Constraining deformable parts

[VW181]

Hi all,

We have a problem with an deformable parts an constraining it in an assembly.
The deformable part is a simple spring created with a helix and an tube-command.

The easiest way to constrain this spring would be to use the coordinate system I think, but all kind of strange things are happening than; it flips, it moves in small steps, all constrains in the assembly are over-constrained, etc....
I tried it with different springs in different assemblies but all with the same result.

I also tried it with a deformable block. Constraining on the faces of the block gives no problem, but constraining on the coordinate system gives the same problem like with the spring.
Adding a extra coordinate system to the part helps a bit, but it's still reacting strange.

So is it possible to use the coordinate system to constrain a deformable part?

I found one old post that called constrains and deformable parts are a bad combination, and that therefore deformable parts can't be used in assemblies.
Deformable parts was called a total useless function in assemblies and that you should use "Reference sets".
Is it really so bad with deformable parts???


Thanks in advance,
Tjeerd


Using NX 8.0 and Teamcenter 8 on Windows 7 (64)

 

[mmauldin]

I have experienced similar issues in the past when using helices and law curves as part of my deformable springs. Over the years I have developed the attached compression spring template that does not use either helices or law curves and have not had any issues when using it in assemblies.

 

[VW181]

Thank you for your answer and your part.
That is indeed a total different way to create a spring. But in this way you have a good control over the beginning and the end of the spring.
I will have a good look at it and will test it.
Thanks again.


Using NX 8.0 and Teamcenter 8 on Windows 7 (64)

 

[JohnRBaker]

If you could move to something newer than NX 8.0 you could use the new Helix Curve function introduced in NX 8.5, which is much more stable than using Law Curves since you can link its orientation to a Datum CSYS, not just the absolute WCS. And you mentioned that you wanted to control what I assume is that so-called 'end conditions' of the helix, which is also very easy using the new Helix Curve, as shown in this picture of a spring modeled in NX 8.5:

The 'helix' used in this example was created as a single curve using only the interactive, and fully parametric feature-based, options found in the new Helix Curve dialog.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Digital Factory
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[VW181]

I hope I can test it in a few months.
We have plans to go to NX9 or maybe NX10 this summer.

Using NX 8.0 and Teamcenter 8 on Windows 7 (64)