(Untitled)

[yhloke]

*blink* And it has been 3 weeks since I have mentioned any significant. Not much to say, just that I've been very busy; not busy enough to be completely isolated, but busy enough to not really think much about posting anything. I do go through these journals and the forums during my daily leisure internet time. I just don't want to mentally put any

Comments

[towerboy]

I used Fluent in a summer internship once, and it wasn't too bad, but I hope you are doing some pretty simple flows, or at least have some pretty loose tolerances because I was running my simulations on a supercomputer, and I think I usually took up 32 processors for it when I was running it. It's been over 2 years since though, so I don't remember much of it any more.

Good luck with the Persian chick ;)

[yhloke]

Yeah, we are doing very basic flows... basically we are trying to simulate 2D around an object in a wind tunnel (the object is supposed to be a simplified 2D model of a LeMans car). So we all really need to do is set up a grid, put it in fluent, set tolerance to 1E-6, run it once, then adapt the grid a little bit more.... we're still getting convergence problems, though. And for some of our curvy models, we're reducing drag a lot but we're also creating positive lift.... so our racing car will basically fly off the track. :)

[towerboy]

On the convergence problems, about all I can suggest is to make sure that right at the surface and especially near corners tighten down the edges so that your vertices are close together, of course, I'm sure you already knew that, but we had one corner that just would not converge for us.

[terrene70]

Wigger's Diagram

Teehee. That sounds funny.