ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 3 [All]   Go Down

Author Topic: GFCI's tripping...  (Read 7106 times)

Jesse Stern

  • Newbie
  • *
  • Offline Offline
  • Posts: 21
    • www.jsaudio.xyz
GFCI's tripping...
« on: October 09, 2018, 07:00:27 PM »

I have often encountered problems with PA systems tripping GFCI protection on generators/pagodas/wall outlets, especially when subwoofers are involved.  Even when the total amps being drawn is well under the breaker threshold.  Why does this happen?  And is there an easy way to disable GFCI protection on a receptacle?

Mike Sokol

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3361
  • Lead instructor for the No~Shock~Zone
    • No~Shock~Zone Electrical Safety
Re: GFCI's tripping...
« Reply #1 on: October 09, 2018, 07:29:15 PM »

I have often encountered problems with PA systems tripping GFCI protection on generators/pagodas/wall outlets, especially when subwoofers are involved.  Even when the total amps being drawn is well under the breaker threshold.  Why does this happen?  And is there an easy way to disable GFCI protection on a receptacle?

Jesse, GFCI's don't trip from excessive over 15 or 20 amps ofcurrent draw due to amplifier wattage. They trip from anything leaking more than 6mA (0.006 amperes) of current. Some amplifiers leak an excessive amount of current to the chassis which can cause problem tripping, but many don't. However, if you have any surge strips with MOV devices (most do), know that a surge strip will often leak at least a few mA of fault current on their own. So if you have two surge strips, that additive current can put your GFCI right at the threshold of tripping.

frank kayser

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1655
  • Maryland suburbs of Washington DC
Re: GFCI's tripping...
« Reply #2 on: October 09, 2018, 07:34:35 PM »

I have often encountered problems with PA systems tripping GFCI protection on generators/pagodas/wall outlets, especially when subwoofers are involved.  Even when the total amps being drawn is well under the breaker threshold.  Why does this happen?  And is there an easy way to disable GFCI protection on a receptacle?


Second question first: No.


Number of amps is limited by circuit breakers/fuse. This protection mechanism in most cases has nothing to do with GFCI (unless GFCI is built into the breaker)


GFCI devices ensure all current in on hot is returned on neutral.  These are very sensitive - I believe the tripping threshold is about 6 milliamps (ma).  10 ma across your chest is considered the lethal limit.


Outdoor GFCI seem to take a beating - rain, heat, humidity, bugs, etc.  As they age, they fail.  It does not take long.  In my city's plaza, I've noticed the GFCI devices have been changed out three times in two years.


Why are the ones you're dealing with are tripping with the PA, especially when using subs, is beyond my knowledge.  The GFCI devices trip due to either internal failure, or have detected a current leak somewhere - maybe a power strip with the surge protection devices, MOVs, are leaking. One power strip plugged into another power strip, the leakages are additive, and maybe the cause of the trip.  Maybe old power amps or other devices with capacitor leaking in the power supply.  Vintage garage sale amps are old, and caps start leaking over time.  Just places to think about. 


frank
Logged

Mike Sokol

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3361
  • Lead instructor for the No~Shock~Zone
    • No~Shock~Zone Electrical Safety
Re: GFCI's tripping...
« Reply #3 on: October 09, 2018, 07:45:59 PM »

Here's something I wrote for PSW on GFCI theory that should help you understand what's happening. https://www.prosoundweb.com/channels/av/no_shock_zone_understanding_and_preventing_electrical_damage_and_worse/

Frank Koenig

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1318
  • Palo Alto, CA USA
Re: GFCI's tripping...
« Reply #4 on: October 09, 2018, 10:08:23 PM »

I'll add that some older, out of spec, GFCIs can trip due to RF hash on the lines from big-dog power amplifiers with switching power supplies. This can happen even though the (low frequency) leakage current is well below the 6mA threshold used for 20A branch circuits in the US. DAMHIK --Frank
Logged
"Nature abhors a vacuum tube." -- John Pierce, Bell Labs

Lyle Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1558
Re: GFCI's tripping...
« Reply #5 on: October 12, 2018, 06:20:18 PM »

The idea that 6mA is the correct trip level is not supported by global experience.  Many countries have GFCI/RCD trip currents much higher, and have fewer electrical fatalities than the US.

