when i solve the problem ,the solver cann't run ahead,I found this words following:

Error: Floating point error: invalid number

Error Object: ()

how can i do. my model: one pipe with length of 20m and diameter of 13mm. I try the scale and units after importing the mesh file from gambit. it's funny when i solve the problem,probly,i try tens of times ,just one time succeed ,I cann't get ideal of it .when i change the diameter into 15mm,I never suceed.Pls help me and give me some advice on model and solving processure .

(Cf. mail of An Modh Coinniolach Fri, 18 Jan 2002, 7:59 a.m.)

hello,

I have the same error message (boundary layer heigth to lower) using Gambit 2.1.2; and I think it dosen't depend on the gambit release : in fact, I've noticed that it was because of the way the boundary layer expands over the face : the direction is almost perpendicular to the edge of the face where the boundary layer is defined (egde1 ]). So if the face has no perpendicular edges (egde1 | and edge2 / ), they can cross the border of the boundary layer (-), and the only solution is to lower the height of the boudary layer.

]------- /---|

]----- /-----|

]--- /-------|

]- /

]/