[MichaelN]

Hi,

I am simulating a blood flow during a cannulation some arteries.
my boundaries are defined as openings and i want to include a certain feature. if the blood is leaving my domain (flowing out of the the opening) there has to be a pressure loss at the border.
But the blood can be sucked into the domain without a pressure loss.

i already tried to include a "loss coefficient", but than the pressure loss occurs in both directions of the flow.

my aim of the simulation is to describe the venturi effect which occurs if my blood stream has got a to high velocity. i already have experimentall results for the exact position, so the simulation has to be wrong.

because i am new to CFD, are there any other simulation properties (e.g. timesteps, turbulence numerics etc) that can change the results significantly?

thanks for your help

greets

[stumpy]

Perhaps you can make the loss coefficient a function of the velocity direction, so for positive values the loss coefficient in non-zero, but for negative values it is zero:
if(Velocity u>0[m/s],1,0)*lossCoeff

[MichaelN]

hi stumpy,

thank you for you quick answer. I tried it today but the solution doesn't make any sense.
again i would be interested how big the influence of the timestep is, regarding the value of the solution.