I am wondering. Shouldn’t the isocurves in between the two profiles be the same size while they sweep through the rial ?
Since I am measuring the in between sections and they are different from the ‘caps’.
Thanks a lot
That might be the case in certain special situations, but it is certainly not guaranteed. --Mitch
The surface is a sweep from the vertical line to the horizontal one.
Shouldn’t be …the same line lenght flowing along the sweep ? I mean in order to be a
proper transition. I understand that the structure should change isocurve lenght If I was
forcing to re- shape into a different thing.In this case …is the same profile and a rial.
Maybe I am not understanding some theory.
I would expect that the lengths would be the same (within tolerance) IF the two rails are exactly the same length and distance from each other and IF the start/end profiles (lines) are both exactly perpendicular to the ends of both curves…
–Mitch
There are nor ‘two’ It is one rail in the middle.
Ahh, OK - in that case, if the start/end lines are the same length, the extracted isocurves should also be the same… (within tolerance)
–Mitch
Could it be the rail is not coming out normal to the cross sections?
Here you can find the file______> https://we.tl/AEIZIKpIR8
Thanks a lot for your help.
As you can see, its an arc, two stright lines… mmm I am wondering what is going wrong ?
I assumed in this case (as per my post above) that both profiles are the same length and in the same orientation to the rail curve. It might also depend on the orientation chosen in the dialog, if it’s other than “Freeform”, I haven’t tested all.
–Mitch
No need to use WeTransfer, just drag and drop your file from its directory into your open post window or use the upload button.
You can see clearly with this print screen.
The surface highlighted is not the same size as the array of profiles…made with the FIRST and END stright lines.
Only the first and last profile are same size as the end and start of the sweeped surface.
Thanks!
Yes, now I’m surprised, I would have thought this would be consistent… but it’s not. In your simple case, as the center section is tilted at 45°, the length is reduced to 0.7071 (one over the square root of two) of the original… Hmm, I don’t know how to guarantee a constant width in this case… 
–Mitch
Me too, I’ll see what I can find out…
https://mcneel.myjetbrains.com/youtrack/issue/RH-42958
-Pascal
In your case, if you are using this with FlowAlongSurface to distribute your profiles, you might need your target surface to look like the attached… (I can describe how I made it later).
FlowSrf.3dm (115.2 KB)
–Mitch
Someone replied it at your post …that ‘its the way it works’’ . I am still confused why…
Rhino, unlike ICEM, Alias or NX, has no option in the surface building tool to ensure equidistance in a case such as yours. The outer boundaries of your surface are generated without you having control over them. C’est la vie.
because mathematically it is impossible. the isocurve in the middle is shorter because your two sweep profiles are the two diagonals of a quadratic geometry in profile view.
watch the video 
1 Like
That is the way sweep1 works. Its not the way everything works.
You could probably use a normal loft to get what you want
SimpleX.3dm (41.6 KB)