IET logo
 
IET
Decrease font size
Increase font size
Topic Title: RCD main switch tripping.
Topic Summary: Help, my brain hurts.
Created On: 15 March 2010 08:38 PM
Status: Read Only
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 15 March 2010 08:38 PM
User is offline View Users Profile Print this message



gnasher

Posts: 162
Joined: 21 September 2005

I have an installation that has a 100A 30mA Crabtree RCD as a main switch on a 10 way board. It is a large house that has had a couple of additions over the years but no changes recently.

A few weeks ago the main RCD started tripping, it would trip 5 or 6 times in quick succession and then settle down. Some time later it would repeat the process.

I was called in to try to resolve the fault, on first inspection I found that the RCD would not reset unless all mcb's were switched off, then reset RCD and switch each mcb individually. Sometimes when an mcb was switched on the RCD would trip but it was not always the same mcb.

I meggered the final circuits and found a fault on the cooker, very low Rins on the oven element, Isolated the oven on a DP switch and left. Unfortunately this did not resolve the problem.

Returned with new element, now oven Rins is okay. Meggered rest of circuits and found fault on Rins on upstairs lighting cct, traced to outside lights, isolated L & N of the outside lights at a j/b in loft and fault disappeared for two weeks.

Thought that the only thing that I had disturbed was the j/b that the outside lights was connected to, removed j/b and re-dressed all connections to new j/b.

Went in to carry out some more testing. tested RCD tripping times/current and got the following results for the 6 tripping time tests (mS) and 2 trip current tests (mA)

2k/2k/39.0/25.2/14.4/10.8mS 21/18mA Tested with one mcb on to the final circuit for kitchen sockets, where I plugged my tester into.
2k/2k/39.8/0/21.4/8.9mS 18/15mA Tested with 5 mcb's on and tested in same socket.
38.3/0/18.2/0/18.4/8.7mS 15/0mA All mcb's on, tester in same socket.

From the above results I surmised that there was a fault on one of the last 5 final circuits to affect the RCD tripping. Meggered all ccts again, L/N to cpc no fault found.

One other thing that I notice. If I use my Dilog test probe and measure between L and cpc on any circuit with all mcb's on, the RCD trips, if I turn all of the mcb's off, then the RCD does not trip. If I test with a DVM then the RCD does not trip under any combination of mcb's. I know that the DVM will have a higher Ohms/volt but the Dilog does not normally trip an RCD.

I have asked the customer if they can recal any pattern to the tripping but the only thing they suggest is that it often, but not always, happens in the early evening.

I have also noticed that they have a lot of computers and suppressed extension sockets.

Anyone offer any suggestions as to what may be the cause of the problem?
 15 March 2010 08:47 PM
User is offline View Users Profile Print this message



spinlondon

Posts: 5490
Joined: 10 December 2004

Too many computers, cumalitive earth leakage is tripping the RCD.
Ditch the single RCD and install RCBOs.
 15 March 2010 08:53 PM
User is offline View Users Profile Print this message



mikejumper

Posts: 2397
Joined: 14 December 2006

Originally posted by: spinlondon
Too many computers, cumalitive earth leakage is tripping the RCD.
Ditch the single RCD and install RCBOs.

My thoughts too.
 15 March 2010 10:08 PM
User is offline View Users Profile Print this message



1652

Posts: 348
Joined: 22 March 2007

Somedays we just hate RCD's.

When testing do you unplug every item in the house or just switch off at the socket wall plate?

Beware of single pole switched sockets! They do not break the neutral.

You may switch off the MCB however current can still flow neutral to earth via filters, suppressors and faults. this total leakage can trip the RCD totally random at times.

Hope this helps.
 15 March 2010 10:36 PM
User is offline View Users Profile Print this message



sparkiemike

Posts: 1630
Joined: 24 January 2008

had a similar problem with a client who had lots of PCs and AV kit. Insulation resistance tests were OK. Changed the board with RCBOs for some selected circuits. Been OK since - nearly a year now
 16 March 2010 07:01 AM
User is offline View Users Profile Print this message



normcall

Posts: 8514
Joined: 15 January 2005

Remove 'surge protected' extension leads and the problem will go away - velly cheep.
They usually have a resistor between earth and neutral - not good for RCD protection.

-------------------------
Norman
 18 March 2010 08:34 PM
User is offline View Users Profile Print this message



gnasher

Posts: 162
Joined: 21 September 2005

Many thanks for the replies. Went back to installation today to make some more tests.

Decided I would try to measure the leakage current to earth with an RMS meter. Disconnected the MEBC at CCU and inserted meter in series with MEC to terminal on DNO's head. Installation is TN-C-S.

With the RCD in the off position I measured 4.5 - 7.6mA. Where is the current coming from?

I tried switching on each MCB to see if one circuit was excessively high, as expected the circuits that had the PC's and AV equipment on was higher that the other circuits, but not greatly (approx 4mA higher).

With all MCB's on and with as many of the appliances as I could get set to standby mode I measured a maximum of 22mA during a 1 hour period. This varied from 10 - 22mA during this period with a constant load. Anyone know why?

My meter has a max hold function and during another period of recording I had the RCD trip out, the transient must have been quicker than my meter because the max recorded was 18mA.

I have instructed the customer to unplug as many of the appliances as he can when he is not using them but he has four children and I can't see that lasting for long.

The CCU is a Crabtree 10 way with RCD main switch and one solution is to replace the MCB's with Crabtree RCBO's and replace the RCD with a main switch, is this practical from a space point of view?

Eric
 18 March 2010 09:05 PM
User is offline View Users Profile Print this message



Zs

Posts: 3811
Joined: 20 July 2006

Originally posted by: gnasher

I have asked the customer if they can recal any pattern to the tripping but the only thing they suggest is that it often, but not always, happens in the early evening.



Anyone offer any suggestions as to what may be the cause of the problem?


A thermostat calling for heat?

Zs
 18 March 2010 09:40 PM
User is offline View Users Profile Print this message



Angram

Posts: 651
Joined: 23 March 2009

You had 22mA trough a 30mA trip; that's enough
As suggested already too many "surge protected" sockets and a need
to split the trip currents across several RCBOs instead of one RCD.
 07 June 2010 11:26 PM
User is offline View Users Profile Print this message



sparkingchip

Posts: 9863
Joined: 18 January 2003

yep, stopped at a "RCD keep's tripping, started a couple of days ago, we have tried unplugging things" on the way home tonight. Testing lead to another Surge protected extension lead.

Andy
 08 June 2010 11:51 AM
User is offline View Users Profile Print this message



slittle

Posts: 3678
Joined: 22 November 2007

We've had loads of these as well. It's one of the first things I tend to ask customers now during the initial set of "what were you doing" questions.

The other problem with the surge protected leads is that they are of course designed to go short circuit when they see a surge. If you happen to IR test at 500v, that could well be high enough to trigger the varistor into s/c mode for the duration of the test. Some recover, some don't...

At home I've got a 30ma RCD incomer (yes,yes I know, but I done the risk assessment and I don't have room for a dual rcd board !!) and including the kid's pc's, I can count 5 pc's, 1 laptop (class 1 psu) and numerous bits of IT kit and we've never had a problem. Don't have a surge protected extension in the place !

Stu
Statistics

New here?


See Also:



FuseTalk Standard Edition v3.2 - © 1999-2017 FuseTalk Inc. All rights reserved.

..