I have performed some flow tests on multiple identical machined plenum parts to determine the pressure drop across these plenums. The lab test results all seem to be aligned with no anomalies. As you can see in the photo below, the part has an inlet and outlet hose connected at either end, with sensors further up each hose measuring the inlet and outlet pressure, with the pressure drop being the difference between the two values. The test was ran at 1, 2 & 3 l/min.

I decided to recreate this test on Solidworks Flow Simulation, using the original model. The following conditions were applied:

• Internal analysis
• Exclude cavities without conditions
• No temperature or gravity
• Liquid is 60:40 water glycol solution
• Flow type Laminar and Turbulent
• Adiabatic wall
• Inlet is fully developed flow 1-3 l/min, the other inlet settings left as default (k, e etc)
• Outlet is environmental pressure
• Global mesh is level 4 with advanced channel refinement at minimum gap size 2mm, mesh independence was observed at circa 300k cells, with the help of adaptive meshing.
• Plenty of global goals set as seen in the screenshot

Results

The CFD results at 1 l/min were very close to the lab test, however as I increase the flow rate, the results start to deviate more. Its not possible to test the parts any higher than 3 l/min in the lab, so these are the only results I can achieve.

Is there any reason as to why the results increasingly deviate? Is there anything I can change in the setup of my simulation to improve the accuracy?