Even a poor earth should suppress any t>> The cabinet came with the flex coming out the top, all ready to
This is a 3-core flex, I presume? That's a rather basic question, but I don't think anyone else has asked...
As I've narrowed this to the socket outlet in the utility cupboard, should I use my DMM to check for continuity between this and the fly lead I have going back to the main earth block? Set to a low Ohms level and probe between fly lead and earth terminal of wall socket outlet?
Yes. it is.
best, deano.
here is that far up the stream, as the mains Earthing in the house has been in situ since I moved here, some 14 years ago and has been tested in anger IIRC at various times with short circuits and fuses blowing but I get that you're covering the bases and going to the worst case scenario, identifying the source.
utility cupboard... I think :)
I'd check the earth wiring from the suspect sockets back to the Consumer Unit earth bar.
If that had become disconnected, then any normal mains filters in the appliances that generate small leakage would cause the socket / cabinet non-earth to float to about 120v, but at high impedance, giving the tingle you describe.
Yes, connect one end of the long lead to the main earth block, or the earth bar in the CU. Then check for continuity between the other end of the lead and the cabinet. If you find an open-circuit then locate the broken connection by working backwards along the route of the CPC until you do get continuity.
As you say, use the multimeter on its lowest resistance range, hopefully you should be able to resolve down to under 1 ohm. An alternative is to use a battery and torch bulb, which can be quite revealing of any poor connections.
OK, so it is meant to be earthed. Class 1 and not Class 2, as many of us tended to assume at the start. You certainly shouldn't be getting tingles off that.
Doing that this morning... need a longer fly lead!
Wow... this is where I'd like to learn more about theory... 120v sounds scary, how does "high impedance" bring that down to just a 'tingle'?
iPhone which doesn't integrate very well with groups.
It doesn't seem to work with threading at all. There are now 12 different threads all with the same subject line! Looks like it creates a new thread every time you post.
Agree and it's driving me nuts!
Deano - the way to fix this is to NOT use google groups. Try using eg Thunderbird and a free news server.
OK, let us know how it goes... (any spare wire will do - so long as you measure its resistance and take that away from your result)
For in depth theory, see:
See the pic on the front of the filter shown here:
Say you take a suppression cap with a capacitance of 2200pF. The reactance will be given by the formula
Xc = 1 / 2 x pi x f x c
So at 100kHz the cap will look like 1 / 2 x pi x 100000 x 100 x 10^-9 = approx 636 ohms. Hence it conducts and snubs the radio frequency quite well. However at 50Hz mains you get and effective resistance of 1.3M ohms.
So if you were to redraw your effective circuit you have your earth wire connected to L & N via a pair of 1.3M Ohm resistors. Since both "resistors" are the same size, if the earth is not connected to anything at all, it will just sit half way between the L & N voltage - or typically about 120V. However even in this circumstance the current that can flow to earth is tiny, if you do V = IR you get 230 / 1.3M = 0.18mA. So lets say your earth is not as good as it ought to be, you come along and stick your mit on the earthed tap, and the cabinet which is connected to the centre point of this notional resistor network, and you will get some voltage across you that you can feel. The maximum current that can flow through you is limited less than 200uA though.
as expected.
Well its slightly better - when know where in the thread you are replying now ;-) Still started another a separate thread though.
or is it your end? No one else has mentioned having problems following this thread... but I used to build websites so I know how many variable can be at play to ensure all see what you want them to see!
my floorplan and location of all the components in play... you should be able to see it here:
Are all those double sockets on the same circuit? Do you know what type of circuit it is? (i.e. ring or radial)
scary, how does
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D\
=A0 =A0 =A0 =A0|
=A0 =A0 =A0 =A0|
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D/
That sure is some concise info there John- thank you for taking the time to write that. I have read it twice now and am starting to get the general gist although the physics loses me a bit for now :)
t this as expected.
I have now switched back to the 'old' Groups UI... does this stop the new thread posting? I have never had this issue before and can't see my topic being repeated as new threads in the group topic listing on either of my laptops!
ith it, or is it your end? No one else has mentioned having problems follow= ing this thread... but I used to build websites so I know how many variable= can be at play to ensure all see what you want them to see!
p show my floorplan and location of all the components in play... you shoul= d be able to see it here:
Yes, they are all on the same circuit.
