Earth Loop Impedance Testing

[Anteaus]

Wire nuts were used on early UK installations, they were called 'Screwits' and were made of ceramic material. They were prone to falling off, no matter how well fitted. Maybe that's why they have a bad reputation here.

 

[PBC_1966]

So the real big question is why don't we loop test here in the USA. Some have said ,it's because it not in the NEC and the instrument manufactures are reluctant to launch the units here for that reason.There is no code that says we have use a multi meters , clamp meters to test voltage or amperage either. It would seem like a really good idea. Many time I have customers that have fires , floods . It would be an excellent way in my opinion to trouble GFCI problems .loop test to located the affected areas.To narrow it down we then then could use a megger to find the melted wires. The loop test would be a terrific trouble shooting tool for finding high impedance ,loose connection and to prevent fires , also takes us of the hook in a court of law that our work was tested at the time of installation .

Loop testing and insulation testing are excellent tools, in my opinion. Back in the 1950's/60's Megger used to publish a lot of little booklets to accompany its range of test meters explaining the benefits of such tests, and setting out how keeping records over a period of time could enable potential issues (deteriorating insulation in motor windings, for example) to be detected before they became full-fledged problems requiring immediate attention.

However the rule book which is not law is universal our BS7671 lays out what we should do with low voltage (below 1kV) and is regarded as being law by many.

The National Electrical Code here is not law in itself either, however just about all jurisdictions adopt it into their state or local building codes, thus requiring it to be followed (although obviously there is a lot of unpermitted work which can be as good or as bad as that found in Britain). The NEC is generally revised every 3 years or so, but does not automatically become the new standard when published, since each state has to adopt it formally. Some states seem to be quick at adopting the new editions, while others sometimes lag two or three editions behind. States can adopt the NEC as is, or can make amendments imposing stricter requirements or relaxing certain requirements. Counties and cities can also introduce local amendments, e.g. while NM-type cable (Romex - equivalent to British T&E) is generally permitted throughout normal residential construction the City of Chicago doesn't allow it, requiring everything to be run in conduit or equivalent.

Will the loop impedance tester trip on test a combination arc fault breaker? Like it will an RCD?

Yes, any combination AFCI/GFCI is liable to be tripped due to the earth current.

Will the tester produce a series arc or parallel arc to test the breaker is function
I have heard arc fault have some limits?

I've not had dealings with AFCI's personally, but I'm not struck on the whole idea. I remember debating with some of my American friends when they starting appearing widely, and many of them had concerns about their usefulness as well. I recall there had also been some tests done which showed they were pretty ineffective with certain types of arcing anyway (I can't remember if it was parallel arcing such as from a damaged cable with lines in contact or series arcing under load such as could happen at a loose connection). Leading back to the original point, I'd much rather employ good wiring practices backed up with thorough testing to ensure that arc faults are unlikely to happen anyway.

Wire nuts were used on early UK installations, they were called 'Screwits' and were made of ceramic material. They were prone to falling off, no matter how well fitted. Maybe that's why they have a bad reputation here.

My feelings are that wire nuts as used here are perfectly fine so long as the correct size is selected for the conductors to be joined and that they are installed correctly. The metal coil spring inside bites into the conductors slightly to give a good, tight connection. They've been in widespread use for over half a century now, so if there were any real concerns they would have been abandoned long ago. The problems come when somebody doesn't line up the conductors properly or tries to get by with one which is too large or too small for the conductors. Or simply doesn't tighten it down properly, of course!

What I would like to see scrapped are the horrible push-in terminations on some of the cheaper receptacles and switches. I'm sure that the few seconds they save on installation must be far outweighed by the time spent replacing them later when the connections start working loose.

 

[Bus bar]

I agree with the wire nut situation
If you install them correctly they make a good connection with coil spring inside. Most of the problems with lose connections are associated with those stab in connectors at the back of the outlets and a real problem with heating up and causing fire. I've been in the trade 30 years and that is one of the biggest problems I encounter the other being loose breaker jaws with certain brands. The loop test would be a great way to track these problems down safely where other sensitive load are connected

 

[PBC_1966]

Most of the problems with lose connections are associated with those stab in connectors at the back of the outlets

I really don't know why they're still being made given the trouble they cause. And how much longer does it take to form the stripped wire into a loop and tighten under a screw anyway - Maybe 10 seconds per terminal?

the other being loose breaker jaws with certain brands

Is there much FPE or Zinsco out your way?

