Discontinuity handling in fit model.

18 views
Skip to first unread message

Mike Vivaldi

unread,
Sep 14, 2025, 3:30:33 PM (8 days ago) Sep 14
to OpenVSP
Hi Rob,

I tried to use the fit model for the point cloud of my fuselage. The fit generated an odd discontinuity when using the general fuse cross section. Is there a way to prevent the pinching in the fuselage cross section?

Thanks,
Mike
 
Screenshot 2025-09-14 122949.png
Screenshot 2025-09-14 122620.png
Screenshot 2025-09-14 122514.png

Rob McDonald

unread,
Sep 14, 2025, 4:26:46 PM (8 days ago) Sep 14
to ope...@googlegroups.com
If I am not mistaken, that isn't a General Fuse XSec, that is the CEdit XSec, with a Bezier curve.

You'll notice that the lower tangent control points (green and yellow) for the left/right sides of the fuse are coincident with the curve control point (black).

I would move those down some (say y=-.3) and then remove the y coordinate of those control points from FitModel's control.  Then, when you re-optimize, it will find the best solution it can without moving those points up.

Rob


--
You received this message because you are subscribed to the Google Groups "OpenVSP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openvsp+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/openvsp/eebffead-759e-4ce9-80b5-2f4341e7985fn%40googlegroups.com.

Mike Vivaldi

unread,
Sep 14, 2025, 4:49:12 PM (8 days ago) Sep 14
to OpenVSP
Hi Rob,

I converted to a Cedit to show the points but nothing had been changed from the fit model as an illustrative change only. I will give your suggestion a try.

Thanks,
Mike

Rob McDonald

unread,
Sep 15, 2025, 12:05:20 AM (8 days ago) Sep 15
to OpenVSP
OK, in that case, LowStr is being driven to zero.  You can set it to something reasonable and then remove it from FitModel's control.

Or, you might try using the CEdit XSec which will give you more control.

Rob
Reply all
Reply to author
Forward
0 new messages