EICR for an installation that cannot be switched off

Talk Electrician Forum

Help Support Talk Electrician Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Status
Not open for further replies.
Joined
Oct 6, 2017
Messages
18
Reaction score
21
Location
kent
Hi all,

Here is a question we got asked today.

We know the HASWA is in place to cover all aspects of safety at work, and can be viewed as the statutory document. And the EAWR 1989 are specific to electrical installations used in the work place.

And we know a way of ensuring conformance with the statutory regulations is to follow the requirements of BS 7671.

But what do you do when your customer tells you he needs an EICR but cannot switch of the power. Not ever. Not even pre-planned, out of hours, etc.?

Is there a know accepted alternative to the standard testing regime? What do you do in this situation?

TIA for your help with this.

 
Are you going to be happy to sign the certificate satisfactory without doing any dead test .

 
But what do you do when your customer tells you he needs an EICR but cannot switch of the power. Not ever. Not even pre-planned, out of hours, etc.?

Is there a know accepted alternative to the standard testing regime? What do you do in this situation?

TIA for your help with this.


Well in my opinion that is about as logical as asking a garage to do an MOT and service on my car, but tell them they have got to do it whilst I am driving to see a client 150 miles away and I must not be held up or delayed.

You need to tell your customer you cannot do an EICR without disconnecting all or part of the installation. You could offer to do a few circuits at a time over a period of a few weeks or months to minimise some of the inconvenience. But If they flatly refuse to allow any disconnection of power the only thing you could offer is a visual check to look for any obvious physical damage or missing labels or notices or inappropriate installation methods hanging lose cables etc..  But you cannot verify or sign to say the integrity of the circuits are electrically satisfactory or safe for continual use, or if any protective RCD's are operating correctly, without removing some power.

If they cannot provide a means to allow appropriate tests and inspections of the electrical installation to be undertaken I would think they are in breach of H&SAW etc.

Doc H.

 
Last edited by a moderator:
I'm guessing they don't want to shut their servers down  ?    Perhaps you could test everything  EXCEPT  the servers ...I used to do that at two different printers.    

Limitations on the server circuits  , nobody seemed bovvered. TBH 

 
Note carefully that an EICR is an INSPECTION & test of the electrical installation to ensure that it is safe for continued use.

“621.2  Periodic inspection comprising a detailed examination of the installation shall be carried out without dismantling, or with partial dismantling as required, supplemented by appropriate tests…”

GN3 reinforces this with the addition of: “The tests are mainly to confirm that the disconnection times stated in Chapter 41 are met.”

Well let's look at what can, be done.

Check and verify previous documentation for correctness against BS7671.

Check and verify labelling against the current version of BS7671.

Visual inspection of complete installation.

Recording of any evidence of overheating or damage from the visual inspection.

In fact, apart from internal inspections of accessories or current consuming devices the whole installation can be visually inspected externally and internally.

Loop impedances on all circuits, and with suitable live working precautions at every distribution board.

Visual inspection of the internals of all distribution boards with suitable live working precautions at every distribution board.

Thermal imaging of the internals of all distribution boards with suitable live working precautions (OK this is not part of the official BS7671 procedure, however, IF the installation cannot be de-energised under any circumstances does offer a useful solution).

Earth leakage measurement at every distribution board, and potentially every final circuit, with suitable live working precautions (OK this is not part of the official BS7671 procedure, however, IF the installation cannot be de-energised under any circumstances does offer a useful solution).

Earth fault loop impedance at every socket outlet that does not have anything connected, and where devices can be disconnected those socket outlets too.

Likely it would be feasible with suitable precautions to undertake Zs testing at all luminaires

R2 wander lead testing across the whole installation to accessible metallic parts.

Any circuits that can be isolated can have dead tests undertaken with suitable live working precautions if working internally in live distribution boards to undertake disconnections and re-connections.

Now that’s quite a lot, OK, no IR testing, if there are 30mA RCD’s on the circuits, they are verifying that the combined leakage and IR does not equate to much more than 24mA, so that would suggest that these circuits are probably OK, combined with the visual inspections.

Remember that an R2 is a valid option as opposed to an R1+R2 measurement, even on initial verification.

So what can’t we do, IR testing on sub-mains, and final circuits unless they can be isolated.

RCD testing, but, the client must do this 3 monthly anyway to prove compliance with HASAWA74, S2 & S3, therefore doing this on an EICR will not be a problem.

Not essential to prove ADS, which is one of the main aims.

R1+R2 testing on circuits, not, really, needed on a periodic anyway.