***I have now found the fault*** The sockets at the top of my diagram (in green) are on a ring circuit. The sockets and FCU at the bottom of the diagram (in blue /red) are spurred off from the 'green' sockets. The CPC of the spur had become detached from its terminal at the green socket and hence no earth connection to either (blue) socket OR to the (red) FCU powering the bathroom cabinet! IIUC- leakage generated by the tumble dryer (even when not running) plugged into any of the (blue) sockets was being 'detected' at the bathroom cabinet/mixer tap, in the form of a 'tingling'.Now I have redressed the cores and remade connections to the (green) sockets in the utility cupboard, I probed between the Earth terminals of each (blue) socket wall plate and my fly lead and got an audible beep which wasn't happening before I remade the connections at "green".
I want to get a reading in ohms as you suggested, but am having some confusion with this... Set dial on DMM to 20k(ohms) and probe leads to COM (black) and V(ohm)mA (red): this gives diminishing readings from eg: 6.xx down to 0.00 then up to 0.01, 0.02 etc- should I be getting a fixed reading or is the DMM pushing a voltage down the line which is dissipating? Is it me or my Raptitest DMM?
Yes, the reply has remained in one of the existing threads ... you do realise most people here are using usenet direct, rather than through google groups' web interface don't you? I believe you can get NewsTap for iPhones ...
Yep, that sounds like the end of it.
That's not really a very low ohms range, e.g. my analogue continuity meter is logarithmic with 50ohms full scale and 4ohms half scale.
Are you testing while other appliances are still plugged into the circuit? Your DMM could be being affected by the same suppression filters that were passing the "tingle" back to earth.
this as expected.
Yup, that's better - this message was threaded correctly.
Its an intermittent problem on gg it seems. If you just thread on the topic then it usually looks ok. However it often leaves out the references headers that are used by proper news clients to to thread stuff.
it, or is it your end? No one else has mentioned having problems following this thread... but I used to build websites so I know how many variable can be at play to ensure all see what you want them to see!
show my floorplan and location of all the components in play... you should be able to see it here:
Well done for finding that. That does seem to explain the symptoms you were seeing.
Yup, as I mentioned in another post, with no earth reference at all, the filter on the TD or WM would end up floating the earth wire at about half mains. Hence compared to your earthed tap, you were getting a 120V shock with a very limited current.
Good. That has found and fixed a serious fault. While there was no immediate danger from the tingle due to the current limiting effects described, had the appliance plugged into that socket developed a genuine earth fault at some point, then you would have been at *very* serious risk of electrocution in the bathroom. With no RCD to save you!
However since your bathroom has a shower, and you don't have a 17th edition compliant CU, you should have supplimentary equipotential bonding in the bathroom. If that was working correctly it should have also suppressed the tingle since either the tap would have been brought up to half mains with the cabinet (hence no potential difference, and no tingle), or the fortuitous earthing path via the tap should have brought the earth connection down to something approaching ground. Hence we can conclude that the earth of the circuit feeding the cabinet is not connected to the bonding of the taps etc in the bathroom - and it should be.
Firstly have you got the electricity turned off? Any current flowing in the wire you are attempting to test could upset readings.
The 20K range is far to high really... does it have a lower range like
20 ohms?Ideally you would expect to get a stable reading. It may take a few secs to stabilise but then will usually stay put. Also check the battery in the meter is not knackered. What happens when you connect the probes together with nothing else in circuit? You would expect a reading approaching 0 ohms.
The physics is actually quite easy to visualise if you think about it from the basic principles.
A capacitor is basically a pair of conductive "plates" separated by an insulator. When you apply a voltage across the plates, you see a some current flow briefly into it as it "charges up" (in reality you are storing a small amount of electrical energy in an electric field created by the plates - all the free electrons migrate round to one of the plates - leaving it negatively charged). You may remember physics demos with a gold leaf electroscope which shows how you can accumulate charge on a plate?
Hence with a DC supply connected to a cap, you get a brief flow of current as it charges, and then the current flow falls away to nothing. Since current can't pass across the insulating gap. If however you now swap the polarity of the battery, then you will again see a brief flow of current as you discharge it and the recharge it with the opposite polarity.
A cap across an AC supply thus looks like it actually conducts because the continual reversal of polarity of the supply allows it to keep passing current as it is charged, discharged and recharged in the opposite direction. How well it conducts is a function of how quickly you reverse the supply polarity (i.e. the frequency of the AC), and the size of the capacitor.
