Learning Forum

[sacharin1993]

I have been simulating flow around a circular cylinder (in both 2D and 3D) using ANSYS Fluent, but I am consistently obtaining drag coefficient values (Cd) that are too low.

I have tried several approaches, including:
• Using a high-quality mesh
• Targeting both a y-plus value of between 0.3 or higher but less than 1
• Running both steady and transient simulations
• Using the k-omega SST turbulence model

I am trying to match experimental Cd values, which are typically around 1.2 for Reynolds numbers between 1×10^4 and 1×10^5. I was able to reproduce these results previously using OpenFOAM.

However, my Fluent simulations are giving me:
• Cd ≈ 0.9 for Re = 1×10^4
• Cd ≈ 0.7 for Re = 1×10^5

These values are significantly lower than expected.

Has anyone been able to obtain closer values and how?

Thanks

[Essence]

Hello,

Did you check the Reference Values?

https://ansyshelp.ansys.com/account/secured?returnurl=/Views/Secured/corp/v252/en/flu_ug/flu_ug_ReferenceValues.html?

[sacharin1993]

I am using these values as reference (as a student version - testing yet - user I don`t think I have access to the link you sent, let me know if I am wrong).

[abhishek.roy]

Hi, to calculate the Cd value, Fluent uses area defined under the reference value. Please refer to the link my colleague has shared. It is the link to the user guide and is available to all users

[sacharin1993]

I am calculating correctly the Cd, but the results do not agree with experimental (standard) data and I would like to improve the results to get closer to them.

 

[Essence]

Could you please share the screenshot of the reference values you are using?

[sacharin1993]

This screenshot shows the reference values used in the simulation.
The simulation concerns flow past a circular cylinder with a diameter of 0.1 m, placed inside a parallelogram-shaped enclosure. The width of both the enclosure and the cylinder domain is 0.2 m. The enclosure has a length of 4 m and a height of 2 m, and the cylinder is located 1 m downstream of the inlet.

The simulation was performed in 3D using both steady and transient approaches. The maximum y+y^+y+ value was 0.6, and the mesh consisted of approximately 10,000,000 cells. At Re=2×104Re = 2 \times 10^4Re=2×104, the resulting drag coefficient was $Cd=0.805$.

Additional mesh refinement did not change the results. I also increased the depth of the cylinder and enclosure from 0.2 m to 0.5 m, but no variation in the drag coefficient was observed.

[sacharin1993]