SOS: the nonpositive k and epsion ,how to get the correct boundary settings
 

hello everybody,when i try to simulation the pantograph wind tunnel experiment,i got an divergent k and epsilon using simpleFoam or pimpleFoam with k-e turbulent modle.

i have looked though the forum a long time for a solution to this kind of divergent probelm, after changing some related settings follow the tips i got,
like checking the model length dimension ,setting the proper magnitude of k and e whenever in the internal field or boundary,setting the relax factor or tolerence
for u and p,changing the numerical scheme for k and e , even modifying the solver for k and e,

however ,all of this didn't change anything :(

the keypoint should be in the boundary settings,because when i turned off the turbulent modeling, the procedure can run smoothly.

those in the attachment is my cas directory , the grid is not in the constant/polyMesh but the boundary dictionary for the file size consideration.



anyone could help me out?

best regards for you,thanks .

abbott

I cannot run your case without the mesh, so it is difficult to spot the error. Can you give the log file? Does it diverge in the first time-step, or only later on? Is your mesh ok (check with checkMesh)?

hello Bernhard thanks for your reply ,
the log file is in the attachments ,maybe you can find some hints from that .appreciate for your help

come on ....
i need you help .
Even when i chang to the rhoSimplecFoam solver,the problem arise the same,and this time ,the "floating points excepting " error came out at the first iterate when solving the pressure eqution,while the simpleFoam or pimpleFoam prompt the bounding turbulent energy and e at the first iterate ,and then the "floating points excepting " error after several iterate.

i will to be crazy ,foamers ,please help me out, thanks so much

Hej,

first of all calm down, you cannot expect people to answer within minutes. this is a forum of volunteers not your microsoft helpdesk hotline.

Second, please check your epsilon and k values, inlet and initial. they seem very high. I don't know what you are simulating, but your values are very large. check the tutorial for values that are more appropriate. k = 0.01 and epsilon = 0.5 and then start varying those values towards what ever you expect to have for your k and epsilon values.

try to run the simulation with "standard" values first, as for example the values that are used in the tutorials

thank for you advice
 

hi
thanks for your advice ，i will follow your suggest.
i'a really a lit impatient, however, i will calm myself down and study the tutorials mor carefully.thanks .