How to create variable pitch helix spring in NX8.5

[hyuni]

Hello.

I need a helix that satisfies the following criteria

total turns = 8
Radius = 80
total length = 560
pitch between start point and end of turn 5 = 40 / length = 200
pitch between start of trun 5 and end of turn 6.5 = variable pitch / length = 60
pitch between start of trun 6.5 and end of turn 8 = 200 / length = 300

I then need to sweep a section along this helix.

Any help would be greatly appreciated.

Best regards

 

[mmauldin]

I'm still using 7.5 so there are may be better ways to do this in 8.0. Generally in this situation, I flatten the helix so that I can explicitly draw the pitch/coils that I need, then wrap the curve onto a cylinder. I'm not too sure I understand the variable pitch requirement for the second section, but attached is a general concept.

 

[carlharr]

Needs the p numbers setting up so its more "parametric", but here you go.

http://www.jcb.com

NX 8.5 with TC 8.3

 

[carlharr]

For mmaudlin, there's a new and improved helix command in NX8.5 which is great. See screenshot attached

http://www.jcb.com

NX 8.5 with TC 8.3

 

[mmauldin]

Maybe one day, not too long from now, we will move up to 8.5. Or I may be retired by then. Could go either way.

 

[carlharr]

8.5 is worth the move.

Just having a go at creating this model using a law curve, but I've got no idea how to get the "transition" section right.

Maybe someone else could have a go...

http://www.jcb.com

NX 8.5 with TC 8.3

 

[JohnRBaker]

No, I think you've pretty well got it nailed. The only thing that I would do is perhaps add a couple of small fillets to your 'Law Curve' so that the resulting curve is a little more realistic in terms of the transition from one section of the spring to the next, as I've done in the modified version of your model attached below.

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

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

 

[carlharr]

John,

I haven't, I've made a mess of it.

You can see with the "original body", my new part using law curve doesn't really have a transition stage.

Pitch jumps straight to 200 from 40, and the result is a spring that's too long.

http://www.jcb.com

NX 8.5 with TC 8.3

 

[carlharr]

*40 to 200

http://www.jcb.com

NX 8.5 with TC 8.3

 

[carlharr]

Further info - it does look like the lengths are wrong in the model made with law curve.

I've coloured my orginal model, in its 3 different sections.

You can see the pitch in my new model starts to increase before the spring has reached 200mm length.

http://www.jcb.com

NX 8.5 with TC 8.3

 

[JohnRBaker]

The original numbers are not correct. It's either going to be 7 turns and 560 mm long or it's 8 turns and 680 mm long. You can't get from a 40 mm pitch to a 200 mm pitch in only 60 mm if there are 1.5 turns. Take the average pitch between 40 and 200, that 120. Multiply that by 1.5 pitches and you get 180 mm. Add 200 + 180 + 300 you get 680 mm.

Also it appears that the Pitch by Law curve may have some problems. Instead I used points 'Linear along Spine' and I get the correct shape (see attached part file). As a check for my solution I've place some Datum Planes and intersection Points along the length of the helix to show the three 'pitch sections'. If you look you can verify the number of turns for each section. To see the final spring, just unsuppress the 'Tube' feature.

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

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

 

[hyuni]

Thank you everyone who answered.

Best regards

 

[carlharr]

John, you're right on the numbers, so I've stuck with 8 turns = 680mm.

And I think I've figured out the law curve.

In a simpler model, I found that changing length of a law curve did not affect the length of a helix.
So I guessed its x-dir must be related to "number of turns of a particular pitch".

So to test, I changed my helix to a "total length" helix, and created a law curve where lengths related to number of turns (I multiplied them all by 10 just to see if it really was a ratio).

And it worked!

No idea how to explain it better so here's the model.

http://www.jcb.com

NX 8.5 with TC 8.3

 

[NutAce]

Hi John, Carl.

Can one of you explain the values used on the sketch which is used as law curve for the pitch?
What happens exactly when I change each value?

I'm trying to use this principle of creating a Spring which can be deformed in an assembly...
I'm having some problems on getting the overall length deformable while keeping the number of turns fixed.

Ronald van den Broek
Mechanical Engineer
Cad Environment Coordinator
Wärtsilä, Propulsion Services
NX8.5.3 / TC9.1.2
HPZ420 Intel(R) Xeon(R) CPU E5-1620 0 @ 3.60GHz, 32 Gb Win7 64B
Nvidea Quadro4000 2048MB DDR5
HP EliteBook 8570W Intel(R) Core(TM) I7-3740QM CPU @ 2.70GHz, 16Gb Win7 64B

 

[carlharr]

Hi Ronald,

Finally, I understand this!

Take my last model I posted, and change Helix (10) to be defined by number of turns, instead of total length, and there is no change in the helix.

So the "lengths" in the law curve are not lengths at all, they are the ratio of the total spring for which each "pitch" applies.

If I increase them all by 1.25 (so it reads out of 100% - attached), there is still no change.

http://www.jcb.com

NX 8.5 with TC 8.3

 

[carlharr]

In your case, if the overall length changes but number of turns remains the same, then pitch must change.

So if yours is a constant pitch spring (along its length), you just want to define your spring by number of turns, and alter the pitch value to make it deform. I think.

http://www.jcb.com

NX 8.5 with TC 8.3

 

[carlharr]

Like this...

http://www.jcb.com

NX 8.5 with TC 8.3