High Torque Gearing for Large Robots

  • I'm trying to understand how gearing works for high torque robots. They often use harmonic drives, which although not backdrivable, can achieve very large reduction ratios. Now, how they trade torque for speed and vice versa is quite confusing. In the motors I have looked at, such as the T-MOTOR V602, torque, RPM and power all increase in unison.


    In research papers, the spec sheets for robots often list both the maximum torque that can be achieved with the gear ratio along with its highest RPM. The motor obviously cannot do the same simultaneously as it would require 2Kw+ per joint. The question is how are they able to trade one for another in what seems to be a single harmonic/planetary drive?


    Image from robot JAXON3-P (doi : 10.1109/Humanoids43949.2019.9035049)

  • AD
  • Go to Best Answer
  • Lemster68

    Approved the thread.
  • Depends on the type of motor. For some motors (BLDCs, for example), just increasing power will increase both RPM and torque. Steppers, OTOH, lose torque at higher speed, but increasing current will increase torque with no effect on speed.


    The harmonic drives are generally installed b/c most servo motors are better at generating speed than torque, and because it allows finer position control. Servo motor systems are generally better at controlling their speed than their position, but by having dozens or hundreds of motor rotations equate to a single degree of axis motion, through gear reduction, basically allows you to control position by controlling speed.

  • Quote

    I'm trying to understand how gearing works

    https://science.howstuffworks.…ngines-equipment/gear.htm

    https://www.youtube.com/result…h_query=how+gearing+works


    Quote

    In the motors I have looked at, such as the T-MOTOR V602, torque, RPM and power all increase in unison.

    of course... the higher the voltage, the higher speed is possible, the higher current is possible, the higher power is possible.


    but for same motor at same voltage one can get different speed or torque by varying duty cycle for example (DC motors) or frequency (AC motors). but the values are in some range.


    to get different range, you need not just motor but also a gearbox.

    1) read pinned topic: READ FIRST...

    2) if you have an issue with robot, post question in the correct forum section... do NOT contact me directly

    3) read 1 and 2

  • The harmonic drives are generally installed b/c most servo motors are better at generating speed than torque, and because it allows finer position control. Servo motor systems are generally better at controlling their speed than their position, but by having dozens or hundreds of motor rotations equate to a single degree of axis motion, through gear reduction, basically allows you to control position by controlling speed.


    but for same motor at same voltage one can get different speed or torque by varying duty cycle for example (DC motors) or frequency (AC motors). but the values are in some range.


    to get different range, you need not just motor but also a gearbox.

    I am aware of how gearing works in that in results in a torque increase with a rotation speed decrease. I plan to use 10:1 quasi direct planetary gearboxes with helical gearing. This will make it highly backdrivable and enable prioperception. What I don’t get however is how they are able to fine tune for torque versus speed after the gearbox has already been implemented.

    • Best Answer

    They don't....


    Ratio does not change, it is a fixed gearbox, not automatic transmission. Axis is sized so that max torque, max speed and max power are never exceeded. You always work with less than axis is capable of, it is just that at some times you may approach one of the limits. And if you not only approach but actually reach or exceed it, something is supposed to sense it and respond. Usually with a fault message.


    Motor can reach some max speed... Or max torque... Or max power. But not all at the same time...!


    Max speed is possible when there is no load. Great but not very useful. We need robot to move some load.


    Adding load reduces possible acceleration rate. Also may also reduce max speed as a consequence of limited motor power.


    And when said load is sufficiently large, axis will no longer be able to move at all. This is stall condition. In this case motor produces the maximum torque and the speed is - zero. Also bot useful to waste energy to not move. We want robot to move, not act as a statue.


    How you want axis to behave will determine type of control. This could be position control, or speed control or torque control..


    You can control one thing only. Others are limits at best...


    You cannot control both positions and speed for example as they are contradictory.


    In most cases axis is using position control. If you tell axis to move to some position, it will do so... If possible.


    Motion planner generates setpoints that axis follows. Or at least tries...


    To ensure desired response PID loop is used. Actually it is often only PI loop - this may not be as capable as PID but at least it is inherently stable (it does not introduce instability).


    So motion planner tells drive what speed limit and target position are, the drive does the rest.

    1) read pinned topic: READ FIRST...

    2) if you have an issue with robot, post question in the correct forum section... do NOT contact me directly

    3) read 1 and 2

  • Motor can reach some max speed... Or max torque... Or max power. But not all at the same time...!

    Thanks for the answer… I suspected this is was the case but I was seriously hoping it wasn’t. From having seen other motors, I knew that Max-Power, Max-Torque and Max-RPM were separate entities but that somehow lightweight drone motors adopted a different behavior. In the case of the drone motors, where they almost always only list Max-Power, is there any calculations that can enable you to find the maximum torque and maximum RPM? Having to manually test each and everyone of them would be expensive and time consuming.

  • There is a "kind of" solution, which is implemented in elliptical trainer machines. The idea is based on the prior knowledge of leverages at particular point, so if leverage is small, we need less torque and vs versa. I believe that similar thing is used in human joints.


    ===

    robotics.snowcron.com

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account
Sign up for a new account in our community. It's easy!
Register a new account
Sign in
Already have an account? Sign in here.
Sign in Now