An isolation transformer of sufficient size will defeat the GFCI, but you should not do this.
Logged

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 17176
  • Hickory, Mississippi, USA
    • Resotune
Re: GFCI's tripping...
« Reply #6 on: October 12, 2018, 06:47:16 PM »

The idea that 6mA is the correct trip level is not supported by global experience.  Many countries have GFCI/RCD trip currents much higher, and have fewer electrical fatalities than the US.

An isolation transformer of sufficient size will defeat the GFCI, but you should not do this.
6 ma +/-  is the US spec for GFCI trips, so no not true for RCDs in other countries. My bench work found it pretty good here.

I do not understand how an isolation transformer can defeat a GFCI as long as the transformer is isolating just line and neutral in series...

If safety ground gets combined with neutral and not kept separate, that is just defeating the purpose of sensing the symmetry of the line/neutral loop. (i.e. don't do that) 

JR
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

Lyle Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1558
Re: GFCI's tripping...
« Reply #7 on: October 13, 2018, 03:11:17 AM »

Deliverying a separately derived supply, so that any leakage on the secondary side of the 1:1 transformer is seen as just part of a normal load on the primary side.   

Yes, I was saying "don't do that!"
Logged

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 17176
  • Hickory, Mississippi, USA
    • Resotune
Re: GFCI's tripping...
« Reply #8 on: October 13, 2018, 09:47:31 AM »

Deliverying a separately derived supply, so that any leakage on the secondary side of the 1:1 transformer is seen as just part of a normal load on the primary side.   

Yes, I was saying "don't do that!"
not sure I follow.

A simple 1:1 transformer should just extend the GFCI current loop....

If leakage current returns to the transformer isolated neutral, that is the same as using no transformer with current leakage to the real neutral.

Current leaking from the transformer output back around outside the GFCI loop to safety ground (or anywhere else) should trip the GFCI at >6mA. A bootleg safety ground to neutral with or without transformer isolation will defeat the GFCI protection.

JR 
 
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

Stephen Swaffer

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2672
Re: GFCI's tripping...
« Reply #9 on: October 13, 2018, 01:30:54 PM »

However, you have to look at the whole circuit-all the way back to the existing transformer that supplies the GFCI.  Properly installed, it has a neutral/grounded conductor (as opposed to grounding conductor) so if you touch the "hot" conductor electricity flows through you to ground-ultimately getting back to the transformer albeit through an alternate path.

If you install a true isolation transformer, there is no neutral/grounded conductor, so if you touch a hot there is no alternate path back to the transformer.  On the surface this is "safer" and some have argued that, but there are several disadvantages to this scheme.  One of the main ones being what happens if there is a short creating an unintentional ground connection.  With an intentional connection, safeguards (gfcis, breakrs, fuses) can work reliably.
Logged
Steve Swaffer

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 17176
  • Hickory, Mississippi, USA
    • Resotune
Re: GFCI's tripping...
« Reply #10 on: October 13, 2018, 02:05:06 PM »

However, you have to look at the whole circuit-all the way back to the existing transformer that supplies the GFCI.  Properly installed, it has a neutral/grounded conductor (as opposed to grounding conductor) so if you touch the "hot" conductor electricity flows through you to ground-ultimately getting back to the transformer albeit through an alternate path.

If you install a true isolation transformer, there is no neutral/grounded conductor, so if you touch a hot there is no alternate path back to the transformer.  On the surface this is "safer" and some have argued that, but there are several disadvantages to this scheme.  One of the main ones being what happens if there is a short creating an unintentional ground connection.  With an intentional connection, safeguards (gfcis, breakrs, fuses) can work reliably.
I do not advise ever touching an energized conductor (transformer isolated or not). I doubt UL would either.

Inside commercial products line and neutral are both insulated, so humans cannot touch them.

