IET
Decrease font size
Increase font size
Topic Title: Nuisance Tripping of RCD - No obvious installation defects
Topic Summary:
Created On: 26 July 2010 11:38 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.
 26 July 2010 11:38 PM
User is offline View Users Profile Print this message



northstarlx

Posts: 3
Joined: 26 July 2010

We have been asked to investigate nuisance tripping of an RCD in a Wylex consumer units fitted to an extensively modified multi-level villa. The property is approximately three years old and is fitted with a NHRS9806 populated with Type B MCB's as follows;

Main switch, 8 un-protected MCB's, WRS 80/2 RCD, 7 x 32A Type B MCB's (2 x radial and 5 x ring circuits) and 2 x 16A MCB's.

We have fitted a replacement RCD which has not cured the problem.

The consumer unit is fed from the mains via a 25mm²split concentric cable, Supply is TNC-S with a Ze of 0.21 ohms at the meter and 0.24ohms at the DB Main Switch. With all the RCD protected MCB's off Zs measured on the load side of the RCD is 0.45ohms.

Insulation resistances for the RCD protected ways average 170 meg ohms with the lowest individual reading of 35 meg ohms All readings were taken between live conductors connected together and earth, appliances connected and a test voltage of 250v (the reason for this was the number of fixed kitchen appliances which would have had to be removed to disconnect)

The overall insulation resistance between live conductors and earth for the RCD protected ways is 16meg ohms.

Test A
With all protected MCB's on and tested from any of the socket outlets or at the load side of the RCD produces the following results;

½ rated trip test - RCD trips (it should not)
1 x rated current - average 38mS trip
5 x rated current - average 16mS trip

Ramp test starting at 12mA - RCD trips instantaneously

Test B
With all protected MCB's off and tests carried out at the load side of the RCD;

½ rated trip test - no trip
1 x rated current - average 38mS trip
5 x rated current - average 16mS trip

Ramp test starting at 12mA - RCD trips at 21-23mA

By switching on any random combination of protected outgoing MCB's the results in test B above are obtained up to and including 5 MCB's in the on position above which the ramp trip level reduces at approx 3mA per MCB and the ½ rated test starts to trip after the sixth MCB is switched on.

To improve Zs, we added a 10mm² additional earth from the DB to the main suppliers terminal, this only reduced the reading by 1 - 2 ohms and had no effect on the RCD tests.

With the property empty we also disconnected the main earth and equipotential bonding conductors and carried out a voltage test between the consumer's earth conductor and neutral to determine if any earth leakage was present the result was 0v.

With regards to the protected ring circuits, the highest R1 + R1 reading was 2.35 ohms with an average of 0.7ohms and Zs readings of between 0.51ohms and 0.98 ohms.

The measured incoming voltage at the time of testing was 247v between 8pm and 10pm.

It is our opinion that there is no obvious defect in the installation and that it has been carried out in accordance with BS7671, there is some dubiety as to when during construction the split load DB was fitted which may mean the installation is not fully compliant with the 17th Ed (conductors in walls etc etc)

We have never encountered such a situation before and wonder if there is some link between the large number of ring circuits in the property and the behaviour of the RCD. None of the ring circuits are heavily loaded and since a centralised lighting control system has been installed there are considerably less socket outlets per room than you would normally expect.

The customer reports that the RCD will trip totally at random regardless of time or day and whether the property is occupied or not.

The test results were check against our second calibrated multifunction meter and were identical

Anyone had a problem like this??
 26 July 2010 11:58 PM
User is offline View Users Profile Print this message



ant1uk

Posts: 2376
Joined: 22 June 2010

disconnect the loads and retest insulation resistance at 500v just to rule out any defects. for a fairly new installation each circuit should be above 200meg ohms.

Regards

Anthony
 27 July 2010 12:33 AM
User is offline View Users Profile Print this message



scottyscott

Posts: 47
Joined: 17 June 2009

Random tripping sounds like possible high protective conductor currents. If you have a clamp meter you could check the current flowing in each circuit with all loads on (CPC) - or test each load cpc to get a clear picture. More than 10mA and you need to take action as in the good book (reg 543.7). If you have already done this then it could be something like borrowed neutral or earth so best to test between live conductors aswell at 250 V. and check that neutrals from RCD protected circuits are in the correct neutral bar.

Scott

edited to add reg. and correct terms (hopefully)

Edited: 27 July 2010 at 12:48 AM by scottyscott
 27 July 2010 12:50 AM
User is offline View Users Profile Print this message



ant1uk

Posts: 2376
Joined: 22 June 2010

May even be one of those connected appliances with a high leakage current which is causing an out of balance of the trip coil within the RCD
 27 July 2010 09:19 AM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 17795
Joined: 13 August 2003

½ rated trip test - RCD trips (it should not)

Ramp test starting at 12mA - RCD trips instantaneously

Test B
With all protected MCB's off and tests carried out at the load side of the RCD;

½ rated trip test - no trip
1 x rated current - average 38mS trip
5 x rated current - average 16mS trip

Ramp test starting at 12mA - RCD trips at 21-23mA

It seems fairly clear that you have earth leakage - quite possibly from appliances rather than fixed wiring. Filter/surge arrestor extension leads are a frequent culprit.

With the property empty we also disconnected the main earth and equipotential bonding conductors and carried out a voltage test between the consumer's earth conductor and neutral to determine if any earth leakage was present the result was 0v.

