Help Modeling Torsion Spring 1

[McDesign25]

I need help modeling a special torsion spring. I want the legs of the torsion spring to be parallel to each other. My current model depicts uses 3 curves (2 straight lines, and a helix) combined into a composite curve, and then I swept the whole thing. This produces a torsion spring but the legs are not parallel to each other. By parallel I mean if you are looking down at the coils, and the legs are in either direction, they are not parallel.

If this is not clear I would be happy to e-mail my model to someone to help clarify the problem.

Thanks.

 

[Theophilus]

The way I've done this before is to sweep only the helix as a first step. Make sure your helix starts and ends 180 degrees opposed by controlling how many revolutions your helix makes.

Then, I select each of the ends of the new sweep, convert the entities on the face (circular, probably) and extrude the length of straight spring I need. They should be parallel.

One thing to note with this method is that instead of creating a new plane "normal" to the end of the helical coil, I use one of the standard planes and make sure my helix intersects it. I place my cross-section on that plane and pierce it to the helical path. This allows my starting and ending faces to be parallel with one of my three primary planes. If you don't do this, but add a plane normal to the curve, your results won't necessarily be parallel.


Jeff Mowry
DesignHaus Industrial Design

http://www.designhaus-i-d.com

 

[SBaugh]

I would make all 3 sketchs (helix and 2 lines) into a composite curve, then add my profile and sweep them into one feature.

Scott Baugh, CSWP
3DVision Technologies

http://www.3dvisiontech.com

http://www.3dmca.com

faq731-376
When in doubt, always check the help

 

[Theophilus]

That sounds like a less-cluttered way of doing things. Do you know if it reduces/simplifies rebuilding? (I would think that is the chief advantage of this method since it eliminates two features--just never tried it before.)


Jeff Mowry
DesignHaus Industrial Design

http://www.designhaus-i-d.com

 

[TheTick]

email me at rol3@liquidschwarz.com and I will send you a couple of my torsion springs. They're a bit involved, but I have a method for designing a spring in closed position that has the flexibility to configure itself into multiple positions, including free position.

 

[JNR]

Interesting information, guys. We have used pretty much all these methods and I think the compound curve seems to be marginally best. We have also had "fun" with regular coil springs - especially those with compressed end ciols and flat ground ends. However, this brings up an issue and I wondered if anyone has looked at it yet. I think it is time now for SolidWorks to consider addressing it, but for now..... Springs are one of those components which are tough to model because you (may) need the free length for a part drawing and several different compressed configurations depending on the assembly it is used in. It often does not make sense to actually keep creating SW configurations (especially when you are using PDM systems and protected library parts).

There are a number of things that could benefit from some form of "variational" geometry that could be manually or even automatically configured form, say mates, in an assembly. Years ago, the old CV CADDS system had some abilities in variational properties of inserted parts (sfigures, nfigures, xfigures).

Any thoughts out there?

 

[StarrRider]

JNR

I completely agree. Most of the springs that I have made use a Design Table to make all of the configurations necessary. Doc – Free Length – Compressed – Compressed to length A – B – C. It doesn’t seem to matter how many compression lengths a spring may already have, the next time I use it I have to add another length or two.
In my opinion, it would be much better to provide (for a compression spring) a special dimension with a Min/Max relationship that would allow the spring to flex. Mating would be much simpler.

Lee

 

[TheTick]

If anyone else is interested, my model can be found t the link listed below. Also below is a brief exlanation. It may seem cumbersome, but it is the result of much "trial and error and frustration". It works for me. I would be interested any feedback, criticism, praise, scorn, monetary contributions or adulation.
====================================================
........(case sensitive) URL is

http://rschwarz.tripod.com/CAD/TorsionSpring--2003-03-10.zip

A few notes about this model.
* The model is oriented "in machine position" w.r.t. it's parent assembly.
* There are quite a few construction features before I lay down any solids.
*pay close attention to equations and linked values

1.) I begin by sketching the diameters, and then I made a sketch of the legs in the closed position.

2.) The next sketch moves the "lid leg" centerline to a position where there would be an even number of turns in the helix.

3.) The "toggle" sketch is used to control leg positions algebraically using equations. I have had poor results using configuration-specific equations. As-saved, the position and toggle sketches are visible. Viewed from the left view, you can see the lengths of the toggles change when you change configs, "switching" the various elements of the control equations. When a toggle's value is "1", then the equation causes the associated term to cancel out by multiplying by zero. "2"="on", "1"="off".

4.) Helix length and number of coils are controlled by equations which tie these properties to sketched positions.

5.) I use a copy of the helix to make the swept body. I find that using the actual helix causes the endpoints of that helix to be unavailable for future operations.

6. I use a patterned copy of an extruded surface to generate the moving leg's centerline. This is more robust than using patterned sketch entities. I also add a perpendicular surface at the base to use for sketching.

This may be more involved than you care to get. It does suit my needs well.

Best regards,
TheTick

 

[TheTick]

Zounds! My link refuses to allow people to download! Stand by for updated information on how to download this file.

 

[TheTick]

Here we go! Like most obtuse endeavors in the CAD world, there was a workaround. The URL is

http://rschwarz.tripod.com/CAD/TorsionSpring.htm

remember, its CAPS SENSITIVE

 

[TheTick]

Direct all future inquiries to the email adress in my profile.

You don't fight destiny, no sir... and you don't eat crackers in the bed of your future!