Re: [CAD_CAM_EDM_DRO] Is a reduction drive necessary for servo motors?

Bob McKnight wrote:

>Organization: Pico Systems
>To: CAD_CAM_EDM_DRO@yahoogroups.com
>From: Jon Elson <elson@...>
>Date sent: Wed, 14 Jan 2004 23:23:59 -0600
>Subject: Re: [CAD_CAM_EDM_DRO] Is a reduction drive necessary
>for servo motors?
>Send reply to: CAD_CAM_EDM_DRO@yahoogroups.com
>
>[ Double-click this line for list subscription options ]
>
>Jon
>
>I guess I don 't know how a Servo works. I thought that
>it was based on current reversals. Energize it with one
>polarity and moves and stays. Reverse the polarity and
>it moves.
>

The simplest servo system uses a DC brush motor. Apply power one
way, it moves CCW. Apply power the other way, it moves CW.
No power, it doesn't move. It gets more complicated when you
have dynamic loads, such as machine tools where the cutting forces
are applied against the table. Then, the servo may have to resist
cutting forces trying to move the table.

So, what really happens is current through the motor produces torque.

> I have not been clear on how it determines which
>way to go.
>

The servo controller reads position from the encoder, compares it to
desired position, and applies power of the correct polarity to drive the
motor to reduce the difference between desired and actual position.

A velocity servo gets more complicated, as a speed value is derived from
a tachometer or the encoder, and a velocity command from the CNC
control causes the servo amplifier to make the motor move at that
commanded velocity.

> So obviously, I have not a full understanding.
>At one time I experimented with Automoile Alternators as
>stepping motors. As I recall, the stock winding on the stator
>drew too much current and I had them rewound with smaller
>wire and more turns. They were very powerful but also low
>resolution.
>

All servo motors have low (or no) resolution. It takes an encoder
to measure position and give you resolution. Servo implies a
measuring device and a feedback loop. You measure position
and feed that back to an error detector, which commands motion to
reduce the error.

> Don Lancaster wrote and article and there was
>a guy using them to do wood carving. It was a very cheap
>way to get a lot of torque. Due to manufacturing tolerances,
>each step was not exactly the same as the other steps. This
>was no problem when you are making an alternator whose
>requirement is to put out a pulsating DC current.
>
>

A car alternator converted like that is really a brushless "DC" motor,
without the use of permanent magnets, so you have to energize the
rotor field. With a proper encoder, it would work fairly well, although
real brushless DC servo motors are optimized for very low torque and
velocity ripple. My guess is the torque and velocity ripple on the
converted alternators would be pretty high.

Jon