Re: [CAD_CAM_EDM_DRO] HELP ! on Westamp servo drives.

"Ward M." wrote:

> I have a Tree Ultra Precision 1000 lathe that I have been using for about
> one year. As long as I've had it, there has been overtravel (overshoot) in
> both X and Z axis. This occurs only during rapids and then can produce an
> overshoot of up to 0.1 " on the Z axis. Turning the federate override to 50
> or 60 % stops this, but then my programs need to call out twice the required
> feed rate for proper cutting.

That may not do any good, as you'll end up with feedrates close to what they
were before. (Some controls allow feedrate override to change rapid rate,
others don't.)

> I believe that the servo amplifiers need
> adjustment, but I don't have any information or schematics for these units.
> They are Westamp Model #30060-25 and are operating on a 90VDC bus. There
> are six multi-turn pots on the edge of the board labeled AUX, SIG, TAC, TC,
> CLM and BAL.

AUX may do nothing, as it is an extra input that is not used in most servo
systems. SIG is a gain adjustment on the velocity command from the CNC
control. TAC is the gain of the tachometer. CLM sounds like current limit.
BAL sounds like the DC offset adjustment to keep the amp from drifting when the

positioning loop is not closed. TC is a little vague, but is probably Tach
Compensation,
or a way to adjust bandwidth of the velocity loop.

> While several of these are a fairly easy guess as to function,
> I'd really like to find some info such as tuning procedure and/or a
> schematic, if available.

This is extremely close to the setup on my servo amps, although I have loop
compensation on the vel and the current loops, and DC balance on both, too.
Do you have an oscilloscope? This is pretty useful for this, and a digital
or other storage scope is even more useful.

First, you want the TAC loop to have as much responsiveness as possible.
If the gain is too high, the system will become unstable, possibly violently
so. So, you need to be where you can get to the E-stop button quickly.
Since you have an axis that works well, you might want to make a fast
start (or stop) on a good axis and see what the peak current is. It is best
to monitor this at a test point on the amp, if one is available. The motor
outputs are usually differential, so you can't ground a scope to either
wire. (Someone else found this out the hard way, and I was helping him
to try to fix his smoked amp.) Then, check the axis that overshoots, and
see if it is getting the same peak current. If not, you may need to turn up
the CLM pot. It should allow the peak to reach the lowest of the motor's
or the amps rating. This may be set lower, however, to prevent the breaking
of belts or shaft couplings.

If the current limit isn't the problem, then you may need to turn the SIG or
TAC settings higher. If these are correctly set, they need to be moved
in unison, but it sounds like they may not be set correctly.

Once you have the peak current set, you would move
the scope to the tachometer, so you are viewing velocity.


The proper procedure is to use a voltage source (not the CNC control,
unless it can do that) to excite the SIG input with a clean DC voltage,
and then adjust the TAC gain until the system becomes unstable. You then
adjust the TC for best stability, back and forth, until you have the TAC
as high as you can get, while still stable.

Then, you hook the Sig input back to the CNC, and use the scope to watch
the beginning of a steady state move just after the acceleration ramp.
If the SIG pot is too low, there will be a slow rise, leveling off after it
finally reaches desired speed. If SIG is too high, it will overshoot the
steady speed, and then wiggle around. If it is just right, the accel. ramp
will change sharply to a steady speed, and then hold it until the end of the
move.

If the flat top of the steady speed move has wiggles, then either TAC is
too high, or TC needs adjustment. This is an iterative process, but you
should be able to get it to work.

There is an analytical approach to this, using control theory, but you need to
know motor and table inertia, motor and servo amp electrical characteristics,
etc. to put into the calculations.