IET
Decrease font size
Increase font size
Topic Title: Very Strange Issue - Multiple RCBOs Tripping in a Commercial Office
Topic Summary:
Created On: 11 October 2018 01:05 PM
Status: Read Only
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.
 11 October 2018 01:05 PM
User is offline View Users Profile Print this message



fzn10

Posts: 20
Joined: 22 September 2013

Hi All,

Im having issue with a relatively new office premise where multiple RCBOs are tripping at random and instantaneously on a monthly basis. The premise is supplied from multiple distriution boards , with RCBOs tripping across all DBS.

Sometimes we have 15 RCBOs tripping at the same time from different DBs! We've done live and cold tests, checked for shared nuetrals, did RCD tests, checked subcircuit leakage currents and all results came back ok.

The premise has every circuit connected to a dedicated RCBO for compliance with a TT Earthing system requirements.

I've contacted the RCBO manufacturer who visited site, and they cant give me an answer.

How is it possible to have multiple RCBOs trip at the same time and from different DBs? Could this be an issue from the supply side? What could be causing this issue? I've discussed this with so many engineers and electricians and none have encountered such anissue.

Here are some other information which might be of interest:-

-Voltage between Neutral and Earth is 3V at the supply side of the DB.

-Earthing system is TT. Typical fault loop impedance figures are 20 ohms which is consdered acceptable.

-The subcontractor advised that this could be due a bad earth by the landlord. But i dont see why this would result in RCBOs tripping.

-Been advised that RCBOs have previously tripped in the middle of the night when the office was unoccupied.

-We've been advised that other tenants in the building changed their RCDs to MCBs due to tripping issues but i dont know how true this is.

-We've changed one of the DBs RCBOs to MCBs and the circuits no longer tripped within the DB, so it cannot be an overload issue.

Any input or thoughts would be appreciated.
 11 October 2018 01:20 PM
User is offline View Users Profile Print this message



mapj1

Posts: 11588
Joined: 22 July 2004

Some thoughts, not neccesarilly all will apply.
I presume the RCBOs are single pole breaking, and do not interrupt the neutral ?
If so then a neutral/earth event will be seen by an upstream device feeding a DB, even if the down stream device is off.
have you checked the L+N to E insulation resistance at 500VDC, or at least 250?
Are there loads on timers (heating is a classic) that may have been off at the time of testing?
Are all the RCBOs the same type? could it be a 'frisky' batch
How does the supply come in? Is it possible that the neutral is higher impedance than it shoud be, e.g. over head lines corroding
Are there impulsive loads (welding or other industrial plant) sharing the substation thta may be injecting spikes into the supplies?

Does a bogey RCBO trip with no load, or just a light? This may be a thing to try.

Do you have any time delay RCD or earth fault relay up front, and if so does that ever trip?
(rules out short spikes if it does ....)

-------------------------
regards Mike
 11 October 2018 01:22 PM
User is offline View Users Profile Print this message



sparkingchip

Posts: 11471
Joined: 18 January 2003

Do the RCD devices have a flying earth lead to give a reference to earth?

Are they type AC or type A.

Are they single pole or single pole with a switched neutral?

What RCD rating are they 30 mA or 100 mA?

What make are they?

Andy B.
 11 October 2018 02:34 PM
User is offline View Users Profile Print this message



fzn10

Posts: 20
Joined: 22 September 2013

Hi Mike, Andy,

Thanks for the questions, responses as follows:-

"I presume the RCBOs are single pole breaking, and do not interrupt the neutral ?"
Yes, they single pole. Does not interupt Nuteral

"If so then a neutral/earth event will be seen by an upstream device feeding a DB, even if the down stream device is off. "
How would this impact other RCBOs within the same and different DBs.

"have you checked the L+N to E insulation resistance at 500VDC, or at least 250? "
Yes insulation resistance tests were all checked at 250V. No issues were noted. We were not able to check every single circuit unfortunately as it is an existing installation. However if there was a fault, shouldnt it be localized? rather then have multiple RCBOs trip accross different distribution board?

