I've got a couple DC servo motors that are in great shape but old
enough that Electro-Craft no longer has any info on them. So I would
like to know if there is a way to figure out what max voltage and
current might be. Can using the motor as a generator be used as a way
to determine the approximate proper voltage and current?
On Fri, 03 Oct 2008 15:45:37 GMT, firstname.lastname@example.org wrote:
I'd first compare the motor to motors of similar size and construction
to get an estimate of torque rating and max RPM.
Then experiment using the relationships
Ke = RPM/volt
Kt = lb*in/amp
to get values for Ke and Kt.
Either rotating the motor at a known speed and measuring the output
voltage, or powering the motor with a known voltage and measuring RPM
will give you Ke.
To determine Kt you'll need to control torque and measure current, or
vice versa. The easiest way would probably be to drive the motor with
an adjustable current source with a torque arm and weight attached to
the shaft. Adjust the current to balance the weight with the arm
Keep in mind that DC motors will have both a continuous and a peak
current rating. The continuous current is mainly a function of winding
resistance and the ability of the motor to dissipate heat. The ratio
between continuous and peak current is generally an indication of
magnet material and the resistance of the magnets to demagnetization
as a result of high current in the windings. You might see a ratio of
3:1 in low end motor, while a high performance, low inertia motor
might be as high as 7:1 or more.
Thanks for the reply Ned. These motors came from a lathe that
originally had a Bandit control. Later these same lathes were equipped
with a Fanuc 5T control and Fanuc motors. If I short the wires on the
Electro-Craft motors and spin the shaft by hand there is a lot of
drag, but the Fanuc motors have very little drag when the shaft is
spun by hand. I'll try powering them both with the same voltage and
compare speed and then I'll measure the current drawn with the torque
arm setup you described.
On Sat, 04 Oct 2008 05:27:58 GMT, email@example.com wrote:
PM servo motors are designed for equal performance in
either direction of rotation. This means that they are
necessarily low speed (<5000 Rpm) machines because the brush
angle cannot be offset to reduce high speed sparking. Typical
rated power is at 3,000 RPM. so the voltage needed to produce
3,000 RPM should be pretty close to the motor rated voltage.
Check the stalled resistance. Ohmeter readings can be very
unreliable - measure the effective resistance by monitoring the
terminal voltage when passing 1 amp or so through a stalled
shows the average relation between motor rated power,rated
voltage and typical winding resistance. It will not of course
give a precise answer because individual motor designs can be far
from average but it will be a LOT better than eyeball estimates.
For a closer figure you need to guesstimate the motor
maximum current. Copper has a temperature co-efficient of 0.4%
/deg C. If you now increase the stalled current (in stages to
allow time for the temperature to stabilise) until the measured
R increase by 30% the winding temperature will be about 100 deg
C which is pretty safe - High temperature servo motors can
withstand up to 150 deg C. This process is not as long winded
as it sounds - you can usually get a pretty close estimate with
no more than two trial currents. For best accuracy use the
brushes to feed the current through the armature and measure the
voltage drop directly on the relevant commutator segments.
Jim - thanks for the great info. Question about the max current
process: if the motor is stalled during this test, you'd only be
heating one winding - is that going to give a good current value that
can be used when all windings are in use (running)? In other words, the
current required to give a certain temp depends upon the cooling and the
cooling will be different when running.
That will tell you rpms per volt. Stall current against a dyno will give
you torque per amp. There's your characteristics.
Now, you have no way to tell what maximums are appropriate, as these have
nothing to do with what the shaft does versus what the wires are doing.
However, the shaft size roughly indicates the max torque, compare what you
have to similar motors in catalogs to get a rough spec. Max voltage has to
do with the insulation and other build features, max rpms and max power
likewise with heat dissipation and the like, you can't measure these things
Many DC motors are 90 VDC or 180 VDC for a max speed of 1800 or 3600 rpm,
so if you know or guess max voltage or speed you can determine the other.
Eric, I'm having the same problem with a motor I'd like to use as a
servo on my lathe feed. I've come to these conclusions:
- the max current is determined by the motor's insulation. It has a
class (A, B, F, or H) which sets the maximum temperature that the
insulation can handle (for a given life). Without knowing the rating,
you (we) would have to assume the lowest (class A), which has a
temperature rating of 105C. The max current, then is that which causes
the motor to heat to 105C (loaded).
The empirical approach might be: apply a voltage that gives a
"reasonable" no-load speed. Apply a load & measure the current. Allow
the temperature to stabilize, monitoring to avoid > 105C. Increase the
load & repeat until 105C is reached. This is the maximum (steady) current.
There is also a maximum peak current, which will demagnetize the
magnets. I have no idea how to determine that.
- I'm having a bigger problem with the maximum voltage. It seems that
the voltage is more of a dependent variable. I.e., a function of the
speed desired and load. There is a maximum speed based upon mechanical
limits & the maximum voltage is what produces that speed, with a given
load. So, under the maximum load (i.e., for the maximum current), the
maximum voltage will be greater than the maximum voltage under no load.
There is a voltage limit based upon insulation break down, but I think
that it is far greater the the speed limiting maximum voltage.
I'm not very confident of these conclusions, so I'd be glad to hear some
Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here.
All logos and trade names are the property of their respective owners.