How to set sweep>swept tangential options in NX5 2

[RobLN]

Hi

I've been trying to sweep a spring shape in NX (due to the quite annoying limitations of not being able to define an associative helix curve...) and would really like the swept form to begin tangential at the section(s) which is being swept.

Default option is for the swept path to be perpendicular to the guide throughout the travel thus leaving a sharp start angle on a swept spring feature.

Within I-deas 12 there was an option to force tangency at the start section, end section (or both) which was nice.

Any advice here would be appreciated.

(NX5.0.4.1)

 

[JohnRBaker]

Could you upload your 'spring' model?

John R. Baker, P.E.
Product 'Evangelist'
Product Design Solutions
Siemens PLM Software Inc.
Industry Sector
Cypress, CA

http://www.siemens.com/plm

http://www.plmworld.org/museum/

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

 

[RobLN]

Sure - here is the model.

I would like to get a smooth transition between the tube and the end of the spring. Ideally without just splicing in a bridge curve (which is a bit of a kludge) maybe a law curve would help. (but they are pretty confusing to set up!)

 

[RobLN]

Here is another model.

I thought I'd try a simple sweep. Not sure how NX has managed to make the sweep such a strange shape. I was after a 'normal' spring

I'm doing something wrong - just not sure what...

Rob.

 

[cowski]

Take a look at: thread561-234523

There is a spring model posted in there by mmauldin (within the first 10 posts) that is excellent.

 

[RobLN]

Thanks Cowski

I looked at the model - I agree it is excellent. How on earth it is made - or the logic behind the law curve generated form the sketch is a different matter. I can't recreate the model. I only get a 2D sine wave style curve. Not sure where I'm going wrong. Help on the law curves in the documentation is thin on the ground. Any assistance is appreciated.

 

[hudson888]

Yep I couldn't trump that version.

Best Regards

Hudson

www.jamb.com.au

Nil Desperandum illegitimi non carborundum

 

[RobLN]

Hi

Brainteaser for you all (certainly been puzzling me...)

I have looked at the model from mmauldin and as far as I can tell replicated it again exactly. (except my model is in mm)

I can re-create the law curve on the original model (PART_A) and I get a helix shape.

If I run the exact same process for the law curve on PART_B I get a straight line. I don't see why though...

Are there settings within the law curve I'm missing?

Any advice greatly appreciated.

 

[mmauldin]

In part b change the upper case T to lower case in expressions xt and yt. Should work then.

 

[RobLN]

Hi Mmauldin

Thanks very much that seems to have solved it, one last thing - If I apply a tube (say a OD value of 0.45mm) to my helix in PART_B once I've fixed the T issue I get a really fat helical extrude. And the helix isn't quite round. Have I missed some more detail somewhere?

 

[hudson888]

I'd just copy the original law curve from the spring file that was posted and then toy with the parameters of it to see what happens.

Best Regards

Hudson

www.jamb.com.au

Nil Desperandum illegitimi non carborundum

 

[RobLN]

Thank you Hudson - default tolerance parameters on the X & Y were way out. All fixed now.

 

[mmauldin]

The reason for using the law curve for the z axis is to provide for variable pitch in order to close the spring ends. The variable pitch could be accomplished using if/then statements but this causes in an abrupt change in pitch and results in a small "kink" in the law curve. By sketching the pitch you can add a couple of small radii to help smooth out the kink.