Low risk or to BS7671 for DIY?

[ericmark]

On this thread we have started to talk about what level of inspection and testing is required, expected, when doing DIY work.

For rented or where there are small children or others with reduced mental capacity there is likely little question we all should follow BS7671.

However if we all follow BS7671 then no point in this forum, so looking at what can be missed with inspection and testing. It seems crazy to me that testers like the old EZ150

only show better than around 1.6Ω when the ring final needs around 1.36Ω to pass, but they will work with radial and FCU and with RCD protection is it really that important.

But we may get the DIY man to spend £75 on test gear, but not £750, so what does the team think is the minimum testing, where if it all goes wrong, a court would say due care was taken, and it was an unforeseen accident.

 

[bernardgreen]

Back to basics

Use a 12 volt battery and a 10 Ohm 1% resistor to force current through the circuit whose resistance is to be measured.

Measure two voltages, (1) across the 10 Ohm resistor and (2) across the circuit being measured.

V(1) = I x 10 ( where I is the current in amps )

V(2) = I x R ( where R is the resistance of the circuit )

I = V(1) / 10

which reduces down to R = 10 x V(2) / V(1)

With about 1 amp flowing through the circuit being tested any stray voltages induced by nearby cables are swamped.

 

[JohnW2]

Back to basics ... Use a 12 volt battery and a 10 Ohm 1% resistor to force current through the circuit whose resistance is to be measured. ... Measure two voltages, (1) across the 10 Ohm resistor and (2) across the circuit being measured.

That's fair enough for measuring the resistance of a 'dead' cable or component(s), but eric is talking about measurement of fault loop impedances, which include the (always energised) path from the installation back to the DNOs transformer!

Kind Regards, John

 

[ajohn]

Use a 12 volt battery and a 10 Ohm 1% resistor to force current through the circuit whose resistance is to be measured.

Power is I^2*R so 14.4w. Way too much for a normal resistor. If it wasn't going to get rather hot 50w wouldn't be a bad idea.

 

[JohnW2]

Power is I^2*R so 14.4w. Way too much for a normal resistor. If it wasn't going to get rather hot 50w wouldn't be a bad idea.

All true - but, as I said, what bernard suggested is not relevant as a means of measuring EFLI (given that the loop is inevitably powered by 230V, such an approach as that would presumable require a resistor of 1-2Ω with a power rating of something like 50 kW )

Kind Regards, John

 

[bernardgreen]

if total resistance Rt is 20 Ohms (10 + 10 in circuit being measured ) then total power dissipated is V² / Rt 12² / 20 = 7.2 watts shared equally between circuit being tested and the reference resistor. 3.6 watts in each.

https://uk.rs-online.com/web/p/panel-mount-fixed-resistors/8932832/

 

[JohnW2]

if total resistance Rt is 20 Ohms (10 + 10 in circuit being measured ) then total power dissipated is V² / Rt 12² / 20 = 7.2 watts shared equally between circuit being tested and the reference resistor. 3.6 watts in each.

Quite apart from the main point I made, you talked of "about 1A flowing through the circuit", which would imply that the resistance being measured was about 2Ω and the power dissipated in your 10Ω resistor about 10W.

Kind Regards, John

 

[bernardgreen]

Sorry but what I posted was a general approach to measurement using basic principles. The fine details can be adjusted to suit the actual situation

With about 1 amp flowing through the circuit being tested

instead of the low ( milliAmps ) current that is often the test current "fully automatic" test equipment.

 

[ajohn]

if total resistance Rt is 20 Ohms (10 + 10 in circuit being measured ) then total power dissipated is V² / Rt 12² / 20 = 7.2 watts shared equally between circuit being tested and the reference resistor. 3.6 watts in each.

https://uk.rs-online.com/web/p/panel-mount-fixed-resistors/8932832/

Power ratingRating is based on chassis mounting area and temperature stability. Proper heat sink as follows: 5W and 10W units, 4” x 6” x 2” x .040” Aluminum chassis; 25W units, 5” x 7” x 2” x .040” Aluminum chassis; 50W units, 12” x 12” x .059” Aluminum panel.

Expect it to get to maybe 80C or more from 25C at 5w mounted as above. That's assuming it's an industrial rating. Commercial might rate it at 10w.

 

[JohnW2]

Sorry but what I posted was a general approach to measurement using basic principles.

I obviously realised that but, as I implied, I don't really understand why you posted it in response to erics's post about measuring ELLI.

The fine details can be adjusted to suit the actual situation

As I've also implied, you would have to change a lot more than 'fine details' to make your method 'suit the actual situation' eric was talking about - which was measurement of loop impedance in a circuit inevitably permanently powered by ~230V AC

Kind Regards, John

 

[ericmark]

Being fair the old BS7671 did say enquiry was a valid method of getting the in coming loop impedance, so yes one could use a low ohm meter to measure the local impedance I think it states some thing like 200 mA must flow, so not a standard multi meter, my old mega does both insulation and low ohms, but the latter meter is insulation only.

