Hinges in the SPH - bug or feature?

[maddog59]

I'm not sure if this is expected behavior or not, and I couldn't seem to find a reference to it anywhere, so I wanted to put it to the community and developers for input:

I built a rover in the SPH; used mirror symmetry to attach two G-11 hinges to the body, so I could stow XL3 wheels forward for launch/landing, and set them to swing out 90 degrees for use  (Min: 0.0; Max: 90.0; Target Angle: 90.0, the 'stowed' position). In the SPH , changing the target angle from 90 to 0 to 90 seemed to work fine to swing them forward and back, so I added the Toggle Hinge function to an Action Group.  But when I tested it on the runway, only the hinge on the left side worked properly ... I could see the one on the right jiggle for a moment, but it stayed in the stowed position.   
The only way I could get it to work properly was to attach the hinges separately using radial symmetry and setting the hinge on the right side to Min: -90; Max: 0; Target: -90. 
That makes a certain amount of sense, I suppose, but wasn't intuitive (at least to me) given the other behaviors of parts in mirror mode, and I lost a few clumps of hair figuring it out (no, I'm not the hottest engine on the shelf),  so  any constructive thoughts are welcome.

[InfernoSD]

It's not expected behavior. Mirror symmetry is ideally supposed to be a perfect mirror reflection of a part or assembly. It doesn't always work out that way, especially with robotic parts. Likewise, it's not intended behavior for anything to behave differently in-flight than it does in construction. In any case, I haven't seen and can't recreate your specific problem with hinges.

What this issue would boil down to would be the exact order of operations you followed to place those hinges, including when you used specific symmetry modes, what chain of grandparent parts are above it, and to which nodes all those parts are attached. A picture or description might help, but even with a picture it may be hard to recreate the exact circumstances again.

I'm assuming you've already checked for obvious things, like checking the PAW in-flight to see if the motor is staying enabled on both hinges.  It's happened to me before where one hinge ran out of electricity but not the other. It's also worth asking if you use the Locked setting for those hinges, which I find has some peculiar behaviors.

[maddog59]

2 hours ago, InfernoSD said:

What this issue would boil down to would be the exact order of operations you followed to place those hinges, including when you used specific symmetry modes, what chain of grandparent parts are above it, and to which nodes all those parts are attached. A picture or description might help, but even with a picture it may be hard to recreate the exact circumstances again.

I'm assuming you've already checked for obvious things, like checking the PAW in-flight to see if the motor is staying enabled on both hinges.  It's happened to me before where one hinge ran out of electricity but not the other. It's also worth asking if you use the Locked setting for those hinges, which I find has some peculiar behaviors.

I suspect you're right in that I ran afoul of something I messed up while building them; the problem occurred right after I'd 'launched' it to the runway as part of my effort to check the functions, and there was always plenty of EC for the wheels and hinges. I was using the Lock setting, w/ an AG to lock/unlock  during the mission; I seem to recall observing some weirdness when I used it. I try to only change one thing at a time when I'm trouble-shooting.

My final solution was that each individual hinge needed to have it's auto-strut setting  checked - some worked when set to the root; one needed to be set to heaviest part, and a couple needed to be set to the grandfather part.

This was my first attempt at using the robotic parts on a rover, so there were a lot of misfires as I built and moved and changed things around.  And since my original post I've built a couple more rovers with similar functionality and the hinges behaved as expected. 

Thanks for taking the time to respond!