Re: [CAD_CAM_EDM_DRO] Servo braking resistor in series with motor armature

vrsculptor wrote:

> Your opinions please. The following was posted on the UHU servo amp
> board discussing the need for braking resistors on certain motors to
> protect the drives in case of hard deceleration and crashes.
>

Same problem with the Rutex drives. It is an abominable way to
solve a major design defect. I had the same problem on my very
first servo amp, and redesigned the overcurrent logic to solve
it. My first amp only reduced on-time when the current limit
was exceeded in the "forward" direction, ie. where the DC supply
was applied in the same directions as the DIR signal was telling
it to go. So, if the motor was being decelerated, the current
would oppose the DIR signal, and the overcurrent would cause it
to short the motor for a longer time, INCREASING the current!

The obvious fix was to make all overcurrent conditions, of any
polarity, turn off the transistors until the next PWM cycle,
then check the current limit again.

But, putting a large power resistor in series with the motor
armature drastically reduces performance, making the motor's
response more "sluggish and softer". It requires a higher
supply voltage to get the same performance you'd get without the
resistor.

I critiqued the UHU design at the last CNC Workshop, and found a
NUMBER of deficiencies. The wrong type of diodes were being
used for freewheel purposes, if any (too slow). The track
layout on the board was poor. The scheme for obtaining dead
time on the FET's gates was laughable (why not put the dead time
circuit on the INPUT to the FET drivers?) And, putting a single
current detector on the supply minus instead of having one sense
resistor on each low-side transistor prevents the detection of
circulating current in the motor. Finally, I think the use of
synchronous antiphase drive is a mistake.

Jon