But it sounds like you didn't disconnect all possible parallel earths ... supplementary bonding, class 1 equipment in contact with extraneous conductive parts (CH boiler, pump, immersion heater, etc etc).

Seems like quite a lot on one RCCB. How about splitting it down further - possibly using RCBOs instead.

- Andy.
 27 July 2010 09:50 AM
User is offline View Users Profile Print this message



Jaymack

Posts: 5571
Joined: 07 April 2004

If the fixed wiring checks out OK, I would carry out PAT for all the appliances, I have known a toaster to give intermittent problems while just plugged in. Appliances with heating elements are always suspect - ovens, kettles, washing machines and tumble dryers etc.

Regards
 27 July 2010 01:51 PM
User is offline View Users Profile Print this message



Jobbo

Posts: 924
Joined: 08 July 2010

northstarlx

Sounds to me that you have reversed polarity somewhere in one of the circuits (neutral/earth), it may be an idea to carry out all the schedule of tests for the complete installation. If there was an insulation resistance failure between neutral and earth that would almost certainly trip the Rcd without any mcb's being in the on position

Try turning on the RCD with the MCBs in the off position, using a low resolution clamp meter, place it around the live and neutral conductors, the reading shown will be the imbalance (you have aprox a 0.03A limit), then one by one turn the MCBs on and see what you get!! Like mentioned above, you could proceed by checking the load in both the live and neutral for each circuit to ensure they are balanced, either individually or collectively in the clamp meter

Ensure that the heating system wiring is checked fully, as this is the most common place for RCD tripping in domestic installations.

 27 July 2010 02:28 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 5571
Joined: 07 April 2004

Originally posted by: Jobbo
Sounds to me that you have reversed polarity somewhere in one of the circuits (neutral/earth)

I don't see that as a cause for an intermittent trip.

Ensure that the heating system wiring is checked fully, as this is the most common place for RCD tripping in domestic installations.

That's one reason, not the most common in my experience.

Regards
 27 July 2010 02:42 PM
User is offline View Users Profile Print this message



OMS

Posts: 22864
Joined: 23 March 2004

I don't see that as a cause for an intermittent trip.


A N/E fault is a prime candidate for intermittent trippng (particularly with any random collection of MCB's being switched on).

As you add load, the voltage between neutral and earth increases until there is sufficent fault current flowing to take out the RCD

Regards

OMS

-------------------------
Let the wind blow you, across a big floor.
 27 July 2010 02:52 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 5571
Joined: 07 April 2004

Originally posted by: OMS
I don't see that as a cause for an intermittent trip.

A N/E fault is a prime candidate for intermittent trippng (particularly with any random collection of MCB's being switched on).
As you add load, the voltage between neutral and earth increases until there is sufficent fault current flowing to take out the RCD

Agreed, I've had a few N_E faults where a fault has developed or where there is an existing fault with a new DB/RCD; but polarity as cause for "nuisance" trips?

Regards
 27 July 2010 03:08 PM
User is offline View Users Profile Print this message



ant1uk

Posts: 2376
Joined: 22 June 2010

It's surprising how many E-N reverses go undetected because the neon plug in testers don't detect this type of palarity reverse so really the only reliable way is R1+R2 or a visual check of connections.
 27 July 2010 03:37 PM
User is offline View Users Profile Print this message



Jobbo

Posts: 924
Joined: 08 July 2010

Yes Jaymack, if the neutral and earth are cross polarity, which is commonly found (especially with multicore cables since 2004 where people didn't know which core to use as earth/neutral)!! Then the load will pass down the earth and the Rcd will trip!! It can be intermittent, because that particular fault will only occur on load, i.e. light switch on timer, socket, timer control etc

Unless you test R1+R2 like mentioned above, no live test will identify neutral/earth cross, as neutral and earth are common (its in the small print on every socket tester, meter etc)!!
 27 July 2010 04:06 PM
User is offline View Users Profile Print this message



ant1uk

Posts: 2376
Joined: 22 June 2010

now that would be a great invention if someone could make a neon tester which eliminates this problem. The E-N star points within the transformer makes it impossible for the neon tester to detect.
 28 July 2010 10:32 PM
User is offline View Users Profile Print this message



northstarlx

Posts: 3
Joined: 26 July 2010

Hi Guys

Just a quick thanks for all of your helpful suggestions to the problem.

I will discuss with the client and report back any findings

Ian
 18 August 2010 12:57 PM
User is offline View Users Profile Print this message



DDAVEHAYES

Posts: 2
Joined: 18 August 2010

Have you checked neutral to earth insulation tests ? Even though the Mcb's are off the neutral conductors remain connected
 18 August 2010 06:42 PM
User is offline View Users Profile Print this message



andardry

Posts: 625
Joined: 02 August 2004

Supply is TNC-S with a Ze of 0.21 ohms at the meter and 0.24ohms at the DB Main Switch. With all the RCD protected MCB's off Zs measured on the load side of the RCD is 0.45ohms.


I would kinda expect Zs to be lower than the Ze?????

-------------------------
Andrew
 18 August 2010 06:54 PM
User is offline View Users Profile Print this message



Pacific

Posts: 738
Joined: 29 January 2005

Any outside lights or other outdoor equipment, I have had two similar faults recently, both were caused by water in fittings
Statistics

New here?

  • To participate in discussions, please log in and introduce yourself.

See Also:



FuseTalk Standard Edition v3.2 - © 1999-2020 FuseTalk Inc. All rights reserved.

 
..