VFD Ground issue?

[BanduraMaker]

I'm using a gecko g540 to control a Hitachi WJ200 vfd. I'm using the 0-10v signal for speed and output 1 for run control. Everything was working fine but today I came down to machine some parts and unplugged a USB cable connecting the Hitachi to the computer, just because i didn't think it necessary. When I went to run my program, the spindle didn't start up and I almost wrecked a bit.

I figured out that it was the USB cable on accident when I started up pro drive next, hitachi's programming software, to see if any settings in the VFD mysteriously changed on me. When I plugged the USB cable back in, the spindle started up!

I'm thinking this must be a grounding issue because there's no control software running on the pc to run the vfd.

My connections between the gecko are

GND, 10v and signal to pins L, H and O, Output one to smart pin 1.

Any ideas as to what may be wrong? Do I need to run a ground between the PC and the VFD?

[Al_The_Man]

You should preferably use a star point ground point, and definitely the VFD ground and motor frame ground should go to this point together with a ground bonding conductor from the PC case.
There have been numerous past posts on issues such as this that involve grounding issues.
e.g.
http://www.cnczone.com/forums/phase_...se_issues.html
Al.

[BanduraMaker]

Update - It seems that the Smart Terminals in the VFD don't share a ground with 0 ~10v signal for spindle control (or perhaps they don't in the Gecko). I ran a ground from my control box (closest thing to the star location I have) to the "L" pin in the smart terminal strip and everything works fine now.

Apparently the issue was with the Ground with the Gecko, not with the PC. Shorting the USB ground with the PC chassis had no effect.