RE: [CAD_CAM_EDM_DRO] RE: Stepper questions

Hi all,

Dug through my files and found this scope photo of a stepper motor with step
pulses at 460Hz. I guess since I have the ELS on the bench (piled in a
corner at the moment), when I pull it out again, I can do some more
detailed tests to show some of the differences in phases and voltages. I
believe this was watching the current on the sensing resistor of the
LMD18245. With a 40 Volt supply and a high inductance size 23 motor rated
at 700mA and 10 Ohms resistance.

http://www.autoartisans.com/cnc/stepcurrent.jpg


What you can see on the above picture is that with the armature turning, the
winding current (yellow trace) driven by the applied voltage of about 40V
(blue trace) doesn't flow until about 1/3 of the step time has passed.
That's because of the back emf from the collapsing field from the previous
step and the induced emf from the motor turning.

Then, as time passes the current starts to build until it reaches maximum
based on the sensing resistor value relative to the reference voltage. At
that point the hardware starts chopping the supply voltage off and on to
keep the current at a steady value.

Finally, the step is over, and it's time to reverse the applied voltage to
the motor again, the current falls to zero.

Notice that full current only flows through the windings for about 30% of
the time. So for 30% of the time the winding will apply full torque to the
armature.

Now imagine the next step pulse happens about halfway through that 460uS
pulse time. Above the 't' in the word Current. Only about a third of the
maximum required motor current is flowing at that point and it's been
ramping up so overall torque is quite low.

Realistically, because the current was lower, the back EMF time is also
lower so the forward current will start sooner. But I think the scope
capture demonstrates why the stepper motor loses torque the faster you step
it.

BTW, using Ohms law I^2R = P where R is the DC winding resistance you can
see that the area under the yellow trace represents the current consumed
over time flowing through the winding and that's what generates the heat.

When the motor isn't moving, the current stays at that maximum current
point (say 1 amp in a 1 Ohm winding is 1Watt). But once you start stepping,
that scope capture shows 1A for 1/3rd the time, and roughly half that for
another 1/3rd. ((I/2)^2R) + (I^2R)/3 = 0.58W. In other words, power has
dropped by almost 1/2.

Cool eh?

John Dammeyer

>
> On Wednesday 03 May 2006 11:17 am, Alan Marconett wrote:
> > That's an interesting point! I don't recall reading
> anything like that, I
> > could be wrong. Basically the windings do not interact as
> far as I know.
> > Actually, one would think that "turning around" a winding,
> as in a bipolar
> > drive would take longer. The field would have to collapse
> before it could
> > build up in the opposite direction. Therefore wouldn't a
> unipolar drive,
> > with its four windings have the advantage?
>
> Ah, but the key here is "rate of change" -- the faster you
> can get that
> magnetic field to change, the faster that motor is going to
> respond. And in
> a unipolar setup, you're looking at either energizing a
> winding in one
> particular direction or not. In a bipolar setup, on the
> other hand, you're
> effectively looking at reversing the direction of the applied
> voltage, which
> means almost having the effect of twice the applied voltage,
> therefore
> tending to increase how fast those magnetic fields are going
> to switch
> themselves around, since higher _difference_ in voltage
> gives you more
> current flow and a faster rate of change in the magnetic field.
>
> Or at least that's my understanding of it. :-)
>
> --