So, all of a sudden with some thought and some creativity almost all, if not all of the schedule of inspections can be completed.

Also, almost all of the test results can be completed.

If they cannot isolate any of the installation at any time, under any situations, what their business continuity arrangements are in the event of a power outage from the DNO.

Do they have their servers on active UPS’s?

In the event of an extended outage, do they have on site generation with suitable change over?

If not do they have an arrangement for generators to be delivered and connected before their UPS’s die?

If they have an ACB on their installation, how do they comply with the maintenance requirements of this without disconnection and removal.

I have only ever heard of one site that has refused total black building, and I have a friend who does a lot of I&T in the city, and even the banks, data centres, trading floors & Whitehall buildings, can be, and are shut down with suitable arrangement.

They are also working on options for this site to be shut down, but that will mean use of data centres outside the UK to take up the capacity and that will require some planning.

Most of the high-profile institutions and organisations realise how much they rely on their electrical infrastructure and therefore realise that it requires adequate maintenance.

@GM at CA UK, perhaps this will help your thoughts?

 
Last edited by a moderator:
One issue if they allow some circuits to be turned off, is circuit identification.

e.g I wanted to change a hand dryer.  I Switched off the circuit labelled "gents toilet hand dryer" then had loads of people running about like headless chickens because an office full of pc's had been shut down while they were working.

 
One issue if they allow some circuits to be turned off, is circuit identification.

e.g I wanted to change a hand dryer.  I Switched off the circuit labelled "gents toilet hand dryer" then had loads of people running about like headless chickens because an office full of pc's had been shut down while they were working.


That I would include against my first two observations, i.e. does the documentation seem to meet the labelling and does that meet the installation layout.

At the end of the day you can do a visual inspection only and it would be a valid EICR, as long it was described as such and agreed by both parties.

It may not be suitable and sufficient, but as long as that is made clear to the client, and the terms agreed in writing, then I would have no issue with it.

Ultimately you can legitimately do an EICR on a single circuit as long as it is all documented as such.

 
Agree with the above from sidewinder, also wish to add that with some prior planning you can request a shutdown thats more likely to be given the go ahead by the client...

Rather than "I'll need to be turning circuits off as I test them", If beforehand you have identfiied circuits (fuse finder helps), Done earth loop tests and visual inspections on them all, R2 to ECPS, then how long does it take to do everything else required on say an average 8 WAY TPN board*? Less time than you might otherwise think!. If you were to ask them for a hour when you can switch it off, even if that means working out out hours... Nip round unplug everything, switch off isolators for fixed equipment, quick global IR, end to end on 8 rings. 8 RCBO tests, back on, plug everything back in.

Now granted you might not get a favorable response everywhere, but most places can manage an hours downtime given enough notice** even if it requires the IT team having to do overtime! (that seems to be tricky bit a lot of the time - they often don't like anything that disrupts their little world!)

*Assuming 24 SP circuits, 8 ring circuits, 8 lighting cirucits and 8 fixed equipment radials

** Once this ended up being the best part of two years!

 
Gents, thanks so much for your detailed answers, I really appreciate the time and effort you have put into this. So it looks like there is quite a lot of fairly comprehensive testing and inspection that can be done live.

The idea of measuring earth leakage as an alternative to insulation testing where a circuit cannot be disconnected interests me.

I'm thinking that if one were to isolate an installation and perform an insulation test between all of the phases (and neutral) connected together and the main earthing point, the test current would take the same route as any leakage current that would flow when the system is energised.

Accordingly, if one then didn't isolate an installation, but rather measured total earth leakage at the incoming supply (by clamping round all phases and neutral) and, using ohms law, divided that into the phase-earth voltage, you would get a resistance value made up of all of the parallel leakage paths to ground.

So lets say there was 0.4mA of measure leakage current and a 230v live/earth potential difference, then 230/0.0004 = 575000 ohms or 0.575M ohms.

I guess expanding on the above, one could measure the earth leakage on each circuit and come up with individual values of resistance for the leakage paths of those circuits, aka insulation resistance.

I appreciate that this is not a 500v DC insulation test, but rather a 230v AC test, but in the scenario above I believe both test currents would take the same path. For sure 500V DC stresses the insulation more, and so may show other issues that would not manifest themselves at the lower mains supply voltage, but maybe a considered alternative, particularly if those leakage values are constantly monitored over time.

