ProSoundWeb Community
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 => Topic started by: Eric_Muller on May 22, 2014, 03:18:44 PM
-
Hello,
I always have trouble tripping GFCIs with a certain power distribution box supplied with a local mobile stage. This year I would like to bring my own. I can't seem to pinpoint the proper feeder connector they describe, even people I have asked say it does not exist. The box looks like a typical yellow box, a "Coleman Extreme" for example.
The feeder connection information I have received is...
"ok, below is from the stage guy ...just received....hope this helps!!
50 AMP, 3 PHASE WITH GROUND AND NEUTRAL TWIST LOCK"
and
"The connection at the end of the feeder is:
50 amp
4 wire
3 phase with neutral
It has neutral, phase 1,2,3 and ground"
Thanks,
Eric
-
L21-50 does not exist. It is listed but no one makes it. The locking part is the real stumper there. It could be a pin and sleeve. A pictures are worth a thousand words here...
Odds are though it is a California connector: http://www.lockingpowercords.com/category/31-hubbell-50a-twist-lock.aspx
-
"The connection at the end of the feeder is:
50 amp
4 wire
3 phase with neutral
It has neutral, phase 1,2,3 and ground"
Wouldn't that be 5 wires?
Steve.
-
4 wire
3 phase with neutral
It has neutral, phase 1,2,3 and ground"
One of these is not like the others.
"It has neutral, phase 1,2,3 and ground" is 5 wires, not 4 as stated.
Here are the configurations available in 50A twist lock. Notice some are high voltage.
http://www.lockingpowercords.com/category/31-hubbell-50a-twist-lock.aspx
-
Wouldn't that be 5 wires?
Steve.
When describing 3 phase systems you never include the ground, a 3 wire system does not have a neutral, a 4 wire system includes a neutral. Ground is assumed.
Now, does the guy on the other end of the phone know this... I don't know... but there is no such thing as a 5 wire 3 phase system...
-
When describing 3 phase systems you never include the ground, a 3 wire system does not have a neutral, a 4 wire system includes a neutral. Ground is assumed.
Now, does the guy on the other end of the phone know this... I don't know... but there is no such thing as a 5 wire 3 phase system...
You can't assume anything. Household wiring such as romex will be described such as 14-2 with ground or 12-3 with ground. In that case the ground is not covered to the same extent as the rest and can't be considered for current carrying. SOOW or similar will be called 12-4. In that case there are only 4 wires INCLUDING ground.
In this case the "feeder" should be SOOW or similar and the full conductor count includes ground. 4 wire cable would indicate 2 hot, a neutral and a ground. If it is 4 wire 3 phase then the ground/neutral are bootleg and that may explain some of the OP GFI problems.
-
"The connection at the end of the feeder is:
50 amp
4 wire
3 phase with neutral
It has neutral, phase 1,2,3 and ground"
The 50A CS "California" style connectors are listed on the plug as "125/250v 3P/4W"...which stands for 3 POLE / 4 wire...NOT 3 PHASE!...now there is a 3 phase version of the 50A CS connector bit it will say "250V 3 Phase" and will not have a neutral.
Most likely he confused the 3 pole with 3 phase and what you have is a single phase 50A connector into a jobsite box that breaks out to 6 20A outlets (with a 50A max per leg)
-
The 50A CS "California" style connectors are listed on the plug as "125/250v 3P/4W"...which stands for 3 POLE / 4 wire...NOT 3 PHASE!...now there is a 3 phase version of the 50A CS connector bit it will say "250V 3 Phase" and will not have a neutral.
Most likely he confused the 3 pole with 3 phase and what you have is a single phase 50A connector into a jobsite box that breaks out to 6 20A outlets (with a 50A max per leg)
This is what I am ASSuming too. Looks like at this weekend's gig I will be using their distro as I still haven't heard any clarification.
Is there anything I can do to mitigate GFCI tripping?
Usually I take 2 circuits out of the supplied stage distro and plug them into my own 2 x 20 amp distro (foh, monitor, stage power).
Then I run each main speaker group off it's own circuit, fed directly from the supplied stage distro.
I use 4 circuits off their power distro, 2 of them are interconnected, but separately breakered within my own rack distro.
Is it interconnection of circuits via power distro or system interconnection that causes GFCI tripping? Varying loads amongst the circuits? GFCI age and condition?
