180-deg twist in sweep

[BHOoi]

Hi, all. I just realized there is a problem in Creo in handling 180 rotation while sweeping along a circular section. I was trying to create a Mobius with 180-deg twist but the sweep feature failed miserably. After many attempts, I managed to model out the geometry through a few different ways:

1. Toroidal bend.

The sweep is done through a straight line before bending it to 360-deg by Toroidal bend.

2. Split sweep.

Sweep is performed twice, along  half circle each time.

3.  Projected sweep

The single trajectory is extracted from a helical sweep created prior to it.

I am sure many of you have come across the similar problem before. I am attaching my proe file here with the hope that I can learn your workarounds  too. Many thanks in advanced for your sharing.

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.

[BHOoi]

As I explore more into this 18-deg twist, I found that the sweep feature will fail as long as the trajectory is a closed contour. For irregular contours, other than circular, Toroidal Bend can't be used to create the twist geometry. Only method 2 & 3 mentioned above are capable, as shown in the examples below:

Anyone has any explanation on why 180-deg twist fails on the closed contours ? Any different workarounds ? Appreciate you can share your thoughts. Thanks again.

[TomD.inPDX]

You highlighted the answer yourself in the first line of your second post.

It can be done with sweep and once you figure it out, you will feel completely dumb

[BHOoi]

I did ask in another forum but got no reply.

[TomD.inPDX]

..."IF the origin path is CLOSED..." 

How can you get a round sweep trajectory with an open origin?  Hint: "round" in orthographic view.

[lnelson]

Is this the type of object that you are looking for

[BHOoi]

I think it's the same way as in my "projected sweep" method. Basically, to break the join, just project the closed section to a helical surface and then use that projected curve as the trajectory. The 180-deg (or any angle) can then be achieved, through "Normal to projection" option in sweep. Refer to my proe file attached earlier for better understanding.

[Patriot_1776]

Sorry, I can't go giving ALL my secrets away, now can I? 

[TomD.inPDX]

No, of course not   But it has to be shown that it "can" be done.

Funny enough, the problem is much the same as in SW when it comes to a closed origin.

I don't know if the trick will work for SW.

But in general, you cannot sweep along a closed trajectory at all in SW.

Edit: SolidWorks would not let me do the trick.  And it also fails at less than a whole rotation.

However, as in Creo, you can make 2 halves.

[Patriot_1776]

It's great that you tried it in SW, yet ANOTHER thing that you can do in Pro/E but not SW! 

[lnelson]

I used a rotational blend ( found under the shapes group)

I needed to follow a few simple rules

1. the rotational blend must have a common axis of rotation for all of the sketches

2. all sketches must have the same number of vertices

3. the rotation for all of the sketches must not exceed 120 deg. between sketches

you can pre-define the sketches ( external to the blend) or define the sketches on the fly (internal to the blend)

under the options tab you can specify if the blend is to be a closed loop

look at the attached file  I took your file and created a fourth option for you to decide on

[BHOoi]

Thanks, Leif Nelson. This could be another to model the geometry. We probably need to fine tune it a little to get more uniform twist.

[TomD.inPDX]

Just to provide a feature reference...

[TomD.inPDX]

Hint: the above is a -closed- 360 degrees loop

[BHOoi]

I just hope PTC will fix this "bug" one day, or even next release so that whatever tricks we have will become redundant.

[dcaucutt]

Have you tried increasing your model accuracy?

[BHOoi]

Steps of method 3....

[TomD.inPDX]

I might have found a simpler way to generate that trajectory, but this accomplishes the -original- goal.

Method 3 is something completely different! 

[Patriot_1776]

Definitely simpler, but the above works......for one turn......... 

Glad to see people dig into it and figure things out themselves.  Bravo!

[BHOoi]

I am really keen to know other tricks or secrets as some may call, appreciate you can share yours.

[TomD.inPDX]

This only solves a your initial question.  How to skin this cat based on a real use case may have you take a completely different approach.

Alternate origins include equation curves or even mechanism trace paths.

Do be careful with the intersect.  The slope should be constant.  Put some comb analysis on those edges to see if they are smooth.

If you think about it, you can see how a variable slope will mess up the "timing" along the 360 degree sweep.  Creo uses "equal distance along" the curve to determine its step offset.

Can you post your "method 3" model?

Bottom line... there are more ways to screw this up than going back to basics... basics: 2 sweeps.  The only thing you have to be sure of is that each sweep is an equal length so you properly "center" the twist at the 90 degree point.

[BHOoi]

Thanks, Antonius. All the 3 methods are shown in my proe file enclosed with my original post above,  "mobius.prt.2.zip".  This file is in creo2 version.

[TomD.inPDX]

....except that method 3 is not circular... this matters!

[BHOoi]

yes, i just want to ensure this serves as a generic approach fro any closed contour. I attached here the actual proe file of that example.