Should Part P get scrapped?

[JohnW2]

610.4 ... 631.1 ... 631.3 ... 633.1 ... So, John, in answer to your questions:
Yes, you would be required to undertake testing in your scenario ....

Thanks. However, that doesn't really answer my questions, because you have answered in terms of the requirements of BS7671, whereas my question was about the requirements of Part P of the Building Regs. As we all know, Part B does not specifically require compliance with BS7671. It only requires that those "reasonable provisions" be made for safety - and, whilst compliance with BS7671 might be taken as one way of demonstrating that such "reasonable provisions" had been made, nothing in the legislation says that this is the only way to comply.

Nevetheless, you have effectively confirmed what I was saying - that many/most electricians would regard reasonably extensive testing as being necessary (for compliance with Part P) even following minor non-notifiable work. As you go on to say, that effectively means that, with a very small number of exceptions (like, I have to say, myself!), the great majority of non-notifiable DIY work is probably unlawful.

...- as a very minimum:
CPC continuity and polarity
Insulation Resistance
Earth Fault Loop Impedance
RCD operation (if the altered circuit is covered by an RCD)
Continuity and adequacy of installation earthing and bonding.

That raises another interesting question - to what extent should one have to seek pre-existing faults when making additions/changes to a circuit? If I take my car to a garage to have its brake pads replaced (clearly part of a safety-critical system), I doubt (and would not expect) that they would undertake inspection and testing of the remainder cof the braking system - and therefore that they could/would hand it back to me complete with any pre-existing faults in the hydraulic system (unless they noticed them by accident). With the added socket, provided that CPC continuity and polarity were correct (and ring continuity satisfactory if a ring has been 'interfered with'), it is all but certain that the resulting circuit will be no less satisfactory/safe than it was before the work was done, even if there were persisting (pre-existing) problems in relation to EFLI, RCDs, earthing/bonding etc., or even IR.

Kind Regards, John.

 

[JohnW2]

Is it more a case of we can test things now that was more difficult years ago therefore we must?

I'm sure that is a a very significant factor - and one which one sees operating in a vast range of fields. In all fields, the problem is that one can never argue with a suggestion that 'testing everying possible every day' will pick up more faults, and/or pick them up earlier. However, one obviously has to temper that with common sense, in relation to how probable it is that one will pick up faults, and how much of a hazard would be created by delayed detection (or non-detection) of such faults. The mere fact that one can undertake a test does not necessarily mean that it is necessary/sensible to do it at every opportunity.

Medicine provides very good examples. We know that early detection of many diseases (like cancers) is desirable, and can lead toi better outcomes. To screen everyone every week for every disease we can screen for would therefore have 'positive' value - but we also have to consider the downsides and (im)practicalities of such an approach!

Kind Regards, John.

 

[Electrifying]

John,

You asked the questions of electricians on the forum.......I gave you an electricians answer.

Regardless of Part P and the building regs, we work to BS7671 (or an equivalent standard)
BS7671 requires that ALL electrical work be inspected and tested to prove that it satisfies the BS7671 regulations - and then one of two certificates completed.

Compliance with the Building Regs is a seperate issue, and of course we are required to do this by law - just as we have to comply with the EAWR.

But you asked if Electricians thought that testing was required on small additions/alterations - the answer is yes, because they are the regs we work to.

An electrician could hardly answer - "No, I don't think extensive testing is required on minor electrical work in order to comply with Part P of the building regs", could he??

He's working to a standard (BS7671) which incorporates (amongst other things), the requirement to prove that an electrical installation is safe to put into service........making the above statement would contravene that.


In answer to your query on existing faults, like I said earlier - read Section 2 of App. Doc P:

 

[JohnW2]

John, You asked the questions of electricians on the forum.......I gave you an electricians answer. Regardless of Part P and the building regs, we work to BS7671 (or an equivalent standard). BS7671 requires that ALL electrical work be inspected and tested to prove that it satisfies the BS7671 regulations - and then one of two certificates completed. Compliance with the Building Regs is a seperate issue, and of course we are required to do this by law - just as we have to comply with the EAWR. But you asked if Electricians thought that testing was required on small additions/alterations - the answer is yes, because they are the regs we work to.

I don't disagree with any of that but, to perhaps be pedantic, it was not an answer to the actual question I asked of electricians, which was:

How about a survey for the electricians here? If, say, you extended an existing socket or lighting circuit (not in special location etc.), would you feel that in order to comply with Part P's requirement for "reasonable provisions to be made ... to avoid fire or injury", that you had to undertake any testing - and, if so, what? - and, hence, whether you feel that a 'standard DIYer' (without any appreciable testing knowledge/facilities) undertaking the same work (without any specialised testing) would be compliant with the requirements of Part P. ??

