CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > ANSYS Meshing & Geometry

[ICEM] Smooth Transition for Tet Prism meshing

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

Like Tree2Likes

Reply
 
LinkBack Thread Tools Display Modes
Old   April 4, 2012, 05:10
Default Smooth Transition for Tet Prism meshing
  #1
Senior Member
 
---------
Join Date: Oct 2010
Posts: 302
Rep Power: 8
saisanthoshm88 is on a distinguished road
In Ansys WorkBench, I find the option "smooth transition" very useful for generating Inflation / Prism layers as it helps in maintaing a good volume transition between the last layer prisms and the tetras.

Could some one please let me know if there is a analogus option for prism meshing in ICEM CFD.
__________________
Best regards,
Santhosh.
saisanthoshm88 is offline   Reply With Quote

Old   April 4, 2012, 07:34
Default
  #2
Senior Member
 
PSYMN's Avatar
 
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,662
Blog Entries: 1
Rep Power: 35
PSYMN has a spectacular aura aboutPSYMN has a spectacular aura about
Sure... The tech actually comes from ICEM CFD first

In ICEM CFD global prism settings, you can set 3 of the 4 prism settings. You can set initial height, number of layers, ratio and/or total height. Setting three gives enough to calculate the 4th.

However, if you leave initial height and total height as 0, a nonsensical number, you have left Prism with enough freedom to float the initial height based on the first cell size. The initial height is adjusted at each column, with respect to the number of layers and ratio, to result in a last prism height that is similar in volume to its adjacent tetra...

If you use this, you actually need to put in the number of layers and ratio that you want. You can't pull the "1 layer and split" trick.

In older version (I think before R13 or so), if you set the initial height on any part or entity, it would stop this from working. But more recent versions are more flexible and allow you to have a global float and specific heights on only certain parts or surfaces.
rikio likes this.
__________________
-----------------------------------------
Please help guide development at ANSYS by filling in these surveys

Public ANSYS ICEM CFD Users Survey

This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)...

CFD Online Users Survey
PSYMN is offline   Reply With Quote

Old   April 4, 2012, 11:36
Default
  #3
Senior Member
 
---------
Join Date: Oct 2010
Posts: 302
Rep Power: 8
saisanthoshm88 is on a distinguished road
Many thanks for the clarification Simon !
__________________
Best regards,
Santhosh.
saisanthoshm88 is offline   Reply With Quote

Old   April 5, 2012, 10:37
Default
  #4
Senior Member
 
Join Date: Dec 2009
Posts: 129
Rep Power: 10
mjgraf is on a distinguished road
icem cfd, i have been doing inflation with the automatic spacing for about half the number the layers I wanted to allow it to mesh faster, split the prism, and then redistribute based on first cell height I desire. so far this is more robust than setting the initial cell height and num layers for some of the more complex geometries we deal with.
mjgraf is offline   Reply With Quote

Old   April 5, 2012, 11:31
Default
  #5
Senior Member
 
PSYMN's Avatar
 
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,662
Blog Entries: 1
Rep Power: 35
PSYMN has a spectacular aura aboutPSYMN has a spectacular aura about
@mjgraf,

Yea sure, that is a good way to do it. Just be clear that the automatic height distribution tool doesn't know you are going to do that and will size the initial height based on the given number of layers and ratio...

The volume transition is still proportional to the base triangle size and you could still get a pretty good result if you start with half the layers and split each in half.

Other users may be trying this with a single layer that they plan to split... It probably won't give very good results in that case. The prism layer will be very thin to keep the volume transition down, and then the users would split that into even thinner slices... The end result is a larger than idea volume transition between the top prism and the adjacent tetra...
__________________
-----------------------------------------
Please help guide development at ANSYS by filling in these surveys

Public ANSYS ICEM CFD Users Survey

This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)...

CFD Online Users Survey
PSYMN is offline   Reply With Quote

Old   April 5, 2012, 12:03
Default
  #6
Senior Member
 
Join Date: Aug 2011
Posts: 114
Rep Power: 6
scipy is on a distinguished road
Send a message via Skype™ to scipy
Sorry to intrude on the topic, but I'd just like to have something clarified in regard to ICEM inflation.

Is "First Aspect Ratio" option that's available in ANSYS Meshing (and for which I thought it was TGrid pre-inflation algorithm) the same as setting the Height Ratio to 5 in ICEM CFD? If not, is there any other way to achieve F.A.R of 5 for boundary layer inflation in ICEM?
scipy is offline   Reply With Quote

Old   May 9, 2012, 17:20
Default
  #7
Senior Member
 
PSYMN's Avatar
 
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,662
Blog Entries: 1
Rep Power: 35
PSYMN has a spectacular aura aboutPSYMN has a spectacular aura about
@ Scipy,

Yes, you can set the Prism height Limit factor to limit the ratio of prism layer height to base... But then every element in the column would be affected.

