Distorted Elements and Mesh Error

[AndyWalker]

Hello everyone,

Hope you can help.

I have made a Volumetric model in Revit and exported to Robot via .sat file.

When i run the calculation i get some problems.

One error is "Distorted Elements" what have i done wrong?

Also get some funny meshing problems. See screenshot attached.

Can i manually delete these stray lines and if so how?

Also what has caused it in the first place?

Many Thanks

Andy

[Rafal.Gaweda]

Send us sat file and robot file

[AndyWalker]

Thanks Rafal,

See link to download files:

https://www.yousendit.com/download/TEhWMFhqb0IwZ21KUmNUQw

Andy

[AndyWalker]

Rafal,

Did you manage to get those files i sent?

Andy

[Rafal.Gaweda]

Yes I am looking at them but meshing takes time.

[AndyWalker]

Spoiler

sorry, i understand this, i just wondered if you were able to download the files, (not finished with a solution)

but i see that you did.

thank you for all your help

[Rafal.Gaweda]

I will wait a little bit more for meshing but I am afraid I have to recommend to use Autodesk Mechanical Simulation for this task.

http://usa.autodesk.com/adsk/servlet/pc/index?siteID=123112&id=13773836

[AndyWalker]

Spoiler

Rafal, thanks for your response.

Playing around with the meshing myself I have managed to get rid of the "distorted elements" error message and reduce the amount of stray mesh down a little in ROBOT (although not completely removed).

I was wondering, what is the theory behind why this happens? How can I reduce it occurring?

I see you recommend another product, this is not really an option for us at this stage im afraid with regards to cost and time constraints.

With the model I have, we managed to actually get it to run but we are are getting some funny results that we would not expect, we also ran a simple "test" load on the model that is easily hand checked and still get something unexpected.

Would you be able to take a look if I send you this model and explanation?

many thanks

andy

[Rafal.Gaweda]

Andy

Playing around with the meshing myself I have managed to get rid of the "distorted elements" error message and reduce the amount of stray mesh down a little in ROBOT (although not completely removed).

This is kind of workaround but be carefull about making "holes" in object \ surface.

I tried to mesh it with mesh size 0.04m.....

I was wondering, what is the theory behind why this happens? How can I reduce it occurring?

Old meshing engine

I see you recommend another product, this is not really an option for us at this stage im afraid with regards to cost and time constraints.

I can imagine. 

With the model I have, we managed to actually get it to run but we are are getting some funny results that we would not expect, we also ran a simple "test" load on the model that is easily hand checked and still get something unexpected.

Would you be able to take a look if I send you this model and explanation?

 OK, send me model.

[AndyWalker]

Rafal

Please see Robot file here:

https://www.yousendit.com/download/TEhYTmZRUzg4aU94djlVag

As a "test" case for the model, we have applied a 100kN/m uniform load to nodes (0.13,0.16,-0.3) and (0.28,0.16,-0.31) in the form of a surface trapezoidal load i.e. p1 = 100kN/m and p2 = 100kN/m.  The length between the nodes is 2.8m thus we would expect the sum of the reactions in the Fx direction to be 280kN.

Looking at the reactions for "test" this figure is 270.73kN  ???

When we consider the actual REAL load case "dlc....spline 1", the difference in output reaction from Robot to that of what we expect is more severe.  For this case we applied the loading in the same way as described above however p1= -9397.224kN/m and p2 = 9144.613kN/m.  The length is the same as above thus we would expect the sum of the reactions in Fx to be -353.66kN.

We actually get -94.67kN which is a VERY big difference.

What is your expert opinion on the above, have we missed something?

[Rafal.Gaweda]

In a minute I will send you model and explanation.

[Rafal.Gaweda · Accepted solution]

Nodes 4272 4273 should be corrected by Detailed Correct to Line created by nodes  4272 4273 .

Corrected model: http://wtrns.fr/EUYoXQ5P3XsYCiX

[AndyWalker]

That is it.

Thank you very much.

So was this error created due to the meshing creating two nodes that were not exactly on the line?

And it can be corrected using the "Detailed Structure Correction"?

What alerted you to this error or was it just experience?

Andy

[Rafal.Gaweda]

So was this error created due to the meshing creating two nodes that were not exactly on the line?

Most likely.

And it can be corrected using the "Detailed Structure Correction"?

As I said, yes. 

What alerted you to this error or was it just experience?

Experience  :-)!