[_Arne_]

I am working on a transient simulation of the flow trough a centrifugal pump for a washing machine. The final goal is to extract the blade loading for structural analysis and also do a multi-phase analysis of a water/air mixture.

As a first try, I used a velocity inlet and a pressure outlet as described in a tutorial I found (I am new to sliding meshes and pumps). After some fine-tuning of the time-step and the maximum number of iterations, this gave a satisfactory result.

Now I am trying to model the BC's more realistically. For the inlet, I was thinking about a pressure inlet corresponding to a few centimetres of water (= +/- 500Pa). At the outlet, I have no information about the flow, so an 'outflow' BC seems appropriate (I have increased the length of the outlet pipe to about 6 times the diameter).

However, these two BC (pressure inlet and outflow outlet) are incompatible. Besides that, the outflow BC cannot be used in multi-phase analyses.

I am finding it hard to formulate accurate BC's for this problem. What would be your choice of BC's for this problem?

Thanks in advance!

[LuckyTran]

A pressure inlet and outflow is underdefined. Consider a pipe with a pressure at the inlet and an outlet. You simply don't have enough information to determine how much flow goes through. Any flow (or no flow) can satisfy this boundary condition pair because all you fixed is the pressure at one location. The outflow BC does not allow you to be ignorant.

You need to specify the velocity / volumetric flow rate through the pump using a mass-flow inlet or mass-flow outlet and this allows you to find the pressure differential.

Otherwise, you can use a pressure inlet and outlet (i.e. fix the pressure rise) and determine the flow rate.

[_Arne_]

Thank you for this clear explanation of the basic physics involved! I already had the feeling that there was information missing, but I was not entirely sure.

I have discussed this with the client and obtained more detailed specifications (mass flow and pressure gain).