Basic installation testing with minimal disturbance

[SimonH2]

I've tenants (hopefully) about to move out of the flat, and now I own an MFT (Alphatek) I'm thinking about what tests would be sensible - beyond a visual inspection.

I don't want to go to a level that involves disturbing lots of connections - basically it's finding a balance between doing some basic tests that should show nothing major is wrong, and disturbing lots of connections which could be as likely to introduce faults as to find them. The board is fully RCBO. I have a set of results from an EIC about 18 months ago.

So my plan is :

Main switch off (& locked)
Visual inspection - in particular see if it looks like the tenant has had anything to bits
Cover off CU, IR test whole installation at 250 or 330V *
Link L&E, do R1+R2 tests on everything that doesn't need dismantling **
Remove L-E link
Do r1, rn, r2 checks at a socket for the RFCs
Power on
Test RCDs
Check Ze/PFC

* 250 is a standard voltage, the tester can set multiples of 10V (from memory). 330V is the peak voltage for 240V AC, so I figure that if anything can take 240V AC it should be good for 330V DC - saves having to disconnect all the RCBOs, PIR lights, etc.
If one test of the whole installation is OK, then I can skip testing individual circuits.

** Sockets/RFCs are OK (got a plug in adapter, can quickly go round the ring), cooker has a socket on the switch, most lights have roses so terminals are easily accessible. A couple of circuits (smoke detector, outside lights) are more awkward so I wasn't going to touch them.


Does that sound sensible ? Anything you'd add ? Anything you'd leave out ?

 

[RF Lighting]

Are you just testing this place so you can have a play with your new tester or are you trying to actually put a safety certificate on the place?

 

[Adam_151]

You'll have to disconnect the white flying leads on the RCBOs when doing your global IR (and remember to tie L&N together and test to earth, not separately to earth)

I wouldn't bother with R1+R2, either do zs or R2

 

[PrenticeBoyofDerry]

You need to follow the correct inspection and test procedures.
This is done by doing a EIRC.
There are lots of important things you have missed and some tests you have mentioned are being done incorrectly and not in the correct order.
What about main earth bonding, polarity (dead&live) Zs RCD trips.
Do you know the correct coding and what you are looking for on a visual? and how can you do this without removal of covers, then there's your lighting circuit, cooker, shower circuits. You need to remove covers to provided the information necessary to assure you have made sure that the installation is safe. Wonder what a judge would think, if you fried a tenant?
Is your MFT calibrated?

I wouldn't bother with R1+R2, either do zs or R2

Why would you not perform a r1+r2?
do either r2 or Zs.?
Seems to be lots of unfilled boxes here?

 

[ban-all-sheds]

* 250 is a standard voltage, the tester can set multiples of 10V (from memory). 330V is the peak voltage for 240V AC, so I figure that if anything can take 240V AC it should be good for 330V DC - saves having to disconnect all the RCBOs, PIR lights, etc.
If one test of the whole installation is OK, then I can skip testing individual circuits.

But you won't know it's OK.

330V is not what BS 7671 requires - for LV circuits up to 500V you must test at 500V DC.

I suggest you read Part 6, GN3 and a decent text book on Inspection & Testing.

 

[SimonH2]

Are you just testing this place so you can have a play with your new tester

Shucks, busted Though I have already had a play with the meter.

or are you trying to actually put a safety certificate on the place?

Not trying to put a safety certificate on it, just to do some basic checks. The alternative is no checks at all - or nothing past a visual.

You'll have to disconnect the white flying leads on the RCBOs when doing your global IR (and remember to tie L&N together and test to earth, not separately to earth)

You reckon they can take 240V AC but not 250V DC ? That's the reason for not using 500V for IR testing as I specifically want to avoid disturbing lots of connections.

I wouldn't bother with R1+R2, either do zs or R2

Makes sense.
Just to check, Zs = PFC measured at the outlet/service rather than CU ? R2 is done with a wandering lead back to the CU earth terminal (and the meter zeroed to exclude it's resistance) ?
Neither of which would detect wrong polarity (L&N swapped) ?

You need to follow the correct inspection and test procedures.
This is done by doing a EIRC.
There are lots of important things you have missed and some tests you have mentioned are being done incorrectly and not in the correct order.
What about main earth bonding

Visual - same as EICs I've had done in the past.

polarity

Won't have changed unless it's been fiddled with, and also detected while doing R1+R2.

RCD trips.

