IET logo
 
IET
Decrease font size
Increase font size
Topic Title: Problem with RCD tripping
Topic Summary:
Created On: 27 May 2007 01:15 PM
Status: Post and Reply
Linear : Threading : Single : Branch
1 2 Next Last unread
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 27 May 2007 01:15 PM
User is offline View Users Profile Print this message



wasasparky

Posts: 100
Joined: 04 February 2005

Hi all, I thought I'd post this message in the hope that somebody might be able to shed some light on a problem I have with an RCD. I've installed a 30mA RCD in the mains tails supplying a Wylex BS3036 CU. Not something I like to do in the norm, however was necessary due to socket outlets in shed and some other installation anomolies that I wasn't impressed with. However, as always IR tested prior to install with 350M tested L/N-E. Powered up RCD, no problem BUT, switch on anything with an ounce of load, and we get RCD disconnection. Never seen this before, no faults apparent proven by powering up OK. Turn on the shower, tumble dryer, freezer (all on different circuits), and out she goes. Any thoughts? One other pertinent thought is that the output from the RCD feeds a DP Conn Block, and there-on to 2 consumer units (both IR tested over 350M L/N-E). It could be possible that there's a neutral connection somewhere between 2 seperate circuits, haven't yet checked this. Does anybody out there know of any other reason?

Thanks
 27 May 2007 01:42 PM
User is offline View Users Profile Print this message



wasasparky

Posts: 100
Joined: 04 February 2005

Thanks for such a quick response,

RCD ramp test gave me tripping current of 21mA, x1 @ 27mS and x5 @ 12.3mS. I changed the RCD which I was suspect about from a Proteus to a Wylex WRS80/2. TN-S earthing arrangement. I have to be honest that this little lot all occured on Friday night so I didn't get a chance to complete a full I&T, I've arranged to carry this out on Tuesday. I did however test the ring main which I modified before leaving the job, and got a max Zs of .85 so connection to earth is verified. I wouldn't expect Ze in this case to be much more than about .35 but I will confirm on Tuesday. Driving me mad is this, I've had N-E faults before which have shown themselves after installing RCD's, but never anything that's not immediately obvious to me.
 27 May 2007 01:46 PM
User is offline View Users Profile Print this message



wasasparky

Posts: 100
Joined: 04 February 2005

Sorry....Zs measured was .65, and I can't even read my own handwriting now !
 27 May 2007 01:58 PM
User is offline View Users Profile Print this message



loftyloo

Posts: 592
Joined: 23 May 2006

change the rcd



lofty
 27 May 2007 02:03 PM
User is offline View Users Profile Print this message



wasasparky

Posts: 100
Joined: 04 February 2005

see message above !!! Changed RCD, still the same
 27 May 2007 02:16 PM
User is offline View Users Profile Print this message



electricman

Posts: 818
Joined: 17 November 2005

hi wasasparky
mmmm a strange one, if you have tested all circuits, could it be a fault on a neutral tail between the rcd and the block/boards

are they metalclad units with no/badly fitted grommits?
just a thought! please keep us up to speed, this one i want to hear

-------------------------
"You're a crackpot till you hit the jackpot!"
Werner Von Braun 1969
 27 May 2007 02:33 PM
User is offline View Users Profile Print this message



wasasparky

Posts: 100
Joined: 04 February 2005

Hi electricman,

One CU is an insulated 3Way Hager unit with a mixture of 1361 and 60898 breakers, the other is a M/C unit. integrity is all ok and there doesn't appear to be any signs of machanical damage to cabling. It's bizarre, because if there was a fault, the RCD just wouldn't reset....Period. I At the moment I've left the RCD and tails disconnected but in situ, and I've modified the meter tails to keep it up and running until Tuesday. I expect that when I do the I&T on Tuesday as planned, that something will turn up and I'll slap myself for missing it.
 27 May 2007 02:46 PM
User is offline View Users Profile Print this message



wasasparky

Posts: 100
Joined: 04 February 2005

Yeah I think you're right alectricity, there's a fundamental flaw somewhere which I'm likely to find when I do the full I&T and I get the chance to seperate the circuitry. But when I think about it, it just doesn't make any sense...why on ANY circuit does the RCD trip only when a load is drawn? There must be leakage somewhere, but it doesn't seem logical. If indeed somebody has 'borrowed' a neutral from one circuit to use in another, then this also wouldn't show up because as far as the RCD is concerned there's still no global imbalance between P&N in the tails. If it was an RCBO covering a single circuit that had a neutral stolen (and I've seen this many times, especially on lighting circuits), then I could understand a constant tripping problem. I know not what to do.

