[sgaggar]

Hello everyone!

I am a relative novice in turbulence modelling. I have made and simulated the flow of air in a pipe with a 90-degree bend using a k-omega SST model on ANSYS Fluent Academic. Through this, it is easy to visualize the formation of Dean's flows at the outlet of the pipe section.

Now, I want to divide the total pressure drop and total friction losses into its components coming from the primary and secondary flow. I looked at the literature but until now, I have been unable to find some CFD analyses or even some theoretical models that might allow me to do this. So I needed your expert guidance on the topic. Specifically, I would like to ask the following:

1. Is there a way to segregate the losses into its components directly or by running a user-defined script in any commercial code (I am working right now in ANSYS Fluent Academic but I would be switching to Star-CCM+ soon)?
2. If not, is there a way to calculate just the pressure drop of the primary flow? I was considering somehow evaluating the Darcy Weisbach equation along the length of the pipe to find the effect due to the primary flow, and then subtracting this from the total pressure loss across the pipe to find the effect of the secondary flow. Is this approach possible (and correct)? Would I be missing something if I took this route? Are there any other possibilities?

I would greatly appreciate if you could help me with the same and please do let me know if I was not clear enough or can provide some additional information.

Thank you for your time!

Kind Regards!

[numore]

Maybe you can give more information why you want two values. An idea that comes to my mind would be to make a second simulation with slip-walls (zero wall friction) as comparison. I would be interested in the results.

Karman-Prandtl friction factor for straight pipes is not valid for curved conduits. For large curvature ratios there is a friction factor by Ito: Ito, H. Flow in curved pipes. Jap. Sch. Mech. Eng. Int. J., 1987, 30, 543–552.

[LuckyTran]

Usually we simulate a case without the secondary flows by inserting some slip walls to break them up and then compare the pressure drop.



You could integrate the wall shear stress along the pipe axis but this is not that trivial. You'd have use the components of the wall shear stress and then project it onto the local coordinate system and local surface normals of each cell.


I don't think you'll find any answers over there but a similar issue occurs in trying to break down contributions of form drag and pressure drag.

[sgaggar]

Originally Posted by numore

Maybe you can give more information why you want two values. An idea that comes to my mind would be to make a second simulation with slip-walls (zero wall friction) as comparison. I would be interested in the results.

This is an interesting approach, thank you! The broader goal is to be able to conduct this analysis for any conduit section with a non-circular cross-section. Through this, we want to understand separately the contributions coming due to the primary flows and the secondary flows within the pipe.

I will definitely try this and post the results here.

Originally Posted by LuckyTran

Usually we simulate a case without the secondary flows by inserting some slip walls to break them up and then compare the pressure drop.

You could integrate the wall shear stress along the pipe axis but this is not that trivial. You'd have use the components of the wall shear stress and then project it onto the local coordinate system and local surface normals of each cell.

This is very interesting as well, thank you for your response! I am not quite sure how feasible it is to add a slip wall at each site of secondary flow creation but I will definitely give it a whirl. I would also try to read more about the second approach you mentioned. Would it be possible for you to give me any references for the same?

Thank you both again!