 

[Bus bar]

So the real big question is why don't we loop test here in the USA. Some have said ,it's because it not in the NEC and the instrument manufactures are reluctant to launch the units here for that reason.There is no code that says we have use a multi meters , clamp meters to test voltage or amperage either. It would seem like a really good idea. Many time I have customers that have fires , floods . It would be an excellent way in my opinion to trouble GFCI problems .loop test to located the affected areas.To narrow it down we then then could use a megger to find the melted wires. The loop test would be a terrific trouble shooting tool for finding high impedance ,loose connection and to prevent fires , also takes us of the hook in a court of law that our work was tested at the time of installation .

Loop testing and insulation testing are excellent tools, in my opinion. Back in the 1950's/60's Megger used to publish a lot of little booklets to accompany its range of test meters explaining the benefits of such tests, and setting out how keeping records over a period of time could enable potential issues (deteriorating insulation in motor windings, for example) to be detected before they became full-fledged problems requiring immediate attention.

However the rule book which is not law is universal our BS7671 lays out what we should do with low voltage (below 1kV) and is regarded as being law by many.

The National Electrical Code here is not law in itself either, however just about all jurisdictions adopt it into their state or local building codes, thus requiring it to be followed (although obviously there is a lot of unpermitted work which can be as good or as bad as that found in Britain). The NEC is generally revised every 3 years or so, but does not automatically become the new standard when published, since each state has to adopt it formally. Some states seem to be quick at adopting the new editions, while others sometimes lag two or three editions behind. States can adopt the NEC as is, or can make amendments imposing stricter requirements or relaxing certain requirements. Counties and cities can also introduce local amendments, e.g. while NM-type cable (Romex - equivalent to British T&E) is generally permitted throughout normal residential construction the City of Chicago doesn't allow it, requiring everything to be run in conduit or equivalent.

We use mostly mc cable around here for commercial installs which is the same as running pipe. Some inspector allow nm cable if not over three floors with wood construction.

Will the loop impedance tester trip on test a combination arc fault breaker? Like it will an RCD?

Yes, any combination AFCI/GFCI is liable to be tripped due to the earth current.

Good to know! Went to a code seminar last week and arc fault Breakers came up
One electrician mentioned that the plug in tester he purchased over the counter from the supply house made for testing arc fault breaker was causing problems. It looks like one of those inexpesive polarity testers. The main concern was the $1,200 cost of installing these breakers on a new house that had a 200 amp 40 circuit panel. Theses combo arc faults cost approx $55 each time x 40 if you used all fourthy
The manufacturers are are pushing these things with the code making committees
For there own interest in my opinion when they could make a single arc fault main instead

Will the tester produce a series arc or parallel arc to test the breaker is function
I have heard arc fault have some limits?

I've not had dealings with AFCI's personally, but I'm not struck on the whole idea. I remember debating with some of my American friends when they starting appearing widely, and many of them had concerns about their usefulness as well. I recall there had also been some tests done which showed they were pretty ineffective with certain types of arcing anyway (I can't remember if it was parallel arcing such as from a damaged cable with lines in contact or series arcing under load such as could happen at a loose connection). Leading back to the original point, I'd much rather employ good wiring practices backed up with thorough testing to ensure that arc faults are unlikely to happen anyway.

Wire nuts were used on early UK installations, they were called 'Screwits' and were made of ceramic material. They were prone to falling off, no matter how well fitted. Maybe that's why they have a bad reputation here.

My feelings are that wire nuts as used here are perfectly fine so long as the correct size is selected for the conductors to be joined and that they are installed correctly. The metal coil spring inside bites into the conductors slightly to give a good, tight connection. They've been in widespread use for over half a century now, so if there were any real concerns they would have been abandoned long ago. The problems come when somebody doesn't line up the conductors properly or tries to get by with one which is too large or too small for the conductors. Or simply doesn't tighten it down properly, of course!