The age of the GFCIs is unkown. They do leave the power distro out overnight on the wet field, my recent investigation tells me that is not good for GFCIs. My gear works everywhere else (I can't believe I am the one saying that, lol).
When this distro started tripping last year I kept 1 circuit from my stage and ran 100 ft of 10 gauge cord to another power distro fed from a different feeder in a different location. All power comes from cement ground boxes, no generators. We limped through it fine on 2 circuits, but I wasn't happy.
I have another gig booked here, for that gig I HAVE to bring a bigger system so I will need more power. I will take pictures of the feeder and distro this weekend and post back here. Thanks for the replies.
Eric Muller
-
This is what I am ASSuming too. Looks like at this weekend's gig I will be using their distro as I still haven't heard any clarification.
Is there anything I can do to mitigate GFCI tripping?
Usually I take 2 circuits out of the supplied stage distro and plug them into my own 2 x 20 amp distro (foh, monitor, stage power).
Then I run each main speaker group off it's own circuit, fed directly from the supplied stage distro.
I use 4 circuits off their power distro, 2 of them are interconnected, but separately breakered within my own rack distro.
Is it interconnection of circuits via power distro or system interconnection that causes GFCI tripping? Varying loads amongst the circuits? GFCI age and condition?
The age of the GFCIs is unkown. They do leave the power distro out overnight on the wet field, my recent investigation tells me that is not good for GFCIs. My gear works everywhere else (I can't believe I am the one saying that, lol).
When this distro started tripping last year I kept 1 circuit from my stage and ran 100 ft of 10 gauge cord to another power distro fed from a different feeder in a different location. All power comes from cement ground boxes, no generators. We limped through it fine on 2 circuits, but I wasn't happy.
I have another gig booked here, for that gig I HAVE to bring a bigger system so I will need more power. I will take pictures of the feeder and distro this weekend and post back here. Thanks for the replies.
Eric Muller
I'm not exactly clear what you are doing, but if you are trying to combine two circuits to feed an L14-30 or something, that will definitely not work from GFCI circuits, and really isn't OK anyway.
GFCIs can potentially go bad, it is also possible that you have something leaky in your rig that you aren't aware of. What happens when you measure continuity (devices unplugged of course) from ground to neutral of male plug end of each of your loads - i.e. unplug the cord from the distro and probe hot to ground on the male plug? the resistance is less than 40K ohms, that means you have 3ma of leakage current. If the number is somewhat higher but you plug several power strips into the same circuit, the leakage is cumulative.
-
It is not clear where the GFCIs are located? If neutrals are shared/parallel/connected to each other or to ground anywhere after the GFCIs you will have trouble with tripping. Basically, the hot wire from the GFCI and the neutral from that GFCI must be connected to the same loads and ONLY to those loads. If that is the case, then as Tom said you may have leakage in equipment, but I would think that should show up at other locations (unless there is a possibility of getting moisture in your equipment overnight here and never anywhere else?)
FWIW, newer GFCIs are "supposed" to reset only if they are working right-not knowing the age it would be hard to tell if this is the case.
-
It is not clear where the GFCIs are located? If neutrals are shared/parallel/connected to each other or to ground anywhere after the GFCIs you will have trouble with tripping. Basically, the hot wire from the GFCI and the neutral from that GFCI must be connected to the same loads and ONLY to those loads. If that is the case, then as Tom said you may have leakage in equipment, but I would think that should show up at other locations (unless there is a possibility of getting moisture in your equipment overnight here and never anywhere else?)
FWIW, newer GFCIs are "supposed" to reset only if they are working right-not knowing the age it would be hard to tell if this is the case.
I think we are talking about a standard Spider Box, with 6-20A circuits broken out from a 50A/220V connection, wherein there is GFCI protection on each circuit, all handled within the Spider Box itself.
-Ray
-
In that case, I would suspect "creative" wiring in the spider box, or bad GFCIs.
If it were me, I might be tempted to "borrow" their connector for my own distro, take a pic and be ready for next year? But I am an electrician so my doing so wouldn't raise many questions.
Thinking out loud-could a poor connection/intermittent neutral in the feeder cause GFCIs to trip? (Essentially putting loads in series across the 2 hots-not good either!) Might pay to check the voltage on each hot to neutral at spider box and at receptacle.
-
I think we are talking about a standard Spider Box, with 6-20A circuits broken out from a 50A/220V connection, wherein there is GFCI protection on each circuit, all handled within the Spider Box itself.
