Discussion Groups

Robot Structural Analysis

Reply
Active Contributor
kope
Posts: 49
Registered: ‎10-27-2010
Accepted Solution

Meshing problem

885 Views, 29 Replies
10-11-2012 09:24 AM

 

it seems that no matter what I try (various meshng methods, size etc.) I cannot get a proper mesh for the walls of the attached model.

 

Moreover If I particularly select one of the walls with meshing issues and then delete its mesh, then I get other walls mesh also deleted... very awkard... 

 

I already spent a lot of time trying to figure it out, so I though sharing this issue here would be helpful.

 

thanks for any input.

 

I'm using Robot 2013 under windows XP

 

Mesh Problem.jpg

Please use plain text.
Product Support
Artur.Kosakowski
Posts: 4,849
Registered: ‎12-17-2010

Re: Meshing problem

10-12-2012 12:09 AM in reply to: kope

I used the repair mode to deal with the meshing issue (please check the Job Preferences settings in the file I'm attaching) and corrected inaccuracies in position of the nodes (e.g. 150 and 68 among others).

 

 

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.



Artur Kosakowski
Please use plain text.
Product Support
Rafal.Gaweda
Posts: 5,189
Registered: ‎04-26-2010
Please use plain text.
Active Contributor
kope
Posts: 49
Registered: ‎10-27-2010

Re: Meshing problem

10-12-2012 03:16 AM in reply to: Artur.Kosakowski

 

 


Artur.Kosakowski wrote:

I used the repair mode to deal with the meshing issue (please check the Job Preferences settings in the file I'm attaching) and corrected inaccuracies in position of the nodes (e.g. 150 and 68 among others).

 



Thanks!! :smileyhappy:

 

Didn't cross my mind that the repair mode would solve it. The help is quite silent about "repair mode" features.

Could you please tell me what exactly the repair mode did?

 

As far as the nodal positions are concerned, you have to do it manually right? I mean... inspecting each node one by one... which is a tedious task.:smileysad: I guess since this model uses inclined axes any auto rounding up would potentially move some nodes away from adjoining elements. For instance, using the app that Rafal suggested, after rounding up, node 68 is moved away from the supporting beam, leaving the two conccurent braces not connected to the adjacent beam. That's why you did it manually, right?

 

thanks for your input.

 

Please use plain text.
Active Contributor
kope
Posts: 49
Registered: ‎10-27-2010

Re: Meshing problem

10-12-2012 03:22 AM in reply to: Rafal.Gaweda

RG_Adsk wrote:

Hint: http://screencast.com/t/imIRxs0dl

 

http://forums.autodesk.com/t5/Autodesk-Robot-Structural/Rounding-up-coordinates-API/m-p/3430533/high...



thanks! :smileyhappy:

Very useful, but not for this case...:smileysad:

The rounding up moves away nodes from adjoining elements.

Please use plain text.
Product Support
Artur.Kosakowski
Posts: 4,849
Registered: ‎12-17-2010

Re: Meshing problem

10-12-2012 03:30 AM in reply to: kope

 

Didn't cross my mind that the repair mode would solve it. The help is quite silent about "repair mode" features.

Could you please tell me what exactly the repair mode did?

 

Checked and corrected the internal consistence of the model.

 

As far as the nodal positions are concerned, you have to do it manually right?

 

Apart from defining them correctly when you are creating the model you can also use the detailed correct option and adjust them to indicated (e.g. skewed) plane. Actually before posting the corrected model I checked the deformations and noticed the problems. Having to deal with only few nodes I moved them manually. Rounding up coordinates for planes that are not parallel to the global coordinate system axes may not solve all the problems but will many of them anyway.


If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.



Artur Kosakowski
Please use plain text.
Product Support
Rafal.Gaweda
Posts: 5,189
Registered: ‎04-26-2010

Re: Meshing problem

10-15-2012 03:48 AM in reply to: kope
Very useful, but not for this case...:smileysad:

 

Just to show you the problem.
Correct only Z by this macro, then remesh it

 



Rafal Gaweda
Please use plain text.
Active Contributor
kope
Posts: 49
Registered: ‎10-27-2010

Re: Meshing problem

10-15-2012 05:11 AM in reply to: Rafal.Gaweda

RG_Adsk wrote:
Just to show you the problem.
Correct only Z by this macro, then remesh it

 


ok. I see... thanks again! :smileyhappy:

I tried previously to adjust the Z using the "Detailed correct" function in order to push the nodes to plane Z=4.2m. Many of them were correctly adjusted but some of them were left with innacurate 4th digit (like 4.2003). I have no such problem using your tool.

Please use plain text.
Product Support
Artur.Kosakowski
Posts: 4,849
Registered: ‎12-17-2010

Re: Meshing problem

10-15-2012 05:23 AM in reply to: kope

When you used the detailed correction you most likely clicked on 3 points to indicate Z=4.2 plane having the coordinate display precision set to either 0.12 or 0.123. In this way you were not able to see what plane exactly you defined. To make sure it is 4.2 edit manually the coordinates for the indicated points e.g.

 

Point 1: x, y, 4.20 (and not displayed 003)

after edition

Point 1:  x,y, 4.2 (delete the last displayed 0 to get rid of unwanted part).



Artur Kosakowski
Please use plain text.
Active Contributor
kope
Posts: 49
Registered: ‎10-27-2010

Re: Meshing problem

10-15-2012 06:28 AM in reply to: Artur.Kosakowski

Artur.Kosakowski wrote:

To make sure it is 4.2 edit manually the coordinates for the indicated points

 


Since you told me that the position of some nodes are not accurate, I increased the display precision up to 4 digits which I guess should be enough.

Initially I did the way you expected i.e. i selected 3 nodes and it was much worse. A lot of nodes were still out of 4.2 plane. Then, I enetered manually the coordinates of 3 non-collinear points at 4.2m elevation. and I got better results but still as I said it didn't fixed all of them.

 

Anyway, I have to be much much more careful developing the model as the object snap is very tricky and it seems that you may end up quickly with misplaced nodes. When creating the objects I rely a lot on local systems and maybe this was the source of error... But, I don't know how else I could have done it otherwise...

 

Thanks again! :smileyhappy:

 

 After fixing the position of all nodes the model is working well now.

Please use plain text.