What I would like to see scrapped are the horrible push-in terminations on some of the cheaper receptacles and switches. I'm sure that the few seconds they save on installation must be far outweighed by the time spent replacing them later when the connections start working loose.

 

[PBC_1966]

The main concern was the $1,200 cost of installing these breakers on a new house that had a 200 amp 40 circuit panel. Theses combo arc faults cost approx $55 each time x 40 if you used all fourthy
The manufacturers are are pushing these things with the code making committees
For there own interest in my opinion when they could make a single arc fault main instead

That was the conclusion my friends arrived at some years ago - That there was little to recommend the devices but that they were being pushed into the NEC by the manufacturers. Some might feel that the same sort of thing is happening with BS7671 in the U.K.

 

[Bus bar]

The main concern was the $1,200 cost of installing these breakers on a new house that had a 200 amp 40 circuit panel. Theses combo arc faults cost approx $55 each time x 40 if you used all fourthy
The manufacturers are are pushing these things with the code making committees
For there own interest in my opinion when they could make a single arc fault main instead

That was the conclusion my friends arrived at some years ago - That there was little to recommend the devices but that they were being pushed into the NEC by the manufacturers. Some might feel that the same sort of thing is happening with BS7671 in the U.K.

Most of the problems with lose connections are associated with those stab in connectors at the back of the outlets

I really don't know why they're still being made given the trouble they cause. And how much longer does it take to form the stripped wire into a loop and tighten under a screw anyway - Maybe 10 seconds per terminal?

the other being loose breaker jaws with certain brands

Is there much FPE or Zinsco out your way?

Most of the problems with lose connections are associated with those stab in connectors at the back of the outlets

I really don't know why they're still being made given the trouble they cause. And how much longer does it take to form the stripped wire into a loop and tighten under a screw anyway - Maybe 10 seconds per terminal?

the other being loose breaker jaws with certain brands

Is there much FPE or Zinsco out your way?

P

Most of the problems with lose connections are associated with those stab in connectors at the back of the outlets

I really don't know why they're still being made given the trouble they cause. And how much longer does it take to form the stripped wire into a loop and tighten under a screw anyway - Maybe 10 seconds per terminal?

the other being loose breaker jaws with certain brands

Is there much FPE or Zinsco out your way?

There is a fair amount of FPE but now the home owner insurance companies make us take it out of doing an up grade
Not sure about zinsco around here I have'nt seen it

 

[ericmark]

The question is why do we measure loop impedance? It may be to insure the magnet part of the MCB will work, it could also be to ensure volt drop is within limits. The use of RCD or GFCI if in US means the loop impedance to earth has very little impact on the protection so in the main it is the volt drop which is important. Yet many testers only test line - earth and omit the line - neutral test. Some do test line - neutral for the PSSC but even that is dependent on which tester is used.

We enter readings onto paper work with little or no thought to what the reading actually mean. In the UK 0.94Ω is likely the limit due to volt drop with a ring final yet for years we took 1.44Ω now 1.37Ω as the limit. I am unaware of USA limits with 120 volt it must be a problem which makes UK problems look rather mild.

The problem for the electrician is if entering the results on paper work lays him open for future claims. We enter readings of better than 1.37Ω and assume we are OK. But the volt drop needs better than 0.94Ω line - neutral. OK we read line - earth and we could claim this does not relate to line - neutral however some maths nut can easy show the relation between line - earth and line - neutral readings and use our own readings to show the ring final circuit never complied with the regulations.

So our minor works or installation certificate could in the future land us with huge bills for remedial work which to put it mildly would bankrupt many firms. Much depends on what is entered on the certificate, but we should be aware readings outside the limits can come back and bite us.

We have seen years after the event how PPE has come back and caused the banks to pay out huge amounts, this could also happen to the electrical industry with massive bills because the loop impedance did not fall within limits.

So it's a case of watch my back and being careful paper work raised does not in the future land one with huge bills where it can be proved the installation never complied with reguations.

 

[Bus bar]

The main concern was the $1,200 cost of installing these breakers on a new house that had a 200 amp 40 circuit panel. Theses combo arc faults cost approx $55 each time x 40 if you used all fourthy
The manufacturers are are pushing these things with the code making committees
For there own interest in my opinion when they could make a single arc fault main instead

That was the conclusion my friends arrived at some years ago - That there was little to recommend the devices but that they were being pushed into the NEC by the manufacturers. Some might feel that the same sort of thing is happening with BS7671 in the U.K.

