Old wife's tail, myths, and other stories we were told as an apprentice, before the days when BS 7671 started.

[SUNRAY]

Really? I was heard it just would not do, because of the electrical unreliability, of the joints. I half remember that it became customary, for the threaded conduit, to include a separate earth wire.

The earth connexion is either a copper clamp or attached to one of the fitting clamp screw.

About 30 years ago the house got flooded and part of the insurance claim was a rewire where the rubber/cotton wiring developed problems, all they did was replace with pvc singles, I was surprised they didn't include earth conductors.

In 1965ish we added a detatched garage and that was also wired that way although I believe that has been completely rewired now.

 

[JohnW2]

..., I know I work out volt drop using the loop impedance,

How do you do that? What is relevant for in an installation (and what BS7671 talks about as 'voltage drop') ids the drop from the origin of the installation, whereas if you measure L-N loop impedance that would tell you the voltage drop (for a given current) from the DNOO's transformer.

... so what 'loop impedance' are you talking about, I wonder?

I did not realise this is in the regulations under inspection and testing, what seems strange is one needs to check the volt drop, but there is no place in the standard form to record the volt drop.

There would surely be no real point in having "a place for recording voltage drop" (on a Schedule of Test Results, I presume you mean?), since BS7671 does not include any regulations which impose any requirement as to what the VD should be - it merely gives 'guidance' figures (not requirements) in an Informative Appendix.

 

[SUNRAY]

Circuit design is basically to provide a working voltage between certain limits at the point of use. In many cases the DNO loop impedance is the biggest part of the VD in the system.
It's no good measuring your supply as 250V then designing your circuit for a VD of 30 if the supply is only 220V at FLC

 

[JohnW2]

Circuit design is basically to provide a working voltage between certain limits at the point of use. In many cases the DNO loop impedance is the biggest part of the VD in the system.

That's the common sense (hence, I would say, 'how it should be') but, as I always say, it really makes nonsense of BS7671's concentration on "voltage drop from the origin of the installation". ...

It makes no sense to me that the guidelines in an Appendix of BS7671 suggest maximum VDs of 3% or 5% (from origin of installation) when there is such a wide variation in 'permitted' supply voltage at the origin. If one worked strictly to those guidelines, then if the supply voltage were 253 V, then a voltage 'at the point of use' less than 241.5 V would be considered 'unacceptable' (for a non-lighting circuit), whereas if the supply voltage (at origin) had been 216.2 V, then any voltage "at the point of use" greater than 204.7 V would be considered 'acceptable' !

[ I have taken the guideline figures of 3% and 5% as being percentages of the nominal voltage (230 V) ]

The practical problem is, of course, that although one can determine the supply voltage (at origin of installation) at one point in time, one cannot know what the supply voltage will be at other times.

 

[ebee]

I think the VD in the installation is a Rule of Thumb for , if you got your bit right it`ll be ok we think sort of back of a fag packet.
Some might mistakenly believe it to mean that DB1 (Consumer Unit) feeds DB 2, Feeds DB3 etc to mean that after DB5 then all the volt drops percentages are acceptable although common sense you that voltage is disappearing rapidly and might cause problems.
Good quick guide to most simple installation's that become a common myth rule for the daft to follow.
It happens in every walk of life, not just electrics.

One I got from an industrial electrician a few years back.
"Why use all this test gear to test an RCD? It`s only got to work 5 times to pass so if it is used a sixth time it might not work but it still passes the tests?"

I think he got the x5 element of personal protection value of RCDs slightly out of context.

Another for a spur on a ring. You need 2 earths on the spur because the sockets on the ring are 2 x 1.5 = 3,0 mm so your spur earths will only be 1.5 therefore too small (now if he was thinking about some of the 2.5 T & E that was allowed a 1.0 erathwire at the time then he might have had some point on a 30A BS3036 ring under the right (well wrong) conditions.

I wish I had a tenner for every time someone told me that a plugtop fuse, when blown, could be repaired by rolling some silver paper around the fuse, stating "A good Electrician told me that.

