CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > CFX

SST turbulence model

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

Reply
 
LinkBack Thread Tools Display Modes
Old   May 6, 2010, 08:54
Default
  #41
Senior Member
 
Attesz's Avatar
 
Attesz
Join Date: Mar 2009
Posts: 355
Rep Power: 8
Attesz is an unknown quantity at this point
Quote:
i think the y+-values calculated in CFX do not only depend on number and height of prism layers but also velocities in impeller and thickness of boundary layer. So i don`t know how you got the value of y+>20 with 5 prism layers?
you should increase the first cell height for example 2 times. I think thats all...

yes you are right, the "real" surge is a totally transient phenomenon. But I found in every paper, that the operational point is easily attainable with steady simulations. My compressor works at 70000RPM and PR1.65, but we reach "surge" at PR1.4. This "surge" means, that there is a very low mass flow in and out of the domain. The separation mentioned above closes the whole blade passage.
Attesz is offline   Reply With Quote

Old   May 6, 2010, 08:59
Default
  #42
Member
 
Susann
Join Date: Apr 2009
Location: Dresden
Posts: 33
Rep Power: 8
Suzzn is on a distinguished road
And how big is the distance between instability/surge-limit and this operational point? Do you have measurements? Often big separations already occur at the design point of a compressor, so that there are strong transient effects even at design point.
Suzzn is offline   Reply With Quote

Old   May 6, 2010, 09:05
Default
  #43
Senior Member
 
Attesz's Avatar
 
Attesz
Join Date: Mar 2009
Posts: 355
Rep Power: 8
Attesz is an unknown quantity at this point
Quote:
Often big separations already occur at the design point of a compressor, so that there are strong transient effects even at design point.
Yes, its a good idea! I'm preparing transient simulations present...

I have measurements only on operating point, because this is a gas turbine engine. We have not enough equipment to measure the compressor stage separately.

Thanks for your advices!
Attesz is offline   Reply With Quote

Old   May 12, 2010, 04:10
Default
  #44
Senior Member
 
Attesz's Avatar
 
Attesz
Join Date: Mar 2009
Posts: 355
Rep Power: 8
Attesz is an unknown quantity at this point
Hi,

can I use Frozen Rotor fluid interface, when my geometry is periodic, and the Pitch Ratio isnt equals 1? There is nothing about it in the help. I have a 60deg impeller, and a 72deg stator, so the Pitch Ratio is about 1.2. I want to take into account the blade interferences...and I think, I cannot use TRS only when I have the same periodicity angles..

Regards,
Attesz
Attesz is offline   Reply With Quote

Old   May 12, 2010, 09:51
Default
  #45
Member
 
Susann
Join Date: Apr 2009
Location: Dresden
Posts: 33
Rep Power: 8
Suzzn is on a distinguished road
Hey,

when you use Frozen-Rotor-Model and the pitch changes, the fluxes are scaled by the pitch change.
Suzzn is offline   Reply With Quote

Old   May 13, 2010, 05:53
Default
  #46
Senior Member
 
Attesz's Avatar
 
Attesz
Join Date: Mar 2009
Posts: 355
Rep Power: 8
Attesz is an unknown quantity at this point
Hi Susann,

I know that, but can it cause high numerical errors? In some papers, I read that in Transient Rotor Stator, or Transient Frozen Rotor simulations, I can use only Pitch Ratio 1. Of course, with Stage option, it is not a problem. What do you think?

Thanks,
Attila
Attesz is offline   Reply With Quote

Old   May 15, 2010, 09:44
Default
  #47
Member
 
Susann
Join Date: Apr 2009
Location: Dresden
Posts: 33
Rep Power: 8
Suzzn is on a distinguished road
Could you maybe give me a link to this papers? I`ve never heard about this before...I would follow the recommendation of CFX-help and I found nothing about a pitch change of 1 for frozen rotor or transient simulations...would be very bad because there is seldom a pitch change of 1 for a compressor with vaned diffusor. Just give it a try with the frozen rotor-model, its even faster than stage model...than you will see whats better in your case after comparing it to the experimental results...
Suzzn is offline   Reply With Quote

Old   May 15, 2010, 10:07
Default
  #48
Senior Member
 
Attesz's Avatar
 
Attesz
Join Date: Mar 2009
Posts: 355
Rep Power: 8
Attesz is an unknown quantity at this point
http://www.ansys.com/events/proceedi...PAPERS/252.pdf

There is really nothing in help about this, but in this paper, it is described.

