Why does final Sliding Mate cause assembly to be overdefined?

roland_hauff

I can drag the plunger and it animates properly but gives the error when I add the final sliding mate.

https://cad.onshape.com/documents/a93d84c4124a05729b1f6469/w/e29bb301f668d024392945d0/e/63ddb5300d695d02781aed2a?renderMode=0&uiState=65d56d9fa6aceb6fbe3e309e

eric_pesty

You have too many constraints stopping things from sliding "out of plane". Get rid of the limits on your cylindrical mates, and switch the first two to "revolue":

https://cad.onshape.com/documents/d27b8fdab628a00a7061b409/w/a6edc118892b03fe0a3c9de4/e/795f4a595e47c8399152a09f

eric_pesty

roland_hauff said:

Ok, I think I get it.  The main point is ONLY 1 rotary mate should limit motion in the Z direction.  I tried making the first Revolute, and the last two Cylindrical with no limit in Z direction, and that works as well.

Thanks #eric_pesty!

Yes you got it, the key is to avoid having multiple mates constraining the same things. You can "sometimes" get away with it when the result is the same (i.e. two things constraining in Z at exactly the same "depth" but it often causes issues as there is always some tolerance in computer math so best to avoid it.

eric_pesty

You have too many constraints stopping things from sliding "out of plane". Get rid of the limits on your cylindrical mates, and switch the first two to "revolue":

https://cad.onshape.com/documents/d27b8fdab628a00a7061b409/w/a6edc118892b03fe0a3c9de4/e/795f4a595e47c8399152a09f

roland_hauff

Well, that is very interesting.  I started out with the first 3 being Revolute with no limits.  Then tried adding limits as forum posts seem to imply that too MUCH freedom can cause problems.  Then switched all 3 to Cylindrical and at least got motion.

How did you choose to make just the first 2 Revolute and leave the third cylindrical?

roland_hauff

Ok, I think I get it.  The main point is ONLY 1 rotary mate should limit motion in the Z direction.  I tried making the first Revolute, and the last two Cylindrical with no limit in Z direction, and that works as well.

Thanks #eric_pesty!

eric_pesty

roland_hauff said:

Ok, I think I get it.  The main point is ONLY 1 rotary mate should limit motion in the Z direction.  I tried making the first Revolute, and the last two Cylindrical with no limit in Z direction, and that works as well.

Thanks #eric_pesty!

Yes you got it, the key is to avoid having multiple mates constraining the same things. You can "sometimes" get away with it when the result is the same (i.e. two things constraining in Z at exactly the same "depth" but it often causes issues as there is always some tolerance in computer math so best to avoid it.