I remember when they came out with ELCB's back when I was an apprentice years and years ago in Ireland
Do they still use them in the UK ? I know RCD's replaced them

The main concern was the $1,200 cost of installing these breakers on a new house that had a 200 amp 40 circuit panel. Theses combo arc faults cost approx $55 each time x 40 if you used all fourthy
The manufacturers are are pushing these things with the code making committees
For there own interest in my opinion when they could make a single arc fault main instead

That was the conclusion my friends arrived at some years ago - That there was little to recommend the devices but that they were being pushed into the NEC by the manufacturers. Some might feel that the same sort of thing is happening with BS7671 in the U.K.

The question is why do we measure loop impedance? It may be to insure the magnet part of the MCB will work, it could also be to ensure volt drop is within limits. The use of RCD or GFCI if in US means the loop impedance to earth has very little impact on the protection so in the main it is the volt drop which is important. Yet many testers only test line - earth and omit the line - neutral test. Some do test line - neutral for the PSSC but even that is dependent on which tester is used.

Why not read line to neutral and line to earth impedance. Approx 95% of the fault current goes back on the neutral from what I understand?Then there is voltage drop question. It would seam that reading the voltage drop is a good ideas to check to see if it is within local standards? Here in the USA branch circuits to stay within 3% and feeders5% per NEC. But, what if the voltage drop is too high after the installation is complete well this would mean correcting the wire size to compensate a little late for that?



We enter readings onto paper work with little or no thought to what the reading actually mean. In the UK 0.94Ω is likely the limit due to volt drop with a ring final yet for years we took 1.44Ω now 1.37Ω as the limit. I am unaware of USA limits with 120 volt it must be a problem which makes UK problems look rather mild.

The problem for the electrician is if entering the results on paper work lays him open for future claims. We enter readings of better than 1.37Ω and assume we are OK. But the volt drop needs better than 0.94Ω line - neutral. OK we read line - earth and we could claim this does not relate to line - neutral however some maths nut can easy show the relation between line - earth and line - neutral readings and use our own readings to show the ring final circuit never complied with the regulations.

So our minor works or installation certificate could in the future land us with huge bills for remedial work which to put it mildly would bankrupt many firms. Much depends on what is entered on the certificate, but we should be aware readings outside the limits can come back and bite us.

We have seen years after the event how PPE has come back and caused the banks to pay out huge amounts, this could also happen to the electrical industry with massive bills because the loop impedance did not fall within limits.

So it's a case of watch my back and being careful paper work raised does not in the future land one with huge bills where it can be proved the installation never complied with reguations.

 

[ericmark]

Yes we use ELCB-c but not ELCB-v but we call a ELCB-c a RCD now. The voltage ELCB was outlawed as too easy for the coil to be shorted out causing them to fail. You still see them used solely as isolators as they were not inside the distribution unit and but using a ELCB-c (RCD) in the consumer unit and routing earth wire direct and leaving old ELCB-v in place one did not need to break seals.

 

[PBC_1966]

We enter readings onto paper work with little or no thought to what the reading actually mean. In the UK 0.94Ω is likely the limit due to volt drop with a ring final yet for years we took 1.44Ω now 1.37Ω as the limit. I am unaware of USA limits with 120 volt it must be a problem which makes UK problems look rather mild.

The smallest branch circuits used are 15A on #14 conductors (just over 2 sq. mm). Bare copper of 14 AWG has a resistance of 2.57 ohms per 1000 ft., so if designing for a full load of 15A with 3% maximum voltage drop that limits the branch circuit length to about 46 ft. for a 120V circuit.

I remember when they came out with ELCB's back when I was an apprentice years and years ago in Ireland
Do they still use them in the UK ? I know RCD's replaced them

Do you mean the old voltage-operated ELCB? They went out of use in the early 1980's in the U.K., but are still around in older installations in many rural areas. Current-operated ELCB's were also employed in some cases, which were just forerunners of today's RCD but with a much lower sensitivity (often up to 500mA).

