Spot The Fault

[ProDave]

I am not disagreeing with this.

I have said no such thing and did not intend to imply it either. I was merely saying that shouting things loudly does not make them fact.

I wasn't shouting.

THIS IS SHOUTING.

 

[apprentice87]

What have i started!!!

I think we can all agree that it is only sensible, regs or not, to earth cable armour, mettalic conduit or the likes at the supply end...

john...

 

[Nicky Tesla]

electricians often only earth the armour at the appliance end when that end is a metallic box.  It is easier (proDave) to just have it earthed at that end and not have to arse about with the plastic box with its structure shape getting in the way and messing around with stupid little nuts and bolts.

 

[apprentice87]

Electricians only earth the armour at the appliance end when they do not know what they are doing..

You are going to have to fit a gland to the end of the cable at the supply end anyway to hold it in the box. The earth is at the supply end, so what is so difficult about connecting the circuit CPC to the gland??

john...

 

[Nicky Tesla]

Experience will tell you that plastic boxes often have extra bits of plastic stiffeners that you have to cut out in order for you to be able to turn the stupid thin nuts that you get with the glands.  Also you need to nut and bolt a connection to the banjo by drilling and bolting through the plastic which also gives complications that metal boxes dont.

   The best invention for glands are "Earth Nuts".

There is also obviously the fact that there is no need to earth the plastic box

 

[ProDave]

Experience will tell you that plastic boxes often have extra bits of plastic stiffeners that you have to cut out in order for you to be able to turn the stupid thin nuts that you get with the glands.  Also you need to nut and bolt a connection to the banjo by drilling and bolting through the plastic which also gives complications that metal boxes dont.

   The best invention for glands are "Earth Nuts".

There is also obviously the fact that there is no need to earth the plastic box

Which is an argument for using a metal box.

But the originator managed to bolt the glands into a plastic box successfully, so it was only a small step from that to include the banjo's and appropriate nuts bolts, terminals and wire to earth them.  Only 2 earth straps needed.

 

[Andy™]

There is also obviously the fact that there is no need to earth the plastic box

try telling this idiot

 

[ProDave]

try telling this idiot

Italian Electrician? trained in a Spaghetti factory?

 

[apprentice87]

If the cable is severed and, as seems certain in this case, re-energised, there are 4 possible scenarios

1. live cable is touching cpc-> the RCD trips

2. live cable is touching neutral-> the mcb trips

3. live cable is touching un-earthed armour-> armour becomes live (assume mostly still enclosed by outer sheath)

4. live cable is not touching anything

With 3 & 4 the rcd should still provide protection if there is a fault current (puddle, floor, finger).

Hopefully the operative will realise that the machine(s) doesn't work and will disconnect the supply and investigate. 

If damage to the cables is likely, perhaps burying or re-routing might be considered. 

I did find a sort-of-relevant reg 523-100, which mentions solid bonding, but it is single core, therefore there is no internal cpc.

It does make sense to earth the armour at the supply just to provide an easier path for fault currents, but whether it is necessary is debatable, some might say that having a larger surface area of live conductors at the cut end allows for more chance of the fault currents being large enough to trip the rcd.

Just my thoughts on this.

Hi Rob,

No, that is incorrect, There is only ONE Zs limit, and that is the one appropriate for the OCPD that protects the circuit against overload and PSCC and PEFC. The RCD is there for two reasons, one, because it will trip at very low currents that, whilst they will kill you, would not result in the OCPD operating, and two, to limit touch voltage to 50V . This is why they say that the max Zs for a 30mA rcd is only 1667 ohms and NOT 7666 ohms, even though that would still operate the rcd.

An rcd provides additional protection, it is not the primary protective device... so there is only ONE Zs limit.

True, you CAN use an RCD as part of a TN system [if needed] to help meet disconnection times that would otherwise be difficult to achieve, but ANY circuit that has a Zs above about 200 needs looking at... You cannot say "I have an rcd, Zs is less than 1667, so that is ok."

An rcd does come into its own on a TT system as otherwise you would be struggling to meet disconnection times with anything but a small OCPD, but even then, you could certainly not have an Ra of 1000 ohms or anything silly like that.

Leaving aside earth loop impedance, nobody seems to care about measuring L/N loop impedance, but i always do, for two reasons. One, to check that the circuit complies from the point of view of volt drop, [how else you going to know] and secondly, to measure what the PSCC will be.

You need to know this for TWO reasons. Firstly, to make sure that the OCPD can break this magnitude of current, and secondly, and just as importantly, to make sure that in the event of a L/N fault, a simple "short circuit" that the OCPD will once again operate in the required time. You need to make sure it does, as otherwise, the cables could get hot enough to be damaged due to the fault current. An OCPD is not there just to protect against earth fault current..

john...

I am no expert on this sort of thing, but i have an idea them that are will soon be along!!!!!