Are there loads on timers (heating is a classic) that may have been off at the time of testing?
I live in a hot climate, no heating. All fan coil units were operational. There may have been a hot water cylender. But again how would this result in multiple RCBOs tripping.

Are there impulsive loads (welding or other industrial plant) sharing the substation thta may be injecting spikes into the supplies?
No, Substaion is dedicated to the building which supplies other office tenants.

"Does a bogey RCBO trip with no load, or just a light? This may be a thing to try. "
When testing, we've injected Risudal currents to subciruits at 50% and they did not trip.

"Do you have any time delay RCD or earth fault relay up front, and if so does that ever trip? "
No time delay on RCBOs

Do the RCD devices have a flying earth lead to give a reference to earth?
Yes it does have a flying lead which gives it a reference to earth.

Are they type AC or type A?
AC

Are they single pole or single pole with a switched neutral?
Single Pole only.

What make are they?
ABB. All Brand new and same type. DS271 AC Type B,C RCBOs Link below.
https://www.voltimum.co.uk/e-catalogue/brand/abb/family/ds271-ac-type-bc-rcbos
 11 October 2018 04:00 PM
User is offline View Users Profile Print this message



mapj1

Posts: 11588
Joined: 22 July 2004

OK. The probelm with single pole breaking is when one DB feeds another - if there is an NE fault, then all the RCBOs between the supply and the fault will trip, not just the one nearest, as all the neutrals are shared, and the fault is not removed when just the phase line opens.

Also if the earthing is shared with othe users,, it may be that some other load is putting occasional pulses down the earth core, maybe from another phase, or in time with yours. Then the N-E voltage will bounce - not normally a problem but the RCBOs may not appreciate large transints between FE and N.
datesheet is not that forthcoming about such details.

-------------------------
regards Mike
 11 October 2018 09:31 PM
User is offline View Users Profile Print this message



peteTLM

Posts: 3751
Joined: 31 March 2005

Although every circuit has in theory an RCD, still needs an upfront device (TD 100ma) on each DB as nothing breaks the neutral...

-------------------------
----------------------------------------
Lack of planning on your part doesn't make it an emergency on mine....

Every man has to know his limitations- Dirty Harry
 11 October 2018 10:21 PM
User is offline View Users Profile Print this message



sparkingchip

Posts: 11471
Joined: 18 January 2003

Originally posted by: peteTLM

Although every circuit has in theory an RCD, still needs an upfront device (TD 100ma) on each DB as nothing breaks the neutral...


Which BS7671 Wiring regulation states that requirement?
 11 October 2018 10:38 PM
User is offline View Users Profile Print this message



Alcomax

Posts: 399
Joined: 12 November 2009

Originally posted by: sparkingchip

Originally posted by: peteTLM



Although every circuit has in theory an RCD, still needs an upfront device (TD 100ma) on each DB as nothing breaks the neutral...




Which BS7671 Wiring regulation states that requirement?


Common sense with a good design should be enough. The set up is asking for trouble , there is no neutral fault discrimination between boards. Bit of a nightmare really.
 11 October 2018 11:24 PM
User is offline View Users Profile Print this message



sparkingchip

Posts: 11471
Joined: 18 January 2003

Originally posted by: Alcomax

Originally posted by: sparkingchip



Originally posted by: peteTLM







Although every circuit has in theory an RCD, still needs an upfront device (TD 100ma) on each DB as nothing breaks the neutral...








Which BS7671 Wiring regulation states that requirement?




Common sense with a good design should be enough. The set up is asking for trouble , there is no neutral fault discrimination between boards. Bit of a nightmare really.


The IET engineers have written a note in the Onsite Guide to BS7671:2018 at the bottom of page 21 to point out that it may not be a requirement.

Andy B.
 11 October 2018 11:34 PM
User is offline View Users Profile Print this message



Alcomax

Posts: 399
Joined: 12 November 2009

Originally posted by: sparkingchip

Originally posted by: Alcomax





Originally posted by: sparkingchip[/I





Originally posted by: peteTLM















Although every circuit has in theory an RCD, still needs an upfront device (TD 100ma) on each DB as nothing breaks the neutral...
