Why not read line to neutral and line to earth impedance. Approx 95% of the fault current goes back on the neutral from what I understand?

It depends on the fault. Short line to earth at an outlet and the whole of the fault current will travel back on the earth, not the neutral (until reaching the point at which the earthing system connects with the supply neutral in the case of TN-C-S).

Then there is voltage drop question. It would seam that reading the voltage drop is a good ideas to check to see if it is within local standards? Here in the USA branch circuits to stay within 3% and feeders5% per NEC. But, what if the voltage drop is too high after the installation is complete well this would mean correcting the wire size to compensate a little late for that?

Hence why the correct cable size for the load should be determined before installing it. It wouldn't be much fun to run 200 ft. of #14 Romex to a 15A load and then realize after testing that your 120V at the panel is down to less than 105V at the far end!

 

[ericmark]

In the UK power is allowed 5% lights 3% volt drop from the supply head. The supply has to be no more than +10 or -6% volt drop so taking both to limits with a 230 volt supply 205.39 volt to 253 volt at the socket.

Where taking the supply from a board which already has a volt drop one has to measure the loop impedance to ensure the volt drop is not exceeded. With our ring final system that means no more than 106 meters of 2.5mm² cable. When I first tried to work this out I could not get the correct results. On enquiry at a IET meeting I found we don't consider the load to be 32A in centre of ring but at 20A and the remaining 12A is considered as even spread so we work it out at 26A not 32A. They also take into account that the two cables are rated at 42A but drawing 26A so correction factors are used. So instead of using 18mV/A/M we use 16.5mV/A/M so the line - neutral impedance incomer of typical 0.35Ω would go to 0.94Ω at the midway point. However the impedance required to trip a B type MCB at 32A was 1.44Ω this has been reduced a little with last amendment to allow for volt drop, but to comply with volt drop we should not reach the limit of the MCB anyway.

Rating factor for grouping Cg and Rating factor for ambient temperature Ca and Installation Method all have an impact on the final result and without having a computer or PDA and a java script or excel program it is near impossible to work out the exact limits on site. So we use rule of thumb so we know on a rewire never to use more than one drum (100M) of cable on a ring.

With 230 volt it's not too bad but with 55-0-55 or with three phase 63-0-63 volt systems the volt drop is a real problem glad I don't live in USA. Our 110 volt system used in building sites is suppose to be safer with call it reduced low voltage but in real terms the amount of fires caused with the lower voltage means it's not really any safer than using 230 volt specially since with 230 volt we use RCD protection.

We have some in-line RCD's for 110 volt but not really a success. Rules say no more than 50 volts to earth but only 55 to start with. We never have a neutral with our 110 volt just to lines and an earth.

In UK neutral is considered as live so we call phase wire line I think you call it the hot wire?

 

[Bus bar]

We enter readings onto paper work with little or no thought to what the reading actually mean. In the UK 0.94Ω is likely the limit due to volt drop with a ring final yet for years we took 1.44Ω now 1.37Ω as the limit. I am unaware of USA limits with 120 volt it must be a problem which makes UK problems look rather mild.

The smallest branch circuits used are 15A on #14 conductors (just over 2 sq. mm). Bare copper of 14 AWG has a resistance of 2.57 ohms per 1000 ft., so if designing for a full load of 15A with 3% maximum voltage drop that limits the branch circuit length to about 46 ft. for a 120V circuit

There seems to be something wrong with the math here? I would start getting concerned with VD with 15amps load on 120 v over 200' our coils of #14 comes in 250' roll
Remember K is 10.4 ohms per mil foot of copper not 2.57ohms per 1000'
based on ED= kIL/CSA
Not sure how you came up 46'

Correction ..according to my math 120volt #14 wire would be good with a 15amp load 80' keeping within 3%

I remember when they came out with ELCB's back when I was an apprentice years and years ago in Ireland
Do they still use them in the UK ? I know RCD's replaced them

Do you mean the old voltage-operated ELCB? They went out of use in the early 1980's in the U.K., but are still around in older installations in many rural areas. Current-operated ELCB's were also employed in some cases, which were just forerunners of today's RCD but with a much lower sensitivity (often up to 500mA).

