CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (http://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [ICEM] Meshing wedge-type flow meter (http://www.cfd-online.com/Forums/ansys-meshing/72247-meshing-wedge-type-flow-meter.html)

negronson January 29, 2010 19:54

Meshing wedge-type flow meter
 
2 Attachment(s)
Hi guys I'm working with a wedge flow meter and i've been having a lot of trouble meshing it with structured hexa mesh in icem. I've maged to succesfully block the geometry and generating the basic block mesh, i'd like to have an ogrid in the inlet an outlet of the pipe and add inflation layers but when i do that the elements get really crappy in quality and even sometimes icem cant make the mesh. i would really appreciate suggestions on the blocking and ogrid subject or if making an unstructured mesh is a better way to go. (i have already made the simulation with tetra mesh but the number of elements to achieve results convergence is over 1,500,000 tetras and the solver time is kind of long considering that is a very simple geometry). I've attached photos of the geom and the blocking.

PSYMN February 22, 2010 20:49

No problem...
 
First, imagine how you would mesh a pipe…
Just a rectangular block with OGrid thru it (faces on the ends).
Next, imagine that there is a baffle at the midpoint of the pipe that passes 2/3rds of the way thru the pipe…
Split vertically to capture the baffle, split horizontally to capture the 2/3ds line.
Next we imagine that the baffle is really a very thin V cut into the pipe and we open it out just a little. How would we block that?
Spit vertically again very near the middle… Then delete the blocks that represent the cut away V and collapse the blocks directly underneath them (see the pipe blade tutorial for an example).
Then we just open up that V a bit until we get your model.
You already have nicely developed boundary layers along the wall, but if you also want them along the split, just 2 more vertical splits on either side, along with some nice edge parameters would take care of it…
I figure this could be done in 1 OGrid, followed by 1 horizontal split, 2 vertical splits, one collapse block and 2 more vertical splits… <5 minutes. Actually, you could do the OGrid before or after (more than one way to get it done)…
If you can give me the measurements of your model I would go thru it and put it on YouTube for you. (after I do the airfoil example that I am already behind on) We could even make that cut angle and distance into parameters in DM and setup some automatic updating.

negronson February 23, 2010 01:36

3 Attachment(s)
Hey PSYMN first of all thank you for your reply and concern... well since my post I managed to make the ogrid and the entire mesh by making the ogrid in the input and output face and it really came along well, with some tweaking in the position of some vertex and edges... though the mesh looked very well and had some good quality (over .4 determinant) when I was validating the mesh simulating some flow of water, I started to see that the differential pressure before and after the wedge was being really dependent to the number of elements taking me to some insane refinement and over 1,500,000 elements which my machine couldn't handle jeje, so it must be something with the angles and distortion of the elements that is making my solution really distant from what I expected in theory... well I'll attach the dimensions of the pipe (In millimeters, sorry for the crappy plane) and some photos of where I'm actually stuck and I really appreciate your help, In the meantime I'll be doing what you've told me and see how it develops....thanks again PSYMN

Attachment 2342

Attachment 2343

Attachment 2344


All times are GMT -4. The time now is 17:37.