-Ray
Yes, GFCIs are in the spider box.
I did ask for clarification on the voltage when I received the original (confusing) information. However, I never got an answer, even after visiting the site office and being told I would be contacted.
My rack distro takes 2 inputs from standard wall receptacles and breaks out into multiple receptacles. it is a centralized, 2 x 20 amp power strip. Nothing fancy, all my speakers are powered so I use combo cables and this rack distro allows me to patch everything from the front. I have been using it for 3 years now.
When I need one circuit (small bar gigs) I just jumper circuit 1 into input 2 and all receptacles are powered.
When I need 2 circuits, I source 2 circuits from the venue and use both inputs.
When I need more power, I source 2 circuits for the rack distro and use it for monitors, backline and mixrack. Then I source more circuits for main speaker power connection.
I provide mainly for smaller shows. 99% of the time I get my power from a standard wall receptacle and I only need 1 or 2 circuits.
-Eric
-
GFCIs can potentially go bad, it is also possible that you have something leaky in your rig that you aren't aware of. What happens when you measure continuity (devices unplugged of course) from ground to neutral of male plug end of each of your loads - i.e. unplug the cord from the distro and probe hot to ground on the male plug? the resistance is less than 40K ohms, that means you have 3ma of leakage current. If the number is somewhat higher but you plug several power strips into the same circuit, the leakage is cumulative.
Actually, there's a better and really easy way to test for ground leakage currents. As I'm sure you know, a clamp-amp meter needs to be connected around a single wire to measure AC current. So if it's clamped around both the neutral and hot the currents will null-out since they're going in opposite directions. However, if the current is leaking somewhere else (to an amplifier on another circuit or whatever), there will no longer be a perfect null, and a clamp-meter around the entire power cord WILL read the leakage currents, even while there's many amperes of normal Hot-to-Neutral current at the same time. Of course, once your external leakage currents get over 5mA you'll be a GFCI trip. This test shows you the external leakage currents to earth or another circuit. If you want to know the actual leakage current of a stand-alone piece of gear, you need to make a simple split out cable and clamp around either the ground wire by itself, or the hot and neutral wires, ignoring the ground wire. Either way will give you a direct readout of hot-to-chassis leakage.
This is a pretty good way to measure hot-to-chassis leakage currents in old tube guitar amps and such, which often have insulation failure in aged power transformers. Of course, the UL ratings for leakage currents with "grounded" gear is 3.5 mA max, and "ungrounded" gear is 0.75 mA, so don't expect zero leakage. Basically, anything hooked to a power line leaks a bit. If you want to have some real fun, plug your smartphone into a wall-wart charger and measure the voltage on the metal body of the phone in reference to earth ground. You'll probably read around 1/2 line potential (60 volts or so).
-
Eric, thanks for the picture.
There are a couple of explanations, the first is a common neutral bus bar (shouldn't be case here, Six2 knows better); another is neutral grounded inside the box (ditto); the final one is leakage in the Edison to PowerCon cables. I've seen shop-built cables trip GFCIs due to less than ideal assembly practices....
You can test for the first 2 by using your ohm meter to check for open or short between the neutral slots of the red and black circuits. The neutral should be the taller slot, typically in the 9 o'clock position. You should find this to be an open. A short indicates that somehow the neutrals from both inlets are tied together. Test for ground bonding by checking for short/open between the neutral and ground pin for both Black and Red circuits. These, too, should test as open circuits.
That takes us back to the cords you're powering the distro with.
-
Actually, there's a better and really easy way to test for ground leakage currents. As I'm sure you know, a clamp-amp meter needs to be connected around a single wire to measure AC current. So if it's clamped around both the neutral and hot the currents will null-out since they're going in opposite directions. However, if the current is leaking somewhere else (to an amplifier on another circuit or whatever), there will no longer be a perfect null, and a clamp-meter around the entire power cord WILL read the leakage currents, even while there's many amperes of normal Hot-to-Neutral current at the same time. Of course, once your external leakage currents get over 5mA you'll be a GFCI trip. This test shows you the external leakage currents to earth or another circuit. If you want to know the actual leakage current of a stand-alone piece of gear, you need to make a simple split out cable and clamp around either the ground wire by itself, or the hot and neutral wires, ignoring the ground wire. Either way will give you a direct readout of hot-to-chassis leakage.
