Re: [CAD_CAM_EDM_DRO] query - 2HP PM DC motor controller - pointers please

tedinoue wrote:

>I recently purchased a 2HP PM DC motor, hoping to build/buy a
>controller so that I could use this on a small lathe. For those
>interested, it's a Leeson motor, designated 120v 14A, 4800 RPM.
>Model C42D340T7A. I plan to reduce the maximum speed then feed it
>into my existing lathe drive system.
>
>Several questions:
>0) the motor doesn't have any designations about reversability. I
>know electrically it works, but I'm not sure if the brushes are
>rated for reverse motion. As I manually rotate the shaft, the
>brushes sound smooth in one direction and click lightly in the
>other. Should I assume that the motor is designed for one direction
>rotation only?
>

Does the nameplate say CW or CCW? IF not, it probably can run both
directions. After running
in the 'click' direction for a while, it will stop clicking, as the tiny
sharp edge on the brushes
will be worn off.

>1) I've seen a lot of postings discussing 90v DC from 110v AC and
>180v DC from 220v AC, but little to nothing on driving a motor rated
>for 120v. I'd get plenty of power out if I ran it at 90v, so I'm ok
>with a controller that does the basic 110AC -> 90DC conversion. Any
>reason why this wouldn't be suggested?
>
>

If you full-wave rectify 120 VAC with a capacitor input filter, you get
about 167 V DC.
Chopping that with an IGBT transistor will provide 0-120 V (with
suitable pulse width
control).

>2) If I do run it at 90VDC, that's 3/4 of 120VDC, so would I simply
>then treat the controller needs as 1.5HP / 90 VDC and 10.5Amps?
>
>

No, you can still use the full 14 A capacity of the motor. A 90 V
controller will limit the
top speed, only, of the motor. Voltage is proportional to speed,
current is proportional to
torque.

>3) I've no problem making a tach circuit to provide closed loop
>speed feedback, and from what I can tell, a PWM circuit will provide
>the best performance. I've made my own PWM DC servo motors before on
>a small scale using encoders and DC gear motors. Is controlling a
>large, PM, brushed motor of this sort simply a matter of scaling up?
>
>

Yes, but there are things to watch out for with bigger drives. IGBTs
are the most desirable
device, but they must never be allowed to run in the linear mode. I
suggest using the International
Rectifier 2110 - 2113 series of driver chips, made expressly for power
FET and IGBT transistors.
A magnet and hall effect sensor can be your tach pickup, feed that to a
555 timer and integrate
the constant width pulses from that with a simple RC, and you have an
analog tach voltage.
Feed that to one side of an op amp, and the other side with a voltage
from the speed pot, and
you have a speed error signal. Feed that to a comparator with the other
input being fed by a
triangle wave, and you have the PWM signal to drive the IGBT driver.

One thing missing from this is a scheme to extract energy from the motor
to slow it down.
So, what is described above is a one-quadrant control. The advantage is
it only has one
expensive power semi in it. You need a double pole double throw switch
to reverse the
motor. A 4-quadrant control can also extract energy from the motor to
slow down, and can
reverse without the switch, but you end up with FOUR power semis. Of
course, a resistor
can be switched in with a relay to slow down, that's probably the
cheapest way to go.

>4) not having experience in higher voltage/power design, I'd
>appreciate a pointer to a web site with some designs. I'm not
>looking for someone to give me a tutorial on this forum, and I've
>done a lot of web searches but they've turned up fairly cold when it
>comes down to real designs for these types of controllers.
>
>

Actually, my PWM servo amp could do this, but it is a bit of overkill. See
http://pico-systems.com/servo.htm
<http://pico-systems.com/servo.html>for some more info. I don't have
the schematics online,
maybe I should do that.

Jon