Valve spring expansion problem

Automotive valve springs are typically cut flat on each end. This shape is fairly easy to make with a NURBs surface. You make a helix curve, use if to form a pipe, and the cut the pipe flat on each end.

Problems arise when you try to create a compressed version of the spring using Scale1D. As you compress the spring the pipe distorts.

To make a spring you can compress, it is suggested to use the ApplyCurvePiping to the basic helix curve. This works immediately, and the resulting spring can be compressed and de-compressed without any distortion.

But here is the problem. The surface of the pipe is not NURBS, and cannot be cut and shaped to form the flattened ends of the valve spring.

A 3dm is attached. The layer “springone” is a NURBS spring with realistic flat ends. The layer “springtwo” is a compressible spring with, I surmise, a mesh surface. The ends are not flattened because they cannot be, in effect, machined like a NURBs surface.

The question is this: How can I make a valve spring that is both flattened at the ends and compressible.

Thank you for your insights. spring expands 11-8-20.3dm (2.9 MB)

except for the flat ends you coud use Record history and apply the Scale1D in the helix curve, then the pipe will update without distortion and after that you can cut the ends to make them flat.

To take Diego’s idea maybe one step further - make the helix, then the pipe, and go ahead and flatten the ends. Then, use CageEdit to manipulate the spring exactly as Diego shows - select the helix as the control object and Scale1D.


Many thanks Diego and Pascal. I followed both of your instructions in sequence. It worked. The techniques compressed a valve spring, flattened at both ends initially, without introducing distortion.

This is not a static model. The ultimate goal is to animate the compression of the spring in Bongo, which has its own inherent Scale procedure. I have been using Rhino’s Scale1D as a proxy for the Bongo scaling procedure because it is quick and easy to try various approaches with Scale1D.

My sense is that the distortion problem is going to follow me over to Bongo, but we’ll see. Thank you again for your insights. Michael