When using ICEM CFD, if you want that "smooth volume transition" to the tetras, just don't set the initial height. Set only the number of layers to 5 and the growth ratio to 1.2 and let it run. It will adjust the initial height for each prism column such that the volume of the last prism is roughly equal to the volume of the adjacent tetra... This has been the way with ICEM CFD for over a decade and would probably give you something close to the 5 to 1 height ratio that Marco Lanfrit recommends for Fluent.
__________________
-----------------------------------------
Please help guide development at ANSYS by filling in these surveys

Public ANSYS ICEM CFD Users Survey

This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)...

CFD Online Users Survey
PSYMN is offline   Reply With Quote

Old   June 17, 2012, 01:14
Default
  #8
Senior Member
 
---------
Join Date: Oct 2010
Posts: 302
Rep Power: 8
saisanthoshm88 is on a distinguished road
@ Simon : Hi Simon,

Could you please let me know how to deal with the tet prism
transition in case we intend a particular first layer height.
This may be required to maintain the y+ in a desired range.

Also could you please let me know if the mesh metric " volume change " actually accounts for the volume change between the pentas and adjacent tetras in a tet prism mesh

Your suggestions were really help full thanks for the tips.
__________________
Best regards,
Santhosh.
saisanthoshm88 is offline   Reply With Quote

Old   June 17, 2012, 06:54
Default
  #9
Senior Member
 
Join Date: Mar 2011
Location: Germany
Posts: 414
Rep Power: 11
cfd seeker is on a distinguished road
hi saisanthoshm88
Simon has clearly expalined your question in his posts and I am repeating for your clarification. In Global Mesh setup set only no. of layers(set at-least 3 or 4 prism layers, don't set just a single layer it causes problem in smooth transition) and growth factor and leave initial height and total height as "zero". This allowS initial height to "float" and total height will be adjusted such that you will get smooth transition b/w last prism and adjacent tetra. After the initial mesh split all or your desired prism layers based on the initial height and recompute only the prim mesh and lastly redistirbute prism layers again based on the initial height and you are done

As far as your question regarding "Penta" is concerned.... Prisms are Penta elements
cfd seeker is offline   Reply With Quote

Old   June 17, 2012, 11:52
Default
  #10
Senior Member
 
---------
Join Date: Oct 2010
Posts: 302
Rep Power: 8
saisanthoshm88 is on a distinguished road
Hi,

Thanks for the reply , but I don't think that the smooth transition attained by floating the initial height will still be retained after the redistribution. The redistribution affects the transition.

I know that pentas are prisms my question was about the quality metric " volume change " . I was asking if this parameter " volume change " also accounts for the volume change between the prisms and their adjacent tetra elements (or) is it like it is calculated only between different prisms and different tetra elements and not actually between tetras and prisms.
__________________
Best regards,
Santhosh.
saisanthoshm88 is offline   Reply With Quote

Old   June 18, 2012, 04:39
Default
  #11
Senior Member
 
Join Date: Mar 2011
Location: Germany
Posts: 414
Rep Power: 11
cfd seeker is on a distinguished road
Smooth transition? smooth transition between prism layers after redistribution or smooth transition from last prim to adjacent tetra? what exactly you are talking about?
cfd seeker is offline   Reply With Quote

Old   June 18, 2012, 11:12
Default
  #12
Senior Member
 
---------
Join Date: Oct 2010
Posts: 302
Rep Power: 8
saisanthoshm88 is on a distinguished road
I'm refering to the smooth transition from last prim to adjacent tetra after the redistribution.

Please correct me if I'm wrong with the notion that this transition between last prism and adjacent tetra would change with the redistribution

Thanking you for the help !
__________________
Best regards,
Santhosh.
saisanthoshm88 is offline   Reply With Quote

Old   June 18, 2012, 14:48
Default
  #13
Senior Member
 
Join Date: Mar 2011
Location: Germany
Posts: 414
Rep Power: 11
cfd seeker is on a distinguished road
yes it does change with redistribution but still it gives acceptably smooth transition, I wonder where you are facing problem?
cfd seeker is offline   Reply With Quote

Old   June 26, 2012, 10:49
Default
  #14
Senior Member
 
PSYMN's Avatar
 
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,662
Blog Entries: 1
Rep Power: 35
PSYMN has a spectacular aura aboutPSYMN has a spectacular aura about
The algorithm simply adjusts the initial height (during the initial prism inflation) in hopes that, that when combined with the given number of layers and growth ratio, the last prism volume roughly equivalent to the adjacent tetra. This is done right off the start with the intial layer, so the only information it has is the size of the base triangle... It does not adjust for curvature or tetra growth away from the wall, etc...

Another approach is to set the prism height limit factor to 0.7... This will use what ever initial height you set, but limit the growth ratio to 1.0 when the prism height over base is approximately 0.7, which is roughly the volume of a tetra with the same size base.

