PA rack - electric shock

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Soeren_DK

Well-known member
Joined
Jun 7, 2004
Messages
526
Location
Denmark
Hi there.
I have a serious problem I can't seems to fix.
When we where out playing a gig last weekend the bass player almost got electrocute when he touched the bass and a radiator simultaneous.
Not good not good.

We do not run our PA-rack with a grounding connection because a lot of our gigs doesn't have a plug with it in. Most of the time we play in small village halls build before World War 1.

I powered up the rack in my storage hall and measure 120VAC with my voltmeter between an radiator and our rack. I have unpluged all machines and tried to eliminate the source but every unit seems to add some voltage to ground. Some of the units has 44VAC between rack and ground and some units have 120VAC.. I have tried to pull all the units out of the rack but nothing helps. When I puts one unit back the 120VAC is showing again.

What am I doing wrong?

Cheers
Søren
 
Hi Søren,

I can't really follow you.

First let's talk about the venue where the incident happened.

I understand there was no ground (earth) connection on the wall socket where you plugged in. Simply Live and Neutral.

Then there is the (heating I assume)  radiator in the venue. Your bass player touched the guitar and the radiator and got a shock.

One possibility is that his amp (valve I assume) had polarity reversed and had live on the chassis. This is normal. He basically provided a path to ground and naturally got a shock.

The second possibility is there was a mains leakage on the radiators. I have seen this in many places where there is no earth connection within the mains installation, and amps of current running on the water pipe developing a voltage drop over it. So, it is possible that your bass player actually provided a path between live coming from the radiator and the neutral through his bass guitar, as opposed to live coming from his guitar.


Now, in your storage hall.

Again you are not clear. If you connect one probe to the live on your rack and the other to the heating radiator you will naturally read  a voltage. But this has nothing to do with your equipment safety.

You have to check whether there is a voltage leakage between the chassis and the neutral of the equipment under test. Or just use a simple good old fashion light pen. If it lights up when it is in contact with the chassis of the equipment then there is leakage. If the equipment has ground connection to the chassis, and there is a ground connection on the wall socket that the equipment is connected to, then either the ground connection on the equipment is faulty, or the ground connection between the equipment plug and the wall socket is faulty, or the ground connection in the mains sytem is faulty.

Basically seeing the radiator as a safety ground mechanism is a wrong notion unless it is connected to the mains ground network.  The radiator and the pipes do not have zero ohm resistance. This resistance will increase when the joints are corroded and there will be a voltage drop over it.








 
Thanks for trying to answer my question with my lack of information.
The bass player plays through a Bo Hansen DI directed feed from the mixer so he is not using any amp.
If the radiator is connected with copper pipes normally the radiator is connected to ground, isn't it?

I tried to measure again today in the storage hall.
I took the rack to a ground connection (earth rod). Without the rack connected to this ground I measured the same 120VAC. I tried with the DI and a jack cable connected. When I touched the shield on the jack and the earth rod I could feel some ticklish so there most be some leakage. I suspect the distribution socket back of the rack.

Does it give any meaning

Cheers
Søren
 
You have to clarify which ground you mean. Earth-rod ground or mains ground?

The radiator being connected to a copper pipe does not mean that it is connected to ground. Yes the water pipe eventually goes under the ground but you have to think of the joint resistances. What happens if the joints are not soldered but simply compression joint? It will eventually oxide and present high resistance. Same for the copper pipe under the ground. If the surface of the pipe is oxided, again it will present high resistance. So that still does not protect you unless it is connected to a proper plate or rod earth grounding scheme.

When you say you are measuring 120V, do you mean between the jack shield and the ground (earth rod)?

If yes, in which case the Neutral on your rack is not connected properly. Check every single connection starting from where your rack connects to the wall socket and all the way the equipment. The neutral connection on the distribution socket could be lose. It will naturally spark and eventually oxide to a degree that it will have tens or even hundred of ohms creating a voltage drop on your equipment chassis. That is probably why you are getting a tickle.

 
Grab a pair of 1kV .1uF line isolation caps and install them in your rack's plug strip.  Tie 1 hot to ground and the other neutral to ground.  Put a bonding jumper between the plug strip ground and the rack chassis.  Put a grounded plug back on the plug strip and use a lifting adapter when premises wiring has no earth pin.  Also add an external ground lug.  Carry a roll of wire with an alligator clip on one end and tied to the ground lug on the other end.  Clamp it to building steel or a cold water pipe.  Just because a receptacle doesn't have a ground doesn't mean the building doesn't have a ground.  Ask your bass player.

When running electronics, especially tube gear, it is not at all uncommon to to have a hot chassis when ungrounded.  There's no substitution for proper grounding but line isolation caps (aka "death caps") will work in a pinch.  I do not recommend line isolation caps as a sole means of protection, but coupled with proper grounding, you now have a belt and suspenders.
 
