Re: [CAD_CAM_EDM_DRO] 3 phase AC motors as cnc servo motors ?

Kim Lux wrote:

>This might be a stupid question, but I'm going to ask it anyway: why
>aren't small (1/4, 1/3,1/2Hp+) 3 phase AC motors suitable for use as
>servo motors ?
>

2 reasons:
1. standard AC motors don't have the cooling required for a servo
motor. An external fan and maybe some opening up of cooling holes
might help this a lot.

2. AC flux-vector servo drives are a LOT more expensive than a DC
brush servo amp. If you happen to come across a bargain on one, you
could put an encoder on an AC induction motor and make it work
quite well. The price of flux-vector drives are coming down, so it is
only a matter of time before somebody designs a low-cost drive for
the small shop/hobby users.

>
>We run several small and medium sized 3 phase AC motors on machine
>equipment with variable frequency drives. Their low speed torque seems
>good. Speed control is generally very good.
>
>Why couldn't a quasi step-direction/variable frequency drive device be
>made to utilize small 3 phase motors as servo motors ? The device would
>accept (step, direction) input from the computer and encoder feedback
>from the motor. It would output power to the 3 phases of the motor.
>
>

This is not a big extension to go from DC brush motors to permanent magnet
brushless motors (commonly called DC brushless). But, it IS a big extension
to go from DC brush to AC induction. The reason is the excitation of the
stator windings excites (by induction) the rotor field, and the rotor field
moves around (called slip) with respect to the rotor's actual position.
So, the
drive has to do some pretty fancy mathematics to figure out where the rotor
field is right now, and what will happen to it if the stator field is
changed.

>At first I thought that the number of poles in the motor might limit the
>"step" resolution, but I suspect that the DC motors used as servo motors
>have the same limitation. Or do they use a very large number of poles ?
>
>

No, they don't. Good DC servo motors have windings designed such that the
armature poles produce almost no detectable torque or velocity ripple,
so that
these effects don't cause errors in the movement. AC induction motors may
or may not be good enough in this respect.

>I was wondering if the slip characteristics of an induction motor might
>have something to do with it,
>

You bet, that is the tricky part. It is totally a textbook thing, now,
as a lot of
research has been done on it, and the details are well documented. But, you
DO have to make the extra calculations for it to work.


Jon