Either way, if you come back later and try to split and/or redistribute prisms, the total height is fixed. It must calculate a new initial height based on your number of layers and growth ratio, or alternatively, a new growth ratio based on your initial height and number of layers... Either way, you are likely to adjust the volume of the last prism and therefore change the ratio with its neighboring tetra... It is just the price you pay for trying this short cut.

I think CFD Seeker's final point is that this difference may still be quite acceptable (worth it for the short cut). Have you actually had a problem with the solver based on the volume jump or is your concern just academic at this point?
__________________
-----------------------------------------
Please help guide development at ANSYS by filling in these surveys

Public ANSYS ICEM CFD Users Survey

This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)...

CFD Online Users Survey
PSYMN is offline   Reply With Quote

Old   June 29, 2012, 05:40
Default
  #15
Member
 
Lorenzo Mazzei
Join Date: Dec 2010
Posts: 34
Rep Power: 6
Mazze[ITA] is on a distinguished road
Hi Simon,

is there the possibility to split the prism layer in ANSYS Meshing?
Mazze[ITA] is offline   Reply With Quote

Old   June 29, 2012, 11:14
Default
  #16
Senior Member
 
PSYMN's Avatar
 
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,662
Blog Entries: 1
Rep Power: 35
PSYMN has a spectacular aura aboutPSYMN has a spectacular aura about
No, neither split nor redistribute prism is available yet in ANSYS Meshing.

Eventually? Maybe. These sorts of automatic options are relatively easy to add (compared to interactive mesh editing).

However, you can take a mesh from ANSYS Meshing (or any mesher) and run this operation in ICEM CFD (using version 14.0 and later).
Mazze[ITA] likes this.
__________________
-----------------------------------------
Please help guide development at ANSYS by filling in these surveys

Public ANSYS ICEM CFD Users Survey

This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)...

CFD Online Users Survey
PSYMN is offline   Reply With Quote

Old   June 29, 2012, 11:18
Default
  #17
Member
 
Lorenzo Mazzei
Join Date: Dec 2010
Posts: 34
Rep Power: 6
Mazze[ITA] is on a distinguished road
Good suggestion, thank you Simon!
Mazze[ITA] is offline   Reply With Quote

Old   February 17, 2013, 22:26
Default
  #18
New Member
 
Join Date: Jan 2013
Posts: 7
Rep Power: 4
jjz2013 is on a distinguished road
This is a nice discussion.
I recently have a question. As the instruction in icem help and suggestions from the aforementioned, I firstly created 3 or 4 prism layers without setting values for initial height and total height. Then in order to satify y+ requirement, I need split the first and second layers into several very thin layers. However, even giving a value 0.001m for the initial height in prism split tab, the finally first layer height is still approximately 0.002m. Note that my first layer before splitting is about 0.005m. It seems that icem prevents the mesher from creating a layer with 0.001m height. I supposed to reduce the triangulation tolerance value in model settings from default 0.001 to 0.0001. But it was still not helpful. Do you guys have any idea of solving this problem. Thanks!
jjz2013 is offline   Reply With Quote

Old   February 18, 2013, 21:59
Default
  #19
Senior Member
 
PSYMN's Avatar
 
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,662
Blog Entries: 1
Rep Power: 35
PSYMN has a spectacular aura aboutPSYMN has a spectacular aura about
@jjz I will need to test and see if I can reproduce this when I have time.
__________________
-----------------------------------------
Please help guide development at ANSYS by filling in these surveys

Public ANSYS ICEM CFD Users Survey

This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)...

CFD Online Users Survey
PSYMN is offline   Reply With Quote

Old   February 20, 2013, 17:23
Default
  #20
New Member
 
Join Date: Jan 2013
Posts: 7
Rep Power: 4
jjz2013 is on a distinguished road
Thank you for your reply. In order to not waste your time, I have to tell you that I fixed this problem. I just simply reduced the triangulation tolerance to 1e-6 and set the initial height as 0.0005m when splitting the first initial layer. Although the final first layer thickness is not exact as the setting, it does decrease less than 0.001m. The difference maybe due to the inaccuracy of the measurement tools. Thanks for your attention.
jjz2013 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
[ANSYS Meshing] Prism meshing mahesh_1402 ANSYS Meshing & Geometry 0 January 30, 2012 01:10
ICEM Prism Layer transition between surface with prism layers and one without TWaung ANSYS Meshing & Geometry 2 October 12, 2009 14:56
prism meshing??? Nitin Dewangan CFX 1 August 1, 2008 18:35
ICEM - problems with prism mesh João Lourenço CFX 1 November 6, 2007 15:41
ICEM 10 smooth transition error Stephen CFX 3 March 13, 2007 10:23


All times are GMT -4. The time now is 12:06.