"Envelope" Body within Part for moving body interference detection?

[Tom G]

Ball valve handles, as well as any other moving component, have a range of motion and corresponding potential clearance requirements as it moves. I have represented these simply with a planar arc sketch for a typical handle swing. I am able to view it from dynamic angles to judge its clearance around other components. It works ok for its purpose. I'm considering an improvement of using solid bodies to check for interference.

What I would ideally like, but do not think is possible, is an additional invisible solid body added to the component that does not add to its mass. There's two ways I can think of to do this. First, a non-merging rotated body within the part which is then assigned a near-zero mass material such as Air, and hidden. This should not affect its mass or center of gravity within any assembly where used?

The other way I can think of is to use Envelope component, which is only available in an assembly, so I would be replacing the valve component with a subassembly that includes the valve and its hidden envelope rotated handle component. I would prefer not to do this because it would create a new subassembly file for each valve part file that I implement this in, as opposed to adding features to an existing design library part file.

How does interference detection consider envelope components, hidden bodies, or hidden components? Am I going about this wrong?

Additional side-topic, I also do not believe that I can use a solid body to generate a rotate boss. Is this true? I can extrapolate its extents (by an extra sketch) upon a sketch plane and rotate that into a body. I could also Circular Pattern the real handle body into multiple overlapping bodies which, if the above is doable, seems like it would also work because the result bodies are hidden where used.

I am open to any ideas. Do you perform interference detection with moving handles, switches, or other operators? TIA.

[berg_lauritz]

I'd love to know more about it. Can you add something to parts so that you can run an interference detection? I'd like to automate that.

[matt]

You guys will go way out of your way to try to make stuff work that doesn't work. I wonder what would happen if you stopped fighting the software and used it the way it was meant to be used?

[Tom G]

I had some time to try it out, to pattern the bodies within the part. Here is my original valve part:

BV original.JPG (18.12 KiB) Viewed 1474 times

I add a Pattern, make it Air material and delete some appearances that came with the new pattern bodies.

BV Patterned.JPG (26.36 KiB) Viewed 1474 times

Place in an assembly, create a simple weldment angle to obstruct the handle. Run Interference Detection. Get results.

Then, I went back and hid those bodies. Interference Detection still found the same list of errors.

[Tom G]

You guys will go way out of your way to try to make stuff work that doesn't work. I wonder what would happen if you stopped fighting the software and used it the way it was meant to be used?

Hack the world.

No wait. I can answer that. I'd use the Routing license that we own so that my designs can automatically fail, like creating a self-intersecting 3 elbow jug handle where the was once only one elbow, or fail critically in its sketch, or lose its drawing notes because it can't well be moved but can easily be destroyed and recreated, or anything else that is actually wrong with the things that they will not fix.

[matt]

Hack the world.

No wait. I can answer that. I'd use the Routing license that we own so that my designs can automatically fail, like creating a self-intersecting 3 elbow jug handle where the was once only one elbow, or fail critically in its sketch, or lose its drawing notes because it can't well be moved but can easily be destroyed and recreated, or anything else that is actually wrong with the things that they will not fix.

So you think that something that's not intended to work is going to work better than something that is intended to work? You're putting a lot of faith in pure accident. I'm sure there's a better answer. You're not going to win fighting the software.

[Tom G]

So you think that something that's not intended to work is going to work better than something that is intended to work? You're putting a lot of faith in pure accident. I'm sure there's a better answer. You're not going to win fighting the software.

I did not place an envelope within a part or break any rules or whatever you're discussing. I made more bodies to represent where bodies can possibly be. I make assemblies. I buy valves. I don't understand either of your responses at all. I showed a simple example where it works. I described simple steps to create this. I'm not fighting anything in this case, but using it to inform my process.

[DanPihlaja]

You guys will go way out of your way to try to make stuff work that doesn't work. I wonder what would happen if you stopped fighting the software and used it the way it was meant to be used?

I guess I don't really understand this either. Is there a better way? If so, what is it? Because interference detection can be seriously flaky, especially in the instance of cam handles that are supposed to push on something at one end, but not interfere at the other end.

Because in the end, it really isn't about interference at all......it is clearance. You need a certain clearance around the handle to allow for hand grip and whatnot.

In the past, I have created a path for the handle, and then swept a good approximation of the clearance that I want around the handle along that path. Then, if the interference between the clearance and whatever it is I want to check is really hard to see, I will make sure that the two parts share a coordinate system, and then do a compare documents (this allows you to see the areas where the two documents share a common volume without changing either component)

I really only do this when it is critical, but it gets the job done.

I don't really know of a faster way that is not flaky.

[matt]

I guess I don't really understand this either. Is there a better way? If so, what is it? Because interference detection can be seriously flaky, especially in the instance of cam handles that are supposed to push on something at one end, but not interfere at the other end.

I can think of a couple of ways. If it's just a library part, having a configuration with a range of motion solid isn't out of the question, especially if you're going to do it more than once. And if what you're really concerned about is clearance, then use the clearance detection thing.

And why you think you have to use one part or one assembly for everything is beyond me. It really just doesn't work very well, as you're finding out. Use a different assembly for this kind of analysis, or use the main assembly as a subassembly in an other assembly just for analysis, or join all the parts in an assembly into a single part and you'll know that any body that joins other bodies intersects with the other bodies. The extra data costs you nothing, and saves you so much. If you use PDM, put the special analysis parts and assembly files in a special area not for release.

It sounds, and not just in this case, but in a lot of questions here, like people are avoiding answers and finding some trivial reason to reject workable solutions. Taking up much more time trying to make the job impossible than by just doing what works. This isn't olympic diving. You've got to get the right answer, but there are no extra points for making it more difficult, and getting it done in less time is a definite bonus.

[berg_lauritz]

And why you think you have to use one part or one assembly for everything is beyond me. It really just doesn't work very well, as you're finding out. Use a different assembly for this kind of analysis, or use the main assembly as a subassembly in an other assembly just for analysis, or join all the parts in an assembly into a single part and you'll know that any body that joins other bodies intersects with the other bodies. The extra data costs you nothing, and saves you so much. If you use PDM, put the special analysis parts and assembly files in a special area not for release.

It sounds, and not just in this case, but in a lot of questions here, like people are avoiding answers and finding some trivial reason to reject workable solutions. Taking up much more time trying to make the job impossible than by just doing what works. This isn't olympic diving. You've got to get the right answer, but there are no extra points for making it more difficult, and getting it done in less time is a definite bonus.

We don't want to change those parts into an assembly, because it would create months of work for too many people. So a band-aid is preferred WITHIN a part.

And please skip the saltiness from

like people are avoiding answers and finding some trivial reason to reject workable solutions. Taking up much more time trying to make the job impossible than by just doing what works. This isn't olympic diving. You've got to get the right answer, but there are no extra points for making it more difficult, and getting it done in less time is a definite bonus.

and

And why you think you have to use one part or one assembly for everything is beyond me. It really just doesn't work very well, as you're finding out. [...]The extra data costs you nothing, and saves you so much. If you use PDM, put the special analysis parts and assembly files in a special area not for release.

next time @matt .