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: Nitin Sidhu on October 31, 2015, 02:32:22 AM

Title: Help trouble shoot speaker voltage leak.
Post by: Nitin Sidhu on October 31, 2015, 02:32:22 AM
Hello!

Im a tad stumped here and would appreciate some direction.

We have with us 4 boxes of the FM12 active wedges.

The FM12's will measure a live voltage in excess of 80-100v across its line in terminals in some venues. Other venues they would be absolutely fine. And will measure the difference without any audio connections made.
This is recurring, as in we know venues where the FM12's work flawless and others where they dont.

We measure all circuits using a multimeter and ensure that we have a good ground, and no leak in the N conductor.
My assumption is that this could be caused due to dirty electricity, but honestly I have no clue.

Electricity here is 220-240volts.


Regards,
Sidhu
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Steve M Smith on October 31, 2015, 06:08:39 AM
Do you mean between pins 2 and 3 of the audio input?


Steve.
Title: Re: Help trouble shoot speaker voltage leak.
Post by: John Roberts {JR} on October 31, 2015, 08:56:15 AM
You need to talk to the speaker manufacturer.

Could be a line-neutral reversed mains, but that still should never energize inputs.

JR
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Nitin Sidhu on October 31, 2015, 01:07:56 PM
Do you mean between pins 2 and 3 of the audio input?


Steve.
Thank you gentlemen for your time.

I have been trying, without success, to have multiple queries answered by the manufacture. No response ever. And I am considering a time span of over a year.

These db boxes were retired and left in the warehouse for almost a yr, well replaced with the stx812's. However, they are great sounding and work excellent. When they do.
We needed them again at a venue where we knew they worked well. So deployed them.

Steve. I am going to find time and do another analysis before I answer your query. Thank you.

Sidhu

Sent from my Moto G using Tapatalk

Title: Re: Help trouble shoot speaker voltage leak.
Post by: John Roberts {JR} on October 31, 2015, 01:32:55 PM
With unit unplugged measure between mains cord plug blades and audio input pins. There should be no measurable resistance, and if your VOM has a capacitance scale check that too.

Note: a sensitive, high input impedance VOM could measure significant voltage while still very low current, so not really dangerous. Perhaps a 10K ohm resistor across the VOM input when measuring the voltage would absorb some harmless leakage.

Both the hot and neutral power inlets should be well insulated from anything the customer could touch. If not take it out of service before it hurts somebody.

JR
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Mike Sokol on October 31, 2015, 04:22:29 PM
Sidhu

Refresh my memory, but India uses British power voltages and pin-outs, correct. That is, 230 volts at 60 Hz with a Brown (hot), Blue (neutral) and Green-Yellow (EGC Ground) wire. Do these boxes have standard XLR connections with pin-2 hot, pin-3 cold, and pin-1 shield? Are you using any kind of pin-1 lifts on these XLR cables, or perhaps any audio isolation transformer in the signal path. Do you have any kind of AC power transformers in the system, for instance to step 230-volts down to 120-volt to run "American" gear?

I want to make sure we all understand exactly what we're measuring before we jump to any conclusions or make dangerous suggestions.

Mike Sokol
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Nitin Sidhu on October 31, 2015, 09:36:24 PM
Yes Mike. British voltages. Red-Hot, Black-Neutral and green Earth. Anywhere between 220-240v
Standard XLR's, no ground lifts or transformers. Most small venues will be shore power, single phase powering both FOH and Stage.

We had a run of outputs on our consoles being burnt and found these boxes to be leaking voltage. Then we found them to leak voltage only in certain venues, and not in others.
I have measured in detail in the past, but as its been a while, im going to remeasure it all and return with what I find. I know they leak voltage in the warehouse. By Thursday I should have it done in two venues, with and without the leak.

There is another box in our inventory. The Alto SXM112a, 2 of which, also give us an issue (like a ground hum when connected to an audio source), but this too only does so at certain venues. I want to see if both the Voltage leak in the FM12's, and the noise with the SXm112a are consistent with the venue.

Again, like i said. We measure our electricity to best of our capabilities at every gig every time. I was almost considering getting an oscilloscope to 'see' the voltage. But then read that I dont plug in mains to an oscilloscope if I dont know what im doing. Which i dont.

Thank you JR, will do so.

Regards,
Sidhu
Title: Re: Help trouble shoot speaker voltage leak.
Post by: John Roberts {JR} on October 31, 2015, 09:53:47 PM
If it's eating console output stages there is a good chance it's faulty... the fact that it might not do that all the time does not mean it's OK...

JR
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Jonathan Johnson on October 31, 2015, 11:47:37 PM
I wonder if you have a reverse polarity bootleg ground in the problematic venues. Standard test procedures where you meter the receptacle only with itself will not reveal this problem; you must meter against a known, tested external reference ground.

If this is the case, pin 1 could have voltage present, but I don't really know how that would measure between pin 1 and the signal pins. If it's not a true balanced input, but pin  3 is referenced to ground/pin 1 then that could explain what's going on.

