Khamsin Virtual Racecar Challenge 2015

[LVDH]

Hi,

I somewhat agree. I few days ago I entirely had the same opinion. Getting a very efficient rear wing is not very difficult. But front downforce comes with a lot of cost in terms of drag. So if you leave the rear wing and balance the car by removing the front downforce you should be able to create an extremely efficient car.

But I am more and more coming to a new conclusion: With the rules of covering the front suspension, a large front wing as pretty much a given. And somehow the ones I create all create a ton of front downforce which has to be matched by my rear wing. So I am more and more wondering how to create a version without rear wing. The TF car should be a good candidate to show us how it could be done.

While I do not like the idea of not gathering any points in the next race, I am very much looking forward to some exotic ideas, yet my car will be very conservative as I do not have enough time. Somehow all my crazy ideas did not work as well as my old proven design, even though I wanted to through it out the window.

[cdsavage]

Couple of questions having looked through the first few and last few pages. Is there a setup guide for the needed software? Also are there recommended versions (the version numbers that the race is run on?)

Also where do you get Khamsin from? A google search takes you to Julien de Charentenay homepage which likes to Hibou Scientific which is now defunct. I looked at a couple more hits on google and they take you to either site:-(

I managed to get SketchUp...

I´m not quite sure, if it is written somewhere, but do you require the cars to be done in a specific CAD software?

Hi, there is no need to use a specific software, only a good quality STL format is required (I use Solidworks, some uses Catia, others use SketchUp, ...).
You can find information about the rules, the files to be submitted and the solver (now called One Click CFD) in the official website: http://www.khamsinvirtualracecarchallenge.com/ or just wait for Chirs and Julien on this forum who will help you.

Answered perfectly One-click CFD is the CFD software recommended for KVRC participants. Installation details on this page: http://www.khamsinvirtualracecarchallen ... -click-cfd

We require submission in the form of STL files, detailed requirements are at the start of the rulebook (available at http://www.khamsinvirtualracecarchallen ... lenge-2015, under "Rule Book and CAD files"). On that page you will also find the extra parts used when simulating the entries - you should include these parts if doing your own testing, but they aren't included with the submitted car.

[machin]

Not yet.

True, but we're probably heading towards a similar regulations package compared to this year. I.e: LMP like cars, more realism, possibly the choice for a front engine layout.

Personally for 2016 I'd like to see things evolve to a sort of "Ultimate F1" like the Red Bull concept... so we'd still see design freedom like we have now, but with an F1 slant....



But we're only half way through this season!

[CAEdevice]

Personally for 2016 I'd like to see things evolve to a sort of "Ultimate F1" like the Red Bull concept... so we'd still see design freedom like we have now, but with an F1 slant...

how did you get that picture? It is my car for the last three races!

[CAEdevice]

This one is interesting too:

[LVDH]

That was pretty much my Idea when I started. I was unable to produce such a car that also matched the rules.
So, from my point of view we should do everything to get the rules to a point where we are producing these kind of cars for next year.
But I guess we should discuss this after the season and hope that the great team behind KVRC choose to continue.

[machin]

how did you get that picture? It is my car for the last three races!

=D>

[CAEdevice]

That was pretty much my Idea when I started. I was unable to produce such a car that also matched the rules.
So, from my point of view we should do everything to get the rules to a point where we are producing these kind of cars for next year.
But I guess we should discuss this after the season and hope that the great team behind KVRC choose to continue.

The same for me: the frontal visibility of the front suspension makes the cars look like a lmp1 protype, but it adds some limitations. Also the outlet cooling surfaces with no possibility to rotate and the long external template are quite a big limit. Anyway I think the realism should be a priority.

[variante]

Not yet.

True, but we're probably heading towards a similar regulations package compared to this year. I.e: LMP like cars, more realism, possibly the choice for a front engine layout.

Personally for 2016 I'd like to see things evolve to a sort of "Ultimate F1" like the Red Bull concept... so we'd still see design freedom like we have now, but with an F1 slant....

http://i.ytimg.com/vi/iaup9MfM2vE/maxresdefault.jpg

But we're only half way through this season!

That's not far away from my design, is it? Covered wheels F1 car, or similar. Current regulations changes, however, are heading towards LMP style (which is an interesting field to explore anyway)

And...being half way through the season is a good moment to start thinking about next season (especially if big changes are expected), as I'd prefer not to make last minute changes again as happened, understandably, just before this season.

[CAEdevice]

Another idea (to avoid rules changes and test races during the season) would be to organise a single test race with the future rules in december-january. The new rules could be about: new templates (ex. more detailed dimensions for cooling device), new car schemas (front engined, lmp1 like, covered wheels f1 with more design freedom but less power than lmp1 like, ...).

I could consider to a new sponsorship for a single event

Last but very important thing: the framework. I'm testing other frameworks (Simflow, CAEdium, Blucape, ...). It would be great if one of them could become a partner providing a Windows porting for Openfoam. The one used during 2015 has difficulties with meshes bigger than 4-5gb (mine is about 8gb). Maybe OpenFoam 2.3 or 2.4 or 3.0 can help.

[graham.reeds]

You can get a windows version for 2.3 from http://www.cfdsupport.com/download-open ... ndows.html. Can't comment on its quality.

[sjns]

If we're talking next year I'll jump in:

Depending the solver I should say we should be carefull. The best thing about this year's solver is it's one click feature. It makes it easy for non cfd people to jump in this challenge. I myself wouldn't mind working more on the cfd side, but maybe there are others that don't. Especially potential new entries.

Another thought I had was the switch between high and low downforce races. This year most people probably built two cars(or almost ). The changes between Round 3 and 4 are huge. I think it would be interesting to have a realistic change of high and low down force tracks. With that in mind we could see two design philosophies during the whole season. Some cars will be better at high downforce tracks others at low downforce tracks. In addition the cars could be more refined.

[CAEdevice]

@sjns: with "solver" I mean the mesher (blockmesh+snappyhexmesh) and the numerical solver (simplefoam). OCCFD is a "preprocessor" and it is working greatly, no need to change it.

[RicME85]

Another thought I had was the switch between high and low downforce races. This year most people probably built two cars(or almost ). The changes between Round 3 and 4 are huge. I think it would be interesting to have a realistic change of high and low down force tracks. With that in mind we could see two design philosophies during the whole season. Some cars will be better at high downforce tracks others at low downforce tracks. In addition the cars could be more refined.

Are you saying to have high and low downforce tracks in a random order rather than all high then all low?
If so, we have done that previously, it was decided to do high then low to make things a bit easier for people to with less time to configure their cars.

[cdsavage]

Another thought I had was the switch between high and low downforce races. This year most people probably built two cars(or almost ). The changes between Round 3 and 4 are huge. I think it would be interesting to have a realistic change of high and low down force tracks. With that in mind we could see two design philosophies during the whole season. Some cars will be better at high downforce tracks others at low downforce tracks. In addition the cars could be more refined.

Are you saying to have high and low downforce tracks in a random order rather than all high then all low?
If so, we have done that previously, it was decided to do high then low to make things a bit easier for people to with less time to configure their cars.

To expand on this a bit - in 2014 we had 5 tracks, and the characteristics of each track were fairly unique, other than Monaco and Nurburgring. The reason we decided to move away from this was that it was necessary to spend a lot of time optimizing the car setup for each round, rather than evolving the design and improving overall efficiency. With the tracks for 2015, there is a greater focus on improving two base designs over multiple rounds, rather than constant re-balancing. It is true however that optimizing the car setup for different tracks is also an important skill, so there is a tradeoff to be made in the calendar choices.