Rotate Component

[Species8472]

A jergens relieved pin needs to be rotated so that the locating surfaces are perpendicular to a line that passes through the centers of the two locating holes. The example in the manual doesn't seem to work as stated. Regardless of what axis you rotate the component round it either doesn't rotate or it fails to rotate to the angle specified. There are no edges or surfaces to aline the part to. Why can't you just select a component, select a point of rotation and enter the angle? If so where? If not WHY not?


Reading this manual it's clear it's written to sell trainings not help users.

[Kiwi]

...Why can't you just select a component, select a point of rotation and enter the angle? If so where? If not WHY not?

I don't know your answer but can you add a feature that you can use as reference to, rotate and then delete that feature.

[JerryFlyGuy]

In these situations I generally use reference geometry and align the part using the reference geometry. However that being said, if there is a specific alignment required then there must be some form of a mechanical interface to lock the alignment.

Do you have pictures of your model?

[Species8472]

WORKHOLDING SOLUTIONS

Jergens #29306

A jergens relieved pin needs to be rotated so that the locating surfaces are perpendicular to a line that passes through the centers of the two locating holes.

It seems that to rotate to a specific angle (37.256°) the component must be free of any mates and will only rotate around the main assm. axis. Then move it back to where it belongs. You can drag around an feature like an axis but for whatever reason there is no way to do it accurately. Using collision detection is really useless because the parts never actually touch there's always a gap of a few thousands. It looks like if you need more then just cosmetic rotation you need to build the component in place.

[JerryFlyGuy]

Species, what would be the issue with using another plane, rotated from either the top or front planes to a specific angle and then mating to that plane?

[Species8472]

I suppose you could compute the angle of one of the facet faces on the pin and then compute the rotation angle delta from that face to the rotation needed to bring the locating faces to the proper angle. Then make a plane and mate that face to it. Is it just me or is that a lot of screwing around just to rotate something 37.256°?

Sorry but that's just plain broke.

[JerryFlyGuy]

Why not just put in a mate to the front plane that is 37.256deg? I guess I'm a bit hamstrung as I don't understand the application of the pin and it's use/install but to set something at a fixed angle is pretty simple.

[Species8472]

Why not just put in a mate to the front plane that is 37.256deg? I guess I'm a bit hamstrung as I don't understand the application of the pin and it's use/install but to set something at a fixed angle is pretty simple.

Like i said. You can delete all the mates then do the rotation around an assm. axis which will drag the component to some random location then move it back to where it belongs. That is the only place where you can enter a precise angle.

My problem was I stupidly mated the pin to its hole in doing so it made impossible rotate it with anything other then the cosmetic drag rotation. Now I got smart I delete all mates and relations after use as I have little or no use for them. If your part has no suitable surfaces to mate you have to use the rotation about an axis.

There should be a simple feature to just select an object, select a point, cylinder or circular edge of anything in the assm. and enter an angle. You sure there nothing like that?

[JerryFlyGuy]

I'd have to see the model to be able to get my head around it Species. I'm clearly missing something.

No, being able to mate an angle to a point isn't possible. Think of the mathematical limitations of that [or the edge of a cylinder etc] there is no 'reference or zero angle' to a point or cylinder. You have to put in an angular mate to another flat surface. The good thing is that every model has planes so you should be able to create a mate using these, I just have to get a 'picture' of your model in my head and what your trying to do.

Trust me, I've mated a fair number of things that were .. 'un-mate-able' in my years of running Sw.

[Species8472]

I'd have to see the model to be able to get my head around it Species. I'm clearly missing something.

You can grab it off Jergens web page, Link is in one of my posts

Think of the mathematical limitations of that [or the edge of a cylinder etc there is no 'reference or zero angle' to a point or cylinder.

None. Why can't something be there without being related to sometime else. That's how you get into this paradox.

You have to put in an angular mate to another flat surface. The good thing is that every model has planes so you should be able to create a mate using these, I just have to get a 'picture' of your model in my head and what your trying to do.

I got a kludge to get it done. thx

Trust me, I've mated a fair number of things that were .. 'un-mate-able' in my years of running Sw.

I find I spend way too much time mucking about trying to mate things. I have made alterations to other peoples assm. and never found a useful mate. In ALL cases you'll have to delete a few mates then make new ones to do what you need. They wasted time to mate them then I have to waste time un-mating and remating just to move something. I'm liking my slash and burn mates philosophy. It really makes SW usable.

[JerryFlyGuy]

Species, I've got the model off their site but don't see any issue w/ putting it into a hole and mating it there, I guess I need to see what model your putting it into to understand.

A point has no angular value, it is a point in space located in the cartesian coordinate [x,y,z] only. If you were to try and dimension between a point and a plane you would get a linear dimension perp to the plane.

Not sure why your having so many issues but guess w/out a clear picture of what your trying to do... thats to be expected..