[hitre]

Hello. New here and also to CFD. I created symmetrical 3d geometry with cylindrical fluid domain and with half of disc inside. I'm trying to measure Cd with laminar flow. during calculation I got periodical residuals and also Cd. I tried with increasing boundary layers and thickness on the front side of disc but still it oscillates. What else can I do? I started calculation with Standard pressure equation and first order upwind momentum. Reference values are ok. But how can I improve mesh more?

[FMDenaro]

What about the Re number? depending on the value you can get laminar but oscillating solution.

[hitre]

Originally Posted by FMDenaro

What about the Re number? depending on the value you can get laminar but oscillating solution.

I have specified RE values to observe CD

[FMDenaro]

Is the solution steady for Re<10?

[hitre]

Originally Posted by FMDenaro

Is the solution steady for Re<10?

yep solution is converged. But i have experimental data with the lowest value of Re = 100. (i need to compare with experimental data)

[FMDenaro]

I strongly suspect that at Re=100 you do not have a physical steady solution

[rajnarayang]

Originally Posted by hitre

I have specified RE values to observe CD

What is the Re for the case that is giving you oscillating solution??

Sent from my Nexus 5 using CFD Online Forum mobile app

[rajnarayang]

Originally Posted by FMDenaro

Is the solution steady for Re<10?

Prof. Denaro,

Why did you specifically ask if Re below 10? Is Re<10 some how critical??

Sent from my Nexus 5 using CFD Online Forum mobile app

[FMDenaro]

I can only assume that the flow around a bluff body is steady for Re=O(1)

[hitre]

Originally Posted by rajnarayang

What is the Re for the case that is giving you oscillating solution??

Sent from my Nexus 5 using CFD Online Forum mobile app

I used Re around 400

[rajnarayang]

Originally Posted by FMDenaro

I can only assume that the flow around a bluff body is steady for Re=O(1)

The critical Re for von Karman vortex street is close to 300. Hence, wouldn't it be safe to assume that the flow around any bluff body is "steady" till that Re value?

Sent from my Nexus 5 using CFD Online Forum mobile app

[rajnarayang]

Originally Posted by hitre

I used Re around 400

Critical Re for von Karman vortex street is close to 300 (as seen in An Album of fluid motion). Your Re maybe in the transition domain.

Sent from my Nexus 5 using CFD Online Forum mobile app

[rajnarayang]

A snapshot from the book

[FMDenaro]

Originally Posted by rajnarayang

The critical Re for von Karman vortex street is close to 300. Hence, wouldn't it be safe to assume that the flow around any bluff body is "steady" till that Re value?

Sent from my Nexus 5 using CFD Online Forum mobile app

Cylinder is not a bluff body

[hitre]

I just played around (wasnt aware about critical reynolds at that time) and reduce Reynolds on 100. It's ok. So the problem obviously was that I without knowing random hit critical Re number

[Lucas_Gasparino]

At certain Re, the stabilized solution really is periodical (I'm assuming your run is transient, otherwise the calculation residuals do not converge well). What you can do is time average the Cd solution, which will give you something very close to experimental data, itself a time average of the experiment. Also, if the solution is periodic, I'd recommend using a full model. Try making the mesh as structures as possible close to the body, and do some refinement studies to evaluate mesh independence. Do you have any reference to base your results?

Sent from my XT1635-02 using CFD Online Forum mobile app