Keep the ideas coming, your help is all very much appreciated.
 27 May 2007 03:17 PM
User is offline View Users Profile Print this message



normcall

Posts: 8111
Joined: 15 January 2005

Loose connections can cause an RCD to do as you describe.
Only problem is that they can be anywhere.

-------------------------
Norman
 27 May 2007 03:19 PM
User is offline View Users Profile Print this message



deeptheory

Posts: 40
Joined: 22 January 2007

just as a side note. we have been on a housing contract and must have second fixed around 15 houses now and 5 of the rcd's have been faulty.

scandlous imo. surely they must go through some quality control?
 27 May 2007 03:40 PM
User is offline View Users Profile Print this message



wasasparky

Posts: 100
Joined: 04 February 2005

good point normcall. If it is a loose connection, then I'd hope that it shows up when I measure indiv circuits R1+R2, at least that may narrow it down to one circuit. Seems logical tho, on 'no load' there'd be no imbalance at the RCD search coil because no current flowing in P or N, if there is a loose connection in any circuit then perhaps this is where the imbalance occurs (not so much earth leakage, but nevertheless a loss of current in the neutral somewhere). BUT......how could this be? If for example there was a loose neutral connection at the shower isolator, then you'd expect the RCD to trip ONLY if the shower isolator is switched on!!! I guess it could be loose neutral tails at one of the boards??? I haven't checked this.
 27 May 2007 03:41 PM
User is offline View Users Profile Print this message



electricman

Posts: 818
Joined: 17 November 2005

hi wasasparky
well if it trips on load then it can only be a n-e fault (the more you load up, the same ratio will run down the earth and neutral)
hope you find it quick

-------------------------
"You're a crackpot till you hit the jackpot!"
Werner Von Braun 1969
 27 May 2007 04:06 PM
User is offline View Users Profile Print this message



wasasparky

Posts: 100
Joined: 04 February 2005

You know the only comfort I have is that there IS a fault somewhere which is affecting the current coming back down the Neutral, and therefore it's fixable. You're right electricman, I need to be looking at where it's disappearing. Ho Hum, I've always said that a technical challenge is worthwhile, but now I've changed my mind..............it's nothing but hassle!!
 27 May 2007 04:45 PM
User is offline View Users Profile Print this message



alancapon

Posts: 5746
Joined: 27 December 2005

I suspect that you are isolating the phase to each circuit, and leaving all the neutrals connected. This will not help your location of the fault. You will need to disconnect all the neutrals too, and reconnect phase and neutral to each circuit in turn to locate which circuit is causing the trip.


Regards,

Alan.
 27 May 2007 05:11 PM
User is offline View Users Profile Print this message



wasasparky

Posts: 100
Joined: 04 February 2005

Hi Alan,

I haven't yet carried out any clinical testing which I fear I'll have to do. Trouble is, I know that any one of a number of circuits when put under load causes the RCD to fail, so I'm not looking to single out an indiv circuit fault, but more to find an effect on the RCD which can be caused by one or other circuits being put under load. All breakers in, all N & E connections made, RCD sits there no problem. Turn on the shower, or the tumble dryer, or anything else that sucks up some juice, and it fails. As a matter of duty I will carry out tests to each of the circuits in turn, which as you have said will mean isolating P/N/E connections for each circuit.....but I can't help but feel that although importantly proving a robust installation, that this little problem will need more investigation that an EIC requires e.g. testing between neutrals of different circuits etc. I'm stumped
 27 May 2007 05:18 PM
User is offline View Users Profile Print this message



pjcomp

Posts: 391
Joined: 28 June 2004

Don't think this is relevant to the case in point, but ...

