sweeping a spline along a helix

don_felton

Hi

I am making a fitting to screw onto corrugated hose. Should be simple. I have a helical path along a cylinder and a curve/endpoint based sketch correctly positioned to enable the sweep. I can test that by using the same sketch to sweep circles rectangles etc around the helix.

BUT if I try and sweep a spline that is the same pitch as the helix (to form a surface or a solid) ... it fails. Lines or rectangles sweep fine. Splines with endpoints that do not coincide form a helix. Splines that are teh right pitch but where the endpoints are at different diameters even seem to fill but obviously leave a "step".

One thing that does appear to work is to insert short straight lines at the "ends" of the spline so the helix "join" is not spline to spline but line to line. I guess I could then shrink these lines so they were very short (0mm?) but its a bit horrible. My suspicion is that while I have bound the spline such that its  should be normal (parallel to the cylinder) at its end points... it is not 100% and that causes it to fail.

Here is my experimental part

https://cad.onshape.com/documents/aad35f83eb632299d2628950/w/4fb8cd1adffeb6064791d223/e/00f5e55dc23cfb2840276095

Thoughts appreciated... though I may go with the very short straight line solution.

Don

S1mon

Sweeping in a single step is creating geometry that Parasolid either considers self-intersecting or non-manifold. If you do two sweeps and boolean merge, it will work. I split the wave in Sketch 2 into two pieces.

https://cad.onshape.com/documents/3f364ff0c954929ccc3ae922/v/39ddb617b96c6a63156b78d6/e/e508965f3da5e24756d7a406

S1mon

Sweeping in a single step is creating geometry that Parasolid either considers self-intersecting or non-manifold. If you do two sweeps and boolean merge, it will work. I split the wave in Sketch 2 into two pieces.

https://cad.onshape.com/documents/3f364ff0c954929ccc3ae922/v/39ddb617b96c6a63156b78d6/e/e508965f3da5e24756d7a406

don_felton

Hi S1mon. Much better than my unreliable short line hack. Turns out that given the hose is flexible and the PLA print is mildly flexible, a manifold break between loops isnt the end of the world. One print has split along the join and it really makes little difference. Thanks for the workaround and I will try it next time.