Which BS7671 Wiring regulation states that requirement?








Common sense with a good design should be enough. The set up is asking for trouble , there is no neutral fault discrimination between boards. Bit of a nightmare really.




The IET engineers have written a note in the Onsite Guide to BS7671:2018 at the bottom of page 21 to point out that it may not be a requirement.



Andy B.



There is no accounting for taste.

If the"engineer's" deemed this opinion so important surely it would be in BS7671? Edited for smartfonefinger

Edited: 11 October 2018 at 11:43 PM by Alcomax
 12 October 2018 12:11 AM
User is offline View Users Profile Print this message



sparkingchip

Posts: 11471
Joined: 18 January 2003

More to the point, is there a regulation that says there has to be an upfront RCD in a consumer unit or distribution board in a TT installation if all the CPDs are RCBOs?

I cannot find a regulation that states a upfront RCD is required and the IET engineers specifically point out in the OSG that an upfront RCD is not always required.

Therefore as far as I can see, there's no requirement for a upfront RCD in some consumer units and distribution boards, even in TT installations, so long as the construction of the equipment is appropriate. So in some TT installations fitting an upfront RCD is an engineering decision made by the designer, not an actual requirement of the regulations.

Whilst we are at it, is there a regulation that states that double pole RCBOs have to be used in domestic TT installations, even is there is not an upfront double pole RCD?

Andy B.
 12 October 2018 12:21 AM
User is offline View Users Profile Print this message



Alcomax

Posts: 399
Joined: 12 November 2009

I think it is not necessarily altogether healthy to be wedded to regulations. The OP related to TT. An all pole RCD somewhere has been used to good effect for more years than I care to remember.
 12 October 2018 08:13 AM
User is offline View Users Profile Print this message



Zoomup

Posts: 5464
Joined: 20 February 2014

Chapter 41 Protection Against Electric Shock. 411.5.2 says that in a TT system one or more of the following protective devices may be used, the former being preferred: (i) An R.C.D. (ii) An overcurrent protective device. The 18th edition uses the same wording.

Also 531.4.1 says that if a TT installation is protected by a single R.C.D. this shall be placed at the origin of the installation.....

Z.

Edited: 12 October 2018 at 08:25 AM by Zoomup
 12 October 2018 08:43 AM
User is offline View Users Profile Print this message



sparkingchip

Posts: 11471
Joined: 18 January 2003

Originally posted by: Zoomup

Chapter 41 Protection Against Electric Shock. 411.5.2 says that in a TT system one or more of the following protective devices may be used, the former being preferred: (i) An R.C.D. (ii) An overcurrent protective device. The 18th edition uses the same wording.



Also 531.4.1 says that if a TT installation is protected by a single R.C.D. this shall be placed at the origin of the installation.....



Z.


I have started a separate discussion, as it going off the subject actually under discussion in this topic.
 12 October 2018 09:14 AM
User is offline View Users Profile Print this message



IronFreely

Posts: 402
Joined: 06 November 2014

Firstly regarding the original post... I recently spent two days chasing a non existing fault around a big building, I scratched my head a lot, we just couldn't work out why various RCDs and RCBOs kept disconnecting seemingly at random (we didn't have the discrimination problems described here as no RCDs were in line with each other) but... Eventually I discovered a high resistance fault on the supply Neutral (I believe it's what the network people call a floating Neutral) I have no idea of the physics involved but the problems with the supply Neutral manifested itself with multiple RCD failures. What I found strange is when I checked supply voltage/polarity with final circuits isolated it all seemed fine also with minimal load on the system but when we switched on a significant number of items the voltage between L and N began to fluctuate between 50v and 250v. The problem was only detectable with a voltage indicator when there was a load exceeding about 10 amps, it was odd but there you go, my advice is check if there is any significant difference in voltage of L-N compared with L-E when the system is under load. I also noted that as we increased load on the system I could measure an increasing voltage between N and E, it usually hung around 10v but would pop up to 50v and I breafly saw it even higher once. If so you might need the supplier to check their Neutral or perhaps that all phases are sound if cross line voltage isn't consistently 415v or so.

