Motor parameters

[petercohen offline]

Hi robot experts, I have some questions about motor specifications.

Generally speaking, given two servo motors of similar specifications, one has higher gear ratio than the other. Which one is better? What are the advantages/disadvantages of having higher gear ratio than lower gear ratio?

What is stall torque in terms of practical use? Give two servo motors of similar specifications, one servo has a lower stall torque than the other. Which one is better? What are the advantages/disadvantages of having lower stall torque?

Thanks.

[PaulL offline]

I'll take a stab at it..

Higher gear ratio means faster speed, but less torque. Better depends on your goals and the intended load. Heavy bots need lower gear ratios which translates to more torque, but slower speed (generally speaking). Lighter bots would use higher gear ratios for faster moves (if, ex, combat is your intention). That said, mass moves mass (ex, what good is a light bot that delivers quick but low mass "smacks" to a heavier opponent?), so there is some optimal point of gear ratio for any given application.

Stall torque is about the only somewhat meaningful spec all manufacturers post about their servos, in order to convey the strength of the servo.

Personally, I've chosen higher torque (lower gear ratio) servos versus lower torque (higher gear ratio) for my application, but I have a significant amount of weight involved (but then, I'm not concerned about combat or speed beyond reasonable need). It depends on what you want / need.

Lower torque as a result of higher gear ratio means higher speed. If you need speed, the torque trade off may be worthwhile, but fast is useless if you can't move the intended load (ex, legs holding up your bot).

This is all in reference to a same power motor using various gear ratios. That said, once you start talking about a different motor with different RPM / torque specs (the motor, not the servo, which is the result of a gear train), two servos become harder to compare on gear ratios, but at least torque is fairly well understood.

I should mention, optimal configuration would be appropriate servos for various aspects - ex, an elbow may not need to use the same servo as that in a leg.

Paul

[petercohen offline]

Thanks PaulL. What if the servo motors have very similar torque but different in gear ratio and stall current? How these two parameters affect the performance in real life use?

[PaulL offline]

If the torque is the same for two servos, but one has a higher gear ratio, both definitely have different motors. Timing on a DC motor dictates its speed, some DC motors even have adjustable timing (ex, Plettenberg, I used to have a few of these, incredible motors). The motor with the higher gear ratio would have a lower speed (less timing advance), where the motor with the lower gear ratio would have a higher speed (more timing advance).

In all, it's difficult to tell which would be better long term, motor performance can vary tremendously (timing, winds, gauge of wire used, brushes, etc, etc). I would tend to lean towards the lower gear ratio if faced with the same scenario - the closer to the motor's RPM a servo operates, the more efficient it should be. But, this means the motor would probably be operating at a higher RPM - this is fine if the motor is built for it, but if a motor is at the far end of its operable limits, it won't last as long as a motor in more conservative usage.

I know of people that have ordered custom sealed-can motors, with special winds, timing, etc, and as they chose wisely, they ended up with decent motors for the application - but this costs a small fortune. The cost gets cheaper per motor if you order more, but who has that kind of money?

The results depend on the motor itself - which are hard to get specs for, are mass produced, and quality can vary greatly.

But as I said, the closer to the RPM a servo operates (lower gear ratio), the more efficient it should be (which should indicate better design, in theory). Steeper gearing is adapting a motor for a task other than what it is best suited for. The ideal servo has no gearing, but that's not practical - it's cheaper to mass produce typical DC motors and gear them. It's cheaper to go with a "default configuration" motor that would be readily available (versus custom ordering specialized motors closer to the RPM, efficiency, power needed) and gear it steeper.

Still, a cheap motor is a cheap motor, even if it uses less gearing. So, it all depends on the motor.

Brushless motors are best (generally speaking - but could be much higher RPM, which would require steeper gearing - the end result could still be decent, depending). Coreless motors tend to run hotter (esp in PWM applications, like hobby servos). Skew-wound motors tend ot have more consistent RPM at low speeds, and tend to start at a lower power level.

Typical motors found in servos are run-of-the-mill 3 or 5 pole brushed DC motors, sealed can, fixed timing, non-replacable brushes, bushings on each end, plastic brush housing, stamped can, typically non-exotic magnets. They're not really made to last, and if operated close to their limits, will tend to fail sooner than later.

If you're on the fence about two servos close in torque, I'd buy one of each, and see how they work, beat them up, and see which one works best (cooler operation, which should mean longer lasting and more efficient).

One more note - a hot motor, pushed way out of its typical range, can heat up and destroy brushes AND magnets. Magnets don't like to get hot, and can be ruined in an overheated motor. Neo magnets don't like heat - I think they become problematic something like 200 degrees farenheit, I forget. All the same, heat is bad.