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/)
-   -   [ANSYS Meshing] Boolean creation error - operation would result in non-manifold bodies (http://www.cfd-online.com/Forums/ansys-meshing/111569-boolean-creation-error-operation-would-result-non-manifold-bodies.html)

thefinalyearkid January 10, 2013 03:34

Boolean creation error - operation would result in non-manifold bodies
 
Hello! So im currently working on a project that requires me to analyze the flow around a vehicle in this case - a renault clio

So i obtained this model and saved it as a parasoild and then imported it into ANSYS, then, i generated a enclosure to depict air as usual and went about creating a boolean to substract the vehicle from the air(enclosure)

This is where the error of 'non-manifold' bodies occurs, i have read some other posts on some inability to substract surfaces and im not too sure on how to go about resolving this issure!

Please help! :)http://img577.imageshack.us/img577/838/img1tx.jpg

Sixkillers January 10, 2013 04:53

Before subtract operation I would definitely unite the hole car into a one body.

diamondx January 10, 2013 10:35

Yes the problem is the surface/solid operation.
Designmodeler cannot perform an operation between solid and surface.
when you import the car, you have to fill the inside with a some solid. There is a feature called FILL in designmodeler that let's you do that.
After the fill is done, delete those surfaces as you don't need them, you can do that with body operation then delete.
Then make that rectangle and perform the Boolean feature.

A more easy way is to import your car as a solid and not as a surface, i guess this is done when exporting you car in the CAD software in which you made the car ? try an extension like IGES , or STEP...

diamondx January 10, 2013 10:50

also see this http://www.cfd-online.com/Forums/ans...29-bug-dm.html

thefinalyearkid January 10, 2013 20:58

Ah yes! Brilliant, so the open surfaces is what causes the problem in DM,

anyhow guys i was doing some research and found this software called blender, and the link shows some amazing features that actually tell you the location of non-manifold body errors and allows for you to very simply cover up the open surfaces


the youtube link is :
http://www.youtube.com/watch?v=H6NXhfw9itI

and the website link is :
http://www.blender.org/download/get-blender/

oewnce932c May 16, 2013 14:38

my solution
 
Hi thefinalyearkid.
i got the same problem. i imported ship hull from IGES(500+ surfaces), just half hull without deck. i wanted to creat a body from it. Done 2 surfaces: deck and mid-ship.
after that tryed with Body Operation: Sew this three objects: 2 created surfaces and surface-body(IGES(500+ surfaces))
with default tolerance was warning and errors:
"Warning: Sewing operation partial failure possibly due to incompatible attributes"
"Warning: not all bodies can be sewn or converted to solid body"
and by boolean: subtract
"Error: Operation would result in non-manifold bodies"

solution was to reduce the tolerance from default value(0.0015m) to 0.015m. after that it works fine!

saeed.marine April 24, 2014 10:42

non-manifold
 
hi.
i have an IGES file of a container ship (KCS-hull). but it is just the surface of it. how can i turn it into an solid body? in order to perform a Boolean operation on it.

oewnce932c April 24, 2014 10:55

load your ANSYS file. for example with this http://dropmefiles.com/en

your use DesignModeller?

saeed.marine April 24, 2014 11:02

it is already on the net.
http://www.simman2008.dk/KCS/kcs_geometry.htm
KSC hull
yes i use DM

oewnce932c April 24, 2014 11:42

1 Attachment(s)
here it is.
i took this file for ship hull http://www.simman2008.dk/PDF/KCS_hull_SVA.zip

couple of steps -> see screenshot

here is DM file for your with SOLID :cool: http://pjotr.ru/data/kcs_cfd_online.agdb

ps do your know any results or cases for KCS with self propulsion tests?!

saeed.marine April 24, 2014 11:59

i haven't worked on it myself. but surely some friends of mine know. they have done several test on it (in towing tank).
i believe you are using a later version of DM. i cannot open it!
mine is 14.0

saeed.marine April 24, 2014 12:08

can you give me an .stp file please?

oewnce932c April 25, 2014 04:40

1 Attachment(s)
oh.. yes my is 15.0. do your know how to save for "older" version?

here are steps, tree in DM is in attach
0. delete two surface in the bow, above main deck
1. create line and surface in prop shaft
2. create mid line of the ship z=main deck
3. create deck surface
4. create surface at y=0, have to select many lines
5. sew ship. Solid=yes. Change Tolerance to 0,1 or less if no solid is created.

here is stp-file http://pjotr.ru/data/kcs4cfd_online.stp

saeed.marine April 27, 2014 04:24

thank you. it was a great help to me.
what do you want to know about KCS hulls?

mjthomp3 June 19, 2014 11:18

Quote:

Originally Posted by thefinalyearkid (Post 401111)
Hello! So im currently working on a project that requires me to analyze the flow around a vehicle in this case - a renault clio

So i obtained this model and saved it as a parasoild and then imported it into ANSYS, then, i generated a enclosure to depict air as usual and went about creating a boolean to substract the vehicle from the air(enclosure)

This is where the error of 'non-manifold' bodies occurs, i have read some other posts on some inability to substract surfaces and im not too sure on how to go about resolving this issure!

Please help! :)http://img577.imageshack.us/img577/838/img1tx.jpg





There is a very useful tutorial from ANSYS website in design modeler that
goes through all these steps but for a Sedan. I found it very usefulas you
might as well.


The basic steps go like this:
1. connect all the edges
2. Tools > merge
3. Tools > Surface patch to fill gaps in surface body
4. Use sew to convert the surface body to a solid body
with Body Operation
and lastly
5. Enclosure with 1 surface plane


All times are GMT -4. The time now is 10:14.