This is a pretty good way to measure hot-to-chassis leakage currents in old tube guitar amps and such, which often have insulation failure in aged power transformers. Of course, the UL ratings for leakage currents with "grounded" gear is 3.5 mA max, and "ungrounded" gear is 0.75 mA, so don't expect zero leakage. Basically, anything hooked to a power line leaks a bit. If you want to have some real fun, plug your smartphone into a wall-wart charger and measure the voltage on the metal body of the phone in reference to earth ground. You'll probably read around 1/2 line potential (60 volts or so).
What kind of clamp meter do you have that can read down to 3mA? Mine is pretty dubious for less than 1/4 amp or so. When leakage is to the ground pin (and not through the environment), it is still in the cord, which is still in the jaws of the clamp, and should still cancel. What am I missing?
-
What kind of clamp meter do you have that can read down to 3mA? Mine is pretty dubious for less than 1/4 amp or so. When leakage is to the ground pin (and not through the environment), it is still in the cord, which is still in the jaws of the clamp, and should still cancel. What am I missing?
Here's a pic of the Fluke meter I use for bench testing of gear leakage which measures down to 1/10 of an mA. It's a bit expensive for field work, but I also a few less expensive clamp testers that go down to 1 mA resolution. I'll look in my field kit later today and let you know the brands/models.
Also, note that I was talking about external current flow (not through the ground wire in the extension cord) being measured with the clamp meter around the entire extension cord. I also noted that if you wanted to find internal current flow (back through the power cord's EGC) you'll need to split out the EGC wire for the measurement as pictured. Maybe that wasn't clear in my previous post, but it should be obvious once you think about it a bit.
And this same idea is what allows you to put a clamp meter around the outside of an XLR mic cable to measure ground loop current flow. Since the ground loop current is only in the shield and loops back through the external grounding system, you don't need to split out the shield from the twisted pair. Just clamp the meter around the mic cable for a direct measurement. If you need more resolution, looping the mic cable through the clamp jaws a few times acts as a resolution multiplier. So 5 loops is 1/5th of the read current, and 10 loops is actually 1/10th of the read current. For example, if you loop a wire 10 times through the amp jaws and your meter now reads 10 mA, there's actually 1 mA of current flow in the test wire.
The reason for all this testing is to identify exactly what fault currents in the field are going where, which can give you a handle on hum current paths and how they act. I've built all sorts of current injecting demonstrations which is what allows me to experiment with hum testing on my test bench at will. I don't know all of it yet, but it's making me a much better field troubleshooter.
-
Here's a pic of the Fluke meter I use for bench testing of gear leakage which measures down to 1/10 of an mA. It's a bit expensive for field work, but I also a few less expensive clamp testers that go down to 1 mA resolution. I'll look in my field kit later today and let you know the brands/models.
Also, note that I was talking about external current flow (not through the ground wire in the extension cord) being measured with the clamp meter around the entire extension cord. I also noted that if you wanted to find internal current flow (back through the power cord's EGC) you'll need to split out the EGC wire for the measurement as pictured. Maybe that wasn't clear in my previous post, but it should be obvious once you think about it a bit.
And this same idea is what allows you to put a clamp meter around the outside of an XLR mic cable to measure ground loop current flow. Since the ground loop current is only in the shield and loops back through the external grounding system, you don't need to split out the shield from the twisted pair. Just clamp the meter around the mic cable for a direct measurement. If you need more resolution, looping the mic cable through the clamp jaws a few times acts as a resolution multiplier. So 5 loops is 1/5th of the read current, and 10 loops is actually 1/10th of the read current. For example, if you loop a wire 10 times through the amp jaws and your meter now reads 10 mA, there's actually 1 mA of current flow in the test wire.
The reason for all this testing is to identify exactly what fault currents in the field are going where, which can give you a handle on hum current paths and how they act. I've built all sorts of current injecting demonstrations which is what allows me to experiment with hum testing on my test bench at will. I don't know all of it yet, but it's making me a much better field troubleshooter.
While that is really interesting and certainly gives more data, I'm not sure all of this is easier than just measuring hot -> ground impedance to find leakage that will trip a GFCI. It may not give you directional flow, but should by process of elimination indicate gear that is leaking - weak MOVs, transformer problems, etc.
-
Update: Here's a pic of the Fluke 322 clamp-meter I use which has 10 mA resolution. As noted above, you can loop the test cable through the clamp jaws a few times to increase the resolution.