JR

PS: I have killed a lot of brain cells debating with myself about the pros and cons of opening a ground path (not exactly the same thing but similar).  I had a relay inside my super-duper artist saver power drop that lifted safety ground if it detected too much ground current flowing from elsewhere (the ground lift was coincident with the GFCI disconnecting mains power.) FWIW I wouldn't even try to persuade UL to bless this while I believe it could be useful in the context of talent exposed to multiple power drops in a single performance space. UL loves their safety ground.
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

Lyle Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1558
Re: GFCI's tripping...
« Reply #11 on: October 13, 2018, 04:15:08 PM »

Defeating a safety mechanism is generally not a good idea.  Upstream GFCI/RCD defeat is required for testing portable GFCI/RCD.

A GFCI/RCD protected arc welder is an interesting example to think about.

Logged

David Buckley

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 565
Re: GFCI's tripping...
« Reply #12 on: October 14, 2018, 10:12:47 PM »

A simple 1:1 transformer should just extend the GFCI current loop....

If leakage current returns to the transformer isolated neutral, that is the same as using no transformer with current leakage to the real neutral.

The difference is that the leaking neutral current back to the transformer secondary does not cause an imbalance in the primary winding oir the wire back to the source and its GFCI.

Logged

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 17176
  • Hickory, Mississippi, USA
    • Resotune
Re: GFCI's tripping...
« Reply #13 on: October 14, 2018, 10:25:25 PM »

The difference is that the leaking neutral current back to the transformer secondary does not cause an imbalance in the primary winding oir the wire back to the source and its GFCI.
For that to happen chassis ground (leakage return path) is connected to the transformer isolated version of neutral.

That isn't allowed for non iso versions for obvious reasons (defeats the GFCI), same thing with transformer iso.... Connect chassis back to real safety ground and the GFCI will still trip.

JR
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 17176
  • Hickory, Mississippi, USA
    • Resotune
Re: GFCI's tripping...
« Reply #14 on: March 28, 2019, 05:56:36 PM »

Had a new different outlet/GFCI failure.

Last night I plugged in my vacuum cleaner downstream from the GFCI outlet and everything plugged into the outlet went dark. The fuse for that branch was good and GFCI LED did not light. Wiggling a plug in the GFCI caused some intermittent operation but it dropped out again from a decent current load, hot water kettle.

I waited until today to swap out the gfci outlet... The screws holding the hot and neutral wires did not appear loose, but i I see obvious evidence of heating at the black (hot) screw and the screw head is discolored presumably from heating.

It seems pretty apparent this is operator error on the part of the installer (me).  :-[ But I repeat the screw did not appear loose when I removed it.

Long story shorter, I replaced with a GFCI outlet I had laying around and this is very tight fit (I need to look at sourcing a few skinny GFCI outlets for my old small junction boxes.)

After removing, replacing, and getting my kitchen back online, I wired a line cord to the suspect GFCI outlet to bench test it...

The test and reset functions work fine, but the trip LED does not light up when tripped by test button...

====
as long as I have the attention of this community, any thoughts on poking wires into the outlet clamps, or wrapping around screws. I am old (school) so wrap wires, but any thoughts on relative robustness?

I made sure the screws were tight, but like i said I thought I did that last time, and they didn't feel loose.

JR
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

Stephen Swaffer

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2672
Re: GFCI's tripping...
« Reply #15 on: March 28, 2019, 09:50:37 PM »

Leviton makes some of the more compact GFCI's-cost a little more but worth it IMO.

Most GFCIs have clamps designed to have wire inserted and clamped-if that is their design, you really shouldn't wrap around the screws.  Wrapping rounf the screws with those is technically a listing violation-practically it makes them more likely to short out in metal boxes.

The trickiest part of these clamps is that the wire that is inserted needs to be straight.  It is easy to twist the wire and work it loose while installing-especially if the wire is not straight.

