I am a wheelchair user and I am trying to build a ball for my dog that I can throw for him at the park, where he will return it 3/4 of the time, but for that other 1/4 of the time, I want to be able to drive it back to me.
I brought a sphero mini, as I was thinking of printing a flexable TPU outer cover with a lattice to create shock absorption. But I can tell the ball wont have enough touque to deal with it. So I thohght I would try the Bolt, that seems like it would have the most power motor. Am I right?
Does anyone have any advice or expertise they could help me out with?
We actually have a cover accessory for outdoor use, but it’s more for traction and some scratch protection than real shock absorption.
- The biggest issue is impact force on landing. Bolt is tested for a 1 meter drop, equivalent to a fall from a table in a classroom environment. Although there are many instances of various polycarbonate shell Sphero models surviving much larger falls, they’re really not designed for being thrown. If you can design a sufficient shock absorbing cover, maybe you can work around this problem.
- Next up is uneven surfaces and the diameter of the ball. The larger you go with the diameter, the smaller the angle formed when it meets an obstacle, as shown by the following diagram comparing mountain bike wheel sizes (Thanks REI for the image). To a ball the size of a Bolt, a tuft of grass is a high angle obstacle. If you’re in dirt it may work well enough, but grass is pretty tough for driving.
- The max torque applied to the shell (and anything you put around it) is determined by the mass of the internal components and the CG offset from the center of the ball. It doesn’t really tie all the way back to the max torque of the motors themselves, as they’re limited by the geometry and mass of the robot in the context of rolling. Bolt does have a higher torque at 90 degrees pitch than Mini, so it would be your best option if you decide you want to try this. The thickness of the cover will affect the overall dynamics by changing the mass and the attack angle, so you can do the math to investigate whether there’s an optimal thickness of your design. If the dynamics change too much the control system tuning may not be a good match for the system anymore, so that’s another unknown.
It’s certainly an interesting project if you can add enough shock absorption. I’m curious about how an extra thick cover would work in practice.