Khamsin Virtual Racecar Challenge 2014

[cdsavage]

I'm not sure if i'm answering your question or not. But i'll try

theres nothing in a skp file you need to set up. The mesh is pre-set and you input the m/s and yaw within the portal, you let it work for a while then you select the openfoam solution. In about 2 hours you then get sent the file which needs to be run with openfoam

It can't be used with khamsin or sketchup

I understand this, but khamsin can output a set of openfoam files like what I assume the portal will do. I'm asking if there can be a sketchup file supplied that will allow us to generate the same output using our own PC.

[astracrazy]

you mean mesh wise?

you'll need to group the front and rear wing as one then the body as the other. Then set the boundary conditions for each?

I need you to pm back as well when you get time

[julien.decharentenay]

@cdsavage: I will need to send you a slightly updated version of Khamsin - as the mesh setting is slightly different. I will prepare it and should be able to send it next week, alongside an updated SketchUp file to use.

@CAEdevice: openFoam does not support GPU out of the box at the moment. There are a couple of extension, which I would love to hear of how they are performing.

Portal: I am working on being able to submit a skp file as well as this will be required for the KVRC challenge.

Video-conference: I am sorry but will not be able to do it this week-end. But I will plan for next week (Saturday 8pm, Brisbane, Australia time). I will send an invite to the pre-registered teams tomorrow.

[CAEdevice]

Sorry to bother you with obviuos general questions, but I was wondering:

- Wich is the definition of Cd.A and Cl.A?
- The value of COP should expected to be near the center of mass: wich is the assumption made for the center of mass?

Thanks!

[RicME85]

Got t he registration email and have signed up.

New team name this year.
No where near ready yet with my car, havent had much in the way of creative juices flowing. I have however spent more time on producing a cleaner model this year on the parts I have started.

[CAEdevice]

Another question from a newbie point of view: the CFD analysis is "steady state" or "transient"? I'm usind a time domain (transient) analysis at the moment (time step 0,5*10^-5 s) because a previous static solution produced strange results.

[astracrazy]

any cfd questions it would be best to e-mail julien, sorry i can't help

[CAEdevice]

A question about the setup for each race. It is possible to upload a new version of the car for each race or only precise components can be modified (es. flaps, wings)?

[astracrazy]

Every race you upload your car using our portal by the submission date (link and details you need to do this will be e-mailed to you). So you free to make any changes you want between races

If you car is illegal in anyway, you will be e-mailed about the changes you need to make. You are then only permitted to make these changes to you car to make it legal (nothing else) and then you need to re-submit by the date which is given to you (which will be 2-3days later). Failing to do so or making other alterations to you car will mean you original submission will be used and a 10sec penalty will be added to your time

if that makes sense

[CAEdevice]

Every race you upload your car using our portal by the submission date (link and details you need to do this will be e-mailed to you). So you free to make any changes you want between races

If you car is illegal in anyway, you will be e-mailed about the changes you need to make. You are then only permitted to make these changes to you car to make it legal (nothing else) and then you need to re-submit by the date which is given to you (which will be 2-3days later). Failing to do so or making other alterations to you car will mean you original submission will be used and a 10sec penalty will be added to your time

if that makes sense

Ok, all is clear, thanks

[julien.decharentenay]

- Wich is the definition of Cd.A and Cl.A?
- The value of COP should expected to be near the center of mass: wich is the assumption made for the center of mass?

Cd.A and Cl.A are the drag area and downforce area, i.e the drag/downforce coefficient multiplied by the frontal area. In short, it is just normalised drag and downforce.

COP: the COP is the COP of the downforce. My understanding is that you get the best of the downforce if the COP of the downforce matches with the center of mass. I am under the impression that the center of mass is given by the F1 regulation and is 45% front and 45% back wheels. Can someone confirm or correct?

the CFD analysis is "steady state" or "transient"?

The CFD analysis is steady state. Can you clarify what do you mean by "strange" results?

The cost associated with transient analysis would be too high - the computational domain is ~ 50m long. With a velocity of 44.7m/s (100mph) and a time-step of 5e-5 seconds, one would require over 20,000 timesteps to do one cycle through the domain. Steady-state simulation are currently achieved in 500 to 700 iterations...

[julien.decharentenay]

I changed https://cfdcloudengine.appspot.com/ to allow for the use of SketchUp file directly (in place of exporting in STL and sending through). Luke and I will be having a meeting tomorrow evening and will decide on a schedule of video conference to provide opportunity for discussion and direct feedback prior to, during and after the season.

Thanks you to the teams that have already registered. I am organising the Paypal account and should be able to send invoices next week (or this week-end if lucky). Let us know if there are any issues.

[CAEdevice]

- Wich is the definition of Cd.A and Cl.A?
- The value of COP should expected to be near the center of mass: wich is the assumption made for the center of mass?

Cd.A and Cl.A are the drag area and downforce area, i.e the drag/downforce coefficient multiplied by the frontal area. In short, it is just normalised drag and downforce.

COP: the COP is the COP of the downforce. My understanding is that you get the best of the downforce if the COP of the downforce matches with the center of mass. I am under the impression that the center of mass is given by the F1 regulation and is 45% front and 45% back wheels. Can someone confirm or correct?

the CFD analysis is "steady state" or "transient"?

Thank you. With "strage results" I mean that the total downforce (actually the floor contribute to the total downforce) does not converge with my CFD code (SWX FlowSim, but similar results were obtained by a colleague who used and old version of CFDesign): the floor downforce oscillates (+/-20%) around and average value. I'm going to try with Kahsmin to, but I could find time to install and learn how to start.

The total length of your domain is about double compared to mine, maybe I have some boundary effects that affects my results. Which is the height of you domain? I have about 5000mm above the car.

COP: How does VirtualStopWatch works about COP? Difference between COP and center of mass (where centrifugal forces are applied) should cause an additional torque, resulting in understeering or oversteering. So a great distance between COP and COM should have bad influence on the laptime. Can you confirm that?

The CFD analysis is steady state. Can you clarify what do you mean by "strange" results?

The cost associated with transient analysis would be too high - the computational domain is ~ 50m long. With a velocity of 44.7m/s (100mph) and a time-step of 5e-5 seconds, one would require over 20,000 timesteps to do one cycle through the domain. Steady-state simulation are currently achieved in 500 to 700 iterations...

[CAEdevice]

Anyone can confirm that the first race will be Monaco?

[astracrazy]

yes

http://www.khamsinvirtualracecarchallenge.com/calendar