The larger the capacitor, or the higher the frequency, the better the conductance (or lower the resistance).
There are basically two common cases where this is going to be relevant in mains electrics. The first you have experienced first hand - capacitors in input filters will leak small amounts to earth. So in your case with no earth, you see bits of metal feeling live to the touch. Alternatively you can start to eat into the leakage allowed by a RCD and get an unexpected trip if you have too many of these things on one RCD.
The second one is with capacitive coupling, where you don't even have an identifiable capacitor as such. Long lengths of wire laying side by side will exhibit a tiny capacitance as well. If one of these is live and the other not connected to anything "solid" like L or N or E, then it will tend to appear to be live as well. This can fool some sensitive digital meters, or cause CFL bulbs to flash unexpectedly in the dark, or sometimes even give a tingle sensation similar to what you were experiencing.
Good, glad that's better
I used to use a dedicated news client years ago (on the mac), but got so fed up with getting used to using one, only for it to be discontinued that I began subscribing using browser based feeds instead... at least they survived longer and remained consistent, however Google Groups now seems to be going through a major shake up which is going to bring with it all sorts of chaos :)
I have just shelled out =A32.79 for Newstap for my iPhone, so lets hope that works how I would like it to, from that device?!
Yep. No more 'tingling'
Ouch!
Yikes! I realise how *lucky* I/we've been, in that the fault manifested itself in this way. Looks like the filters on the TD/WM acted as accidental warning systems!
Yes, I understand this now and have to admit, the pipework feeding the downstairs shower room was not bonded to earth because (underground) the feeds are run in plastic but now, with all the appliances in the utility cupboard, there is a lot of potential for the surface run 'copper' pipework to become energised with mains voltage, from a fault within the utility cupboard, thus putting the plumbing fittings in the shower room 'at risk'!
I have now run a 10mm Earth cable from the Main Earth block under the stairs, to the pipework in the utility cupboard and have daisy chained it to the hot and cold feeds using the appropriate bonding clamps and tags.
Q: In supplementary bonding situations, how does one fashion a mechanical bond between pipework and electrical circuits in the same vicinity? Do you take an earth lead from the pipe clamps into a wall box or junction box (in the floor/ceiling) and just connect it to an earth terminal of each independent electrical circuit? i.e. Instead of running the lead back to the Main Earth block (near the CU) where would one connect this earth/bond in localised situations? Is there a dedicated fitting for this or as I mentioned, do you run it in the wall to a pattress box? I'm wondering how you would add supplementary bonding to an already decorated/ tiled bathroom?
Also, (and maybe a subject for another topic?). I have been thinking of updating my CU to 17th edition, with an RCD protected, split load and after this episode will be doing this sooner rather than later as it brings a much higher level of safety to domestic installations. From a tradesman's POV though, I have seen [and heard the swearing] many electricians, on site, who are constantly having to reset RCD and all digital clocks in the house (ovens, dvd players etc) after tripping them whilst working on an "isolated" circuit, simply because they have cut through a TW&E with side cutters and shorting the 3 cores when this is done, with even a tiny amount of current (static??) cause the RCD to trip. Do you know how this can be practically and 'safely' avoided whilst working on "isolated" circuits on a RCD protected board? I know RCBOs are the way to go but this solution is still significantly more expensive than a split load board that still meets the regs! Am interested in your take on this as it seems to be a real issue and I have heard of some dodgy (and dangerous) ways round it happening!
No! :o Oops
Not sure... its a Rapitest MAS830L and the Ohm segment of the dial has...
200m; 2M; 200k; 20k and 2k [ with this symbol underneath: ->|- )photo of the dial here...
With the probes plugged into COM(black) and V[Ohm]mA(red) and held together, the dial to 2k->|- the display reads 1. then .xxx finally resolving to .000
Thanks once more for your [and everyone's] time and assistance in this matter. And sorry for annoying everyone with my Google groups gremlins.
I expect that symbol marks the range to use for testing diodes.
range and would be best for continuity testing.