 
And remember that it is just as important ( if not more so) to list what you HAVE NOT tested)... otherwise some bewigged nonce from our esteemed legal ambulance chasing profession will want your clock-weights as a rear view mirror embellishment 

"well,Mr Sparky, you say you performed an EICR. At no point does it say you did NOT test the circuit that resulted in my money grabbing clients injury. Did you" .I rest my case m'Lud 

 
Gents, thanks so much for your detailed answers, I really appreciate the time and effort you have put into this. So it looks like there is quite a lot of fairly comprehensive testing and inspection that can be done live.

The idea of measuring earth leakage as an alternative to insulation testing where a circuit cannot be disconnected interests me.

I'm thinking that if one were to isolate an installation and perform an insulation test between all of the phases (and neutral) connected together and the main earthing point, the test current would take the same route as any leakage current that would flow when the system is energised.

Yes it would, so that is why there is some mileage in the earth leakage test, but it is not as yet an official test.

Accordingly, if one then didn't isolate an installation, but rather measured total earth leakage at the incoming supply (by clamping round all phases and neutral) and, using ohms law, divided that into the phase-earth voltage, you would get a resistance value made up of all of the parallel leakage paths to ground.

So lets say there was 0.4mA of measure leakage current and a 230v live/earth potential difference, then 230/0.0004 = 575000 ohms or 0.575M ohms.

Yes that's the theory, but as the minimum insulation resistance is 1M Ohm, then we are looking for values < 0.23mA, the sensors are not there yet.

Challenge 1! ;)

I guess expanding on the above, one could measure the earth leakage on each circuit and come up with individual values of resistance for the leakage paths of those circuits, aka insulation resistance.

The trick is being able to encircle all live conductors, with a suitably sensitive clamp!

I appreciate that this is not a 500v DC insulation test, but rather a 230v AC test, but in the scenario above I believe both test currents would take the same path. For sure 500V DC stresses the insulation more, and so may show other issues that would not manifest themselves at the lower mains supply voltage, but maybe a considered alternative, particularly if those leakage values are constantly monitored over time.

You're getting there! ;)

Also, a 250V DC test is IMHO acceptable for all live conductors to earth but, the minimum insulation resistance is still 1M Ohm.

It was allowed in the 16th but, disappeared in the 17th, we'll have to see if it comes back in the 18th...  It was in the DPC, but only under special circumstances.

Remembering that 230V AC nominal is probably higher than that and the pk voltage is Root 2 above this so 325V ac approx up to maybe 357 V ac.


Well kind of, it's not an official test, but, it is an indicator, and is useful for fault finding for sure.

I think it would be good to try a CA 8336 with a B102 or an MN93/93A to try and monitor leakage in an installation, this could be done on single or three phase.

Don't have the clamp though, and I'm not sure what other options on clamps there are that have that sort of sensitivity.

It wouldn't matter what the data name was in the instrument, that can be sorted later, it's gathering the data that is important.

IYKWM.

Chapter 4.9 in GN3 does recognise thermal imaging as a useful test, even though it is not a recognised formal test under BS7671.  It is an industry accepted test method for fault finding, and when correctly applied is a very useful test.

Also in GN3 now Ch 2.6.14 recognises stake and clamp and clamp on stakeless testers as a means of testing for earth electrode resistance.

Leakage current measurement is recognised in GN3 but only for IT systems, as in Impedance/Terra, not Information Technology.

This came from the meeting you went to which meant you could not make Elex Sandown didn't it! ;)

 
Well kind of, it's not an official test, but, it is an indicator, and is useful for fault finding for sure.

I think it would be good to try a CA 8336 with a B102 or an MN93/93A to try and monitor leakage in an installation, this could be done on single or three phase.

Don't have the clamp though, and I'm not sure what other options on clamps there are that have that sort of sensitivity.

It wouldn't matter what the data name was in the instrument, that can be sorted later, it's gathering the data that is important.

IYKWM.


I'll see what large diameter flexiclamps we have that would be suitable to connect to a Qualistar and will send you one to play with. It may be that we have a better solution as a stand alone current measurement device. Ill check that out too. Ill also look to see what the best sensitivity is that we can produce for earth leakage based on the need to measure fractions of a mA.

Chapter 4.9 in GN3 does recognise thermal imaging as a useful test, even though it is not a recognised formal test under BS7671.  It is an industry accepted test method for fault finding, and when correctly applied is a very useful test.

Also in GN3 now Ch 2.6.14 recognises stake and clamp and clamp on stakeless testers as a means of testing for earth electrode resistance.

Leakage current measurement is recognised in GN3 but only for IT systems, as in Impedance/Terra, not Information Technology.

This came from the meeting you went to which meant you could not make Elex Sandown didn't it! ;)

It started there, but was a genuine question that came to us at our office after the event.