Also, make sure the wire is actually under the clamps.  I have a pic somewhere of a breaker where the wire was not under the clamp-never figured out how it ran as long as it did!
Logged
Steve Swaffer

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 17176
  • Hickory, Mississippi, USA
    • Resotune
Re: GFCI's tripping...
« Reply #16 on: March 29, 2019, 10:52:12 AM »

Leviton makes some of the more compact GFCI's-cost a little more but worth it IMO.
I just ordered a couple skinny Levitons, notably thinner and shallower than cheap GFCIs. My old house has small junction boxes and when I put a fat GFCI in my bathroom the narrow box became energized, and energized the safety ground.  ::)  (My outlet tester that I stopped pursuing detected the hot safety ground.)
Quote
Most GFCIs have clamps designed to have wire inserted and clamped-if that is their design, you really shouldn't wrap around the screws.  Wrapping rounf the screws with those is technically a listing violation-practically it makes them more likely to short out in metal boxes.
These appear designed to be used both ways... I did retighten the screws twice, but i thought they were tight the last time.  ::)
Quote
The trickiest part of these clamps is that the wire that is inserted needs to be straight.  It is easy to twist the wire and work it loose while installing-especially if the wire is not straight.
These several decades old wires will never be perfectly straight, and aren't long enough to comfortably re-strip.  :o
Quote
Also, make sure the wire is actually under the clamps.  I have a pic somewhere of a breaker where the wire was not under the clamp-never figured out how it ran as long as it did!

Any thoughts about the red status LED not working? The clearly overheated GFCI outlet is in the trash already, but curious. Perhaps an internal solder connection opened up from the heating. Not sure I am curious enough to take it apart.

JR 
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

Stephen Swaffer

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2672
Re: GFCI's tripping...
« Reply #17 on: March 29, 2019, 12:41:51 PM »


Any thoughts about the red status LED not working? The clearly overheated GFCI outlet is in the trash already, but curious. Perhaps an internal solder connection opened up from the heating. Not sure I am curious enough to take it apart.


Any number things perhaps-but one things is clear:  a safety device has an obvious malfunction, it needs to be replaced.  Just passing electricity doesn't make it usable!
Logged
Steve Swaffer

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 17176
  • Hickory, Mississippi, USA
    • Resotune
Re: GFCI's tripping...
« Reply #18 on: March 29, 2019, 01:08:20 PM »

Any number things perhaps-but one things is clear:  a safety device has an obvious malfunction, it needs to be replaced.  Just passing electricity doesn't make it usable!
Just to be crystal clear I already replaced it.... and will have better (Leviton) replacements waiting on hand JIC.

I only have 3 GFCI outfitted outlets in my entire house (kitchen, bathroom, laundry room). That is three more than many of my neighbors.

JR
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

Stephen Swaffer

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2672
Re: GFCI's tripping...
« Reply #19 on: March 29, 2019, 05:19:58 PM »

I assumed you had-just a blanket principle (carry over fromo my day job in industrial maintenance).  I'd rather have a plain receptacle than a non functioning GFCI giving a false sesne of protection.
Logged
Steve Swaffer

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 17176
  • Hickory, Mississippi, USA
    • Resotune
Re: GFCI's tripping...
« Reply #20 on: March 29, 2019, 06:07:03 PM »

I assumed you had-just a blanket principle (carry over fromo my day job in industrial maintenance).  I'd rather have a plain receptacle than a non functioning GFCI giving a false sesne of protection.
Not to flog this dead horse (GFCI already replaced) BUT I bench tested it before discarding and it is still functional (responds to test button) BUT, the red "tripped" LED does not light.

I rather have fully functional GFCI outlets near water/plumbing in my casa.

JR
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 17176
  • Hickory, Mississippi, USA
    • Resotune
Re: GFCI's tripping...
« Reply #21 on: April 01, 2019, 01:17:15 PM »

The new improved (skinny) GFCIs arrived and surprisingly about same size as my old ones.   ::)

I disassembled the old faulty GFCI because i could, only 4 screws. I found some odd loose wires floating around inside and more symptoms of heat damage, but the evidence still points to loose (line) mains wire attachment. So mea culpa.   :-[

JR
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

ProSoundWeb Community

Re: GFCI's tripping...
« Reply #21 on: April 01, 2019, 01:17:15 PM »


Pages: 1 2 3 [All]   Go Up
 



Site Hosted By Ashdown Technologies, Inc.

Page created in 0.028 seconds with 22 queries.