Tough Spring Like Modeling Problem

[slashct]

I figured that I would throw this out there since I was able to model this but needed to createa whole mess of datum points axis and planesto do so and feel that there has to be an easier way. If you could use a wrap on a solid it would be cake. Anyway here is the part in a nutshell:
We make this part by taking a piece of square wire and punching notches in the length every .20 mm than wrap it 5 times like a spring (helical sweep). The notches are cut on the long surface, and than it is formed. Issue is with a helical sweep it is difficult to make these cuts since they are evenly spaced before the forming and does not follow the trajectory. Any Ideas? Here are some pics of the wire to give you an idea, cannot post the actual part for corporate reasons:


Wire before notches:


View attachment 2462


After the Notches


View attachment 2463

 

[CPiotrowski]

You can make a curve array thru your helical sweep and use a pattern along curve to create the notches.

 

[Dell_Boy]

I manage to do the following with 3 datum planes, 1 datum axis and 0 datum points but still is pretty big and slow.

View attachment 2466


It doesn't match your dimensions but I think it has pretty much all of the required elements. I wasn't sure whether you wanted the cuts on the top, inside or outside but I figured that putting them on the inside or outside would allow you to cheat with a rotational cut around the perimeter where as with on the top you can't.

DB

Edited by: Dell_Boy

 

[slashct]

DB,


The cuts are on the outer face of the spring, it is used to grip.Dimensions not an issue, it is the idea... Possible to see the part?


Thanks


JW

 

[Dell_Boy]

having the cuts on the outside should make it a whole lot easier and the model smaller and faster. Can you re-post the second picture rotated more to confirm the side profile of the cut

DB

 

[slashct]

Sorry for the delay in the reply here are some shots of what the section and would like like on one coil:


View attachment 2494


This is the section that is patterned on the wire to be coiled


View attachment 2495


This is the wire coiled with the cuts on the outside, it forms teeth so to speak..

 

[slashct]

View attachment 2496


And this is the side of one coil...

 

[Dell_Boy]

Now I can see the true profile I realise the method I used won't work for your requirement because the sides of the cut aren't vertical. The posted picture was created by helically patterning a helical cut up the spiral. A chamfer was added to the group before patterning. Several relations were written to correctly proportion the rise and run of the cuts and locate the cuts w.r.t. to wire.

Nevertheless you can get what you want without a mass of datum features. The principle is to use relations to tie the cuts to the wire.


I have 3 suggestions to do what you need. The choice depends on how accurate you want to have your model at a cost of increasing difficulty and decreasing speed.

1. Rotationally pattern an extruded through-all cut. The angular pitch of the cut will have to be 360/n which may not agree with your EXACT requirements but you should be able to get close enough. Also the cuts will not be exactly perpendicular to the wire

2. Rotationally pattern a helical full length cut. Imagine the long pitch riflings in a gun barrel. To be perpendicular to the wire the pitch of the cut will need to be equal to π

 

[elab.theken]

Have you considered trying the Spinal Bend feature. If you can model the wire easily in its straight condition and model the curve it needs to wrap to, then this should work.


I made this model with 7 features (1 pattern) in the model tree after the default datums.


View attachment 2509

 

[Guest]

Great. I think you will make a new way ueing Pro/e