From what I heard (when I wasn't at ELEX Sandown) the guys did a "prescribed sequence" of tests that included ultrasonic listening to distribution boards, measurement of voltage drops across connections, thermal imaging of panels, and even analysis of PQ for spikes and harmonics, etc. We were there because of our single phase PQ analysers (one of which they use).

So, sadly I only got to present our session, and not listen to any others which would have been great, but talking to one guy in the evening I mentioned that they appeared to do some interesting tests that I hadn't heard of being used in this application before (ultrasonic sniffing to identify micro arcing poor connections for example) but never appeared to do other tests that might be more obvious (earth leakage).

And the very next day we got called up and asked the question of what we would recommend testing wise, in the event the customer could not switch off, and specifically how to satisfy them if they wanted an EICR. So I'm guessing there are a variety of ideas out there and its all still evolving.

Thanks again for time you've spent answering this - Ill get on to looking for a clamp for you :signthankspin:

 
Note carefully that an EICR is an INSPECTION & test of the electrical installation to ensure that it is safe for continued use.

“621.2 Periodic inspection comprising a detailed examination of the installation shall be carried out without dismantling, or with partial dismantling as required, supplemented by appropriate tests…”

GN3 reinforces this with the addition of: “The tests are mainly to confirm that the disconnection times stated in Chapter 41 are met.”

Well let's look at what can, be done.

Check and verify previous documentation for correctness against BS7671.

Check and verify labelling against the current version of BS7671.

Visual inspection of complete installation.

Recording of any evidence of overheating or damage from the visual inspection.

In fact, apart from internal inspections of accessories or current consuming devices the whole installation can be visually inspected externally and internally.

Loop impedances on all circuits, and with suitable live working precautions at every distribution board.

Visual inspection of the internals of all distribution boards with suitable live working precautions at every distribution board.

Thermal imaging of the internals of all distribution boards with suitable live working precautions (OK this is not part of the official BS7671 procedure, however, IF the installation cannot be de-energised under any circumstances does offer a useful solution).

Earth leakage measurement at every distribution board, and potentially every final circuit, with suitable live working precautions (OK this is not part of the official BS7671 procedure, however, IF the installation cannot be de-energised under any circumstances does offer a useful solution).

Earth fault loop impedance at every socket outlet that does not have anything connected, and where devices can be disconnected those socket outlets too.

Likely it would be feasible with suitable precautions to undertake Zs testing at all luminaires

R2 wander lead testing across the whole installation to accessible metallic parts.

Any circuits that can be isolated can have dead tests undertaken with suitable live working precautions if working internally in live distribution boards to undertake disconnections and re-connections.

Now that’s quite a lot, OK, no IR testing, if there are 30mA RCD’s on the circuits, they are verifying that the combined leakage and IR does not equate to much more than 24mA, so that would suggest that these circuits are probably OK, combined with the visual inspections.

Remember that an R2 is a valid option as opposed to an R1+R2 measurement, even on initial verification.

So what can’t we do, IR testing on sub-mains, and final circuits unless they can be isolated.

RCD testing, but, the client must do this 3 monthly anyway to prove compliance with HASAWA74, S2 & S3, therefore doing this on an EICR will not be a problem.

Not essential to prove ADS, which is one of the main aims.

R1+R2 testing on circuits, not, really, needed on a periodic anyway.

So, all of a sudden with some thought and some creativity almost all, if not all of the schedule of inspections can be completed.

Also, almost all of the test results can be completed.

If they cannot isolate any of the installation at any time, under any situations, what their business continuity arrangements are in the event of a power outage from the DNO.

Do they have their servers on active UPS’s?

In the event of an extended outage, do they have on site generation with suitable change over?

If not do they have an arrangement for generators to be delivered and connected before their UPS’s die?

If they have an ACB on their installation, how do they comply with the maintenance requirements of this without disconnection and removal.

I have only ever heard of one site that has refused total black building, and I have a friend who does a lot of I&T in the city, and even the banks, data centres, trading floors & Whitehall buildings, can be, and are shut down with suitable arrangement.

They are also working on options for this site to be shut down, but that will mean use of data centres outside the UK to take up the capacity and that will require some planning.

Most of the high-profile institutions and organisations realise how much they rely on their electrical infrastructure and therefore realise that it requires adequate maintenance.

@GM at CA UK, perhaps this will help your thoughts?
I teach this stuff, and your explanation of Periodic Inspection and testing is by far the best explanation i have ever read about the subject, would you mind if i nick it?

Locked to prevent being resurrected (again)
 
Last edited by a moderator:
Status
Not open for further replies.
Top