[kds]

Hello
I was simulating a simple two pipes joined by 90 degree elbow where a volume flow rate of 15gpm was given. I tried to simulate by letting th inlet flow as mass flow rate but for this case the pressure drop i got from my calculation is way different than the CFD analysis. So I used velocity inlet using the average velocity i calculated analytically. For this one my CFD pressure drop became 103 pa while the analytical one is 150 Pa am just wondering if this difference is okay.
Thanks

[me3840]

For pipe flows you should be able to get a lot closer. What are the boundary conditions you are using?

[kds]

Originally Posted by me3840

For pipe flows you should be able to get a lot closer. What are the boundary conditions you are using?

I used inlet velocity of 0.46 m/s. For the flow model i used turbulent with k-epsilon. The outlet is pressure outlet.

[me3840]

0.46 m/s is pretty slow, are you sure the flow is turbulent? What does the mesh look like?

[kds]

Originally Posted by me3840

0.46 m/s is pretty slow, are you sure the flow is turbulent? What does the mesh look like?

The fluid type is water with a volumetric rate of 15 gpm so i changed this to mass flow rate which is 0.95 kg/s so i used the basic equation to calculate the velocity v= m/(A*rho) which gives a velocity of 0.468 m/s. I calculated the Re number using R = rho*V*D/dynamic viscosity which gives me about 26683. the diameter of the pipe is around 50.8 mm. I used polyhedral mesh.

[me3840]

Okay, the mesh is probably the problem here then. Can you post pictures of it?

[kds]

Originally Posted by me3840

Okay, the mesh is probably the problem here then. Can you post pictures of it?

I was able to figure out the problem. It was my mistake I used a different diameter when I draw the 3d model. Now it seems alright I got 156 pa on the cfd simulation. Which one do you think is the most trusted one with this aspect? The CFD solution or the analytical one?

[me3840]

That's not a question I can answer for you. You need to evaluate what assumptions your analytic solution has and how much these impact the real-world compared to the assumptions in your CFD model.

6 Pa doesn't seem to be much to fret over though.

[kds]

Originally Posted by me3840

That's not a question I can answer for you. You need to evaluate what assumptions your analytic solution has and how much these impact the real-world compared to the assumptions in your CFD model.

6 Pa doesn't seem to be much to fret over though.

Ok thank you!