...ok, MIS TAKE.
I had made a sweep with the notch profile. And then a cut perpendicular to the spline and thought: Looks like a circle > is a circle. WRONG...hmmm´grmpfl...
My idea now is to work with an array. Starting from the two already rotated profiles in a clean side view.
Problem: I would like to place the array profiles where the spline points describe the tight curves. To use them in the sweep. How do I achieve this?
re:
> Problem: I would like to place the array profiles where the spline points describe the tight
> curves. To use them in the sweep. How do I achieve this?
That's kind of weird thing to do for a couple of reasons.
One is that the control points of a curve are a polyline like this:
Note that some control points are on the curve and some are away from the curve. So placing profiles at each point is not really productive.
Another thing is that it is already built in to the sweep command that it adds more internal profiles for bends. So you're kind of trying to manually reproduce what sweep already does internally.
It could help if I understood why you want to do this, what problem are you trying to solve?
Hi Michael,
Thank you very much for your energy and commitment in dealing with my constructive problems.
More details:
I would like to achieve the following. At the point where the two green profiles are, the sweeps should change or meet, specifically from horizontal to vertical.
This means that the profiles arriving from both sides rotate along their splines. This rotation should begin at the “bend point” when they come from the side with the narrow radius, for example. That is why I would like to place the profiles there.
Is that reasonably understandable?
Mala
P.S.:
I tried to recreate it for illustrative purposes.
What irritates me a little, beyond the actual problem of profile placement at the bending points, is that these sweeps don't work...hmmm....
Hi Mala, I'm sorry I'm still not really understanding what you're trying to do but I will keep looking at it.
re:
> What irritates me a little, beyond the actual problem of profile placement at the bending
> points, is that these sweeps don't work...hmmm...
Can you show me a specific sweep that doesn't work?
You've got 12 different profile curves and 4 different rails in your file and it's difficult for me to know which combination in particular is not working.
...I am myself looking for a solution for a constructive idea. I have already been able to solve this idea with a few nodes. But it doesn't work here. Perhaps the concept is simply not applicable here.
I want the profiles to rotate so that they meet vertically. However, they start horizontally and vertically. The longitudinal rotation should take place in the bend.
I don't think that's possible. Regardless of where exactly the profiles are placed for the sweep...
Hi Mala,
here is another option.
Cut out section where the twist is required
.Select corresponding edges and copy to create curves to blend,
Use the blend curves as guides for sweep.
Cheers
Barry
Hello supporters,
Thank you very much for your principles and ideas.
In principle, the solutions show something of what I wanted to achieve. When I look at it this way, I realize that it looks different than I had imagined. The closest is the last option from Barry-h. (even though I don't quite understand the workflow yet). Because there, the axial rotation happens over a long distance. I need to try it out some more and tinker with it. I'm still too far away from the ideal design.
Best regards, Tom
Hey Tom,
Here is a short video showing what Barry is describing. I couldn't tell whether you wanted the twist to be longer or shorter, so I took my file and created a shorter "harder twist"..
It should explain what is going on and how he described to generate the curves for the twist sweep. If you wanted the twist to be longer, You would do it over a further distance... This should have it make sense.
PROFILE ROTATION PROB_2 MOI FORUM.3dm