Or to double a couple of strands of 15A fusewire to make a 30A fusewire. Or to have 2 x 15A fusewires from different fuseways connecting a ring.
Hmmm I would not like to work out what currents might indeed flow under different conditions.

Or indeed, how does an ELCB (or a RCD) limit the voltage to 50V?
No it does not, it merely limits the duration of faults above the 50V plus ??? limit being sustained indefinitely

The (now old) push button wylex breakers will save your life but a BS3036 fuse will not!

 

[JohnW2]

I think the VD in the installation is a Rule of Thumb for , if you got your bit right it`ll be ok we think sort of back of a fag packet.
Some might mistakenly believe it to mean that DB1 (Consumer Unit) feeds DB 2, Feeds DB3 etc to mean that after DB5 then all the volt drops percentages are acceptable although common sense you that voltage is disappearing rapidly and might cause problems.
Good quick guide to most simple installation's that become a common myth rule for the daft to follow.

Agreed, but for reasons I've explained, I think that even the 'rule-of-thumb' is pretty daft/useless in this case - simply because there is so much permitted variation in supply voltage

As I wrote, it makes no sense to me that in one installation a voltage 'at the point of use' of, say, 241V, would be 'unacceptable' in terms of the rule-of-thumb, yet in another installation a voltage of 205 V 'at the point of use' would be 'acceptable'

Kind Regards, John

 

[Harry Bloomfield]

Agreed, but for reasons I've explained, I think that even the 'rule-of-thumb' is pretty daft/useless in this case - simply because there is so much permitted variation in supply voltage

As I wrote, it makes no sense to me that in one installation a voltage 'at the point of use' of, say, 241V, would be 'unacceptable' in terms of the rule-of-thumb, yet in another installation a voltage of 205 V 'at the point of use' would be 'acceptable'

I once installed some big automatic pumps, at the far end of a long 11Kv supply, daisy-chained from a tractor factory. When the kicked in, they would bring the heavy machining work to a stop at the factory. They eventually had to install a separate supply.

 

[JohnW2]

I once installed some big automatic pumps, at the far end of a long 11Kv supply, daisy-chained from a tractor factory. When the kicked in, they would bring the heavy machining work to a stop at the factory. They eventually had to install a separate supply.

That makes sense - and emphasises that what matters is the voltage supplied 'at the point of use', regardless of how much 'voltage drop' there has been relative to some point in the suppy path!

 

[ericmark]

Many years ago I was doing some PIR's now called EICR's and I was worried I would miss some thing which was latter picked up on, and one items was the volt drop in the installation, which had just changed, it was 4% and then became 3% lighting and 5% for the rest, I knew the volt drop could be calculated from the line - neutral loop impedance or prospective short circuit current, however the calculations seemed rather complex, with corrected mV/A/m and it used square roots to work out the correction factor Ct. So I sat down an made a java script program so I could run some known figures through it to see if there were any errors.

Try as I did, I could not get the 106 meters of 2.5 mm² I knew was permitted for a ring final, so at an IET meeting (I was a member back then) I asked, and was told we consider 20 amp at centre of ring final, and 12 amp even distributed for the design current for circuit Ib, once I started using 26 amp all dropped into place after @JohnW2 and @EFLImpudence identified some blatant mistakes, and I have used the program ever since to work out if there is room to add to a ring final. It looks like this.

The new PC seems to hide some of the figures, but for a ring final 0.35Ω incomer the mid point of the ring should show 0.94Ω which shows a 5% volt drop (11.5 volt) using 2.5mm² cable at a length of 106 meters, with a design current of 26 amp, using installation method 100 with a max temp of 70ºC. It is that long ago when I wrote the program I can't remember the java script used.

I am sure today there are on line programs to do the same, however once written it became clear a loop impedance tester is really not accurate enough to show there has been an error, unless rather blatant, if one changes the reading from 0.35Ω to 0.34Ω and 0.94Ω to 0.95Ω the results show 11.85 volt drop, and reverse 11.07 volt so one digit out and we have a 0.8 volt error. Take a loop impedance meter and repeat the tests and one can expect to see 0.02Ω variation so the results can easy be 1.6 volt out, so before one could say an error had been made the volt drop would have to have been well over the permitted 11.5 volt.