 

[Andy™]

Italian Electrician? trained in a Spaghetti factory?

i done a full thread on this idiot a few years ago, re-wired the lighting circuit & CU change and bodged the lot. basically, condemned lighting circuit due to no earth. he then re-wired it in 3c&e, snipping the earths at every point... oh, and he wired the shower submain in 1.5 3C&E

 

[Rob_the_rich]

Thanks apprentice87

Surely the cpc in the cable ensures that the Zs limit for interruption of supply will not be exceeded in PEFC. If there is a fault current through the armour, then something has gone wrong. This is exactly why the rcd is needed for the additional protection, and it should work.

Or am I missing something?

 

[Andy™]

if something goes through the armour and into a line, then the fault current will go through the armour, but if its got to go all the way to the end, then back again through the earth, then it may be that Zs* is too high. Zs being not end of circuit, but end of circuit and all the way back to start...

 

[Rob_the_rich]

if something goes through the armour and into a line, then the fault current will go through the armour, but if its got to go all the way to the end, then back again through the earth, then it may be that Zs* is too high. Zs being not end of circuit, but end of circuit and all the way back to start...

Additional protection again.

Another point to note is that if the earth fault occurs close to the supply, there is a shorter R1 component.

 

[Andy™]

you wouldnt need to rely on 'additional protection' if you done the job correct in the first place

 

[Rob_the_rich]

you wouldnt need to rely on 'additional protection' if you done the job correct in the first place

Good point

Just saying

 

[steptoe]

Hi Rob,

No, that is incorrect, There is only ONE Zs limit, and that is the one appropriate for the OCPD that protects the circuit against overload and PSCC and PEFC. The RCD is there for two reasons, one, because it will trip at very low currents that, whilst they will kill you, would not result in the OCPD operating, and two, to limit touch voltage to 50V . This is why they say that the max Zs for a 30mA rcd is only 1667 ohms and NOT 7666 ohms, even though that would still operate the rcd.

An rcd provides additional protection, it is not the primary protective device... so there is only ONE Zs limit.

True, you CAN use an RCD as part of a TN system [if needed] to help meet disconnection times that would otherwise be difficult to achieve, but ANY circuit that has a Zs above about 200 needs looking at... You cannot say "I have an rcd, Zs is less than 1667, so that is ok."

An rcd does come into its own on a TT system as otherwise you would be struggling to meet disconnection times with anything but a small OCPD, but even then, you could certainly not have an Ra of 1000 ohms or anything silly like that.

Leaving aside earth loop impedance, nobody seems to care about measuring L/N loop impedance, but i always do, for two reasons. One, to check that the circuit complies from the point of view of volt drop, [how else you going to know] and secondly, to measure what the PSCC will be.

You need to know this for TWO reasons. Firstly, to make sure that the OCPD can break this magnitude of current, and secondly, and just as importantly, to make sure that in the event of a L/N fault, a simple "short circuit" that the OCPD will once again operate in the required time. You need to make sure it does, as otherwise, the cables could get hot enough to be damaged due to the fault current. An OCPD is not there just to protect against earth fault current..

john...

I am no expert on this sort of thing, but i have an idea them that are will soon be along!!!!!

thats not strictly true John,

yet another miscocenption from a misinformed author

Ive worked on some installs where an Ra [Ze if you must] would be nigh impossible to get under 300 never mind the holy 200 ohms

 

[apprentice87]

Hi Steps, Yes that is fair enough, you would not think; "201 ohms" cannot be having that" But it would be nice to investigate i would think, and if at the end of the day, it was higher than that, what can you do, there is only so much you CAN do, but it would be worth trying to go deeper with the rod if possible. Having said all that, as you correctly say, the exact figure does not actually matter too much, what does matter is trying to make sure that the figure you DO achieve is stable. Deeper is better!!!! Makes you laugh though when you read of people trying to pee on the rod or otherwise wet the ground just to get a "better" figure to put on their cert.

john..

 

[Steve3948]

Hi Steps, Yes that is fair enough, you would not think; "201 ohms" cannot be having that" But it would be nice to investigate i would think, and if at the end of the day, it was higher than that, what can you do, there is only so much you CAN do, but it would be worth trying to go deeper with the rod if possible. Having said all that, as you correctly say, the exact figure does not actually matter too much, what does matter is trying to make sure that the figure you DO achieve is stable. Deeper is better!!!! Makes you laugh though when you read of people trying to pee on the rod or otherwise wet the ground just to get a "better" figure to put on their cert.

john..

John,

The reading on the rod will change with climate conditions, and deeper is not always better, as you may just be going into rock bed or gravel, so a mat, tape or more rods would be a better option.

 

[Andy™]

must be lucky here, even with a single rod half way in, its rarely over 100. fully in is usually 25-70

 

[apprentice87]

As i said, [and is confirmed in the info provided by Canoe. page 37] deeper is better. According to megger, twice the length of rod "generally" gives about a 40% reduction.. Also, deeper is more stable as it is less affected by climatic conditions, but agreed, sometimes you might not be able to go deeper, might hit rock, so yes, mats or more rods would be the way to go then, but LOADS more hassle!!!

john...