Just going to weigh in on the Alomax/Sparking chip conversation. I've queried this myself once or twice and no one has ever been able to show me an actual requirement number stating a TT system must have a DP RCD on all circuits and the OSG even has a picture of a multi SP RCBO board with no upfront RCD as one of its recommended setups for TT systems since we started using metal boards again. There is a requirement to have DP isolation but this does not state automatic disconnection, only that there must be a linked main switch which does not need to be an RCD. Perhaps we should draw the line between requirements and best practices here?
I've said it before and I'll say it again, DP RCBOs are coming down in cost at the moment and undoubtedly are going to become more popular in the future (not seen any typ A or F ones yet)
 12 October 2018 09:32 AM
User is offline View Users Profile Print this message



sparkingchip

Posts: 11471
Joined: 18 January 2003

I have struggled to find out why a RCD was occasionally tripping in a house with a TT installation and could not find anything wrong in the house, eventually the RCD tripped and would not reset. Waving a volt stick about indicated that the neutral tail was live with the RCD off due to a failed DNO overhead cable neutral joint on top of a pole in the garden, which also supplied the adjoining house.

So I'd agree that it needs to be investigated if there is a bad neutral joint upstream of the RCBOs, possibly out of the installation.

Andy B.
 12 October 2018 11:05 AM
User is offline View Users Profile Print this message



mapj1

Posts: 11588
Joined: 22 July 2004

Further to that, as these devices have a functional earth (FE) as well as the neutral, I'd be wondering if something that is common to all users on the transformer is either bouncing the CPC, or the neutral, but not frequently enough to be obvious what.
It could, but may not be, at the transformer earth end of course. Do you know how HV and LV earths are being handled ?
To verify the neutral resistance,
1)Does a PSSC check stack up with cacble sizes and distance to transformer, and if you have 3 phases, can you compare the phase to phase volt drop on a known heavy load with the phase to neutral drop ? Neutral should rise by the same amount phase drops.

-------------------------
regards Mike
 12 October 2018 12:36 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 17451
Joined: 13 August 2003

+1 for a dodgy N (flylead PE RCBOs are meant to trip if N is lost but L is intact).

Other possibility is radio interference - there have been reports of hand held radios (being used during an EICR) tripping electronic RCBOs. Local taxi firm downstairs perhaps? or close to some "defence" establishment?

If the installation is really TT (seems odd if it has its own dedicated transformer), there are multiple DBs and the only RCDs are RCBOs on final circuits, then what's providing ADS for the submains?

- Andy.
 12 October 2018 11:42 PM
User is offline View Users Profile Print this message



sparkingchip

Posts: 11471
Joined: 18 January 2003

I have just learnt something I didn't know about ABB DSE201 RCBOs.

DSE201 is also equipped with an overvoltage protection feature in order to monitor the voltage level of the network. When the voltage exceeds a preset threshold (330V) the DSE201 trips, shielding loads from damage. In this way, more safety is ensured to the installation.

ABB DSE201 brochure.

Does that explain anything?
 13 October 2018 12:06 AM
User is offline View Users Profile Print this message



mapj1

Posts: 11588
Joined: 22 July 2004

Nice in some ways, I wonder if it is detecting between L-N or L-E? (still thinking noisy N-E voltage.)
Slightly oddly, the full datasheet makes no mention of that, but does say you can couple it up to a separate over-voltage release module, the S2C-OVP which makes little sense if one is already built in, unless I suppose you wish to do the rare thing and monitor one part of the installation for over-voltage, but trip off another.

-------------------------
regards Mike
Statistics

New here?


See Also:



FuseTalk Standard Edition v3.2 - © 1999-2018 FuseTalk Inc. All rights reserved.

 
..