Shock from switch

Peppa Cig

Hi, I am not a an electrician But comfortable enough looking. we have metal light switch in bedroom And get mild shock from it when we touch it. I have looked at back of switch. All connections look good. I see that metal back box is "live" using a crappy phase tester Suggesting to me that "current" coming from earth cables? Why would this be? Could it be the switch? This switch is near bed it is connected with switch at door. Could door switch/wiring be the culprit?

I can get a multi meter from friend if that would answer any questions....
Cheers.

aido79

Peppa Cig wrote: »

Hi, I am not a an electrician But comfortable enough looking. we have metal light switch in bedroom And get mild shock from it when we touch it. I have looked at back of switch. All connections look good. I see that metal back box is "live" using a crappy phase tester Suggesting to me that "current" coming from earth cables? Why would this be? Could it be the switch? This switch is near bed it is connected with switch at door. Could door switch/wiring be the culprit?

I can get a multi meter from friend if that would answer any questions....
Cheers.

What type of light fittings are they switching? The problem may be with one of the earth connections in the light fittings.

Peppa Cig

aido79 wrote: »

What type of light fittings are they switching? The problem may be with one of the earth connections in the light fittings.

hi, Pendant ceiling light, 12v downlights over bed, bedside table lights which are plugged into 5amp socket.

It's not a new problem was always there as far as I know. I am thinking a plastic back box will solve it or do I need to identify root cause?

aido79

Peppa Cig wrote: »

hi, Pendant ceiling light, 12v downlights over bed, bedside table lights which are plugged into 5amp socket.

It's not a new problem was always there as far as I know. I am thinking a plastic back box will solve it or do I need to identify root cause?

Plastic back box will not solve the problem..unless you change the switch to a plastic switch as well. Better to get to the root of the problem. You'll need a multimeter to check how much voltage there is in the back box.

Peppa Cig

aido79 wrote: »

Plastic back box will not solve the problem..unless you change the switch to a plastic switch as well. Better to get to the root of the problem. You'll need a multimeter to check how much voltage there is in the back box.

Cheers Aido. I'll check voltage on back box if someone kind enough to tell me how I.e setting on multi and where to place probe. I guess one on box and one on earth?

aido79

Peppa Cig wrote: »

Cheers Aido. I'll check voltage on back box if someone kind enough to tell me how I.e setting on multi and where to place probe. I guess one on box and one on earth?

That won't do any good because electrically they are the same point. Ideally you would use a flylead and put one end on the main earth bar of the fuseboard. The multimeter would then be used to measure the voltage at the back box by putting one lead of the meter on the end of the flylead and the other on the back box using the ac voltage setting. If you are not comfortable opening the fuseboard then you may be able to use the bonding from a copper pipe in your hot press as your earth.

Peppa Cig

aido79 wrote: »

That won't do any good because electrically they are the same point. Ideally you would use a flylead and put one end on the main earth bar of the fuseboard. The multimeter would then be used to measure the voltage at the back box by putting one lead of the meter on the end of the flylead and the other on the back box using the ac voltage setting. If you are not comfortable opening the fuseboard then you may be able to use the bonding from a copper pipe in your hot press as your earth.

Fuseboard a very long distance away. How about the earth on a nearby double socket?
Thanks

zega

How would this not trip the MCB if backbox is live and also earthed?

Peppa Cig

zega wrote: »

How would this not trip the MCB if backbox is live and also earthed?

I say "live" but "shock" is just a tickle but a tickle all the same.

Bruthal

zega wrote: »

How would this not trip the MCB if backbox is live and also earthed?

If the back box was in fact properly earthed, there would be no voltage relative to earth/neutral present on it in a domestic setup. But an earth wire connected to the box does not certainly mean the box is properly earthed.

confusedeire

There's a good chance that the voltage could be from another appliance and just travelling through the earth. Most common culprits are heating elements from immersion, washing machine, water heater, dishwasher or even a kettle or iron. They would usually trip an rcd but if you have an old fusebox without an rcd the current might not be enough to blow the fuse. If it's a new board with an rcd test to see is it working.