Since the standard forms do not record the Line - Neutral loop impedance or PSCC it is impossible to say if any error is down to the designer making an error. Or some thing which has changed since the original design.

So

612.14 Verification of voltage drop
Where required to verify compliance with Section 525. the following, options may be used:
(i) The voltage drop may be evaluated by measuring the circuit impedance
(ii) The voltage drop may be evaluated by using calculations, for example, by diagrams or graphs showing maximum cable length v load current for different conductor cross-sectional areas with different percentage voltage drops for specific nominal voltages, conductor temperatures and wiring systems.
NOTE: Verification of voltage drop is not normally required during, initial verification.

may be in the regulations, however it would have to be well out of spec before one could say there was an error in either the installation or the design.

However until I had designed and made the program to show volt drop using the line - neutral loop impedance I did not realise how hard it was to evaluate volt drop from measuring the circuit impedance.

However before adding to a ring final I still try to find near centre of ring final and measure either the loop impedance or PSCC to see how close to the wind I am sailing.

As to @JohnW2 question, do you not remember evaluating my program all those years ago?

 

[JohnW2]

Try as I did, I could not get the 106 meters of 2.5 mm² I knew was permitted for a ring final, so at an IET meeting (I was a member back then) I asked, and was told we consider 20 amp at centre of ring final, and 12 amp even distributed for the design current for circuit Ib, once I started using 26 amp all dropped into place after @JohnW2 and @EFLImpudence identified some blatant mistakes, and I have used the program ever since to work out if there is room to add to a ring final. It looks like this.

I don't know (and/or can't remember) exactly how your program works, but, although it is entitled "Volt drop and Cable length from loop impedance Calculator", the calculation of voltage drop (within the installation, from the circuit's origin) is surely just based on the cable length and CSA of the live conductors, isn't it, with 'loop impedance' having nothing to do with the calculation?

There's really no way that one can (or would need to) determine that voltage drop from "earth fault" Loop Impedance (Zs) since, even if one subtracts Ze, one then theoretically gets (R1 + R2) of the circuit, whereas VD is dependent upon (R1 + Rn). Only in the case of 1 mm T+E is (R1+Rn) the same as (R1+R2), so it's only with that sized cable that one could estimate VD from Zs & Ze.

As to @JohnW2 question, do you not remember evaluating my program all those years ago?

To be honest, whilst I can certainly remember your showing us your program (at least, the results it produced) I ahve no recollection of having 'evaluated' it. I'll have a delve back and see if I can find something to refresh my memory!

Kind Regards, John

 

[SUNRAY]

... Or to have 2 x 15A fusewires from different fuseways connecting a ring...

Twas a very common arrangement back in the early days of converting 2 radials to a ring.

 

[ericmark]

The idea of having two 16 amp linked MCB's so that there was no chance of overload of one legs was some thing I considered, but when we went to RCBO's this idea was knocked in the head.

 

[plugwash]

I remember at my parents house we used to have a socket in the loft fed off the immersion heater circuit. This never caused a problem, the immersion heater was rarely used and the loft socket normally only powered a network switch and a TV signal booster.

When we had our extension built (sometime in the late 1990s) the electrican was insistent that a socket on the immersion circuit was not allowed, and in the end we ended up with a dedicated circuit just for the loft socket.

 

[Jurassicspark]

, and in the end we ended up with a dedicated circuit just for the loft socket.

That's a bit absurd. Couldn't he spur off a socket? He could have even got the supply from a lighting circuit if that was the only load.

 

[Harry Bloomfield]

When we had our extension built (sometime in the late 1990s) the electrican was insistent that a socket on the immersion circuit was not allowed, and in the end we ended up with a dedicated circuit just for the loft socket.

Ridiculous, was he desperate for extra work?