The problem is that nearly every piece of modern equipment has an XY capacitor. That's the combination of one cap between Live and Neutral, one cap between Live and Ground and another cap between neutral and ground. These caps have nothing to do with safety, they are there just for interference protection.
The problem is that, when you run such an equipment without a ground connection, the capacitors form a voltage divider, which normally would  result in half the mains voltage. I bet this is what happens because your mains is 240V, am I right?
The fact that you measure only 40V or so with one piece and increasing voltages when you increase the number of pieces is probably because the voltmeter you use is not very sensitive and loads the circuit.
It is absolutely clear that using this equipment on non-grounded receptacles is not only illegal, but also dangerous.
However, I understand that you need to find a solution. One of the solutions would be to have a long ground wire that you would use to connect to a fixture that you know is grounded. I understand it's not very practical, but that would solve the problem both in therms of legality and in terms of safety.
Now an alternative solution would be to disconnect (remove) the XY caps in all the equipment. That would not be a very nice solution, because ther is always a leakage capacitance in mains transformers, but it is generally low enough to not be lethal. that would still be illegal, but certainly safer.
Many american equipment made before the 80's were using a different approach to this subject. Since non-grounded receptacles were the norm, the equipment had a cap that could be switched to either side of the plug, to allow connecting it to Neutral (the other side of the cap was connected to the chassis ground. You may have found such arrangements in Fender amps. Although this solution allows in theory to solve both the safety and hum problems, it is important to note that, put on the wrong side, the switch would connect the chassis to Live. in the US, it was a "minor" problem, because mains voltage was 115V. In Europe, this was a potential recipe for disaster. I had a friend who was electrocuted this way; he got such violent contractions, he broke the bones in his spine. he is now in a wheelchair.
All things considered, I think the only viable solution for you is to buy an isolation transformer. These are meant for the equipment of public lavatories. They are guarantedd having less than 120pF leakage capacitance, which is considered safe, even with 240V. These are costly and bulky, but life has no price...
 
Butterylicious said:
>These caps have nothing to do with safety
I beg to differ.
http://www.justradios.com/safetytips.html
From that link:
Line filter capacitors also help to keep you safe from nasty (possible fatal) electrical shocks, by isolating the “hot” side of your 120 volt power line from your radios chassis.
Connecting a Class Y capacitor between mains live and the chassis is a funny way of isolating it! Someone seems to be confusing interference suppression with safety. Abbey Rd is correct - unless the chassis is grounded properly these are a safety hazard, especially with 220 or 240V mains.
 
Butterylicious said:
>These caps have nothing to do with safety
I beg to differ.

http://www.justradios.com/safetytips.html
What is correct is that you can't use just any old cap between mains and ground, and it's agood thing that there's been a standardized type for this application.
The rest, what John R says.
 
In countries where there is no safety ground in the installations the Neutral is tied to the earth pin on the wall socket.

I read the article in the link and it already contradicts itsel with the line filter (Y) capacitor. Anyhow, as Abbey and JohnR indicated it has nothing to do with safety. In contrary it can be a hazard.

Think about the Y cap as JohnR mentioned and think about the high resistance issues due to the pipe joints that I mentioned. If the pipe installation is not grounded properly you'll have voltage drop on it. So not only you are not safe but you also put other people at risk because somebody may touch that radiator or the water pipe and get electrocuted.





 
People tend to get complacent about Class Y capacitors because they have a relatively high impedance at 50 or 60Hz. Unfortunately high speed transients will see it as a much lower impedance. If these transients are common mode (ie. the same on live and neutral) a chassis that isn't grounded properly could rise to the full transient voltage.

In medical equipment it's illegal to put a capacitor between live and ground. I use medical grade filters on my own audio gear, partly because of safety concerns, but also because I don't like noise currents in the ground conductors. Likewise I chop out the MOVs between live and ground in any "surge protected" mains adaptors or multi-way outlets I use.
 
johnR said:
.... Likewise I chop out the MOVs between live and ground in any "surge protected" mains adaptors or multi-way outlets I use.

I remember Bob Katz saying exactly the same thing.
 
I bet this is what happens because your mains is 240V, am I right?
Yes exactly.

One of the solutions would be to have a long ground wire that you would use to connect to a fixture that you know is grounded. I understand it's not very practical, but that would solve the problem both in therms of legality and in terms of safety.
I think this is the best solution. I don't want to remove any thing inside our machines.

You have to clarify which ground you mean. Earth-rod ground or mains ground?
I don't exactly know what you mean but here we don't have a mains ground. The power supplier/power planet does not support our homes with ground. Every household has to make a ground connection. The electrician makes a earth rod in the back yard and connect it to the house installation also called an TT-system-ground

Old houses does use the neutral as ground. This system is called TN-C-system-ground.

But its the same. The neutral is tried to an earth rod in the giant transformer station.

Likewise I chop out the MOVs between live and ground in any "surge protected" mains adaptors or multi-way outlets I use.
What is MOV's? I don't fully understand this.

Cheers
Søren  
 
This is a very interesting read, also because I have had a similar experience at home.

I live in a house which is a little more than 20 yrs old. I received a horrible tingly feeling through my arm when I was leaning over to adjust the mixer and made contact with the metal edges of my equipment rack; it felt like that feeling you get when you get nettle rash!!

I stuck a multimeter across the two points (metal headphone socket on the mixer and metal edge on the rack enclosure) and measure over 120V - I reckoned it was from the IEC Mains Filter on my mixer and assumed I had a bad path to ground.

After checking the IEC cords, and the resistance of the rack edge to the earth pin of the socket I thought I must have a grounding issue at the socket.

I borrowed a socket tester (you know the type with LEDs on the front where you plug it in and the lights indicate fault or OK on each pin) - all OK.

If I get the electrician in, they'll probably suggest replacing the entire consumer unit first (regulations have moved on a lot here in the UK) and I don't own the house - it's rented.
 
solderspongebob said:
I borrowed a socket tester (you know the type with LEDs on the front where you plug it in and the lights indicate fault or OK on each pin) - all OK.
these testers just check the voltages. If the ground lug is not connected, the tester will see zero voltage and conclude it is correct, but as soon as you connect any piece of gear with a leak between Live and Ground (such a s an XY cap), the voltage won't be zero any more.
Installing a ground connection is not very difficult. even if you can't drive a wire through the conduit, you can run it externally, may not be very nice, but it will be safe.
You say you rent the house, I think the regulations are the same all over Europe, the landlord has a duty to have the electrical installation updated.
 
Back
Top