Or it might be from the transformer for the down lighters

2011

Somewhere within the electrical installation a conductor is making contact with something conductive (such as one of the earth cables) that is electrically connected to the metal box and / or switch.

To diagnose this problem I would measure the voltage between the switch and a known earth (possibly by running a cable from the main earth bar to one of the multimeter leads). This would have to be done with a professional multimeter such as a Fluke. Then I would get a friend to switch off MCBs one at a time. When the voltage disappears I would know that I had identified the circuit causing the issue. I would then disconnect points along this circuit / carry out a visual inspection / insulation resistance test (meggar test) to pinpoint the issue. Naturally this work should only be carried out by a qualified electrician.

Bruthal

Id test between neutral and back box myself. If there is a problem on the whole earthing system, the earth bar may be above 0v.

Or at least test between the neutral and earth bar to ensure 0v as part of the testing.

If the house is neutralized, the earth bar and neutral are locked together anyway.

If it is not neutralized, then the earthing system can float above 0v/N if there is a fault, and the switch box may be at the same potential as the earth bar. In this case, the person carrying out the tests will see 0v between main earth bar and switch box, and may assume there is no problem.

The other possibility is the lighting circuit earth is not correctly connected to the earth bar at all. A test between L at the switch and the back box should show 220 - 230v. If the back box is not wired back to the earth bar properly, this reading will be usually be well below 200v, indicating an open earth. A 230v reading doesnt mean there is a good earth. But well below it indicates a non connected earth.

Cerco

Peppa Cig wrote: »

I say "live" but "shock" is just a tickle but a tickle all the same.

Are you sure it is not static?
What type of tester are you using? Is it a standard phase tester?

2011

Cerco wrote: »

Are you sure it is not static?

That is a good point!

Bruthal

Static wont keep a phase tester lighting. But people do get static shocks and think things are live.

2011

Bruthal wrote: »

Static wont keep a phase tester lighting.

Your typical DIY type "phase tester" is not a reliable test instrument. i would not pay much attention to it TBH.

It has been known to generate static charge in different ways such as by walking on certain carpets when wearing certain shoes. Touching something that is earthed (such as a metal light switch) can cause this charge to discharge. This can feel like a shock as described by the OP.

Cerco

Bruthal wrote: »

Static wont keep a phase tester lighting. But people do get static shocks and think things are live.

It is also possible the Op is using a cheapo non-contact voltage tester. These pick up a voltage from adjacent cables.

http://www.amazon.co.uk/dp/B004DG9LGQ/?tag=goedko-21

Bedrooms are likely places to find static due to nylon carpets and bare feet...

Irish Steve

HAd a scenario here (long time ago) where an earth wire was not making connection in a joint box in the roof, so there was a long run parallel to power cables, and the proximity was enough over a sufficient length to generate a small voltage in the earth cable, we found out about it when the plastic switch was replaced with a brass switch plate.

Once the poor connection on the earth wire was found ( a number of earth wires twisted together and taped over, yeah!) the problem was solved.

Carpet static is another possible, was a regular problem at one of the computer sites I used to support, if we were using the system for a long period, it was more comfortable to take our shoes off when walking around the room, as if we built up too much static and then touched the wrong place on the console, the entire computer would hang, which wasn't exactly helpful!

Bruthal

2011 wrote: »

Your typical DIY type "phase tester" is not a reliable test instrument. i would not pay much attention to it TBH.

It has been known to generate static charge in different ways such as by walking on certain carpets when wearing certain shoes. Touching something that is earthed (such as a metal light switch) can cause this charge to discharge. This can feel like a shock as described by the OP.

Static is probably as likely as anything else.

A non contact tester may light when placed on a switch near the rocker.

A phase tester may not be a reliable installation test instrument, but if it lights on contact with a metal switch plate, its not from static. And means the plate is not earthed properly. They will light from induction/capacitance from nearby live wires to the non earthed earth wire. This can be seen on 2 way light fitting pins dimly lighting a phase tester when its switched off, due to capacitance from one strapper to the other.