Mike is the expert on RPBG, I'll defer to him.
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Nitin Sidhu on November 01, 2015, 02:00:16 AM
I wonder if you have a reverse polarity bootleg ground in the problematic venues. Standard test procedures where you meter the receptacle only with itself will not reveal this problem; you must meter against a known, tested external reference ground.

We measure to ensure that the Hot-Neutral wires are not swapped. So...
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Steve M Smith on November 01, 2015, 03:41:49 AM
Refresh my memory, but India uses British power voltages and pin-outs, correct. That is, 230 volts at 60 Hz

50Hz.


Steve.
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Nitin Sidhu on November 01, 2015, 04:12:29 AM
50Hz.
Steve.

Sometimes.. :)
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Mike Sokol on November 01, 2015, 06:56:27 AM
We measure to ensure that the Hot-Neutral wires are not swapped. So...

You really need to measure ground-to-ground for each of the outlets and check for any voltage. I take a basic 100 ft extension cord with me and run it from the console power outlet up to the stage. Then I drag the end of the extension cord over to each stage outlet and meter between the ground of the extension cord and ground of the stage outlet. Anything more than 1/10th of the volt can cause ground loop hum for gear with the "pin-1 problem". If you measure around 1/2 of line voltage (60 volts in the USA and 100-ish volts in your venue) then there's probably a floating ground wire. If you measure full line voltage (120 in the USA and 240 elsewhere) then there's a direct connection between the hot and ground wires. If this is a true RPBG then connecting a 100 watt light bulb between the two "grounds" will have the bulb light up at full brightness. A true RPBG will have full circuit breaker current available so you could even light a 1,000 watt PAR light connected ground-to-ground. That's what makes it so scary.

Do not assume that your hot and neutral wires aren't reversed unless you measure it yourself. And has JJ hinted, you must have an external source to measure against. I used to run a test wire to a copper water pipe for measurements, but now that there's so much PVC pipe that doesn't work very well. So running a test wire back to the service panel ground is the best way to test outlets if you can manage it. 
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Geoff Doane on November 02, 2015, 11:38:59 AM
I know you mentioned that you have a good ground, but it sounds a lot like the venues' grounds are not actually connected to anything and are floating.  The capacitance in the windings of the speaker's power transformer will cause its chassis to float up to about half the line voltage, in your case, 100V.  There shouldn't be very much current available, but the high static voltage might be able to destroy output circuits connected to it.

GTD
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Mac Kerr on November 02, 2015, 12:22:33 PM
I know you mentioned that you have a good ground, but it sounds a lot like the venues' grounds are not actually connected to anything and are floating.  The capacitance in the windings of the speaker's power transformer will cause its chassis to float up to about half the line voltage, in your case, 100V.  There shouldn't be very much current available, but the high static voltage might be able to destroy output circuits connected to it.

And hopefully by now you understand that a ground rod or a cold water pipe is not a ground for electrical safety or circuit protection. That demands a ground safety wire run from the outlet back to the service panel to provide an alternate path for fault current to reach the breaker and blow it.

Mac
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Jonathan Johnson on November 02, 2015, 12:39:02 PM
And hopefully by now you understand that a ground rod or a cold water pipe is not a ground for electrical safety or circuit protection. That demands a ground safety wire run from the outlet back to the service panel to provide an alternate path for fault current to reach the breaker and blow it.

Mac

To expand on Mac's point, the equipment grounding conductor (EGC) must be bonded to the grounding electrode conductor (GEC; leads to the earth ground point such as a ground rod, concrete-encased rebar, or metallic underground water line) AND the neutral conductor in the service entrance. This should be the ONLY point of common bond between ground and neutral on the customer side of the utility meter.

Without this ground/neutral bond, there is no reliable return path to trip a circuit breaker or GFI. It might be argued without the bond, the system is floating, but that's not reliable either because in some utility distribution scenarios (such as is common in the United States), all secondary neutrals -- even ones from different utility transformers -- are bonded together and to ground. So while you may be missing the ground/neutral bond in your service entrance, there may be one at the utility transformer or even at your neighbor's utility transformer. So there will be a return path of some sort, though it might be high-resistance. If the resistance of that return path is higher than the resistance of the return path through a human body or through equipment, bad things can happen.
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Nitin Sidhu on November 02, 2015, 02:33:58 PM
I know you mentioned that you have a good ground, but it sounds a lot like the venues' grounds are not actually connected to anything and are floating. 

By good ground I mean a Ground-Live measurement that is within a couple of volts of the L-N difference. If the ground is floating, I have seen we measure a voltage in the Ground wire using a voltage tester..
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Mike Sokol on November 02, 2015, 03:49:21 PM
By good ground I mean a measurement that is within a couple of volts of the mains. If the ground is floating, I have seen we measure a voltage in the Ground wire.

