Discussion Groups

Simulation CFD

Reply
Member
Posts: 4
Registered: ‎01-11-2013

Reynolds number and drag force slightly high

283 Views, 4 Replies
01-11-2013 12:44 PM

I am just beginning to use the software and conducting some sample similations. I have a 3D model of a  3.94 in diamater sphere in a square column of water with a horizontal velocity of 39.37 in/sec.  I would like to chek the software output for the  Reynolds number and drag force.  Using water density of 9.33204e-5 lb-s^2/in^4 and Dynamic Viscosity of 1.45342e-7 lb-s/in^2, I calculate the Re to be 99,597.2 and drag force to be 0.3742 lbs.  The wall results and summary file show drag force to be 0.637 lbs and Re to be 159876.

 

Why are the simulated results slighty higher?   

 

Thanks

Please use plain text.
Product Support
wildej
Posts: 643
Registered: ‎08-25-2011

Re: Reynolds number and drag force slightly high

01-14-2013 01:29 AM in reply to: jcwldy

Hi jcwldy,

 

These analyses are typically highly mesh dependant, have you tried refining it and investigating the impact on the results?

Also what Advection scheme are you using? Try 5.



Jon Wilde
Please use plain text.
Member
Posts: 4
Registered: ‎01-11-2013

Re: Reynolds number and drag force slightly high

01-14-2013 12:29 PM in reply to: jcwldy

Yes, I have ran many simulations with different mesh sizes. 

 

I am using the wall calculator to view the drag forces.  Is this the correct mathod to view the drag forces? 

 

The drag forces should be dependent on the density, fluid velocity, projected sphere area, and the drag coefficient.  Is there a way to output or view a drag coefficient?

 

I tried simulations using a ADV scheme of 5 and the forces are still about double of what I expect to see.

 

Thanks

Please use plain text.
Product Support
Royce_adsk
Posts: 512
Registered: ‎08-24-2011

Re: Reynolds number and drag force slightly high

01-16-2013 07:00 AM in reply to: jcwldy

Hi jcwldy,

 

Could you share the support share file?

 

You may also want to consider using a development flag, resid_bdry_force_calc, which is being used to test a different way to calculate the wall forces.  I have seen improvements with this flag.  You will need to use ADV5 in conjunction with that flag to get the new outputs.  The flag needs to be set at the scenario level and the default level.

 

Have you tried adaptation?

What Re number are you using from CFD?

It could be that the k-e/RNG turbulence models are difficient in this simulation.

 

-Royce



Royce Abel
Please use plain text.
Member
Posts: 4
Registered: ‎01-11-2013

Re: Reynolds number and drag force slightly high

01-16-2013 09:22 AM in reply to: jcwldy

Royce,

 

Thanks for your reply.

 

I have never implemented a development flag, but I will begin to research how.  What will the development flag do for the analysis.  Can you provide more information on this.

 

I will conduct a few runs using the adaption feature you suggested.

 

I am not using a Reynolds number from the simulation, I am calculating the Re using the fluid density, fluid velocity, sphere diameter, and dynamic viscosity, and using that to find the drag coefficient and calculate the drag force.  The Reynolds number calculated from the analysis is higher.  It is using the area of the continuum and the flow rate from the fluid velocity to calculate the Re. 

 

How can I solve this?

 

I would be happy to share the support share file, but when I try to upload the file, it errors out because the file is too large.  Contact me and I will link it to you.

 

Thank you again for your reply.

Please use plain text.