[oevermann]

Hi everybody,
I am experiencing some strange behaviour with inflow BC'c. I computing the simplest case of an inviscid straight chanal of length 1m with inflow velocity 1m/s (10x1x1 grid pints). Outflow boundary condition is zero gradient for U and p=0, slip condition along walls for U, and zero gradient for p at the inlet. The solution should be constant U=1 everywhere! Solver is essentially turbFoam with turbulence switched off. The solution looks like this (obtained with the sample programm)

x U_x
0 1.00176
0.111111 1.00848
0.222222 1.035
0.333333 1.06522
0.444444 1.0575
0.555556 1.04222
0.666667 1.0397
0.777778 1.03272
1 1.0012

But that means I have a mass error of 6.5% at x=0.333. The runtime output tells me, however, a
continuity error of
sum local = 2.86984e-43, global = 2.86984e-43, cumulative = -1.12757e-16
which is very fine!

Is there an explanation for the velocity profile?

Thanks for any hints and help

Michael