Sweep failure

Next
 From:  medi (MALD)
3246.1 
Hi Michael.

At the moment I am modelling a Sherman tank in MOI. As part of the modelling I tried to model a protective pipe guard for part of the tank by sweeping a circular profile along a single curved rail. Unfortunately, the result is unsatisfactory. Sometimes a pipe is formed along only a quarter of the curve; at other times the whole pipe appears to be formed but sections of it will shoot off at a tangent to the curve. Have enclosed a test file. Any thoughts would be appreciated.

Regards
Mal
Attachments:

  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:  Ralf-S
3246.2 
Hi Mal,

My English is to poor, to describe it properly.
It works fine.
Please take a look:
(60 is only an example...)
Image Attachments:
Size: 400.1 KB, Downloaded: 45 times, Dimensions: 1031x905px
Size: 406.5 KB, Downloaded: 33 times, Dimensions: 1031x905px
Size: 402.9 KB, Downloaded: 75 times, Dimensions: 1031x905px
  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:  medi (MALD)
3246.3 
Thanks Ralf.

I tried rebuilding the curve as your images show but the same problem still occurred(tried 20, 50 and 60 points).
However I did note that the distorted pipe seemed to be mainly at the middle top of the curve and deleted the middle point there.
Tried again and it worked! Don't know why.

Anyway, thanks for your help.

I would still like to know why the original curve didn't work however.

Cheers
Mal
  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:  Ralf-S
3246.4 
Hi Mal,

If you want you can look into my "MOI V2.0 beta Dec. 2009" file.
Attachments:

  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:  Michael Gibson
3246.5 In reply to 3246.3 
Hi Mal,

> I would still like to know why the original curve didn't work however.

It's because the segments of the curve are not not quite smooth to one another, there is a small angle (about 8 degrees or so difference) between their ending tangents instead of having equal ending tangents.

On a rail curve that has sharp corners like that, the current v2 beta will attempt to do extensions and mitering of the sweeps in those areas, but that is not working completely properly yet for rails that have curved segments instead of straight line segments. That is pretty much the last remaining area I have left to fix up before v2 will be all completed.

If you align the 3 control points where segments meet, or delete the control point where 2 segments meet to fuse them into a single smooth segment, then that solves the problem.

Here's a zoom in to one of the segment join areas of your curve to illustrate how it is sharp there and not actually smooth:





- 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

Previous
 From:  medi (MALD)
3246.6 
Thanks for the explanation Michael and thanks for your help Ralf.

Cheers
Mal
  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