[ICEM] Blocking difficulties

CFD_SIM · October 22, 2012, 14:18

The simulation aborts after 20 iterations...at the beginning the convergence looks good, but after a while it starts to diverge.

I will look into the problem tomorrow, because i had some other things to do. Hopefully i will find a solution, or the blocking and the guideline of energy382 will solve my problem :-).

CFD_SIM · October 23, 2012, 02:48

Never mind...problem solved, but i can't delete my post^^.

Far · October 23, 2012, 03:22

How did you solve your problem? was it is error due to mesh or solution parameters? Did you try the energy382's method? I would be interested to see the meshing by this method, as mesh looks promising.

CFD_SIM · October 23, 2012, 03:38

Hey Far,

the problem that i was refering to was a problem with extending the faces. This one is solved :-).
My other mesh did not converge as good as i want it to be, so i decided to try energy382's method.
Iam not quite finished and have another appointment today, so i hope that i will be ready by the evening.

I will post a picture of the result and if you like the project-file when iam ready.

energy382 · October 23, 2012, 05:24

I've a customer meeting and will not be at offfice today. I'll make some pics tomorrow. I've to mesh my volute first, but I'll get back to you as soon as possible

You've to choose the outer faces of solid blocking, not the inner ones. (to extrude fixed distance)

CFD_SIM · October 23, 2012, 05:30

Hello Energy382,

the problem with the extended faces is solved :-). I just had to figure out which combination works well.

I do have a meeting soon too, so i will get to my blocking in the afternoon. Sadly i am stuck at the part with the "create blocking from faces". I adjusted some more splits, but i do not get a cut at the top. So i do not have a face at the top of the rip_outlet.

I will post a picture later on.

CFD_SIM · October 23, 2012, 10:08

There has been a change in my geometry, so i will not finish my blocking today...

I will upload my blocking later. Iam still stuck at the part were i connect the c-grid to my rib.

http://dl.dropbox.com/u/30926816/Pipe_New.zip

The new geometry will have a longer Outlet, and in the middle of the outlet will be a parabolicle body (axially symmetric) which goes down to nearly the ribs (not touching it!).

I will prepare the geoemtry today and hopefully will find a way to block it right tomorrow, but iam not sure that this will be so easy. I will try to upload it tomorrow so you can look at it.

CFD_SIM · October 25, 2012, 15:19

Sorry for the delay, but there are some major problems with my geometry. As soon as they are fixed i will give you guys an update on how the progress with the blocking is.

kind regards

John

Far · October 25, 2012, 15:36

Your geomtry isn't complex!!!

CFD_SIM · October 25, 2012, 15:47

Quote:

Originally Posted by Far

Your geomtry isn't complex!!!

I actually can't remember when i said something like this.
The major problem has nothing to do with the geometry itself :-). Sorry if it sounds like this.

Far · October 25, 2012, 16:04

So it is confirmed that angle less than 20 does not work for CFX?

CFD_SIM · October 25, 2012, 16:16

My problem has something to to with the additional part, which was missing in the beginning. It hasn't been manufactured yet, nor has it been drawn. So i have to wait some time, till i get the rest of the necessary geometry.

The angle problem hasn't been solved yet, because of the missing part. But i read a lot of literature about the problem with the angle and i think i will wait and look how my solver will solve the mesh.
What i can say is, that cfx does solve meshes with a angle quality under 20° (My worst was down to 9°). But i guess it depends a lot on what kind of geometry iam using.

