The weight of the switch will not matter as long as you’re just looking for the perfect distance, because when balanced, it will be directly below the pivot. If you are interested in calculating the range to get close enough, it will act like a robot of 93 lb * 26 in / 48 in = 50.4 lb at the center. If you do try to do the “close enough” calculation, it will also matter if each robot is hanging straight down or clamping the pipe tightly, and if clamping tightly, how far down its CoG is.
Here’s a geogebra model I made:
If the robot is hanging straight down ("pendulum’), rather than being clamped tightly (“fixed”), does the location of the CofG matter? The model that gideonrab made and linked to is interesting, but I don’t believe it makes any distinction between a “pendulum” bot and a “fixed” bot. I am having difficulty with how a “pendulum” robot versus a “fixed” robot would affect the balance.
No, I don’t see how it matters if a pendulum. This is what I meant by:
It does matter (significantly!) whether or not the robot is fixed or free hanging. If all the robots are fixed, the COM of the assembly does not change as the structure rotates. On the other hand, if one of the robots is free hanging, the COM does change as the structure rotates since the free hanging robot moves with respect to the assembly. In particular, the geogebra simulation assumes all the robots are fixed to the bar.
Another interesting result of this is that with a fixed bot, as the bar swings, the COM of the robot moves closer to below the pivot, so fixed bots will be better at leveling the free hanging bots. Also, having a bot whose center of gravity is closer to the ground is better.
If you’re using SamFuch’s simulator, enter a height offset of 0 to simulate a bot which is free hanging with a fixed pivot.
I think when @BCR-Jim says “pendulum” he means “free hanging”, as opposed to “fixed”. As you say, “hanging straight down”, as opposed to “clamping the pipe tightly”.
That’s exactly how I read @BCR-Jim’s post. If hanging free/pendulum style, the vertical distance between the hook* and the CoG doesn’t matter. In your phrasing, for a pendular/free-hanging robot, the height offset is zero independent of the separation between the hook* and the CoG.
*hook - this may not strictly be a hook, but a hook with a non-slip coating is the simplest case of this type of hanger.