Random GFCI tripping, Data Rack

5232 Views 5 Replies 5 Participants Last post by PaliBob, Jul 11, 2009

theatretch85

Discussion Starter · Jul 10, 2009

As part of my basement remodel I replaced the main panel (old FPE, New Square D QO) and had to install some GFCI outlets in an unfinished area of the basement. I am currently building a server room for a data rack where I am routing all data/telephony/CATV cables to. Eventually I will finish off the room and run dedicated 20 and 30 amp circuits into, but for now the equipment in the data rack is plugged into a standard outlet that is protected up-stream by a GFCI outlet.

The rack will stay powered on for days on end, and at random the GFCI trips, killing power to the entire rack. Usually it just means resetting the GFCI outlet and then its fine for several more days until it trips again. Last night the outlet the rack was plugged into had the GFCI protecting it trip, and then plugged the rack into another outlet (protected by a different up-stream GFCI) and by morning it had tripped again.

Like I said its not a huge deal as I will eventually be running the dedicated 20 and 30 amp circuits to this server room, just curious if anyone has any ideas why the GFCI would trip at random? Also, this circuit is protected by a 20 amp AFCI breaker and that has never tripped (which I figure with the computer and data equipment plugged in the AFCI would be more prone to nuisance tripping).

Save

1 - 6 of 6 Posts

Gigs

· #2 · Jul 10, 2009

Don't plug computers into GFCIs. The capacitors across the line and ground in a computer power supply add up to too much leakage current.

It's kind of like:
|------------cap-------------|
Hot---cap---ground---cap---neutral

Save

Gigs

· #3 · Jul 10, 2009

http://images.electronicsinfoline.com/View/Photo/10001/Atx_power~0.gif

Save

Yoyizit

· #4 · Jul 10, 2009 (Edited)

theatretch85 said:
The rack will stay powered on for days on end, and at random the GFCI trips,
then its fine for several more days until it trips again.

". . .we just don't understand it (yet), and the events therefore seem random to us."

Probably means 4 mA or slightly less, of mostly capacitive leakage current. You could:
put a 7-1/2w incand. lamp in series with the ground wire that serves the cable downstream of your GFCI.
If the voltage across the bulb reads from 0.6 vac to up to 120vac you've found your leakage path to ground. A normal reading would be less than 15 mVac.

I'm surprised it doesn't trip more often.

In the top left corner of the posted schematic, what component is Z1? Maybe a surge suppressing gas tube?

Save

jerryh3

· #5 · Jul 10, 2009

Yoyizit said:
". . .we just don't understand it (yet), and the events therefore seem random to us."

Probably means 4 mA or slightly less, of mostly capacitive leakage current. You could:
put a 7-1/2w incand. lamp in series with the ground wire that serves the cable downstream of your GFCI.
If the voltage across the bulb reads from 0.6 vac to up to 120vac you've found your leakage path to ground. A normal reading would be less than 15 mVac.

I'm surprised it doesn't trip more often.

In the top left corner of the posted schematic, what component is Z1? Maybe a surge suppressing gas tube?

Z should be a filter.

Save

PaliBob

· #6 · Jul 11, 2009

MOV
http://www.phonda.com.tw/varistor.htm

Save

1 - 6 of 6 Posts

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.

Top Contributors this Month