To really know if you have a low impedance ground, you need to provide some momentary fault current and measure it. So, for instance, testers such as the Amprobe INSP-3 and Ideal SureTest Analyzer work by creating a single line cycle "fault" between the line and the EGC "ground" in the receptacle. They then measure how high the ground wire voltage rose to, then calculate an impedance of the fault path. In the US I believe that code calls for less than a 1 ohm impedance of the fault return path back to the G-N-E bonding point. If that same test measures very close to zero ohms, then a bootleg ground in the receptacle/outlet itself is suspected, and some testers such as the SureTest will light up a big F for "False" ground. 
Title: Re: Help trouble shoot speaker voltage leak.
Post by: David Buckley on November 04, 2015, 06:06:11 AM
From an audio perspective, it doesn't matter if the ground works (in the electrical, safety sense) or not; what is important is that all the ground wires of every device are connected together.  So can the OP confirm that they have a proper distro setup, and that power to the console is run from the distro, alongside the multicore(s)?

Title: Re: Help trouble shoot speaker voltage leak.
Post by: Nitin Sidhu on November 05, 2015, 02:32:38 PM
Hello!

As advised, today I metered ground-ground between all receptacles on stage. No voltage. H-N was 231v. H-G was 229, and N-G metered 0 across all. I will assume that this is a single phase distributed through the stage. And we have proper ground.

The speaker is powered up, no audio link.
The FM12 across pins 2-3 would light up my contact voltage tester (one contact neon test light), and cause the speaker to buzz.

The FM when measured with a multimeter (VOM I think for the west) measured 112v across any of the 3 XLR pins on the FM12 and the ground on the power drop. Mike says this could be because of a floating ground. Within the speaker Electronics ? (Note, across all 3 pins)

*Edit : I will ensure our IEC power cables are good tomorrow. Afterthough.

I also did a resistance measurement between the 3 audio pins and the IEC power socket of the speaker. Infinite resistance.

David, when we can, we use our distro, and ensure all is running as should be. However, in many situations, we provide additional tech and gear in venues where installs are available. And do not deploy all our gear. However, we still measure to the best of our education to ensure all is good. This concern is regarding a speaker which I suspect to be leaking high voltage and taking out our console outputs, but only in certain venues. And from our basic measurements, all venues measure well. I really need to get my hands on one of those fancy
tester that Mike suggested... Waiting for someone to take a vacation to India from the US!

Again, thank you all for your time.

Tomorrow I should verify all these measurements again.

Regards,
Sidhu
Title: Re: Help trouble shoot speaker voltage leak.
Post by: David Buckley on November 05, 2015, 10:41:58 PM
There are things not adding up here.  When you are in the venue tomorrow, hold one probe of the DVM in your fingers, and the other probe in the ground pin of the outlet, and note the voltage.
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Mike Sokol on November 06, 2015, 02:25:48 AM
The speaker is powered up, no audio link.
The FM12 across pins 2-3 would light up my contact voltage tester (one contact neon test light), and cause the speaker to buzz.

The FM when measured with a multimeter (VOM I think for the west) measured 112v across any of the 3 XLR pins on the FM12 and the ground on the power drop. Mike says this could be because of a floating ground. Within the speaker Electronics ? (Note, across all 3 pins)

*Edit : I will ensure our IEC power cables are good tomorrow. Afterthough.

Is there some sort of ground lift switch on the speaker's built-in amplifier? Either that, or the IEC cable has an open ground pin. That's the only thing that makes sense for this measurement. If either is the case, then the chassis of the speaker amp will try to float up to 1/2 line voltage if it's not connected to the mixing console ground via pin-1.

So do you have a pin-1 lift in your XLR cable feeding the speakers? Do you hook up all speakers first before powering up everything? That's best practice on most stages for initial setup, but during festival changeover I just make sure individual power switches on speakers are off before plugging in speakers to extra feeds. Heck, we sometimes yank XLR cables out and move to different console outputs while the speakers are still powered up. In the middle of battle things get hectic and shortcuts are sometimes taken. But I could see that doing hot patching could cause a voltage pulse back into the outputs of the mixing console. Is that what's causing your console outputs to fail? Are you hot ptaching without powering down your monitors?
Title: Re: Help trouble shoot speaker voltage leak.
Post by: Peter Morris on November 16, 2015, 06:27:31 AM
I know you mentioned that you have a good ground, but it sounds a lot like the venues' grounds are not actually connected to anything and are floating.  The capacitance in the windings of the speaker's power transformer will cause its chassis to float up to about half the line voltage, in your case, 100V.  There shouldn't be very much current available, but the high static voltage might be able to destroy output circuits connected to it.

GTD

I think you are on the right track.

Usually when you see around 120 volts appearing in strange situations like this it’s caused by a missing earth.

For example where the earth wire is disconnected at the start of a cable it can act as a centre tapped capacitor – think in terms of the Active being on plate of a capacitor, the Neutral another,  and the Earth as a middle plate.  You can try it with an extension cord – disconnect the earth at the plug end and then measure the voltages at the socket end.
There will be (about) 120V between the earth wire and active and the earth wire and neutral.

If your speaker has the metal parts/ input connector earthed, the supply has 3 wires and the earth is not connect at the start of the cable this can happen.  When you plug it into your mixer (which is earthed) you will find 120V with enough current to give you a noticeable shock.

I would be looking for something like this.