But that wouldn't rule out static.

Bruthal

Irish Steve wrote: »

HAd a scenario here (long time ago) where an earth wire was not making connection in a joint box in the roof, so there was a long run parallel to power cables, and the proximity was enough over a sufficient length to generate a small voltage in the earth cable, we found out about it when the plastic switch was replaced with a brass switch plate

How did you discover this voltage?

Ging Ging

Can you determine what circuit this 5amp socket is on by switching off MCB's at fuse board? It may be on its own one, it would be worth doing a visual inspection of wiring in this socket. Find and fix the root cause anyway.

2011

Bruthal wrote: »

A phase tester may not be a reliable installation test instrument, but if it lights on contact with a metal switch plate, its not from static.

Agreed.
But the issue itself could be static and the phase tester could be lighting for an unrelated reason.
The phase tester illuminating or not should be disregarded as it is meaningless IMHO.

Id test between neutral and back box myself. If there is a problem on the whole earthing system, the earth bar may be above 0v.

That would be one way of doing it alright. When fault finding it is quick and easy to test that the neutral and that the earth are both what they should be.

Bruthal

2011 wrote: »

Agreed.
But the issue itself could be static and the phase tester could be lighting for an unrelated reason.
The phase tester illuminating or not should be disregarded as it is meaningless IMHO.



That would be one way of doing it alright. When fault finding it is quick and easy to test that the neutral and that the earth are both what they should be.

I'd have to admit though, if it is static, and if a phase tester Lights from touching a metal switch, it's not meaningless in my opinion.

A switch plate could be live, and also discharge static. Don't disregard anything, regardless of how someone got an indicator, if that indication usually happens when something is live.

2011

Bruthal wrote: »

I'd have to admit though, if it is static, and if a phase tester Lights from touching a metal switch, it's not meaningless in my opinion.

A switch plate could be live, and also discharge static. Don't disregard anything, regardless of how someone got an indicator, if that indication usually happens when something is live.

Did you ever hear the expression "even a stopped clock is right twice a day"

My advice would be to use proper test instruments so that the readings given are credible.

These DIY €3.99 type phase testers are not to be trusted. IMHO there is a 50/50 chance that this phase tester was picking something up. There will always be some "noise" on the earth, even this has been know to cause some of these devices to light.

Peppa Cig

Thanks for all the input. It is clearly a troubleshooting task for a qualified sparks.

As mentioned, you would always get a tingle (shock) from this metal light switch. Regarding static possibility the floor is tiled not carpet. the switch is next to bed and you can sometimes get a tingle from the bed as well as standing. The tingle is continuous at times. I think static would be once off?

I used a cheap and nasty phase tester. The screwdriver type that lights up And only cost a few Euro. I was curious and found that screws going into back box would light up this tester. Some times intermittently.

is there any immediate safety concerns? It's a newish house with rcd's etc.

If it was a plastic switch I would not be aware of the issue unless I touched the recessed faceplate screws.....

Bruthal

2011 wrote: »

Did you ever hear the expression "even a stopped clock is right twice a day"

I have an idea about that alright.

But how long are they correct for, now there's a question

Risteard81

2011 wrote: »

Did you ever hear the expression "even a stopped clock is right twice a day"

My advice would be to use proper test instruments so that the readings given are credible.

These DIY €3.99 type phase testers are not to be trusted. IMHO there is a 50/50 chance that this phase tester was picking something up. There will always be some "noise" on the earth, even this has been know to cause some of these devices to light.

Agreed. I have a non-contact voltage detection pen that signals voltage when I rub it along my clothes! Far better with a two-pole approved voltage indicator.

Bruthal

The thing is though, the point is nothing remotely about whether proper instruments are better than a phase tester. It's about a person asking about a phase tester lighting when touching a switch plate with it.

I wouldn't say to that person that you can disregard the phase tester lighting. This is not because I believe a phase tester is a marvel of instrumentation, but because a screwdriver type phase tester lighting is not a good sign on a metal switch plate, regardless of how much better a multi meter is.

