Boundary Conditions, Turbulent model
 

I am trying to simulate the flow through the aorta. I already did some simulations assuming rigid wall and no branches at all. (one inlet, one outlet) I simulate three cardiac cycles of 1 second each and with a time step of 1 ms. I used the SST model - Gamma theta turbulence model. The simulations converged quite quick, and I was happy with that results.

The problem is that now I want to simulate again the same flow but including 6 branches in the descending aorta, so I will have in total 7 outlets (1 inlet-6 outlets). I have assumed a 0 pressure at the outlets and I am using the SST turbulence model with the Gamma Theta model again. The convergence criteria is 1e-5. This simulation does not converge. It runs prety quick until 1.3 seconds (1 cycle and 0.3 seconds of the second) but then it diverges and the simulation stops.

at this point I have got the idea of run some simulations without branches and without turbulence model and the simulations converge very good.

Then I run another simulation with branches but without turbulence model and then it stops again around 465 times steps (0.46 seconds)

I think the problem of my simulations is that i am assuming a 0 pressure at all the outlets. But at the same time I know i am assuming rigid walls. Can it be a problem of quality of the mesh. but then why the simulations without branches converges?

As you can see i am pretty confused. Any ideas to improve my simulations?

what code are you using? how the inflow flow rate distribute in the outflow branches? you can get several solutions for different distribution of flow rates....

I am using Ansys CFD, I have got a waveform for the inlet, and because the pressure at the outlet is zero the program distributed the flow rate by the areas of the branches. all of them have the same area. in one of the simulations also define the flow rate but it also stopped.