Helix sweep, flat end

[mretherford]

I am trying to create a screw type part. The best way to imagine it is a flat rectangular bar that is twisted. I know how to make a screw, but i always end with edges i dont want and then cannot find a way to clean it up nicely.

This thread is inactive and closed by the PTC Community Management Team. If you would like to provide a reply and re-open this thread, please notify the moderator and reference the thread. You may also use "Start a topic" button to ask a new question. Please be sure to include what version of the PTC product you are using so another community member knowledgeable about your version may be able to assist.

[MauroStefanovic]

I think there are better solution, but this is without any equation.

[TomU]

Might want to look at this: How to use get uniform pitch in warp/twist

You may need to model it slightly longer and then just cut the ends off.

[Patriot_1776]

As always, the answer is.....it depends.  From what I see, it's a constant helix simply cut off at both ends.  Helical sweep, or a VSS (Variable Section Sweep) will easily do that.  Simple.  IF, however, you need the twist to change and it to end up tangent to the long axis on both lead-in and lead-out, then you need to use either the Warp command (tough to get exactly what you want, but very powerful) or a VSS with the proper trajectory.

Without more accurate info on EXACTLY what you want, it.....depends. 

[Inoram]

The DNA comment had me thinking. So I was messing around, but if you zoom in everything is pretty distorted, disappointing actually.

[Patriot_1776]

Well, that shouldn't be surprising considering that you're twisting existing geometry.  I would expect that.  It's taking circular cross-sections of the rails normal to the longitudinal axis, and twisting the geometry around that, leaving the sections round.....normal to that axis.  Which "smears" the circular section into an ellipse if viewing it along the helical trajectory, and it smears the rungs as well.  The way to probably use a warp like that, would be to use a flat surface, pattern points along an edge, twist that, then use the edges as trajectories for the rails, and then pattern the rungs from the points.

[BHOoi]

see if this helps:

[Patriot_1776]

Ok, I made a quick model.  You can play with "offsetting" the marque on both ends (use absolute, then the offset on both ends is in "units"), to minimize the tangency, but that's not really a cure because you might lose some twists.

Anyways, like I said, without knowing the exact desired geometry...... 

[rohit_rajan]

proe wildfire 4.0 file attached,please have a look.

[TomD.inPDX]

In the proper conditions, even if you could control the twist in a constant pitch IRL, the "section" of the part would warp into a twist.  So any straight cross section in the actual part would not be a true representation of the native material's shape upon close inspection.

That said, duplicate the exact shape and treat the ends for further processing.  With the constant pitch, that shouldn't be difficult.

This is the technique that has proven quite reliable;

make a VSS of the center surface; thicken it about the middle; and add a full round on each edge.

Now you will see the true and warped section at the ends when you twist metal in this manner.

[emathivanan]

In sheet metal we have twist option. that may be useful for your need.

[Patriot_1776]

IF it does not need subsequent features that will prevent it from being a sheetmetal part.  That's my main issue with sheetmetal.  There should be sheetmetal FEATURES that you can use at any time in the middle of a regular part.

[blim]

I think it will be like when you use a blend rather than a helical sweep.

[TomD.inPDX]

The sheet-metal twist feature acts much like the warp feature.

You also have to make up for the offset side where thicken in a non-sheet-metal part can offset equally on both sides.

And a few other undesirable requirements for this case.

I played with spinal bend in Creo 3 last night.  It too has some interesting bugs by not closing the ends by default.

Fortunately I got it to work but it had me concerned for about a half hour.

[Patriot_1776]

There are also other ways to control end conditions.  This one starts and ends tangent to the longitudinal axis, but reverses direction.

[Inoram]