CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > Siemens > STAR-CCM+

Drag Coefficient Flow over a Block

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   March 5, 2015, 13:59
Default
  #21
Senior Member
 
Matt
Join Date: Aug 2014
Posts: 947
Rep Power: 17
fluid23 is on a distinguished road
Validating experimental drag results is one of the most difficult things to do in CFD. For the most part, your biggest limitation in CFD is your turbulence model, the biggest limitation to the turbulence models are the near wall aproximations,and the near wall approximations are what drive your surface shear stress. If you need very accurate Cd numbers then you probably want to run LES or DNS.

The reason you aren't getting the correct Cd is probably that your wake isn't modeled well enough. Look at the image I posted in a previous reply comparing RANS and LES wakes. You should immediately understand the difference between the two. If you aren't aware, you are currently trying to do a Reynolds Averaged Navier-Stokes analysks (RANS). These typically use 2 equation turbulence models (k-e or k-w) althogh there are others too...

Wake has a huge effect on external aerodyanmics.

How is the drag coefficeint that you are going to extract applied to your research?
fluid23 is offline   Reply With Quote

Old   March 6, 2015, 12:21
Default
  #22
New Member
 
Erik
Join Date: May 2014
Posts: 23
Rep Power: 11
egwuab is on a distinguished road
I understand the limitation. I already refined my wake region on that model but the Cd value is still lower than the experimental result. I wont be able to use LES or DNS because it is very extensive and time consuming. Do you have any advice so I could get a decent Cd/forces result from RANS model?

My whole research is about vehicle and bridge interaction when induced by the wind. Getting the right forces on the vehicle and the bridge will be crucial for my research. Thats why I need to get approximately good forces distribution across vehicle and bridge.
egwuab is offline   Reply With Quote

Old   March 6, 2015, 12:49
Default
  #23
Senior Member
 
Matt
Join Date: Aug 2014
Posts: 947
Rep Power: 17
fluid23 is on a distinguished road
Sad to say, I don't think that you can avoid the LES or DNS analysis. If you are really that concerned about body forces and effect of wake being super accurate then you shouldn't really use RANS. If I were analyzing an aircraft in RANS I wouldn't expect to get accurate drag values, they would be close (within 10% or so) but never match wind tunnel data. You might get lucky sometimes, but that is usually the exception to the rule.

Perhaps someone with more experience can give you better advice on improving your RANS drag results. If this is for a thesis or technical paper you can probably live with whatever drag values you get from RANS and just use those in your analysis. The difference between 1.2 and 1.4 isn't really all that much if you are just trying to validate methods, models, etc... If you are relying on those values for design inputs into an actual functioning piece of equipment then I would invest the time in the LES runs.
fluid23 is offline   Reply With Quote

Old   March 6, 2015, 12:59
Default
  #24
New Member
 
Erik
Join Date: May 2014
Posts: 23
Rep Power: 11
egwuab is on a distinguished road
Thank you for clarifying that out. Im trying to validate our forces data from experimental to FSI simulation. Is there anybody has a good thumb rules to reduce this error for RANS model?
egwuab is offline   Reply With Quote

Old   November 2, 2016, 22:50
Default
  #25
New Member
 
Mason
Join Date: Sep 2014
Posts: 22
Rep Power: 11
Mason liu is on a distinguished road
Erik,


have you solved you problem? i.e. use RANS turbulence models to predict the drag force.

I also run into this issue when I want to get Cd on some simple profiles, like Square / rectangle / semi-cylinder. Following the best practice, I tried k-e / RNG k-e / SST models, but they give obvious different Cd values, on some profiles k-e give close Cd, but on some other profiles SST give the close Cd. I want to find a universal turbulence model to predict Cd on some other profiles(also simple profile), so I firstly want to validate on some classical profiles, that why I'm doing the jobs.

thanks a lot.
Mason liu is offline   Reply With Quote

Old   December 6, 2016, 12:54
Default
  #26
New Member
 
fei
Join Date: Oct 2016
Posts: 5
Rep Power: 9
egw.uab is on a distinguished road
Hi Mason, from my study, RANS model won't give good result for this type of problem. I am trying to learn to use DES and LES for now.
egw.uab is offline   Reply With Quote

Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Flow past a cylinder at Re 1e05 using LES, drag force coefficient is to low Scabbard Main CFD Forum 21 June 19, 2018 14:58
Drag Coefficient in flow over a cylinder 30na OpenFOAM Running, Solving & CFD 14 May 19, 2014 12:21
Drag coefficient of flow past cylinder vs time pedroxramos FLUENT 0 January 14, 2013 13:39
Non-steady flow simplified for use in Vissim steamerandy Main CFD Forum 0 October 31, 2011 22:08
Drag Coefficient of Flow across a Circular Cylinde Zhihua Li FLUENT 3 April 20, 2004 02:01


All times are GMT -4. The time now is 21:57.