Re: Finding servo motor voltage without specs.

--- In CAD_CAM_EDM_DRO@yahoogroups.com, Jon Elson <elson@p...> wrote:

>
>
> tompiatt wrote:
>
>> How does one go about determining the voltage and amperage
>> of a servo motor with no specs available?
>
> As for electrical specs, you can attach a low voltage
> battery or DC supply, say 12 V, to the motor and read
> the speed through the encoder or even with a piece
> of tape on the shaft, and count how many times it slaps
> your finger. This establishes the Kv, or Volts/1000 RPM.

You can also determine Kv, by spinning the motor at a known
speed and measuring the voltage that it generates. A lathe
makes a convenient "motor spinner", the 3-jaw or a collet
is an easy way to connect the motor. I've also used a drill
press for smaller motors. Measuring the open-circuit voltage
this way is more accurate because there is no current, and
thus no I*R drop in the winding and/or brushes.

> You can make extremely rough guesses at the current
> rating of the motor by measuring the armature resistance.
> These measurements are very hard to do with a DVM, due to
> the low resistance, and the flaky connections of the brushes.
> You need to take many readings and keep the lowest one, usually.
> (A better measurement is made under load.)

Armature resistance can also be measured by spinning the motor.
Connect an ammeter in across the motor terminals (shorting the
winding). Measure the current as the motor spins, as well as
the terminal voltage (will be quite low). Subtract the shorted
terminal voltage from the no-load terminal voltage at the same
speed, then divide by the measured current to get the resistance.

Do the resistance measurement at a moderate speed - if the motor
is spinning at its rated speed, shorting it will cause a large
current to flow - possibly well above the rating.

>
> The Kt, or torque constant, is pretty easy to determine. You
> attach an arm to the shaft, and place weights on the arm.
> You apply current until it lifts the weight. You can then
> calculate the torque units/amp. A fish scale can also be used.
>

Actually Kt and Kv are related, and if you know one you can
calculate the other. The conversion factor depends on the units
in use, and I can never remember it. I usually derive the answer
when needed, as follows:

Assume I have already measured Kv. I spun the motor at 500 RPM
and measured 12.5 volts. (This is what you would get with a 90V
3600 RPM motor.) Based on the measurements, Kv is 25V/1000RPM.

Assume a power level, completely arbitrary, I use 1 horsepower
because I know the constants that go with HP. It doesn't matter
whether the motor is as big as a house, or a tiny toy motor, this
is all theoretical.

To get the assumed 1HP input to the motor, we need 746 watts input.
Assume that this is accomplished by applying 746 volts at 1 amp.
Again, the assumption may be ridiculous, but it's only theory, and
the 1 amp current makes things easier later.

Since Kv is 25V/1000RPM, that can also be expressed as 40 RPM/volt.
(1000/25 = 40). So at 746 volts, the motor would be turning at
746*40 = 28940 RPM.

So we've got this motor theoretically spinning at 28,940 RPM, and
putting out 1 horsepower. The constant for HP that I mentioned
earlier is 5252. Speed in RPM times torque in ft-lbs divided by
5252 equals power in horsepower. "RPM * torque / 5252 = HP", which
can be re-arranged to "torque = HP * 5252 / RPM". Plug in 1 HP
and 28940 RPM and you get 0.1815 ft-lbs torque. That torque is
at our assumed current of 1 amp, so Kt is simply 0.1815 ft-lb/amp.

There are other ways to convert Kv to Kt, and Mariss and others
probably have more direct methods that work in oz-in instead of
ft-lbs and horsepower. My method relies on me remembering
the one number, 5252, which I have stuck in my head from my
work experience with horsepower sized motors. It works for me...

John Kasunich