More on Re: Insulating D-shell from chassis?

To the best of my DVM's ability, the potential between the PC chassis and
driver box chassis is .000V; the potential between the driver box chassis
and its isolated DB25 shell is .000V; and the potential between the PC
chassis and the isolated DB25 shell at the driver box is .000V. Both PC
and driver box are using switcher power supplies. The DC ground of the
driver box is not connected to chassis ground (unless it is through the
Xylotex board...)

The spindle motor is a Sherline permanent-magnet DC motor with their drive
electronics (KB Electronics KBIC PWM driver.) I always turn the speed knob
down all the way before switching power to the spindle. The spindle drive
is not under G/M code control at all.

The random steps were only while the spindle motor was running--I could
feel them by placing my fingers on the handwheels on the stepper motor
backshafts. This would be while the gantry was idle, or when making a
single-axis move I could feel steps on the other axes. Before adding the
filter caps to the Xylotex board, the random steps would happen even
without the spindle motor powered. After adding the filter caps, the steps
were only while the spindle was running. Isolating the DB25 shell at the
controller box totally eliminated the random steps.

I understand grounding from a safety standpoint. I understand star vs.
daisy-chain grounding. (Especially so, dealing with the TUV representative
in my day job of designing laboratory equipment!) I have also been taught
to only terminate a cable shield at one end. I refuse to believe that
isolating the DB25 shell at the driver box is any kind of hazard (otherwise
Amp would not make plastic-shell D connectors.)

While I appreciate everyone's input in the coulda, woulda, shoulda about
grounding, I was just asking opinions on isolating options.

Best regards,

Randy



Randy Gordon-Gilmore ,----.___________ ______________ _________________
ProtoTrains // = = === == || == == == = || == == == = == =|
Rio Vista, CA, USA /-O==O------------o==o------------o==o-----------o==o-'
zephyrus@... http://www.prototrains.com