Have a nice weekend!
Attesz is offline   Reply With Quote

Old   May 17, 2010, 03:43
Default
  #49
Member
 
Susann
Join Date: Apr 2009
Location: Dresden
Posts: 33
Rep Power: 8
Suzzn is on a distinguished road
I know this paper...but i find the same description of the interface-models as in CFX-help...i cannot find something about pitch change 1...sorry but can you give me a hit where in the text you find that?
Suzzn is offline   Reply With Quote

Old   May 17, 2010, 05:01
Default
  #50
Senior Member
 
Attesz's Avatar
 
Attesz
Join Date: Mar 2009
Posts: 355
Rep Power: 8
Attesz is an unknown quantity at this point
Hi Susann,

Quote:
First the geometry of computational section
must be adjusted to minimize the area difference at the interface between the impeller and the diffuser.
Because there are 31 impeller passages and 22 diffuser passages the single passage computational geometry
used in the steady state simulations had an area difference of more than 40%. This means that the area of
the diffuser at the impeller-diffuser interface is 40% greater than the area of the impeller. In the unsteady
simulation this is not acceptable because of the significant circumferential variations in the flow at the
interface. The ideal setup would be to simulate the entire compressor, but this was judged to be too costly
for the computational resources available. Instead the computational domain was defined to include three
impeller and two diffuser passages which resulted in an area difference of 6.45%. The CFX solver stretches
the solution at the interface during the calculations to remove the area difference and allow the simulation
to converge, but this obviously introduces some modelling errors.
Here is what I mean. You are allowed to use not only Pitch 1, but here, they speak about a 6.25% area change which means a pitch change very close to 1, and they mentioned that it can cause errors. It is valid for Frozen Rotor.

So I correct myself, you can use not only pitch ratio 1, but using frozen rotor, it can cause as big numerical error as big the area difference is.
Attesz is offline   Reply With Quote

Old   March 17, 2014, 01:17
Default Turbulence model in Rotor 37
  #51
New Member
 
Manpreet
Join Date: Jan 2014
Posts: 14
Rep Power: 3
manpreet is on a distinguished road
Hello Guys,
I am working on project Flow field analysis through rotor 37. Cd anyone please let me know about which turbulence model is better to get results and why ?
I really appreciate .
Thanks
Manpreet Singh
manpreet_singh_er@yahoo.co.in
manpreet is offline   Reply With Quote

Old   March 21, 2014, 04:45
Default
  #52
swm
New Member
 
Song Weimin
Join Date: Dec 2013
Posts: 7
Rep Power: 3
swm is on a distinguished road
Quote:
Originally Posted by ghorrocks View Post
For a steady run, once the residuals are converging nicely you should increase the physical time scale to quite large compared to the fluid timescales. This should still converge quickly, but should allow the flow to "average" as best as possible.
what u mean by "increase"is continuing to calculate the original results by larger time step?
swm is offline   Reply With Quote

Old   March 23, 2014, 16:26
Default
  #53
New Member
 
Manpreet
Join Date: Jan 2014
Posts: 14
Rep Power: 3
manpreet is on a distinguished road
Hello Guys,
Cd anyone please let me know detail about Timescale factor and Convergence RMS. What's significance of these two? In addition, Which solver method is responsible for solving N-stokes equations in CFX. I really appreciate . Thanks
With regard
Manpreet Singh
manpreet_singh_er@yahoo.co.in
manpreet is offline   Reply With Quote

Old   March 23, 2014, 17:49
Default
  #54
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 10,929
Rep Power: 85
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
Time scale factor is used by the solver to march towards a solution. It is part of the numerical method. The residual is a measure of the accuracy of the solution of the equations.
ghorrocks is offline   Reply With Quote

Old   March 24, 2014, 00:06
Default
  #55
New Member
 
Manpreet
Join Date: Jan 2014
Posts: 14
Rep Power: 3
manpreet is on a distinguished road
Thanks..ghorrocks.
Cd u please tell me in detail which numerical method CFX used for soling N-S equations such as time marching or anything else. Where can I find detail about this.
Thanks
Manpreet Singh
manpreet is offline   Reply With Quote

Old   March 24, 2014, 00:07
Default
  #56
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 10,929
Rep Power: 85
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
It is all in the documentation - see the theory manual.
ghorrocks is offline   Reply With Quote

Reply

Thread Tools
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 On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Huge Heat Transfer Coefficient with SST turbulence model andoss CFX 4 January 9, 2010 08:40
Use of k omega turbulence model john_w OpenFOAM Running, Solving & CFD 2 September 22, 2009 05:15
SST transitional turbulence model Ben Akih CFX 3 June 8, 2006 15:52
Sinclair Model + secondary turbulence Yi FLUENT 0 October 26, 2001 13:37


All times are GMT -4. The time now is 17:40.