Hi all
I have a problem with tangency constraint for a smoothed volume. It's not new to me and I've always managed to get around it but with SW 2018 I stumble!
This is what it is all about: when creating a volume by smoothing, it is possible to constrain the tangency of the volume to the guide curve and in particular by imposing a normal on the sketch plane of these curves.
In the attached example, I have three closed sections and three guide curves. Each section arrives perpendicular to the front plane. In this front plane I have drawn three guide curves. I want my volume to arrive perpendicular to the front plane, so I chose the "normal to plane" tangency option. But this option is refused on the grounds of ambiguity...
Do any of you have an explanation? Thank you and have a good end of the day
probleme_tangence_lissage.pdf
Hello Pierre,
Look in the menu "Constraints of Departure/Arrival" maybe you have Tangent to the profile activated
Hello Pierre,
No, I don't have anything activated for the start/arrival constraints 
With SW2015 I just had the error message but the tangency was usually realized.
If the problem is encountered in volume, in surface with the same curves and profiles it works very well ....
Good evening Pierre :-)
Can you post your file (I'm in 2019 but I can read the 2018) that I'm looking at a bit.
Kind regards
Hello Zozo,
Here is a file reproducing my problem. I want the yellow side of the smoothed volume to arrive on the red side according to its normal, but the "normal to profile" condition for the guide curves is refused. However, the two end profiles (semicircle and half ellipse) normally arrive at the red face. In the surface, with a smoothed yellow surface, it goes by itself....
Thank you for your opinion
pb_lissage_tangent.sldprt
Hello Pierre
I don't see anything abnormal in your last file compared to the smoothing function with only a starting and an ending sketch.
If you want to have a normal with a red face between the two ends, you have to make other intermediate sketches (with the desired profile (like the one at both ends). So two or three depending on the surface tension .
look at the attached image :-)
What you need to understand about the smoothing function is that it means smoothing BETWEEN two (or more) profiles and one or more guide curves.
After SMOOTHING means that it will soften, smooth as if a fabric (or latex) were stretched between the two profiles.
So if you want less taut shapes, you have no choice but to put intermediate underwires as in an igloo tent or a taut curtain (the profile sketches then acting as an underwire).
I think it should not be confused with the surface because it doesn't seem to me to be the same logic. That's the power of having different tools for the maximum possibilities:-) :-) After you have to move on to NURBS it's starting to look like SW 2019
Kind regards
lissage_entre_differents_profils_-_pompe_peristaltique.jpg
1 Like
Oh Oh Nurbs with SW 2019, that's going to be interesting.
In the meantime, of course, if I add profiles like on the crushed hose of your pump, the situation improves. I made a tutorial on the site called "modeling a fish" which presents smoothing and it is true that I put several profiles. But the fact remains that this refusal to take into consideration this constraint of "normal to the profile" for the guide curves saddens me. Especially since the more profiles you put in, the greater the risk of "chaocity".
For the water lines of a sailboat in the old days, three points were enough to determine them (bow, beam and transom) after which they played on tangencies and curvatures. Transverse sections (couples) were deduced by intersection.
That being said, thank you and have a good evening