chip3 actuator design updates: rough initial sketches and planetary calculations [updates]
So I spent some time today sketching out the first initial designs of the actuator I'll be designing over the summer just to get ideas down.
Basically, the actuator is going to be a single Falcon 500 motor with a single-stage planetary gearbox on the end of it. The reason we're going with the Falcon 500: https://www.vexrobotics.com/217-6515.html
Is for a few reasons. First, this motor is a brushless motor that outputs 4.69Nm of torque at stall. That's a nuts amount of torque. Second, this motor has an integrated motor controller and CAN-Bus communication network, so that's less space the system will take up and less weight too.
Now I just got an interesting idea, what if we used the shaft of the falcon 500 as the start gear of the planetary that gives it a much smaller diameter and a higher ratio, a ratio of about (0.275591+5)/0.275591 = 19:1... that's actually a really good idea, I didn't even realize.
Maybe we'll try that out, that would give us a max torque of a little less than 100Nm, and it reduces the complexity, not bad. Here's the video I used to calculate planetary gear ratios.
With the 19:1 ratio the final actuator specs will be:
Now this system has significant advantages over the old chipTwo system. That system required the motor, the motor controller, wires to the encoder/etc, so the wiring was so complicated on that thing. Second, this system is much lighter, it doesn't need the heavy VersaPlanetary gearboxes we have the lighter 19:1 ratio (potentially) 3D printed designed actuator. And the ROS node for this actuator should be very easy to develop, the system is already compatible with a Jetson Nano, we have a nano, the code is mostly written we just need to fix it a little.
Finally, it's cheap. The Falcon 500, being for FRC, is really not too expensive of a motor, and with a 3D printed housing and rest of the actuator, the whole actuator will probably cost 150 dollars. Now not to mention we don't need the expensive 100:1 VersaPlanetaries anymore, so the bulk of the cost of the robot will be 150*12=1800 dollars, +100 for the Jetson Nano, we're mostly done at ~2k. Estimating already that chip3 will be 2500 dollars which is 1/2 of what chiptwo costs and, guess what, 1/10 of what spot costs.
Oops! :) Now let's see if we can actually design an actuator like this!
And here's the formulas for the gear calculations: https://www.engineersedge.com/gear_formula.htm
See the link for variable definitions and further explanations of gear math.