Err, part of "Test RCDs"

Do you know the correct coding and what you are looking for on a visual? and how can you do this without removal of covers, then there's your lighting circuit, cooker, shower circuits. You need to remove covers to provided the information necessary to assure you have made sure that the installation is safe. Wonder what a judge would think, if you fried a tenant?

Coding is irrelevant. I know what the installation used to look like and should still look like. I'll have to take some fittings off, but I want to minimise disturbance and the risk of introducing a fault.

Is your MFT calibrated?

Yes actually.

Seems to be lots of unfilled boxes here?

Yes, it's not intended to be a full EIC - just a quick check. IMO a full EIC would involve such a level of disruption that the risk of introducing a fault is more than the risk of not finding a fault that's subtle enough not to be detected by the abbreviated test.

Like I say, I'm open to suggestion as to where to draw the line - it's a compromise between disturbing the installation unnecessarily, and what's most likely to detect the most likely faults.

 

[ban-all-sheds]

You reckon they can take 240V AC but not 250V DC ? That's the reason for not <doing a proper test>.

I want to minimise disturbance and the risk of introducing a fault....IMO a full EIC would involve such a level of disruption that the risk of introducing a fault is more than the risk of not finding a fault that's subtle enough not to be detected by the abbreviated test.

If the installation is so flaky that the "disturbance" from a proper test is going to cause faults then forget all the testing and get the whole b****y lot rewired before the next tenant goes in.

 

[SimonH2]

But you won't know it's OK.

330V is not what BS 7671 requires - for LV circuits up to 500V you must test at 500V DC.

True, I won't know that it's good for 500V, but then I would know that it's good for 330V which is the peak of what it would normally be exposed to. But, the key things is that it would be the lowest risk test by virtue of not having to disconnect (IIRC) 8 circuits at the CU, plus at least 4 (again IIRC) items from different circuits.

So, what's the chances of modern (<20 year old) PVC cabling deteriorating to the point that it will fail a 500V test but not fail a 330V test ? Quite low, I'd think vanishingly small.
What's the risk of removing multiple items from circuits so the circuit can be tested at 500V ? Still quite low assuming whoever does it isn't a numpty and takes care to put everything back together properly - but I reckon considerably higher than above.

Of course, there's always the option of taking a quick look round, pressing the test buttons on the RCDs, and leaving it at that.

 

[SimonH2]

If the installation is so flaky that the "disturbance" from a proper test is going to cause faults then forget all the testing and get the whole b****y lot rewired before the next tenant goes in

You reckon that you can do all the disconnecting required, and guarantee 100% not to introduce a fault ?
You've got to remove all the RCBO earths or disconnect all the circuits from the RCBOs. Can you guarantee that (say) an earth won't slip out, or a neutral won't quick be properly in the awkward terminal, without you noticing ? Tiny risk, but not zero.

 

[EFLImpudence]

Just to check, Zs = PFC measured at the outlet

Zs = PEFC
Zn = PSCC

Higher of these = PFC.

 

[securespark]

You mentioned Ze but not Zs's anywhere?

 

[ban-all-sheds]

But, the key things is that it would be the lowest risk test by virtue of not having to disconnect (IIRC) 8 circuits at the CU, plus at least 4 (again IIRC) items from different circuits.

Why do you keep going on about risks? There shouldn't be any.

What's the risk of removing multiple items from circuits so the circuit can be tested at 500V ? Still quite low assuming whoever does it isn't a numpty and takes care to put everything back together properly - but I reckon considerably higher than above.

Ah - is that why you don't want to disconnect things?

I thought you were worried about the risks from someone who wasn't a numpty disconnecting and reconnecting.

Of course, there's always the option of taking a quick look round, pressing the test buttons on the RCDs, and leaving it at that.

Leaving it at that as far as your involvement is concerned, OK.

You'll still be getting a proper electrician to do a proper test, as you should, surely?

 

[ban-all-sheds]

Zs = PEFC
Zn = PSCC

Zs/Zn and PEFC/PSCC have different dimensions....

 

[ban-all-sheds]

You reckon that you can do all the disconnecting required, and guarantee 100% not to introduce a fault ?

I would hope so.

But I'm not the one doing the guideline tests on a rented property at change of occupancy.

 

[JohnW2]

Zs = PEFC
Zn = PSCC

Zs/Zn and PEFC/PSCC have different dimensions....

Indeed they do, but only because they are related by a constant which has dimensions (volts) - so it's a pretty trivial issue.

Kind Regards, John.