An electrician could hardly answer - "No, I don't think extensive testing is required on minor electrical work in order to comply with Part P of the building regs", could he??

Well, (s)he might if (s)he was answering a question that was specifically about what was required in order to comply with Part B of the Building Regs!

He's working to a standard (BS7671) which incorporates (amongst other things), the requirement to prove that an electrical installation is safe to put into service........making the above statement would contravene that.

Again, the clue is in the wording of my question. I was not asking what the electrician would do in order to comply with the regulatoions to which (s)he was working but, rather, what would be necessary to comply with Part P. What you haven't said, but I would have expected you to say, is that, even without considering BS7671, the requirements of Part P for "reasonable provisions for safety" probably do require some testing which is almost certainly not undertaken by most DIYers.

However, as I said, I'm not arguing with any of this. The answers you have given are as I would have expected and confirm the view that most DIY non-notifiable work almost certainly does not comply with Part P, since (whether one invokes BS7671 or not), the 'reasonable provision for safety' required by Part P probably is not satisfied without at least some testing which the vast majority of DIYers can't or don't undertake.

Kind Regards, John.

 

[maltaron]

Perhaps I could add to this discussion with my recent dealings with the LABC. I would point out here that whilst I do not have the piece of paper I am fully competent, know most of the requirements of the 17th and have necessary test gear. (being an old codger and there were not grandfather rights for Pt P compliance)
However, the story. I recently had an extension built to my bungalow and put everything including electrics through building regs, with the intention of electrics being done by a qualified and registered spark. When the time came for first fix the spark eventually turned up for a meeting after 3 phone calls, said he would start on Wednesday morning, he turned up on Wednesday PM, did 2 hours work and said he would be back tomorrow. He never turned up and after 2 days I e-mailed him not to bother again. I the mean time I asked the building inspector if I could do this myself and the reply was “yes OK, ring me after first fix and our electrician will check it, no extra charge as it is included in the full regs charge”. I first fixed over the weekend and called the inspector on Monday morning, the council spark arrived Monday PM, asked what I had done, looked at nothing other than cable capping and said call me when you have finished. I asked if he wanted to test before I connected the new circuits. The answer – no, when everything is working. I tested all my new circuits and powered up. All was OK so I called for final test, the council spark returned, plugged in his tester and confirmed OK, checked one switch and said the certificate would be sent to the LABC and copied to me on completion of the job. Sorry for the long post but this shows that Part P, whilst necessary, is useless as it stands. The LABC electrician knew nothing of my competence (or lack of it) but signed the work off with rudimentary testing and little visual inspection

 

[JohnW2]

Sorry for the long post but this shows that Part P, whilst necessary, is useless as it stands. The LABC electrician knew nothing of my competence (or lack of it) but signed the work off with rudimentary testing and little visual inspection

That has always been my suspicion, and I've heard many similar stories. As you say, Part P as it is currently implemented is often not just useless, but actually dangerous, since it could easily result in incompetent work coming with the paperwork of an 'official blessing'.

Kind Regards, John.

 

[JohnW2]

In answer to your query on existing faults, like I said earlier - read Section 2 of App. Doc P:

I forgot to respond to this bit....

I wasn't saying that regulations/legislation/guidance documents do not require/suggest testing to detect pre-existing faults afecting a circuit being extended - but, rather, in a more philisophical sense, was saying that there was a question as to how far such regulations/legislation/guidance should go in requiring such testing. In terms of analogies, should someone replacing brake pads be required to inspect and test the entire hydraulic braking system, or should someone being treated for a minor everyday infection be subjected to a battery of tests to determine whether they might possibly have some underlying disease which predisposed them to the infection?

Since you've brought Approved Document P into the equation, in relation to the current discussion it is perhaps worth reminding people that it's pre-amble says:

Thus there is no obligation to adopt any particular solution contained in an Approved Document if you prefer to meet the relevant requirement in some other way.

...and that section 0.2 (General Guidance) says:

A way of satisfying the fundamental requirements would be to follow ... the technical rules described in the body of BS7671:2001

Kind Regards, John

 

[SimonH2]

Is it more a case of we can test things now that was more difficult years ago therefore we must?

I'm sure that is a a very significant factor - and one which one sees operating in a vast range of fields. In all fields, the problem is that one can never argue with a suggestion that 'testing everying possible every day' will pick up more faults, and/or pick them up earlier. However, one obviously has to temper that with common sense, in relation to how probable it is that one will pick up faults, and how much of a hazard would be created by delayed detection (or non-detection) of such faults. The mere fact that one can undertake a test does not necessarily mean that it is necessary/sensible to do it at every opportunity.

