Re: Hot servo motor

Hello Jon...

Appreciate your comments. The motor ran hot on other drivers. I did
not check to see if there was residual idle current. The axis was Y,
and by hand turned fairly easily without binding. All of the servos
were identical and all of the encoders were identical. I went back
to retune the Y stepper and in that process the a large amount of
smoke escaped from the driver (dang). Checked the motor and did not
find any open windings or shorted windings...all measured 2.2 ohms
per motor spec. Anyway, I have retired the servo motor and now
looking for an equivalent replacement.

Regards...Richard in Michigan.

>
>
> k8zre wrote:
>
> >That is a good thought. Will check. However, the servo gets hot
> >just idling (no movement).
> >
> The problem is that the servo drivers have finite gain. If the

axis and

> motor
> can move freely, it will move the motor to within one encoder count

or

> so of the commanded position, and the motor current will drop to

nearly

> zero. If the axis is stiff for some reason, the Rutex will deliver

less

> current
> as the motor gets closer and closer to the commanded position. At

some

> point
> the friction will prevent any further motion, and it will stall

there.

> If the
> friction is substantial, then the encoder will continue to indicate

some

> distance from commanded, and the motor will be sent a steady

current,

> but not enough to break the static friction. If this current is a

few amps,

> then the motor will get hot.
>
> > The Y axis seems to run smoothly without
> >apparent binding by the hand wheel (servos off). I will be

swapping

> >drivers to the board to see what happens...maybe scope the motor
> >outputs.
> >
> >
> Scoping the motor outputs is difficult. Definitely don't connect

the

> scope ground
> to either terminal. You will see a 100 V square wave with a

varying

> duty cycle.
> You can connect a 10 A ammeter in series with the motor to see the

average

> current. It should be very close to zero at idle. You can apply a
> little torque
> to the shaft and see the current rise as the drive resists your

attempt

> to move
> it off the null position. The current should return to zero when

you stop

> deflecting the motor. I'm guessing under some conditions that the
> current does
> not drop to zero. If the DC current does appear to drop to zero at
> idle, then
> the drive must be dithering at a very high rate. This usually is

audible on

> low-freq PWM drives. Most Rutex drives run at 19.5 KHz, so the

highest

> dither frequency would be 9.75 KHz, which I would certainly hear,

and find

> maddening.
>
> These PWM drives that use the 50% duty cycle scheme can send a fair

amount

> of PWM carrier ripple current through the motor, and it can cause

heating on

> low inductance motors. but, if your motors are identical, then the

ripple

> current should be the same in all of them.
>
> Jon