Hot ground Help!

[moxy]

Hi all,
I have been running 3 geckos for some time using 2 36v switching psu's without problems. (or so I thought) I noticed that my X axis had lost power and found that one of the psu's had died and I was just getting 36v output. It appears that these things don't like being in series! To sort this out I used a variac which cranked up to 74v and fitted a rectifier and caps etc. All well and good, until I touched my mill and the controller box at the same time. Result-------200v kick! On measuring I found 200v plus on my controller box chassis Includes heatsinks as they are directly mounted to chassis I tried grounding everything correctly and thought this would sort it. All I achieved was to trip the mains earth leakage. I disconnected all 4 gecko's (original 3 plus new one for dividing head) and this voltage dissapeared. I replace the psu with the 36v switcher...... no problem! How can one psu setup do this and another not. The variac psu does display some ripple, can this cause such problems!
Best regards
Les.
PS Best wishes to everone for the festive season.

[Al_The_Man]

#1/ It is considered dangerous to use a Variac for power supply as it does not afford any isolation and it is possible to have a 'Hot Ground'
A variac is an auto transformer so the supply is always connected to the 'Secondary', and unless you take extra care to ensure the incoming supply is connected properly. a chassis can be live potential WRT Ground.
Al.

[GrahamIT]

You can use a variac as you've found, as long as it can supply a high enough current without glowing in the dark, but for the sake of safety use a 1:1 wind mains isolation transformer as well.

[moxy]

Hi Guys,
Thanks for the quick replies. What you say makes sense so I guess I'll ditch the variac idea. What did surprise me was that the frequency of the hot ground was not mains freq. It was very unstable but was in the region of 200 HZ. Also I was only putting 75 volts into the thing and getting a potential of around 200v on the chassis? Checking the chassis voltage with a scope I found that the signal was really messy (not a nice sinewave) and had short periods of very high freq. but generally very unstable.
Anyway, as I said, I'll drop the variac and look for a nice isolation mains transformer.
All the best over christmas and the new year
Les: