[PopeySailor]

Hello,

I need help with decision about which evaluation method to use for determining preassure drop.

I run transient, incompressible, k-Omega sim of pipes (with multiple curves, shape changes etc.) to compare old/new geometry. I found two ways to evaluate pressure loss (using Paraview):
a) integrate variables on inlet, gained pressure on inlet equals pressure drop

• problem of this method is in more complex cases, the pressure drop can be negative (found that it is caused by running incompressible by removing to many DOF, so following method should be used

b) using find data, max value of pressure in solved volume should be find, and that is also the pressure difference between inlet and outlet (= loss)

I expected to these values be the same, but... Values are different (for example a) 450 Pa, b) 800 Pa ). After comparing old vs new state of pipes, improvement is observed in both cases but different (method a) 30% b) 60% of original pressure loss).

My question is, how to choose whitch method should be trusted?

Thanks a lot for help.

pressureloss.PNG
Picture of original pipe geometry (green point is location of found max pressure)

[Tobermory]

If I follow your posting correctly, for method b you are using ParaView to find the maximum P (Pmax) in the domain, and then I assume that you have a fixedValue pressure boundary at the outlet, with p=0, so you are assuming that the pressure loss is equal to Pmax? Well that method will not be very accurate since the pressure can rise above the inlet pressure in a stagnation area.

A better approach would be to get the solver to calculate the area-averaged pressure on the inlet and outlet boundaries. You can do this with a function object either during the run or after the run. Check out the following, for example: https://cpp.openfoam.org/v8/classFoa...e.html#details

PS are you remembering that the pressure in the incompressible solvers is typically kinematic, i.e. p/rho?

[PopeySailor]

Hi, thanks for reply.

My problem is, that I am not skilled in programming, and thus I use Helyx OS (GUI for OpenFOAM). As you said, method of averaging pressure on inlet/outlet should do this, and I think that it should be similar as "integrate variables" in Paraview (method a) ) - dont you think?

(yes, I am aware of kinematic pressure )

[joshwilliams]

Originally Posted by PopeySailor

Hello,
a) integrate variables on inlet, gained pressure on inlet equals pressure drop

• problem of this method is in more complex cases, the pressure drop can be negative (found that it is caused by running incompressible by removing to many DOF, so following method should be used

b) using find data, max value of pressure in solved volume should be find, and that is also the pressure difference between inlet and outlet (= loss)

I would use method (a). Using max - min (which is what I think you do in b) is not reliable as these are just point measurements. Also, since you have a kind of complex geometry, flow separation causes increased wall pressure in this region. If there is some doubt, I would say to compare to some experimental data and or even better, an analytical solution and add complexity. I think there is likely experimental data on pressure drop in pitzDaily case? Which also has separated flow like your case.