A continuity test shall be made. It is recommended that the test be carried out with a supply having a no-load voltage between 4 V and 24 V, d.c. or a.c., and a short-circuit current of not less than 200 mA.

A quick google for an insulation tester and they start at around £35 remember for the DIY not worried about calibration, but to read resistance or impedance to two decimal points using 200 mA finding a cheap meter I have found hard. And do you really think a DIY guy could rig up a battery with a low internal resistance or voltmeter and ammeter to push 200 mA down a cable to measure the resistance?

Yes as electricians we can make up a rig to measure resistance or impedance, better to measure impedance at 50 Hz, but we are not talking about electricians they will likely have all the calibrated meters they need, what I am talking about is the DIY chap who clearly is not fully conversant in what is required or he would not be asking on this forum.

So first the big question, should you tell him to go into a consumer unit which can not be isolated else where?

My maths was not to the standard required for level 6, so I went to do an 'A' level to brush up on my maths, one course was around £120 and three courses were £10, so I had to do another 2 courses to get the price down, one I selected was Physics 'A' so in with a class of students who have all passed 'O' level or GCSE as it is called now, and we are given multi-meters to do an exercise, around 20 in the class split into pairs, so 10 multi-meters, and 5 had the fuses blown by end of lesson where they had selected the wrong range.

'A' level stands for advanced level, and 5 fuses blown, so what do you think the chances are to get some one to follow instructions safely on a forum?

So plug in a EZ150 fair enough, not much they can get wrong, but play with a multi-meter, are you sure, I know I bought my son a clamp on meter as it did not have a wired amp range, and when he needed a wired amp range I lent him my AVO Mk8 because I know it has a cut out. Now it is very different he is likely a better electrician than me, but at 14 years old when he passed his RAE I was not very happy with him playing with mains power. I did fit RCD protection to all circuits, but still careful in the meters I got him.

So back to question, what should you tell a DIY guy to do, and what is a reasonable test set that he can use which will reduce danger rather than increase it?

 

[JohnW2]

So back to question, what should you tell a DIY guy to do, and what is a reasonable test set that he can use which will reduce danger rather than increase it?

We've been over this many times, and there is no simple answer, particularly when the advice is being given 'in public'. It's not as bad as it used to be, but there are still some people around who could be critical if we 'advised' a DIYer to do things which were not strictly compliant with the regs (or even the law), yet if we limit our replies to say things which are totally compliant with the regs (and the law) then, as you say, we would very often effectively be telling almost everyone not to do any DIY electrical work - which would make nonsense of this ('DIY Electrics') forum.

Essentially, I agree with what seems to be the implication of the title of this thread - i.e. that it is better for someone to be given sensible advice as to how to do things pretty safely than to imply that they should not be doing it (hence very probably will do it without any advice). However, as above, doing that 'in public' can sometimes result in problems, so I personally quite often resort to giving 'advice' privately by PM!

Kind Regards, John

 

[ericmark]

I agree with @JohnW2 being in the public domain is a problem as some one other that the starter of the thread may try to follow the guidance but miss a vital bit. However a PM also has problems, as if you make an error, no one else can correct your error.

 

[ajohn]

I'd say an insulation tester is a must to check existing cables and a continuity tester to check for breaks. Rather than cable resistances use the site guide which gives max cable length tables that can be used with and without an rcd or rcbo. Without RCD is a bit pointless when one is fitted. Ze gets important when there isn't one.

Insulation testing is fun. Unplug / disconnect everything, remove bulbs etc and switch the lights on. The chinese tester is ok.

Doing it that way assumes that the main supply will be within a certain resistance range. Only way around that is to measure it. I'm not sure if any of the cheap testers will measure that but there are some for measuring Ze. When an rcd is fitted that value is not so important. Perhaps the ability to light a 240v bulb between L and E would do for a very rough check that it can take current. Ours is way out of spec, 25R but would still light a bulb. Say a 60w bulb as it's takes about 1/4 amp.

The biggest problem using a multimeter is crap probes and connections. They tend to be pretty problem free on expensive meters.

 

[JohnW2]

I agree with @JohnW2 being in the public domain is a problem as some one other that the starter of the thread may try to follow the guidance but miss a vital bit. However a PM also has problems, as if you make an error, no one else can correct your error.

Yes, that's an issue, given that we are all capable of mistakes/errors - so, when 'in private', one has to be as careful as one can not to make mistakes. I've spent much of my professional life 'giving advice/opinions' in a manner which is often essentially 'private' (i.e. with no-one 'checking' what I have said), so I suppose I have learned to be very careful, in the knowledge that I am solely responsible for what I have said/written!

Having said that, it's commonly not the actual technical advice that gets 'given' by PM. It's much more often something like "Although what XYZ has said about the regulations is strictly true, if I were in your position I would be happy to do as you have proposed" !

Kind Regards, John