Angled Helix

[jsantangelo]

How do you go about setting up a VSS to do an angled helix? I've tried a couple things but can't seem to get it right. I've tried the method described in the "Creating Surfaces with Variable Section Sweeps" article. I've also tried the Leo Greene method.






Edited by: jsantangelo

 

[jimshaw]

Use a VSS with a planar circle as your trajectory. Since trajpar in it's normal use would give you a constant pitch (an "unangled" helix if you will) use evalgraph, and usethe graph to create a relationship between the angle used for trajpar (to create the helix itself) and the trajpar variable as the x and y axes respectively.


I imagine the graph asa line from (0,Pi)to (360,Po) where Pi is the inner pitch and Po is the outer pitch.


I've never tried this before, but theoretically it should work, right?


Sorry no graphical examples

Edited by: jimshaw

 

[Starosta]

Hi

I use VSS and special graph
Dimension on graph: 360 - 1 revolution, 471.239 - Pi*R R-General radius of spring

Maciej

 

[Jacoolo]

simple and cute

that is the place where evelgraph rocks

I tried to achieve that within ordinary relation sin+trajpar

but I failed

 

[jsantangelo]

Is it possible to do something like this with trajpar?

 

[jbuckl]

lots of things are possible.
Keep trying.
One thing for sure... other cad systems cannot produce graph driven helical sweeps like Pro/E does....look at the section drawing, and some of the strange shapes encountered along the way!!!


2009-03-10_202109_spring_drw_2.zip

 

[anand_m_s]

Hi Angelo,


First of all you can notcreate the VSS using entire circle and DTM1 as the reference direction. However, you can use a sector of circle correspondition a single turn of the helix and this datum plane as the direction reference. Then you can pattern them to get the complete helical ring. But there is a problem here also; you will not get tanget continuity between adjacent sectors. I have tried this method and the snapshot is given below:





So this can not be the method of getting the angled helix.


Regards,


Anand

 

[jimshaw]

First of all you can notcreate the VSS using entire circle and DTM1 as the reference direction.

If you look closely at the original picture (at about 10:00 or so), it looks like an open spring, so the trajectory should be open (and valid) for the VSS. So instead of a full circle, use a 359

 

[jsantangelo]

So I think I got it kind of figured out a bit.

Here is a regular helix with 24 turns.





Here is the angled one. Again 24 turns. It looks similar to the image in the
original post.




Here is an image with both in the same view. Green = Angled Helix, Red = Regular
Helix




Here is what I did. I created a sketch with a circle for the VSS (the blue circle in the
above images). I then did a 24 turn helix using trajpar. I created another VSS but
instead of the circle sketch as the origin I used the helix edge and then the circle
sketch as Chain 1. The X-trajectory had to be selected or the result looked funky.




Here is what the sketch for that VSS looks like.




Then create another VSS (#3), use that surface edge from VSS 2 and sweep a
circle. Now you have a parametrically controlled angled helix.

The only issue was in VSS #3 it wouldn't sweep all the way around. So I did half
then did another VSS to complete the sweep.






Thoughts?







Edited by: jsantangelo

 

[jimshaw]

Wow. Nested VSS's. Minimized feature tree. Visually intuitive (unlike evalgraph)


Excellent!

 

[anand_m_s]

Hi Angelo,


I tried your method and it works out be GREAT !!. But there was one difference in the procedure followed; I did not make the Circle as X trajectory in VSS 2. When I did make it the X trajectory, VSS failed and when I removed the X trajectory tick mark against the chain 1, I got some funny shape. So I had to redo the VSS command without selecting the X trajectory for the circle.


Regarding the issue in your VSS #3; When you select the edge of conical helix, only half the edge gets selected. What you have to do is toselect the remaining edge holding the Shift key down.Then entire edge gets selected and you need not repeat the VSS command for the other half.


Any way,

 

[jsantangelo]

Regarding the issue in your VSS #3; When you select the edge of conical helix, only half the edge gets selected. What you have to do is toselect the remaining edge holding the Shift key down.Then entire edge gets selected and you need not repeat the VSS command for the other half.

Thanks everybody for your feedback

I know that trick using shift to continue the edge selection of the chain. When I do that it won't build the surface

If I just choose one it works. I'll try playing with it more stuff tomorrow to see if it was just Pro/e being quirky.

Cheers!

 

[jsantangelo]

So upon further investigation there seems to be a non-tangent portion of the

angled VSS, thus the reason it won't create a complete sweep. See image below.
It also doesn't look as smooth as VSS #1. Looks like there is some wavyness to
the surface. I've also attached my Pro/e WF4 file for those that would like to
inspect the model.











2009-03-12_115030_wf4_angled_helix.prt.zip



Cheers!