Iderown

Peppa Cig - if your (cheap and cheerful) phase tester lights CONTINUOUSLY when you touch the switch metalwork or if you feel a continuous tingle when you touch the metalwork then there is a problem with the earthing. Static charge would be fast discharged.

The problem is likely to be a disconnected earth connection somewhere back along the the line of cabling. Voltage can get coupled to such a stretch of (unearthed) conductor by capacitive coupling from live cables running parallel with it. You may have to inspect every junction between the affected wall switch and the consumer unit (fuse board). Some of the junctions will be in ceiling units above pendant lights. Start with the easy ones - and don't forget the earth connection at the consumer unit.

It's better to get this sorted now that you know about it. Earth connections are for the safety of folk.

Bruthal

2011 wrote: »

Did you ever hear the expression "even a stopped clock is right twice a day"

The thing people dont think of is, that that is more often than a running one is right:pac::)

wait4me

Peppa Cig wrote: »

..........the switch is next to bed and you can sometimes get a tingle from the bed as well as standing. The tingle is continuous at times. ....

If that wasn't so serious, it would be funny :eek:

2011

Plenty of food for thought.
Time to get an electrician.
This should not be difficult to solve and should not be ignored.

Irish Steve

Bruthal wrote: »

How did you discover this voltage?

When the plastic switch plate was changed to a brass plate, the induced voltage was enough to tickle!

From much painful experience in a couple of places in the UK, my first thought was static, but suitable discharge measures etc didn't solve the issue, so I went looking a bit more closely, and connected a "flying earth" cable to a known good earth that could then be used to check what was happening in the box, and an analog meter was capable of detecting the voltage being generated, as did a more accurate digital meter.

Fortunately, it wasn't enough to be dangerous as such, as there was nothing local to earth to, and it wasn't in a danger area.

At that point, I went looking more closely, and found a plastic juntion box in the roof that had a number of earth wires going in to it, but there was no connector on the wires, they'd been twisted together and then taped, and at some stage, they'd been pulled and one of the earth wires had separated from the rest of the bunch, so it was floating. Onve the wire was reconnected, and an appropriate junction termination made, the problem went away. It was a long 2 way light run, and the earth and live were parallel over about 20 Ft in conduit.

Bruthal

Irish Steve wrote: »

When the plastic switch plate was changed to a brass plate, the induced voltage was enough to tickle!

From much painful experience in a couple of places in the UK, my first thought was static, but suitable discharge measures etc didn't solve the issue, so I went looking a bit more closely, and connected a "flying earth" cable to a known good earth that could then be used to check what was happening in the box, and an analog meter was capable of detecting the voltage being generated, as did a more accurate digital meter.

Fortunately, it wasn't enough to be dangerous as such, as there was nothing local to earth to, and it wasn't in a danger area.

At that point, I went looking more closely, and found a plastic juntion box in the roof that had a number of earth wires going in to it, but there was no connector on the wires, they'd been twisted together and then taped, and at some stage, they'd been pulled and one of the earth wires had separated from the rest of the bunch, so it was floating. Onve the wire was reconnected, and an appropriate junction termination made, the problem went away. It was a long 2 way light run, and the earth and live were parallel over about 20 Ft in conduit.

It would be unlikely to be a danger anyway, from an induced voltage into an open ended earth wire, as the circuit would be high impedance because part of the circuit path is an open circuit. And even if the person were to contact both ends of it, its too short to have enough energy to do any damage.

It is often through capacitance that this voltage appears, such as on 2 way lights where it can be detected at the pendant fitting with bulb removed. There is no load on the strappers, so no induction, but capacitance.

I suspect that lightly touching a metal switch plate in such a scenario as you describe where you do perceive a tingle, that a tighter or heavier contact with it may reduce the sensation. This may happen in the op case also.

When testing a voltage in a high impedance circuit, the analogue meter may give a lower reading than the digital, because the analogue ones are much lower impedance meters. In series with the high impedance part of the circuit, they see less voltage than the digital one.

