CAD Forum

I am looking for recommendations on how to model this part.

This part must be manufacturable on a 3 axis mill, or lathe with live tooling. It also must have a sharp corner in the bottom for the pigtail of a spring to sit, thus the reason for the flat at the end of the helix.
I was able to make this part using surfaces and helix, but I do almost no surfacing, so I'm probably not using the most efficient approach. The other problem is that the helix feature can't use equations or reference geometry, so any change requires a bunch of hand adjustment and still results in a "close enough" solution.

You can make the ramp shape with a surface and then extrude a pair of nested circles up to the surface. Or you could extrude a cylinder, make a hole in it, sweep a rectangle along a helix to make a solid cut, and then do a little extrusion on the bottom to make the flat. There are a thousand ways to make it with solid or surface. Doesn't really matter what's efficient, does it? Done is done. As long as the geometry is correct.

I've tried several of the ideas you've suggested, and struggled with all of them. The problem is the flat is square (as opposed to a wedge that would be created with a sweep operation.) What looks like a simple solution becomes complicated in a hurry. Here is a screenshot from another attempt. You can see that the top of the ramp doesn't stay parallel to the bottom.
I only ask about efficiency because I would like to be a better surface modeler, when the need arises.

You can use height/revolutions from the helix and your radius/diameter of the sketch to control everything on it in equations.

I prefer to make one surface and extrude up to it, then build from the solid verses making everything from surfaces.

Here is my take on it.

Is that file 2023 or 2024? I'm upgrading to 2023 tomorrow, so hopefully I can check it out then.

What are the required helix parameters? The numbers in your model seem manually tweaked as you said.

The more I think about this, the more I think I have the geometry wrong. I think I want the flat on the bottom to undercut the start of the helix. That way the intersection point is still in the correct location.

The original design was 3d printed, so it had a full helix with a 20mm step.
I should be able to take that design, add the flat in the bottom and eliminate all of the complication I created for myself.

Truth be told, the end of the springs aren't very good, and I probably don't need to be this concerned with the geomoetry, but there's still a right way to do this and that is my goal.

If you put the spring inside a tube with a cap that is flat bottom, you don't have to worry about the helix. The tube will keep the spring aligned. You just need the tube to encompass the entire length of the spring or the coils will catch on the top of the tube.

That wouldn't be representative of how the spring is constrained in the real world, which is the point of the test. Also, the spring rate is changed quite a bit by only supporting the tip vs. the majority of the first coil. This part is used in a machine that measures the rate of the spring by compressing it.

SPerman wrote: ↑Thu Mar 28, 2024 2:44 pm That wouldn't be representative of how the spring is constrained in the real world, which is the point of the test. Also, the spring rate is changed quite a bit by only supporting the tip vs. the majority of the first coil. This part is used in a machine that measures the rate of the spring by compressing it.

Ahh, didn't realize that you were testing it. Well then....
Solidworks (and probably most CAD programs) generally have a hard time between the transition of a helix and a flat....at least on how to mimic how an end mill would machine it.

The only place you can "transition" from a helix to a flat is directly along a radial line. Any line that is not exactly a radial line cannot have a transition from a true helix to a true flat without a step. Anything you bodge together with surfaces may sorta look like a helix, but it won't be one.

josh wrote: ↑Thu Mar 28, 2024 5:26 pm The only place you can "transition" from a helix to a flat is directly along a radial line. Any line that is not exactly a radial line cannot have a transition from a true helix to a true flat without a step. Anything you bodge together with surfaces may sorta look like a helix, but it won't be one.

Exactly. To create two equal height helixes that start on one line and end on a second parallel line, the two would have to have different pitches. You can see this clearly by using two equation driven curves.

Once I wrapped my head around the proper geometry it is rather simple. I was trying to force helix of 2 different pitches and angle of rotations to accomplish the geometry I thought I needed. It kind of worked, but was the wrong approach either way.

I need to do research into equation driven curves. I think that would have accomplished what I was attempting. (NX created helix using Law curves.) I don't see any way to control the helix feature using Global Variables or equations, but maybe I'm missing something.

SPerman wrote: ↑Fri Mar 29, 2024 7:47 am Once I wrapped my head around the proper geometry it is rather simple. I was trying to force helix of 2 different pitches and angle of rotations to accomplish the geometry I thought I needed. It kind of worked, but was the wrong approach either way.

I need to do research into equation driven curves. I think that would have accomplished what I was attempting. (NX created helix using Law curves.) I don't see any way to control the helix feature using Global Variables or equations, but maybe I'm missing something.


SpringHelix2.SLDPRT

The equation driven curve is nice because once it is generated, you can then drag it around and add relations to position it and you can reference variables in the equation.

SPerman wrote: ↑Thu Mar 28, 2024 11:24 am Is that file 2023 or 2024? I'm upgrading to 2023 tomorrow, so hopefully I can check it out then.

It's 2023

SPerman wrote: ↑Thu Mar 28, 2024 10:00 am The other problem is that the helix feature can't use equations or reference geometry,

Quick tip: Remember that you can get a helix using the Wrap command, using a triangle or just a diagonal line in the sketch. And in the sketch, equations and parameters or reference are not a problem.
Additionally, I would personally make two pieces on a lathe. One is the bottom with a flange, the other is the internal "ramp". This procedure ensures faster production. It is easier and faster to use a CNC lathe with an additional axis than a milling machine

Doesn't it seem that instead of the pocket, the retaining wall of the spring could be tilted to provide better support?

Have you tried surface loft? I would draw the beginning, middle, end, 3 lines and inner and outer path. I'm not at my PC so I can't try it.