I will be happy to give some more informations on this topic, as soon as i have an working mesh and some results :-).

energy382 · October 26, 2012, 07:11

I've attached blocking (right_side). try the left side on your own!

after blocking, you've to move some vertex and set edge paramters etc. to improve mesh quality

right_side_pipe.zip

CFD_SIM · October 26, 2012, 07:57

Thanks a lot! I've done the blocking and did the association, but the mesh still looks kind of funny (See attachment). I'm not sure how to make it "smooth", but i think that the problem has something to do with the associations.

energy382 · October 26, 2012, 08:22

Quote:

Originally Posted by CFD_SIM

Thanks a lot! I've done the blocking and did the association, but the mesh still looks kind of funny (See attachment). I'm not sure how to make it "smooth", but i think that the problem has something to do with the associations.

is that your blocking or mine? looks like there are still some missing blocks or there are wrong associations. click "edge" "show association" to check.

energy382 · October 26, 2012, 08:22

is that your blocking or mine? looks like there are still some missing blocks or there are wrong associations. click "edge" "show association" to check.

upload the current blocking

CFD_SIM · October 26, 2012, 08:29

Sorry, but the project is to big to upload it directly on this forum. I can send you later a link via dropbox if you need it.

Here are some more pictures which might give you a clue.

I took your blocking and tried to block the other side like you did. It looks all right i think, so i guess it has something to do with the associations.

thanks

John

energy382 · October 26, 2012, 08:38

Quote:

Originally Posted by CFD_SIM

Sorry, but the project is to big to upload it directly on this forum. I can send you later a link via dropbox if you need it.

Here are some more pictures which might give you a clue.

I took your blocking and tried to block the other side like you did. It looks all right i think, so i guess it has something to do with the associations.

thanks

John

you should enable all surfaces

you could also delete SOLID blocking

CFD_SIM · October 26, 2012, 08:51

Hmmm...the surfaces were enabled. After i deleted the solid part the meshing did not really improve

.

http://www.file-upload.net/download-...pe1.1.zip.html

I hope you can open it.

energy382 · October 26, 2012, 09:17

Quote:

Originally Posted by CFD_SIM

Hmmm...the surfaces were enabled. After i deleted the solid part the meshing did not really improve

.

http://www.file-upload.net/download-...pe1.1.zip.html

I hope you can open it.

use another file hoster. this one doesn't work....at least to me

October 22, 2012, 14:18		#21
CFD_SIM Member Join Date: Oct 2012 Posts: 46 Rep Power: 13	The simulation aborts after 20 iterations...at the beginning the convergence looks good, but after a while it starts to diverge. I will look into the problem tomorrow, because i had some other things to do. Hopefully i will find a solution, or the blocking and the guideline of energy382 will solve my problem :-). Last edited by CFD_SIM; October 22, 2012 at 14:33.

October 23, 2012, 05:30		#26
CFD_SIM Member Join Date: Oct 2012 Posts: 46 Rep Power: 13	Hello Energy382, the problem with the extended faces is solved :-). I just had to figure out which combination works well. I do have a meeting soon too, so i will get to my blocking in the afternoon. Sadly i am stuck at the part with the "create blocking from faces". I adjusted some more splits, but i do not get a cut at the top. So i do not have a face at the top of the rip_outlet. I will post a picture later on. Last edited by CFD_SIM; October 23, 2012 at 09:09.

October 23, 2012, 10:08		#27
CFD_SIM Member Join Date: Oct 2012 Posts: 46 Rep Power: 13	There has been a change in my geometry, so i will not finish my blocking today... I will upload my blocking later. Iam still stuck at the part were i connect the c-grid to my rib. http://dl.dropbox.com/u/30926816/Pipe_New.zip The new geometry will have a longer Outlet, and in the middle of the outlet will be a parabolicle body (axially symmetric) which goes down to nearly the ribs (not touching it!). I will prepare the geoemtry today and hopefully will find a way to block it right tomorrow, but iam not sure that this will be so easy. I will try to upload it tomorrow so you can look at it. Last edited by CFD_SIM; October 23, 2012 at 12:05.

