I was trying to simulate a hydraulic system, however, the simulation result is always 30% higher than testing result. I'm guessing there might be something wrong with my fluid property setup.

I did simulate and test just a constant diameter tube. The results still don't match.

I'm using "pure substance" option with the correct density and dynamic viscosity at the testing temperature. I left the molar mass the same as water (18.02 kg kmol^-1). Is there anything else I should do?

Did you run your tube as Laminar flow or turbulent?

No, I have not. I probably should. However, I did run the same tube with the same physics, except with water. The results were correct even with SST turbulence model. I'm sure if it was laminar with oil, it won't have much turbulance with water.

The viscosity of water is different, so your Reynolds number will be different and hence the effect of turbulence as well. If you are comparing to laminar flow data, you should turn off the turbulence model (i.e. run laminar).

-CycLone

I did run with laminar. The result was closer to reality. However, I did notice that with a inlet Mass Flow Rate setup, the flow took over 2ft to develop. The velocity profile over the 2ft long pipe was changing all the way through. That did not make sense.

Here's some detail of the setup: Volume flow rate: 12 gpm pipe id: .484 in pipe length: 2ft.

I try to run it with the viscosity of Mobile dte 24 which has a dynamic viscosity of 31 centipoise @100°F

Lab test showed 3.7 psi Pressure drop. CFX with laminar method has a result of 6.9 psi

Try setting the inlet as a constant total pressure and and the outlet as a mass flow rate. A mass flow at the inlet will have a constant velocity profile with a variable total pressure, which is not what you would have in your experiment.

-CycLone