Not only that, but bear in mind that there is a positive risk in performing some (most ?) tests.

A good example is an RFC. To test it fully you must disturb some connections - so there is a risk that in performing the test, you may actually create a fault because you've disturbed some connections, flexed a solid wire or six, and then re-made the connections.

If you do IR tests, then you need to disconnect anything that may be adversely affected. Though I wonder if it might be better to just use a lower test voltage.

For testing the light circuits, the minimum disturbance is probably to remove the cover on a ceiling rose - with the risk that you'll damage it, or the personal risk that you'll fall off your steps.

Taking the medical analogy, there is a school of thought that some tests are better left undone. Either there is a high chance of false positive - which then involves the patient undergoing other (potentially intrusive and/or risky procedures/tests). Or the test itself may involve some risk - a classic one being where something may be detectable by X-ray, something to which exposure is normally minimised as much as possible because it has it's own health risks.

 

[acmedone]

Apart from basic wiring faults - screw terminals that don't grip the conductor (open circuit), shorted conductors, swapped conductors - which can be checked by inspection and simple instruments, I am willing to bet that the cases of faults (other than damp tracking) on a new installation only found by a stress test is vanishingly small. Even then, if such a fault is found, what are the long term problems if left uncorrected?

Is it more a case of we can test things now that was more difficult years ago therefore we must?

To carry on mine and John's example of adding a socket to a ring final circuit - a very simple job within itself.

So the DIYer taps into an existing socket, runs 6 metres of cable and installs his new double socket outlet.

He now has to get three 2.5mm conductors into two of the existing socket terminals - he squeezes them in and tightens as hard as he can to keep them all in place.

Not realising he's crushed one of the existing 'Line' conductor ends to the point of breaking, he forces the socket face back onto the back box...........snapping the 'line' conductor where it enters the terminal.

If he doesn't do a RFC continuity test and leaves this fault undetected, then you could have an overloaded leg of the ring final.........all over one tiny oversight and then not testing your work!!

That's just one example - all electrical work should be tested in compliance with the regulations - otherwise, why bother following the 'Regs' at all...........might as well just do as you please.

Of course it is possible to break a connection... but whenever I have changed socket covers it has always been obvious that a conductor is at breaking point. Even so, I doubt that in these days of central heating, most rings that do not serve the kitchen ever get loaded close to capacity so inadvertantly converting a ring to two spurs is unlikely to cause trouble. (Though it is still necessary to wonder about the seriousness of the situation of having the washing machine, dishwasher, toaster and kettle on a 20A cable with 30A fuse.) The main purpose of wiring once upon a time was a single socket in each room for a heater; with central heating now, it is to try and reduce a proliferation of extension leads for the multitude of electronic appliances we have (which are mostly class 2, anyway).

Here I am posting from a room with 2 single wall sockets - but it has 22 outlets. Three class 1 devices are in extensions off an extension. It's very difficult to get paranoid about the house wiring.

 

[JohnW2]

Not only that, but bear in mind that there is a positive risk in performing some (most ?) tests.

Indeed so, which is what I meant by 'downsides' in what I went on to say after the bit you quoted, namely:

Medicine provides very good examples. We know that early detection of many diseases (like cancers) is desirable, and can lead toi better outcomes. To screen everyone every week for every disease we can screen for would therefore have 'positive' value - but we also have to consider the downsides and (im)practicalities of such an approach!

A good example is an RFC. To test it fully you must disturb some connections - so there is a risk that in performing the test, you may actually create a fault because you've disturbed some connections, flexed a solid wire or six, and then re-made the connections.

Exactly. I've often expressed my concerns that any testing which involves disturbing the installation is inevitably going to result in the introduction some faults.

If you do IR tests, then you need to disconnect anything that may be adversely affected. Though I wonder if it might be better to just use a lower test voltage. ...For testing the light circuits, the minimum disturbance is probably to remove the cover on a ceiling rose - with the risk that you'll damage it, or the personal risk that you'll fall off your steps.

Agreed again!

Taking the medical analogy, there is a school of thought that some tests are better left undone. Either there is a high chance of false positive - which then involves the patient undergoing other (potentially intrusive and/or risky procedures/tests). Or the test itself may involve some risk - a classic one being where something may be detectable by X-ray, something to which exposure is normally minimised as much as possible because it has it's own health risks.

