Distorted Loft

Hello, I was hoping someone might be able to say why my loft gets distorted from the profiles used.
Each profile was created sequentially in the same manner, e.g. same number of control points, same order, etc.

I even tried the same using a 2 rail sweep and get the same result.

It just seems odd since the shape of the profile is only incrementally different than the one before and the one after - No differences larger than any of the other profiles elsewhere along the sequence of profiles that get connected as expected.

Distorted_Loft.3dm (4.6 MB)

Hello- try:

Select all but one curve and Rebuild, using the last curve as the ‘master curve’

Then loft.

Hi Pascal,
Thanks for the suggestion.
However by doing a Rebuild it changes the shape of the profile curves.
Those profile curves(The edges of) are serving as references to other geometries.(So the profiles need to be keep as is)
These profiles are being imported as igs files from another program and I need to have the ‘surfaces’ created to run through the profiles(original).
Maybe there’s something about the loft command and the 2 rail sweep command that I don’t know which can perhaps prepare me to construct the profiles differently.
It just seems odd how on just the one profile the both commands just start creating differently.

Hello- in this case as far as I can see rebuilding to a master curve results in no loss off precision.
Minimum deviation = 0
Maximum deviation = 9.46493e-13


Actually it’s quit a bit.
Measuring at a point between the old and new profile curve I can find a difference of 0.02mm
Also at a position chosen on the old then new curve - a tangency extended results in a very big difference.

Yeah… the one curve seems to be and odd ball somehow - - try exploding and joining all the curves at once, then Loft.

@BabaJ if I use the odd curve as the master curve to rebuild to, I cannot, so far, find any curves that are not pretty much identical with the inputs.


Thanks Pascal…by working with your suggestions I’ve realized what I need to do to avoid these ‘issues’.

By exploding and then re-joining is the ‘solution’. And yes, the profiles don’t change their shape from original either.

In doing so I noticed that each profile is broken up into 5 segments.
When I imported the geometry each profile was in 4 segments each. Then in Rhino I joined them, then lofted which led to the problems.

So I think the issue is Rhino needs a curve ‘composed’(joined) in a certain way otherwise it ‘looks’ at the curve differently depending on how it was joined - Or something like that.

In the program where I created the profiles, I tried joining each profile first then exported them for Rhino.
So in Rhino when I imported the profiles, instead of doing a join then loft I only had to do a loft.
And it resulted in the same expected(desired) result as if I imported the separate 4 curve segments, joined, exploded, re-join then loft.

So the solution for the least amount of work is to just create the one segment closed profile before exporting - but if in a ‘pinch’ I can just join, explode then re-join.

Thanks again.