surface quality check

[kshowdown]

Is there a way to check the quality of the surfaces in a model that is already a solid? I am having problems with moldflow imports of a solid pro/e model.


The moldflow design link module uses ptc's actual kernals while importing, but does not show what is the model defect is after trying to fix it upon import.


Basically the model needs to be "water tight" for moldflow,which is the same in pro/e inorder to solidify a model to a solid,

 

[Jacoolo]

well, if your model lacks violet or purple edges while displaying the geometry in wireframe mode, then it is solid for sure.

point is - and you are right - while exchanging the geometry to another software some gaps can appear due several reasons, mainly by the use of relative accuracy in the original model.

To check so called quality of the model, you can use Modelcheck. Of course it has to be set up right first, but it can give you wide feedback about potential problems.


The way my previous company handled the problem with exchange to catia was to create DMU(Digital Mock-up) by the use of Save as > Shrinkwarp > Solid.

The model crerated in this way was finaly chcecked by Model Check with extra setting regarding surface demands introduced by customer.

 

[krow72]

what I have done in the past is to try and create an stl file. If that fails, it will highlight the area that is failing and that area is usually the problem. Then you can focus on that area to fix.

 

[cpoirier]

Export the model to IGES and re-open in pro-e, repeat with STEP and then ACIS(this will be the worst). This will show you the ugly right away. Each format is a bit different and may be able or not able to make a solid, but I found doing this will eventually show you the problem area quite readily with minimum effort. I used to push models back a fourth to SW and Trace Pro all the time. Model accuracy, less aggressive geometry, or a fancy surfacing trick to patch over the bad geometry is the way to fix the problem, rounds and sharp angles between surfaces can be a pain. I have also found that SW2006 can have mysterious crashes and problems after it has made some good looking but strange surfaces. activating a cross-section through this geom is like lighting a fuse in SW. Converting to IGES and reimporting in SW helps find this as you get to see the real surface boundary and not the pretty graphics.

The point is Pro sometimes makes good geom only inside Pro.


Good luck,
Chris

 

[design-engine]

define quality.I often export into Alias for G2 checks.

 

[kshowdown]

Cpoirier......I also use the iges export and re-import method. It works pretty good for showing cracks etc.


The designers here use way to many rounds and it seems they are all over the model tree instead of at the bottom, but you can't tell them how to model things as they know it all!no one has been accountable for a crappy model so they don't know any better.


Modelcheck is a great tool to make users design to company standards and quality, but it has to be set up correct.


As far as quality I was looking for maybe a mesh or something of your model to show intersections or problem areas??


It looks like moldflow 2012 has new import upgrades that will convert poorly designed models now .....so maybe the problem designers are off the hook
Edited by: kshowdown

 

[cpoirier]

Hi Bart,
This was my observation. (at least what I was thinking when I wrote my post)
Mold flow is used for mold fill analysis. Underneath is an approximation made from a mesh of finite elements. MF can not automate the mesh generation or work out boundary conditions without knowing the solid volume. So in my interpretation of quality would say that the model needed to be water tight and the export/import trick will show where pro may be handling some issues internally in an advanced way that becomes a problem in a down stream application that is less advanced. (typical)

As for G2 check you got me on that one, mind as well tell me you checked the R2D2 and it beeped just fine. I am still learning about surfacing and curvature in fact I have a few posts I'm going to go read later tonight.


BTW I have read many of your posts, thanks for all of your efforts and contributions.

To anyone who reads this.Thanks for sharing both techniques and nuts and bolts of Pro

Regards,
Chris