This example is not a good way to model a spring. For example, the correct cross section of a spring is only seen tangential to the helix of the spring. This method simulates winding with elliptical section wire. Also, keeping the number of coils via relation is good, but as a spring compresses, its diameter increases - Remember, the volume of material in the spring coil remains a constant. Alternatively, if its diameter is forced to remain constant, then the number of turns will change.
The learning exchange method is also 'inflexible' as a way to make a helical spring. It is very difficult to add end finishes to a spring made this way.
A better way to make a spring model is to start with a revolved surface that represents the winding of the spring, wrap a datum curve that has the ease-ins at the ends, and then create a swept feature, normal to curve. This accurately represents the spring, as this is the way a spring is wound. Relations are still required to coordinate changes of compression and extension.
Models like this should be part of the Pro/E distribution.
I have only the commercial example version at LMS learning.
Do you have maybe the student version. The student license does not accept commercial parts.
Take a look at http://communities.ptc.com/docs/DOC-3063 for an alternative method.
You may be correct regarding true spring geometry; I will defer to you as I am definitely not a spring expert. It has been my experience over the last 20 years that most customers only need a good, easy to create representation of a spring to put in their assembly. A tooling designer for example is not typically concerned that the geometry of a purchased spring component is exact. As an example, at both Ford and Maxtor we often did not even use the helical sweep but instead represented springs with only cylindrical extrusions that had attachment rings at each end. When in Mechanism mode, any solid spring geometry is removed and replaced with a Spring connection to better represent the true nature of the design.
Regarding components being part of the “Pro/E distribution”, I do not have details but have heard good things about future standard component functionality being worked on. I would encourage you to post a spring created using your method so that if the helical seep method is not sufficient for others in the community, they can give yours a try.
The helical feature you showed is perfect for screw threads and other features where the cutting profile is in plane with the axis of rotation. The solid spring simplification is a good one for semi-static uses - where the mass distribution of the spring is not a factor, as in tool and die work.
Anyway, I created a document on making wound springs in PlanetPTC communities. What it misses in directly controlled constant volume it more than makes up for in flexibility.
One method to make wound springs
The datum curve wrapping method works on right cylinders for constant spiral; it is a constant volume when the curve length is fixed and only the height of the curve or diameter of the cylinder is changed. It does have the disadvantage that Pro/E (CREO too?) tracks the maximum excursion for geometry for extents, so a long piece of wire wound into a thin spring will have create unnecessarily large extents.