Exactly - that's precisely what I meant about 'downsides'. In addition to those you mention, there's also a potential problem of true 'positives' from screening tests which relate to very minor/early disease which, at least in the past, has led to detrimental and probably unnecessary treatment. An example of that was seen in the early days of screening for cervical cancer - which resulted in many women being subjected to surgery because of 'pre-cancer' ("Stage 0" cancer) which we now realise probably did not warrant such treatment. There has, of course, been a lot of technological advance in terms of non-harmful medical testing but it remains the case that vast numbers of people suffer at least some undesirable effects of medical tests - ranging from the trivial (bruising or discomfort as a result of blood tests) to death. I have certainly seen some such deaths in my time.

For such reasons, in medicine (at least, in the UK) there has been increasing encouragement that one should think, and only undertake testing if one feels that the potential benefits outweigh the potential risks - and I can't help but feel that there would be some wisdom in applying the same philosophy to electrical testing.

The benefit/risk balance obvioulsy moves in favour of not testing if one is essentially going to ignore the result of the test - for example, testing for the presence of diseases which one wouldn't treat even if found (e.g. because of the age, frailty or other diseases of the patient). In terms of electricity, I often wonder.for example, what purpose (other than generating a figure to write on a form!) is served by Zs measurements in a TT installation. They may, of course, be low enough to result in acceptable disconnection times as a result of bonding and parallel paths to earth. However, if they are not low enough, one will conclude that this is what one would expect in a TT installation, and simply record the result!

Kind Regards, John.

 

[EFLImpudence]

How many electricians do minor work like adding a socket and do not test the new cable other than a quick polarity and CPC continuity check.

Then they shouldn't be calling themselves electricians..

It would not normally be just the new cable that required testing.

If they extend and add a socket to, say, a Ring Final, then it would be necessary to do the full Ring Fnal Continuity Test.........what if they inadvertently cause a break in the existing Ring Final whilst adding their new cable and socket?........they would have no idea by just testing continuity on their new section of cable.

I've been wondering about this suggestion.

Obviously an electrician would have tested the Ring (before the addition) to ensure everything was in order.

Is it suggested that the Ring is tested again (after the addition)?
Apparently not trusting himself to reassemble the conductors correctly.

If so, I presume the best place would be at the board but that, surely, has inherent dangers of disturbing many things unnecessarily.
If done at another socket outlet then we are back to square one.

Is not an EFLI test sufficient to ensure a satisfactory conclusion to the work or is that what was meant?

 

[JohnW2]

Is it suggested that the Ring is tested again (after the addition)? Apparently not trusting himself to reassemble the conductors correctly.

It seems to me that precisely that is being suggested - and similarly after any work is undertaken.

If so, I presume the best place would be at the board but that, surely, has inherent dangers of disturbing many things unnecessarily. If done at another socket outlet then we are back to square one.

Exactly - per my recent message, I'm not at all convinced that the 'benefits' of such testing would outweigh the 'risks'

Is not an EFLI test sufficient to ensure a satisfactory conclusion to the work or is that what was meant?

That's certainly not how I read it - and, in any event, (again per my recent post) what if it is a TT installation?

Kind Regards, John.

 

[Electrifying]

So where do you want to draw the line??

We either follow the 'Regs' guidance and run the appropriate tests on any circuits that we alter - or we second guess the IEE (as some posters are doing), and decide for ourselves what we should or shouldn't test.

If we're going to do the latter, then why follow any of the 'regs'??

As I stated in my previous posts - other than straight-swapping accesories or luminaires, ALL electrical work is required to be inspected, tested and certificated by BS7671........no matter how minor or whether it's notifiable.

Going by some poster's suggestions, a DIYer could extend a downstairs Ring Final to incorporate 8 new double sockets upstairs - and only test Zs!!

Really?

If not, then where would you like to draw the line?? :

Adding one socket - no testing required

Adding two sockets - test Zs

Adding 3 sockets - best get R1 + R2

Adding 4 sockets - better check the supply characteristics as well

Adding 5 or more - better check the RCD functions now


Get real!!

 

[EFLImpudence]

To reiterate, in case you missed or misread my post,...

...i would, as a matter of course, test the Ring to ensure all was in order before adding any spurs, accessories etc.

My confusion was where to do any tests, after the work, without being in the same position after those tests.
There will obviously be a final part to be reassembled - how can it be assured that this has been done correctly without a further test somewhere else which then becomes a subsequent final part?

 

[JohnW2]

...i would, as a matter of course, test the Ring to ensure all was in order before adding any spurs, accessories etc.
My confusion was where to do any tests, after the work, without being in the same position after those tests.
There will obviously be a final part to be reassembled - how can it be assured that this has been done correctly without a further test somewhere else which then becomes a subsequent final part?

Exactly. As you say, it would become an endless process. Any test (like ring continuity tests - at least when undertaken in the usual way) which involves any disturbance of the circuit would create the need for a further test, which would create the need for a further test, which would create .... ad infinitum.

Kind Regards, John