Very much so. Equipotential bonding is a very good safeguard, but it
*has* to be complete, or else it won't do its job[1]. Normally a earth fault on a washing machine without a functional earth connection would leave the casework of the machine live indefinitely. This could be nasty of you touched it - but is unlikely to be fatal in itself unless you also had a good connection to earth via your feet or something. The fact that then was linked through to the bathroom cabinet, made it far more dangerous, since the chances of touching the earthed tap and cabinet at the same time were quite high, and that would have been full mains from hand to hand across the chest. Almost guaranteed to kill you. [1] Had your supplementary bonding have been complete, it would have actually masked the problem of the missing earth on the socket (i.e. no tingle). However it would also probably have provided enough fortuitous earthing to the washing machine socket via the bonding to have rendered that safe as well. Not an ideal situation - but better than the alternative.
The thing to do is look at each conductive item in the bathroom, and ask "could this bring a potential into the room?" (and potential in this case includes an earth potential). So pipes that are plastic into the room are ok, but metal ones that enter and go to a touchable tap etc need bonding. Same goes for waste fittings if they are metal and leave the room. Also importantly (as this example illustrates very nicely) the earth of *every* circuit that is used to feed something in the bathroom must also be joined to the supplementary bonding. So in this case the utility socket circuit earth since its connected to the cabinet and probably extractor, and the lighting circuit. Other bits of metal in isolation (say a bath if there was one) can be ignored since they can't bring a voltage into the room.
Assuming your main bonding is ok (this ought to be from the main earth terminal (MET) to where the cold main enters the house, and also to the gas pipe if you have one), then the bonding in the bathroom only needs be supplementary. So it does not a need direct connection back to the MET (although it will in effect get one via the CPC of the bonded circuits)
Supplementary bonding can be done in 4mm^2 wire. This ought to be small enough to enable a connection to an earth terminal inside the FCU feeding the cabinet - that will take care of the CPC for that circuit. The cabinet is adequately bonded via the CPC in its flex. You have now (over) done the pipes etc by the sounds of it.
Yup that's the idea. So a convenient accessory (like a flex outlet, socket, FCU etc), or if one is not handy then you can add a JB somewhere.
Yup, no need actually to go back to the MET. (although you are allowed to make bonding connections in an adjacent room rather than inside it where that is more convenient - and in your case I get the impression that you MET is to all intents adjacent to the bathroom anyway).
Here is where doing it outside can help. If you CU is close to the bathroom, then a connection there would cover the lights etc. I would still run one to the FCU feeding the cabinet though since that represents an independent conductive path into the room.
Yup, a neutral to earth fault (even a temporary one from side cutters) will often trip a RCD. (by the time there is a bit of load on the neutral somewhere in the house, chances are it will rise little in voltage. Hence a connection to earth will cause a current flow from neutral to earth and the RCD will see this).
The "minimum" spec 17th edition CU:
If working on one of these, either accept that there is a chance you are going to trip half the circuits in the house, or open the CU and disconnect the neutral for the circuit you want to work on as well as turning off its MCB. (you can leave it earthed)
;-)
diode symbol under it (that suggests if you want to test a diode, use that 2K range rather than the lower one, because the lowest one probably does not use enough voltage to forward bias it - hence it would look knackered whichever way round you connect it!).
Try on the 200 range. You should get similar, although you may find you end up with a reading of 0.2 or something rather than 0 (since the probes will have a fraction of an ohms resistance). Once you know what this is you can just subtract it from your reading.
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D\
=A0 =A0 =A0 =A0|
=A0 =A0 =A0 =A0|
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D/
Hi John.
Not sure if you're still following this thread? Been v busy and also trying to sort out usenet on my iPhone!!!
A bit off topic but can you (or anyone) recommend an independent news server that allows posting as well as reading? For Free? Or a veryveryverysmallcharge? My phone provider is orange and they no longer offer a news feed... my home broadband is virginmedia and they don't allow access to their news server via a foreign network (i.e. via my phone) they use my IP as authentication, not login/pass!
So I have my great NewsTap, news client app on my iPhone but can't post replies or start threads! Bloody nuisance!
Back on topic...
I just wanted to say a big thank you for your help with this thread... I sorted the problem AND learnt quite a bit while at it :) So cheerz to you and all.
I still haven't had a chance to test the resistance of the CPC using my DMM (as you directed) and the fly lead is still running the length of my lounge, but I will do this soon... especially if this damn rain keeps up and keeps me off my roof job!
by4now, deano.
Aaahhhgg - snip the post mate ;-o
It's best to start a new topic with a new post (I know this was a mixed query).
But to answer the question -
Cheers,
Tim
Seconded:))...
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.