Both are correct for the circuit conditions they present though.

Cerco

Bruthal wrote: »

It would be unlikely to be a danger anyway, from an induced voltage into an open ended earth wire, as the circuit would be high impedance because part of the circuit path is an open circuit. And even if the person were to contact both ends of it, its too short to have enough energy to do any damage.

It is often through capacitance that this voltage appears, such as on 2 way lights where it can be detected at the pendant fitting with bulb removed. There is no load on the strappers, so no induction, but capacitance.

I suspect that lightly touching a metal switch plate in such a scenario as you describe where you do perceive a tingle, that a tighter or heavier contact with it may reduce the sensation. This may happen in the op case also.

When testing a voltage in a high impedance circuit, the analogue meter may give a lower reading than the digital, because the analogue ones are much lower impedance meters. In series with the high impedance part of the circuit, they see less voltage than the digital one.

Both are correct for the circuit conditions they present though.

Not doubting there is an element of capacitance but when you measure the strappers your meter's impedance is a load albeit high.

2011

Cerco wrote: »

Not doubting there is an element of capacitance but when you measure the strappers your meter's impedance is a load albeit high.

The "ideal" voltmeter has infinite impedance and the "ideal" ammeter has zero impedance. Obviously this is impossible but modern voltmeters have very high impedance resulting in only a tiny current flowing. This current would be larger when an analog meter is used (due to its inherently lower impedance) resulting in a lower voltage reading.

Bruthal

2011 wrote: »

The "ideal" voltmeter has infinite impedance and the "ideal" ammeter has zero impedance. Obviously this is impossible but modern voltmeters have very high impedance resulting in only a tiny current flowing. This current would be larger when an analog meter is used (due to its inherently lower impedance) resulting in a lower voltage reading.

in theory, yes. But depends on the scenario imo. I'd trust, or at least prefer a solenoid type tester on mini pillars and transformers etc than a high impedance Digital meter.

High impedance is good for electronic circuit testing or anywhere where an impedance presented to the circuit can change the circuit properties. Not need for testing power circuits.

If a voltmeter had infinite impedance, it would read 230 volts when testing from a live point to a piece of wire connected to nothing. And read 230v for everything near an ac circuit. Including it's own leads probably. Is that good? I wouldn't think so.

Already, Digital meters impedance is so high that the result for the above would be well over 100 volts, where as a solenoid one would read 0.

The high impedance meter showed 100v on a USB socket as seen on a thread recently. Everyone came to the conclusion that it was not there for all intents and purposes.

2011

Have a read of
www.phys.ufl.edu/courses/phy4802L/spring09/labs/Lab2_DC.pdf
where page 2 states

An ideal voltmeter has infinite resistance: It is an open circuit.

Also check out the link below where it states:

Obviously, the higher the voltmeter resistance, the less loading of the circuit under test, and that is why an ideal voltmeter has infinite internal resistance.

http://www.allaboutcircuits.com/vol_1/chpt_8/3.html

Bruthal

2011 wrote: »

Have a read of
www.phys.ufl.edu/courses/phy4802L/spring09/labs/Lab2_DC.pdf
where page 2 states


Also check out the link below where it states:




http://www.allaboutcircuits.com/vol_1/chpt_8/3.html

So you honestly believe an infinite impedance meter is the ideal meter? No analysis of your own?

Testing a power transformer needs the highest possible impedance in a meter? Rubbish I say.

2011

Bruthal wrote: »

So you honestly believe an infinite impedance meter is the ideal meter? No analysis of your own?

Yes.
Analysis later, busy right now.

What is the advantage of having a lower resistance voltmeter?

2011

Once a voltmeter is connected to a circuit it draws a current. This influences the circuit.
This can not be described as "ideal".

I guess that your point is that it is often such a small impact that it is irrelevant.

Bruthal

2011 wrote: »

Once a voltmeter is connected to a circuit it draws a current. This influences the circuit.
This can not be described as "ideal".

I guess that your point is that it is often such a small impact that it is irrelevant.