Doing an inspection, checekd the meter box for earthing, and the earth wasn't wired in to the PME conenction ("it's always been like that" says householder). Getting an earth reading off the CU so must be something somewhere ("we always get a tingle from the shower" says householder - non-electric shower so presumably the "earth" is via the water main, though can't see a conenction anywhere).

Anyway, fix the earth into the PME and pop goes the RCD. Trace it to one of the lighting circuits, though I've got to go back to pin it down to the actual fault.

So there's an earth of sorts, but not good enough to trigger the RCD, but improve the earth and bang, away it goes.

Had a similar one a few months back - brought the earth up to scratch and the lights went out. Eventually found some dodgy wiring in a bathroom light switch, of all places.

PJ

 27 May 2007 05:33 PM
User is offline View Users Profile Print this message



alancapon

Posts: 5746
Joined: 27 December 2005

wasasparky, it could still be a single fault. When you energise one of the circuits that trips the rcd, most of the current (I hope) is returning through the neutral, through the rcd. The rest of the current, the bit that is causing you the problem, could be leaking to earth on the circuit you energised. It could also be finding a route via the neutral bar in the consumer unit, and down one of the other connected neutrals, and leaking to earth through a fault on this other circuit. Unfortunately, this is one of the penalties we have for using single-pole mcbs.


Regards,

Alan.
 27 May 2007 08:23 PM
User is offline View Users Profile Print this message



wasasparky

Posts: 100
Joined: 04 February 2005

Alan,

I think you might have found me a solution ! Current could be being lost via a route through the Neutral bar and into another defective circuit which is leaking to earth, so it wouldn't matter which circuit is energised. Now that's got to be the most logical explanation so far; but wouldn't the RCD fail to reset at all if there's leakage to earth on any circuit? maybe it's such a small leakage that the RCD doesn't notice. I can see that if there are no circuits loaded, then nothing would be expected to come back through the Neutral so it could well stay in, then a significant load applied would make the difference in what's coming back through the neutral to the RCD more significant, and the RCD notices it and trips.

Alectricity's last posting could be right then, the IR test could have missed it, i've still got to test the individual circuits so hopefully it'll show itself.

Thanks Alan
 27 May 2007 08:50 PM
User is offline View Users Profile Print this message



alancapon

Posts: 5746
Joined: 27 December 2005

You will probably find that the voltage between the neutral and earth terminals is very close to zero with no load. In order to get a current to flow to earth from the neutral, you need not only a connection between them, but a voltage difference as well.

If you switch on an electric shower (for example), you will get a voltage drop across all the cables that the current is flowing in, due to their resistance. Half of this voltage drop will be in the phase conductor, the other half in the neutral conductor. If you draw the circuit out, it will be supply --> phase conductor --> consumer unit --> phase conductor --> shower unit --> neutral conductor --> consumer unit --> neutral conductor --> supply. Measuring with respect to earth, the voltage on the phase conductor will drop, but the voltage on the neutral conductor will rise. It is this difference in voltage between the earth and neutral conductors, coupled with your neutral to earth fault, that will cause the rcd to trip. A long explanation I am afraid, but hopefully worth it.


Regards,

Alan.
 27 May 2007 09:31 PM
User is offline View Users Profile Print this message



wasasparky

Posts: 100
Joined: 04 February 2005

I've just about got my head around that one, but it was well explained - thankyou. No load, means no p.d between earth and neutral (both at near zero). Current flows on load creating the rise in voltage in the Neutral conductor thus creating a p.d between neutral and earth. The fault creates a direct connection between two conductors with a p.d between them, and the RCD drops out....

Did I get all that down correctly?

Crikey, I'm really having to muster up old memory's of electrical science lectures now, it's amazing how you leave all the technical lessons at college when you head out with your newly attained JIB card to go tear up old houses for a living. I've always enjoyed the technical side of electrical theory, but when you're under pressure to get something fixed, it makes it rather foggy.

Great stuff Alan, thanks
IET » Wiring and the regulations » Problem with RCD tripping

1 2 Next Last unread
Topic Tools Topic Tools
Statistics

See Also:



FuseTalk Standard Edition v3.2 - © 1999-2014 FuseTalk Inc. All rights reserved.