Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums > AC Power and Grounding
Tripping GFI
Stephen Swaffer:
GFCis are pretty sensitive-they only take .006 amps to trip-so it could have just been moisture from the rain and damp. It certainly wouldn't have to be laying in a puddle to trip.
The other place I would be suspicious is the GFCIs themselves. New ones shouldn't reset if they are defective-but I received some new ones this spring that were defective and would trip when a load was applied. They are designed to fail safe-so more likely to trip when they shouldn't than not to trip when they should. The fact that it happened on more than one circuit makes this the less likely possibility here.
At least you didn't just bypass them-GFCI's tripping is not necessarily a bad thing. It could just save a life.
Tim McCulloch:
--- Quote from: Stephen Swaffer on July 07, 2016, 01:12:58 PM ---GFCis are pretty sensitive-they only take .006 amps to trip-so it could have just been moisture from the rain and damp. It certainly wouldn't have to be laying in a puddle to trip.
The other place I would be suspicious is the GFCIs themselves. New ones shouldn't reset if they are defective-but I received some new ones this spring that were defective and would trip when a load was applied. They are designed to fail safe-so more likely to trip when they shouldn't than not to trip when they should. The fact that it happened on more than one circuit makes this the less likely possibility here.
At least you didn't just bypass them-GFCI's tripping is not necessarily a bad thing. It could just save a life.
--- End quote ---
This.
I've had GFCI tripping from plugging the extension cord I use for my lawn mower when there is dew on the grass and the connection is lying on the lawn.
For Robert: the GFCI is tripping because something is wrong. That the problem moved with the cord is your clue, and that different GFCIs tripped indicates that a faulty GFCI is unlikely.
Mike Sokol:
One more thing to consider. In old tube guitar amps and microwave ovens, the insulation in their power transformers often start deteriorating from all the heat and constant loading. I've seen as least several of each that developed a live-to-chassis leakage of a few 10's of mA. Certainly enough to trip a GFCI, but not enough to trip a circuit breaker when plugged into a conventional grounded receptacle. Since most affordable clamp-on ammeters don't go down below 10 mA resolution, you can make a current multiplier by wrapping a split-out ground wire in an extension cord around the clamp ammeter jaws 10 times. Now your 10 mA resolution ammeter just became a 1 mA resolution meter. Plugging in your gear one at a time to this ground current sensing cord should give you an indication of anything leaking current to the EGC. I'll build up one of these in the shop tomorrow and post a picture, but it's really simple to find small leakage currents with this booster trick.
John Escallier:
--- Quote from: Mike Sokol on July 07, 2016, 12:40:11 PM ---Yes, everything leaks a little. And it's all additive. So a single piece of gear on a GFCI may not trip it. But put 2 or 3 things on a GFCI and the leakage currents can easily add it to the 6 mA tripping threshold. The fact that in this instance the GFCI's would trip under heavy current draw, but not under more moderate current draw suggests some intermingling of the Ground and Neutral. First thing I would do is check your own gear to make sure there's no accidental G-N bond or G-N swap in any of your equipment or extension cords. That's the only thing I can think of that would create a load-dependent GFCI random trip.
--- End quote ---
A GN swap should instantly trip out a GFCI. The inrush far exceeds the time/current trip spec, and any idling transformer will still draw current from the line due to magnetizing inductance even though it's 90 degrees out. GFCI's do not examine the leakage current phase.
A GN bond, that is another horse. That's a front runner.
Also, if too many Y's are on the line, the leakage can reach the GFCI levels. If it's close, audio leakage off the supply rails will be shunted to ground as well.
John
Steve Garris:
--- Quote from: Stephen Swaffer on July 07, 2016, 01:12:58 PM ---GFCis are pretty sensitive-they only take .006 amps to trip-so it could have just been moisture from the rain and damp. It certainly wouldn't have to be laying in a puddle to trip.
The other place I would be suspicious is the GFCIs themselves. New ones shouldn't reset if they are defective-but I received some new ones this spring that were defective and would trip when a load was applied. They are designed to fail safe-so more likely to trip when they shouldn't than not to trip when they should. The fact that it happened on more than one circuit makes this the less likely possibility here.
At least you didn't just bypass them-GFCI's tripping is not necessarily a bad thing. It could just save a life.
--- End quote ---
I just came to this forum to get an answer on my question, and here I see this post at the top! Same exact thing happened to me last night.
I had a modest setup, under a gazebo but things were getting "moist". My 10 ga cable was plugged into a 20 amp GFI circuit, then split into a 3-way with power surrounding the stage. PA barely on, as the gazebo was horrible for sound. The GFI tripped just once, but I left my cable in that same circuit and re-set it. I unplugged (1) 70w Par front flood, and turned it down just a tiny bit. We made it through the rest of the gig without any power losses.
I did this same venue last summer (weekly shows), and never had a problem. The difference being the rain, and my new SRX system draws more than my old PRX that I used last year. I didn't think it it was that, and wondered if in fact enough moisture was present to cause the GFI to trip. Another possibility was a bad GFI. These shows are usually sunny and dry, but the next time it's wet, I'll do a better job of routing and covering cables.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version