Direct Drive Four Motor System

[Matt Adams]

Originally Posted by Collin Fultz

I am looking at possible drive trains for next season. One thought I had was to run each CIM motor directly into each wheel (after running through a transmission). With a 12.75 reduction, you get about 9 ft/s with 6" wheels.


If you're having problems visualizing, think crab drive system without the wheel assemblies rotating and the motors running the same direction the wheels are turning (sorry I can't explain it more).

Does anybody have any experience with this type of drive? What were your problems?

Thanks!

Collin:

I have one pretty significant concern about using direct drive on each wheel - load sharing. In an ideal world of equal force and weight distribution, direct drive isn't a bad idea and it's not a bad idea in the 'real world' either, others will testify that it has worked for them!

However, take for example a robot that doesn't have weight evenly distributed, say it's 70% in the back and 30% in the front. You will need to set your ratios based on the largest amount of torque you'll see - in this case, the back wheels. Since you want to ensure the front and back wheels are spinning at the same rate (roughly, it's an open loop system)... you'll have to gear the front wheels to this same ratio. You will essentially be running slower than you would with a combined motor system.

To put this in a better frame of reference (numbers!)...

Let's say it takes 70 inch pounds of torque peak to turn the the back wheel and 30 inch pounds peak on the front wheel, and your peak current draw is at 10 inch pounds of torque from your motor. You will end up gearing both sets of motors at a 7 (plus some) to 1 ratio because you don't want your breakers to trip in a pushing match.

However, if you had these two motors combined on a single shaft that was linked via chain to two separate wheels, things change. Though each motor would still put out peak torque at 10 inch pounds, their combined torque would be 20 inch pounds at peak current. The max torque this pair of motors would see is 100 inch pounds. Thus, you'd gear the motors at 5 (plus some) to 1 ratio. This is a significant increase in speed (the difference between slow and average or average and fast!!)

Another case is assuming that your robot is lifted by another so it's only on it's back wheel- you now have two motors in the air doing no work, and you have the two rear motors seeing almost twice the load they were geared for - you'll be tripping breakers. That's no fun either!

For the two reasons above, I don't design direct drive robots. I also don't like the idea of creating 4 individual gearboxes, but it's mostly the load sharing benefit that keeps me sticking with motors in parallel. Others do direct drive and it works for them.

Fighting Words: It comes down to a mater of weight, and I think that if you're smart about your chain size (#25) and sprocket materials (aluminum) you'd be hard pressed to say the weight difference is enough to over come the disadvantages. I don't considering creating a good chain tensioning system upfront and spending a few minutes to tighten it every few matches at the first regional to be terribly inconvenient vs. the tolerances and assembly requirements of building 4 robust gearboxes.

Just some thoughts and fighting words,

Matt