October 26, 2012, 07:11		#33
energy382 Senior Member Christoph Join Date: May 2011 Location: Germany Posts: 182 Rep Power: 17	I've attached blocking (right_side). try the left side on your own! after blocking, you've to move some vertex and set edge paramters etc. to improve mesh quality right_side_pipe.zip Far likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[ICEM] Blocking and Symmetry	BrolY	ANSYS Meshing & Geometry	32	August 24, 2012 03:13
[ICEM] Some hints required on blocking geometry	air_engineer_arsenal	ANSYS Meshing & Geometry	6	August 16, 2012 11:58
[ICEM] Using Blocking with ICEM (infinite blocking lines appeared)	Touré	ANSYS Meshing & Geometry	2	August 5, 2012 17:27
[ICEM] Blocking For Swept Wing	air_engineer_arsenal	ANSYS Meshing & Geometry	13	July 24, 2012 11:52
[ICEM] Blocking strategy	BrolY	ANSYS Meshing & Geometry	0	July 22, 2010 04:46

October 23, 2012, 02:48		#22
CFD_SIM Member Join Date: Oct 2012 Posts: 46 Rep Power: 13	Never mind...problem solved, but i can't delete my post^^.

October 23, 2012, 03:22		#23
Far Super Moderator Sijal Join Date: Mar 2009 Location: Islamabad Posts: 4,553 Blog Entries: 6 Rep Power: 54	How did you solve your problem? was it is error due to mesh or solution parameters? Did you try the energy382's method? I would be interested to see the meshing by this method, as mesh looks promising.

October 23, 2012, 03:38		#24
CFD_SIM Member Join Date: Oct 2012 Posts: 46 Rep Power: 13	Hey Far, the problem that i was refering to was a problem with extending the faces. This one is solved :-). My other mesh did not converge as good as i want it to be, so i decided to try energy382's method. Iam not quite finished and have another appointment today, so i hope that i will be ready by the evening. I will post a picture of the result and if you like the project-file when iam ready.

October 23, 2012, 05:24		#25
energy382 Senior Member Christoph Join Date: May 2011 Location: Germany Posts: 182 Rep Power: 17	I've a customer meeting and will not be at offfice today. I'll make some pics tomorrow. I've to mesh my volute first, but I'll get back to you as soon as possible You've to choose the outer faces of solid blocking, not the inner ones. (to extrude fixed distance)

October 25, 2012, 15:19		#28
CFD_SIM Member Join Date: Oct 2012 Posts: 46 Rep Power: 13	Sorry for the delay, but there are some major problems with my geometry. As soon as they are fixed i will give you guys an update on how the progress with the blocking is. kind regards John

October 25, 2012, 15:36		#29
Far Super Moderator Sijal Join Date: Mar 2009 Location: Islamabad Posts: 4,553 Blog Entries: 6 Rep Power: 54	Your geomtry isn't complex!!!

October 25, 2012, 16:04		#31
Far Super Moderator Sijal Join Date: Mar 2009 Location: Islamabad Posts: 4,553 Blog Entries: 6 Rep Power: 54	So it is confirmed that angle less than 20 does not work for CFX?

October 25, 2012, 16:16		#32
CFD_SIM Member Join Date: Oct 2012 Posts: 46 Rep Power: 13	My problem has something to to with the additional part, which was missing in the beginning. It hasn't been manufactured yet, nor has it been drawn. So i have to wait some time, till i get the rest of the necessary geometry. The angle problem hasn't been solved yet, because of the missing part. But i read a lot of literature about the problem with the angle and i think i will wait and look how my solver will solve the mesh. What i can say is, that cfx does solve meshes with a angle quality under 20° (My worst was down to 9°). But i guess it depends a lot on what kind of geometry iam using. I will be happy to give some more informations on this topic, as soon as i have an working mesh and some results :-).

October 26, 2012, 08:22		#36
energy382 Senior Member Christoph Join Date: May 2011 Location: Germany Posts: 182 Rep Power: 17	is that your blocking or mine? looks like there are still some missing blocks or there are wrong associations. click "edge" "show association" to check. upload the current blocking

October 26, 2012, 08:51		#39
CFD_SIM Member Join Date: Oct 2012 Posts: 46 Rep Power: 13	Hmmm...the surfaces were enabled. After i deleted the solid part the meshing did not really improve . http://www.file-upload.net/download-...pe1.1.zip.html I hope you can open it.