# generate 3 phase delta?

Discussion in 'Electronic Basics' started by BobG, Dec 16, 2005.

1. ### BobGGuest

I have a small ac motor.. I looked up the part number... globe
75a120-2.. 3 wires... datasheet says its a delta connection. I'd like
to understand how to drive this thing... seems like it would be simpler
if it had 4 wires and was connected wye.... but its not, so here I am
asking for ideas. I guess I need 3 'power amps' that will put out 3
phase sin waves. I guess the clever part is generating 4 or 5 bit sin
waves using pwm? I guess this is just like a synchro... will work at
10hz or 1hz or .1hz? The magnet just follows the field around?

2. ### John PopelishGuest

You will need some source of 3 phase power (3 sine waves, each shifted
120 degrees with respect to the other two) at the appropriate volts
per hertz (higher frequency requires higher voltage) capable of the
required current. Each winding runs on the voltage difference between
two of the phases.
With the correct voltage to frequency ratio, it can run over a wide
range of frequency. I don't know if that model is a permanent magnet
synchronous type or an induction type.

3. ### Bob EldredGuest

This is a four pole hysteresis synchronous mortor, B-2700 series. It runs at
1800 RPM on 60Hz single or three phase, 115VAC. It gives one ounce-inch of
torque at 1800 RPM, 12 watts.

To run single phase connect the AC line to the Blue and Green leads. Connect
a 2uF, 200VAC capacitor between the White and Green leads. To reverse
direction, connect the capacitor between the White and Blue leads.

A hysteresis synchronous motor does not have permnanet magnets nor is it an
induction motor. The rotor is iron and assumes a permanent magnetic flux
when energized. There are no fixed poles on the rotor. The rotor will lock
onto sychronous speed of the rotating field, in this case 1800 RPM. These
motors are used where accurate rotational speed is required but they don't
give much torque.
Bob

4. ### BobGGuest

BE:
This is a four pole hysteresis synchronous mortor, B-2700 series. It
runs at
1800 RPM on 60Hz single or three phase, 115VAC. It gives one ounce-inch
of
torque at 1800 RPM, 12 watts.
To run single phase connect the AC line to the Blue and Green leads.
Connect
a 2uF, 200VAC capacitor between the White and Green leads. To reverse
direction, connect the capacitor between the White and Blue leads.
===================================
Hi Bob. I saw where one could run it from single phase with a cap....
but I wanted to see if I could drive it with a microcontroller
somehow...wondering how to generate 3 phase. I noticed it put out a sin
wave between any pair of wires.... that's why I thought it had a perm
magnet in it. Sounds like a lot of work. There's a dozen or so of em at
Skycraft.

5. ### BobGGuest

BE:
This is a four pole hysteresis synchronous mortor, B-2700 series. It
runs at
1800 RPM on 60Hz single or three phase, 115VAC. It gives one ounce-inch
of
torque at 1800 RPM, 12 watts.
To run single phase connect the AC line to the Blue and Green leads.
Connect
a 2uF, 200VAC capacitor between the White and Green leads. To reverse
direction, connect the capacitor between the White and Blue leads.
===================================
Hi Bob. I saw where one could run it from single phase with a cap....
but I wanted to see if I could drive it with a microcontroller
somehow...wondering how to generate 3 phase. I noticed it put out a sin
wave between any pair of wires.... that's why I thought it had a perm
magnet in it. Sounds like a lot of work. There's a dozen or so of em at
Skycraft.

6. ### Bob EldredGuest

Yes you could generate three phase signals in a microprocessor. You'd have
to run the motor on higher voltage, however, 115 volts for 60Hz. You could
run at higher speed (frequencies) with higher voltages and lower speeds with
lower voltages. The volt-time product per cycle is held constant for
variable frequency drive. The frequency varies the speed and the constant
volt-time product keeps the torque controlled without allowing the motor to
overheat at low speeds. Usually such drives use pulse width modulation to
generate sinusoidal currents for the windings. Again a microprocessor could
easily develop the required three PWM switching signals for a three phase
system. The PWM signals would also incorporate the necessary constant
volt-time product characteristic. The drive elements for the motor could be
FETs or IGBT's or power transistors all driven from the processor. It sounds
like a big deal though just for a one ounce-inch motor. Such variable
frequency inverter drives are used on all kind of motors from fractional HP
on up.
Bob

7. ### Rich GriseGuest

Generating 3-phase in a uC isn't hard at all - just have three look-up
tables of 3 sine waves 120 degrees out of phase from each other, and
run 3 DACS to give you the three phases. Or, you could use one LUT and
three scrolling indices 1/3 of the way apart from each other, but that'd
be nasty ugly because your LUT would have to be 3N entries long. ;-) You
probably wouldn't need very big look-up tables; it's surprising how
quickly harmonic content drops off with just a few samples.

That also gives you absolute speed control, without relying on a
cap, but mind what John Popelish said about making the drive voltage
inversely proportional to the frequency. (at least, I _think_ that's
what he said - it's been a couple of posts ago. ;-) )

I don't know how to turn three, ground-referenced outputs into a
delta, however. I wonder what would happen if you just used three
half-H's, which, I guess, amounts to 3x totem-pole outputs? My intuition
says, "of course", but I haven't done the grunt work yet.

Good Luck!
Rich

8. ### Jasen BettsGuest

possibly there's no magnet but a rotor that picks up a current from the
magnetic field of the windings and behaves a bit like a magnet - google for
induction motor,

Bye.
Jasen

9. ### Jasen BettsGuest

it may be possible to drive only two of the inputs (and ground the
third). the two driven inputs should have a phase difference of 60 degrees.

bear in mind that with this setup the driven inputs have nearly twice the
voltage from the undriven input than they would have from ground if being
driven from a three phase supply... so check the winding to frame insulation
in that datasheet before trying that with the undriven input grounded.

Bye.
Jasen