10 Mega ohm voltmeter connected to 230 volts is around 20 micro amps. It will affect nothing but micro controller type circuits. What it does do though, is give misleading results.

Is it ideal if one probe is connected to live, and second connected to ask 2 meter piece of floating wire, and 230 is displayed?

Bruthal

2011 wrote: »

Yes.
Analysis later, busy right now.

What is the advantage of having a lower resistance voltmeter?

When I used to test mini pillars, the solenoid type tester was far more intuitive than a high impedance Digital meter.

That same tester would have shown almost nothing on that USB scenario discussed recently.

2011

When I type "ideal voltmeter resistance" into Google the answer is consistent from all of the reputable sources.

Bruthal

2011 wrote: »

When I type "ideal voltmeter resistance" into Google the answer is consistent from all of the reputable sources.

Yea that may be. But just take an example.

Digital meter with infinite input impedance. It would seem ideal. The idea behind that is simple, and obvious. Its taught to apprentices, so why would anyone question it.

Now take a socket. Live wire intact. Earth also. Neutral open circuit at a junction box.

Digital meter with theoretical infinite impedance will show 230 from live - earth, and live - neutral. So the tester sees no problem.

Use a solenoid type tester and it becomes very clear the neutral has a problem. Same with analogue meter.

Use current Digital meter and anywhere up to the 230 is seen to the floating wire, usually a bit lower, making many think it's a bad connection.

And we seen the 100v USB thing, when you yourself posted that the 100v is not really there. Anyone with understanding of that will realise the very high impedance meter showed that, due to the nature of the circuit it was testing, and meter impedance.

Bruthal

Cerco wrote: »

Not doubting there is an element of capacitance but when you measure the strappers your meter's impedance is a load albeit high.

A bit about it here

http://www.fluke.com/fluke/uses/comunidad/Fluke-News-Plus/ArticleCategories/Electrical/DualImpedance.htm

Ghost voltages occur from having energized circuits and non-energized wiring located in close proximity to each other, such as in the same conduit or raceway. This condition forms a capacitor and allows capacitive coupling between the energized wiring and the adjacent unused wiring.
When you place your multimeter leads between the open circuit and the neutral conductor, you effectively complete the circuit through the input of the multimeter. The capacitance between the connected, hot conductor and the floating conductor forms a voltage divider in conjunction with the multimeter input impedance. The multimeter then measures and displays the resulting voltage value.

Interesting that fluke make dual input impedance meters.

2011

Bruthal wrote: »

Its taught to apprentices, so why would anyone question it.

Exactly, it was taught to you, me and all other apprentices.
As a generaly statement is is correct.

Bruthal

2011 wrote: »

Exactly, it was taught to you, me and all other apprentices.
As a generaly statement is is correct.

I guess you just accepted it. Nothing wrong with that. My point was that apprentices will just accept it, and never question it, in practically all cases. I just dont accept that the higher the impedance, the better, in all scenarios.

Infinite impedance? Endless misleading readings will ensue. Thats a fact really.

In closed, properly functioning circuits, the higher the impedance, the better, no problem there.

It is when there is an open circuit, and the meter is across that, that readings can be misleading.

I seen the exact scenario mentioned in the fluke meter thing above, in work last year. Circuit de-energised, but showing around 12 volts. Their top men down talking about ohms law 12v and if there was a short circuit, it could draw high currents. Absolute cluelessness. I mentiond to them that it was capacitance and or inductance from the circuit beside it, and the digital meter will pick that up in a misleading way. But it has no real energy in it. Might as well have been talking to the wall.

So when I see many sites saying that the higher the impedance the better, im not surprised. I seen them before.

Bruthal

Anyway another thing about the fluke thing.

Fluke

Cerco

2011 wrote: »

The "ideal" voltmeter has infinite impedance and the "ideal" ammeter has zero impedance. Obviously this is impossible but modern voltmeters have very high impedance resulting in only a tiny current flowing. This current would be larger when an analog meter is used (due to its inherently lower impedance) resulting in a lower voltage reading.

Agreed, which is why I said albeit high (impedance)