP, pressure differentials

I agree...

By the way: I have the impression that many teams are nearer to "efficiency" than "extreme df" setup.

I also think that the new cooling rules, even if they could/should be refined before the 1st race, helped to obtain more realistic and interesting design.

[RicME85]

I too am surprised by the JJR car, I would of thought the 2016 car would of been an evolution of the 2015 car, can only assume JJR has had a McLaren 2012 situation and come to the end of the line with development and needed to try something else. Either that or the first race was a fact finding mission to see how a different solution would stack up against the competition. Or its just a red herring

[JJR]

Hallo,
my 2015 car was only good on high downforce tracks .
I have rebuild 2016 car completly with later tracks in mind.
Actually new car has more downforce and it is more efficient than last year car in high downforce config.
I hope that it will working well on low drag circuits.
It looks like that 2016 competition will be tight.
I wish everybody good season.

[LVDH]

Wow, cool designs.
I am most surprised by JJR. I highly favor the closed pod design. I think it performs very well and looks much more pleasing. Now with one of the top guys going for open, I will have to take another look into this.

I think to have discovered two S-ducts. Probably the one on the Pure Power Racing is more of a diffuser.

For me clearly the most elegant car is the Kineuton.

Also there has been discussions about helping Chris out before he even complained about anything. Maybe he enjoys the work he is doing? As this is not F1, we should not loose our heads and start changing good rules. Looking at the cars it seems that some competitors have issues with submitting cars of good CAD quality. That Blindboom car does not even look like abstract art anymore. Some others at least have a F117 style going on. Maybe we could help out the competitors by creating a good guide on how to make good quality stl files. If the car quality gets better, maybe the KVRC staff will have less work and automatically we get our results faster. And I am sure that smoother cars get better aerodynamic performance. Not only in real life but due to meshing issues on the hard edges, the results of smoother cars will be much better.

Mantium is still using the low diffuser, which surprises me since we are in the high downforce rounds! I'm very curious to see how it performs

Well, like I said I did not have time to make something new. So the car is only the best of 2015 with a new cooling package for the new rules. Comparing my car with the others, I think I did get something wrong with the engine intake. I guess I will have to check once more.

I have started on a "new" car though. But as I still do not have that much more time it is only an improvement of the pre-race car. It turns out that when I touch the parts I created by using optimization, performance always deteriorates. This is quite annoying as I am running out of ideas and should clearly start on a completely new car. But as so far I cannot touch anything you might see that diffuser again next race, it works quite well.

[CAEdevice]

Yes, decent geometries would make the rule check and the pre processing faster.

I don't think it is only a issue with stl export, probably the geometry native models is quality poor. I am convinced that using SketchUp gives more problems than advantages: tools like Blender or FreeCAD would require more time to learn, but would make the job better.

[RicME85]

I think its mostly modelling ability, Blindblom has always had these issues although this car is better than past entries

[variante]

If that's the problem, maybe a few tips might help: http://www.sketchup-ur-space.com/2014/j ... tchup.html

[CAEdevice]

If that's the problem, maybe a few tips might help: http://www.sketchup-ur-space.com/2014/j ... tchup.html

Very good guide! I appreciate your skills (much better than mine) with SketchUp, but tools like Blender or FreeCAD (I am not talking about commercial tools) open a different perspective. Take for example the sidepod built with interpolating sections!

Ps: Honestly, looking at some cars, SketchUp is not always the real issue

[julien.decharentenay]

Apologies I came down with a bad cold last week and it dragged on this week. I had to take some time off to recover. I am now back to some level of functioning and will have a look at the past posts. I wanted to let you know that I will be pushing the next race back by 2 weeks. I will be releasing a new version of occfd in the next couple of days - that will hopefully solve some of the parallel issues and have a starting point for the heat-exchanger.

[CAEdevice]

Hi Julien! I hope you are feeling better.

Two weeks is quite a long time, but if it is not possible to be ready in a shorter time, its ok. Anyway, I am not plannig a revolution on my car for the extreme downforce races (I don't like them very much: to design an efficient car is much more interesting, and nearer to the lmp spirit). So I'm only going to correct a issue that caused balance problems in the pre season race.

Great news about the parallel solver! Thanks! Is the solution only related to a specific OF compilation (BlueCape) or it will work with other OF porting/virtualizations (CFDsupport, SimFlow, ESI, ...)?

I have to say that this year I am quite happy with the consistency of the results between local and remote solver.

[machin]

Try installing paraview 4.4, and this

https://www.microsoft.com/en-us/downloa ... x?id=36045

If this doesn't solve the issue, I'm not sure what the error could be related to

Thanks for the help: I downloaded both of those but no luck: I get the same issue. Interestingly both versions of Paraview open the "Body" mesh OK, so I can make my own surface pressure plots, but opening the "internal mesh" causes it to crash. I suspect that it is because I am using a very low power computer (A netbook). The computer I do my modelling on is even older and less powerful: 15 years old and has never been connected to the internet!

It is a shame, but it looks like I won't be making any Surface LIC plots: I will have to base my development on the official results and intuition [-o<

[LVDH]

Your problem is 99% a Paraview problem. The *.foam file is (I guess even on Windows) a 0 byte file, so it would be difficult for OCCFD to do something wrong with it as long as it is located correctly. But from your probable amount of RAM it sounds like Paraview is running out of memory. So for once not a Paraview bug.

[Alonso Fan]

Try installing paraview 4.4, and this

https://www.microsoft.com/en-us/downloa ... x?id=36045

If this doesn't solve the issue, I'm not sure what the error could be related to

Thanks for the help: I downloaded both of those but no luck: I get the same issue. Interestingly both versions of Paraview open the "Body" mesh OK, so I can make my own surface pressure plots, but opening the "internal mesh" causes it to crash. I suspect that it is because I am using a very low power computer (A netbook). The computer I do my modelling on is even older and less powerful: 15 years old and has never been connected to the internet!

It is a shame, but it looks like I won't be making any Surface LIC plots: I will have to base my development on the official results and intuition [-o<

No problem sorry I couldn't solve the issue. I have paraview 4.4 and all the other recommended software from the kvrc occfd page installed on 2 pcs, (relatively new pcs) and both work fine.

Perhaps you need a new pc?

[machin]

Almost certainly!!!

[CAEdevice]

Any news about the pre season race?