Shelling oddity...

Next
 From:  mcramblet
5650.1 
I had an interesting experience with shelling, that I thought I'd share and see if anyone could enlighten me. I'm working on creating what will be a thermoformed clamshell. Yesterday I was working on the attached piece, I selected several edges and applied a fillet, and immediately thought that I'd better see if it could be shelled, I had a concern because of the relatively complicated surfaces . I applied the shell with an amount of .015", and it shelled without any problems what so ever, which surprised me a little. I opened the file this morning to continue work on it and it wasn't shelled, I assumed that I had forgotten to save the file after I had shelled it. I selected the back surface to shell it again, and this time it won't do it. I'm a little stumped by it working fine yesterday but not today. Any thoughts or reasons on this?

Thank-you,
Michael Cramblet
Packaging Design
Phone: 616-574-6271
 
Attachments:

Image Attachments:
Size: 62.6 KB, Downloaded: 14 times, Dimensions: 456x483px
Size: 67.6 KB, Downloaded: 14 times, Dimensions: 342x522px
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  mcramblet
5650.2 In reply to 5650.1 
One item to note; the first time it was shelled, I used the direction "normal", and that's what I tried to use again, earlier. I tried shelling using the "flip" direction and that mode worked on the object, see attached image. I'm just not clear as to why one method would work one day and the following it doesn't. Please note: I'm fully aware that this may 100% operator error, based on my limited (and still learning) knowledge of MoI. If this is the case, I want to understand what I'm doing wrong so I can work out my bugs with the way that I'm using and building it the application.
Michael Cramblet
Packaging Design
Phone: 616-574-6271
 
Image Attachments:
Size: 106 KB, Downloaded: 7 times, Dimensions: 1223x808px
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
 From:  Michael Gibson
5650.3 In reply to 5650.1 
Hi Michael, when MoI saves out to 3DM it involves a type of translation from the geometry library's own native data structures into the 3DM ones. This can potentially change a few kinds of things like the order that things might be listed. Under normal circumstances these things are not really supposed to make any difference but it is possible for that to have some impact on particularly sensitive operations.

One thing you might try in general is to not put on the smaller fillets until after you've done the shell, especially the corner juncture areas can be particularly difficult for the shelling mechanism to handle, it has to try and construct offsets of all those pieces.

It's unfortunately quite difficult to give you a specific reason why it worked for you before and not now - most likely there is some kind of bug involved but it will be pretty difficult to figure out exactly what.

It does not look to me like you've done anything wrong at all, other than maybe if you know you want to shell don't put on the smaller fillets until after the shell instead of having them before the shell. Although just to make life interesting there are also other particular situations where putting on a fillet before can actually help, if it helps to avoid the kind of offset situation where one vertex in the base surface would need to diverge into more than one in the generated offset, that's the particular kind of situation that the offset mechanism in the geometry library does not handle currently.

Basically shelling is not a very strong area of the geometry library that MoI uses, it can be good to have ViaCAD to use along with MoI to do some kinds of shells that MoI can't figure out.

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged
 

Reply to All Reply to All