Hi

I am trying to simulate the transient fuel injection process in ic engines with axisymmetric model and expanding/contracting domain that extends from 0 to R m in radial dirn and 0 to Zpist in axial direction. Injector is located at (0,0) If there are lets say 20 time steps , fluent injects only in some time steps saying for rest of time steps, "unable to find injection" If the injector location is fixed and is very much in the computational domain, such a problem should not have come. Please can somebody solve this problem ?

Thanking you.

I've had problems before with injections positioned exactly on the boundary/symmetry plane.

I think it is best to cheat slightly and pick a point very slightly inside the domain, and/or have a slight radius for a conical injector.

Rob

thanks for this important feedback. Also have u come across such warnings as - Can't figure which cell particle is going to -- ABORT

??? Please can u tell me the remedy you got for overcoming the above.

Actually since it aborts many droplets, the final air-fuel ratio reduces considerably compared to what actually i should have got !

Well I am also experiencing the same problem, and unfortunately cheating doesnt help - nozzle is located inside of domain. Tried to vary the location - still the same error (cant figure blablabla). Anyone to help with this? Thanks in advance!