Altering part properties breaks assembly?

joshua_foxworth

I am building an app that is integrated with onShape and the app pushes values to onShape variables. I have used those variables to set the length of a strut in a test truss. I built the truss in an assembly and everything was fine. I then updated the value of the length of the struts and somehow the mates on the truss became over defined.

Any ideas as to why this would happen? I can't even figure out what's wrong with the mates.

https://cad.onshape.com/documents/97434099131350e534d87e91/w/f983dd69e4287aef89e7ff21/e/f5129967f8d66322d1536a7f

philip_thomas

Hello Joshua!

I took a look at your assembly. There wasn't anything specifically wrong with it - but i made a few changes.
Notice the much cleaner assembly structure and the significantly fewer mates at any assembly level. In addition, you will see that there are far fewer components in any assembly (by reusing sub-assemblies).
With regards to your mates - for the mechanism you have built, there is a single positional solution for any strut length (or ring diameter etc). There will be no motion period - its simply not possible. Now if you replace the revolutes with spherical (ball) mates, it starts to get interesting (Branch 'Top and Bottom Free').

Have fun!

https://cad.onshape.com/documents/76bbdb2478177d2a4195c627/w/a79be76bf737c24cb51ec13e/e/d8e1f03873bb39758bd70382

joshua_foxworth

Thanks for the help. Are you saying that if you use fastened mates then onShape creates a single positional solution and therefore changing any parameter - like length of the strut - breaks that mate?

philip_thomas

@joshua_foxworth - I am not sure exactly what you're asking.

For the mechanism you have designed - for any given strut length and revolute mates, there is only a single solution (ok strictly speaking two, but now we are talking about perfectly spherical chickens in a vacuum). There is no motion in this situation. If ONE of the struts was a different length, many of the mates would break because the system is unsolvable. To introduce motion or two allow struts of different lengths, you would need to introduce additional degrees of freedom into the mates (spherical). When mates 'break', it's either because their reference is now missing or the system is unsolvable.