Hi, I created my grid file in I-DEAS, and then imported into Fluent for computation. Just after importing the grid, the "Grid check" command didn't find anything wrong, but after I finished the initialization, there was a warning saying non-positive volumes exist, and this time the grid check also failed, stating that many elements are lefthanded and many nodes lie below the axisymmetric axis. (my computation involved a mass-flow-inlet and a pressure-outlet, and I chose mass-inlet as initialized point.)

I can't figure out where I've gone wrong, if anyone could give me some hint or help, I'd be very grateful.

For the nodes lying below axisymmetric axis..try a grid translate option in fluent. If ur problem is axisymmetric about X..use Y translate =0.00001 (basically a small value). I had the same problem once with this error and it worked.

For lefthanded cells

do this..

> grid
> modify zones
> repair face handedness

Try this might work..!

I alos had the problem with non positive volumes. They only seem to materialise after I initialze. I don't experience any face nadedness problems though. Do you know why fluent suddenly found or created volumes with non positive sizes???

I've had this problem, too, when I was generating the grid using Gambit for both hex- and tet- elements. I do not know any reason why, but there is also a grid check in Gambit which is very helpful in finding out which volume in your model is wrong---most of the time I try to change the interval assignments and remesh the volume.

Thanks a lot, Sandeep, I solved this problem by translating my grid a little bit along Y direction. and it seems that if there're nodes lying below the axis for the axisymmetric problem, then non-positive volumes will arise by revolving such nodes, and at the same time cell lefthandedness appears also.

thanks again, all.

..well sometimes little things like this just happen..!I am so glad we have this discussion forum..!