DPRG List  

[DPRG] Servo Torque

Subject: [DPRG] Servo Torque
From: Ralph Tenny rften at swbell.net
Date: Wed Jan 14 23:05:02 CST 2004

If you can build a slip clutch, you can rig a beam balance that you
drive with the output of the clutch, and measure the torque directly.
That is, set the clutch at some degree of loading and run the servo.
With too light a load, the servo doesn't notice. If you set the load too
high, the servo stalls. Back down from that loading in small increments
until the servo just carries the load and you are close.

Is that enough, or do I need to elaborate on some details?

-----Original Message-----
>From: dprglist-admin at dprg.org [mailto:dprglist-admin at dprg.org] On Behalf
Of JDrumm9015 at aol.com
Sent: Friday, January 09, 2004 9:16 AM
To: Eric Sumner; dprglist at dprg.org
Subject: Re: [DPRG] Servo Torque

  This may not be "scientific", but it is an observation.
When I program a servo to move full range and hold it back with my
fingers, not stalling it, but just restricting it, it
feels like the same push all the way to the programmed end point.  I
realize my finger is not calibrated.

  I think that observation makes sense with what I know about servos,
feedback, and PWM.  The circuitry is trying to move the servo until the
output sense pot provides the feedback that the motion is complete.  My
observation of RC servos is they don't overshoot, and they change very
quickly.  All good signs of a correctly implemented control loop.  I
think what
my finger "observations" and a little hypothetical thinking
tell me, is the servo rapidly develops the maximum torque when
programmed to move.


John Drummond

DPRGlist mailing list
DPRGlist at dprg.org

More information about the DPRG mailing list