-
While that is really interesting and certainly gives more data, I'm not sure all of this is easier than just measuring hot -> ground impedance to find leakage that will trip a GFCI. It may not give you directional flow, but should by process of elimination indicate gear that is leaking - weak MOVs, transformer problems, etc.
I think both test methods have validity. I like to measure actual leakage currents in the field so I can duplicate them on the bench. That's because I need to know the actual amperage/voltage values as well as which way the current it going. In that way I can correlate any theoretical models with empirical data and know if I'm on the right track. Your method is quick and dirty (which can be good under pressure), and while my method takes more gear and setup time, it yields more data (which is sometimes more confusing).
-
Eric, thanks for the picture.
There are a couple of explanations, the first is a common neutral bus bar (shouldn't be case here, Six2 knows better); another is neutral grounded inside the box (ditto); the final one is leakage in the Edison to PowerCon cables. I've seen shop-built cables trip GFCIs due to less than ideal assembly practices....
You can test for the first 2 by using your ohm meter to check for open or short between the neutral slots of the red and black circuits. The neutral should be the taller slot, typically in the 9 o'clock position. You should find this to be an open. A short indicates that somehow the neutrals from both inlets are tied together. Test for ground bonding by checking for short/open between the neutral and ground pin for both Black and Red circuits. These, too, should test as open circuits.
That takes us back to the cords you're powering the distro with.
Tim, I had an employee once.... once. He never assembled cabling though... more of a self proclaimed "A1 mix god" after 25 shows.
The distro feeder cables are made by me. Photo attached. No stray strands or obvious (to me) issues. No continuity between terminals.
I recently went through my pre built CBI powered speakers cables too. Some need some light fixing like re stripping to get the connector strain relief back on the cable housing, but they were all functional and there were no blatant mis-connections or "only 1 strand still connected" situations.
The distro has separate neutral bars for each circuit. The red circuit's neutral bar has only 2 taps, 1 coming from the input and one going to the first red receptacle. Then the neutral daisy chains from red receptacle #1 to red receptacle #2. On this distro, the 2 front red receptacles are the only receptacles fed by input 2.
There is NO continuity between circuit 1 and circuit 2 neutrals (white wire/silver terminal). As expected, there is continuity between the ground/green terminals as the ground bus bars are connected at the chasis ground screw located on the left, behind the circuit breakers. See pic for test setup.
-Eric
-
bigger pic of the distro guts in my dropbox...
https://www.dropbox.com/s/g5nwk061kmw4q7i/power%20distro%20big.jpg
-Eric
-
bigger pic of the distro guts in my dropbox...
https://www.dropbox.com/s/g5nwk061kmw4q7i/power%20distro%20big.jpg
-Eric
I don't see anything obvious wrong with your distro equipment. Keep in mind that GFCI tripping usually happens because of equipment, not the distro. Guitar amps can be problematic, as can surge protectors with MOVs that are failing. I would suggest investigating the things you plug into your distro next.
-
I don't see anything obvious wrong with your distro equipment. Keep in mind that GFCI tripping usually happens because of equipment, not the distro. Guitar amps can be problematic, as can surge protectors with MOVs that are failing. I would suggest investigating the things you plug into your distro next.
Yep, I wanted to rule out the front end of this first. If connecting the distro (without loads connected) makes a GFCI trip, the problem is in the distro or the cables/connectors feeding it. When we know the distro/cables/connections are 100% good, he can start connecting stuff until the GFCI trips.
-
Eric says his gear works "everywhere" else-does that include other venues with GFCIs or not? If the same gear, hooked up the same way trips one GFCI but not another I would suspect a bad GFCI-though the ba d one could be the one that does not trip! On the other hand, if they have a bad GFCI. you'd think they would have other complaints and would hopefully have found the problem.
-
Of course after all the discussion and worry the power caused no problems this year.
I kept subs, mains, monitors, band backline and my mix gear on their own circuits/GFCIs so I could trouble shoot if needed.
It was nice that the venue representative sent me a picture of the side of the feeder cable connection the day before the event (I contacted them a month before) ::)
I now know the venue's power connection specifics so at least I can buy my own power distro as they have been storing their "weatherproof" distro outside for years and I still don't want to rely on it for the heavier gig coming up.
I really don't know why they wouldn't let me see it earlier, apparently the (wrong) info they gave was supposed to be enough.
Thanks for all the help and information. I read all of it and will read it again.
Eric