Why not read line to neutral and line to earth impedance. Approx 95% of the fault current goes back on the neutral from what I understand?

It depends on the fault. Short line to earth at an outlet and the whole of the fault current will travel back on the earth, not the neutral (until reaching the point at which the earthing system connects with the supply neutral in the case of TN-C-S). The point I was making is the fault ultimately returm to the xtrm xo via the neutral

Then there is voltage drop question. It would seam that reading the voltage drop is a good ideas to check to see if it is within local standards? Here in the USA branch circuits to stay within 3% and feeders5% per NEC. But, what if the voltage drop is too high after the installation is complete well this would mean correcting the wire size to compensate a little late for that?

Hence why the correct cable size for the load should be determined before installing it. It wouldn't be much fun to run 200 ft. of #14 Romex to a 15A load and then realize after testing that your 120V at the panel is down to less than 105V at the far end!

Knowing the wire sizes, cir mil sizes,loads involved
Length of wire, temp considerations 200'should not be a problem



I suppose the reason for checking vd would be if you suspected a problem based on the operating conditions

We enter readings onto paper work with little or no thought to what the reading actually mean. In the UK 0.94Ω is likely the limit due to volt drop with a ring final yet for years we took 1.44Ω now 1.37Ω as the limit. I am unaware of USA limits with 120 volt it must be a problem which makes UK problems look rather mild.

The smallest branch circuits used are 15A on #14 conductors (just over 2 sq. mm). Bare copper of 14 AWG has a resistance of 2.57 ohms per 1000 ft., so if designing for a full load of 15A with 3% maximum voltage drop that limits the branch circuit length to about 46 ft. for a 120V circuit.

I remember when they came out with ELCB's back when I was an apprentice years and years ago in Ireland
Do they still use them in the UK ? I know RCD's replaced them

Do you mean the old voltage-operated ELCB? They went out of use in the early 1980's in the U.K., but are still around in older installations in many rural areas. Current-operated ELCB's were also employed in some cases, which were just forerunners of today's RCD but with a much lower sensitivity (often up to 500mA).

Why not read line to neutral and line to earth impedance. Approx 95% of the fault current goes back on the neutral from what I understand?

It depends on the fault. Short line to earth at an outlet and the whole of the fault current will travel back on the earth, not the neutral (until reaching the point at which the earthing system connects with the supply neutral in the case of TN-C-S).

Then there is voltage drop question. It would seam that reading the voltage drop is a good ideas to check to see if it is within local standards? Here in the USA branch circuits to stay within 3% and feeders5% per NEC. But, what if the voltage drop is too high after the installation is complete well this would mean correcting the wire size to compensate a little late for that?

Hence why the correct cable size for the load should be determined before installing it. It wouldn't be much fun to run 200 ft. of #14 Romex to a 15A load and then realize after testing that your 120V at the panel is down to less than 105V at the far end!

We enter readings onto paper work with little or no thought to what the reading actually mean. In the UK 0.94Ω is likely the limit due to volt drop with a ring final yet for years we took 1.44Ω now 1.37Ω as the limit. I am unaware of USA limits with 120 volt it must be a problem which makes UK problems look rather mild.

The smallest branch circuits used are 15A on #14 conductors (just over 2 sq. mm). Bare copper of 14 AWG has a resistance of 2.57 ohms per 1000 ft., so if designing for a full load of 15A with 3% maximum voltage drop that limits the branch circuit length to about 46 ft. for a 120V circuit.

I remember when they came out with ELCB's back when I was an apprentice years and years ago in Ireland
Do they still use them in the UK ? I know RCD's replaced them

Do you mean the old voltage-operated ELCB? They went out of use in the early 1980's in the U.K., but are still around in older installations in many rural areas. Current-operated ELCB's were also employed in some cases, which were just forerunners of today's RCD but with a much lower sensitivity (often up to 500mA).

Why not read line to neutral and line to earth impedance. Approx 95% of the fault current goes back on the neutral from what I understand?

It depends on the fault. Short line to earth at an outlet and the whole of the fault current will travel back on the earth, not the neutral (until reaching the point at which the earthing system connects with the supply neutral in the case of TN-C-S).

Then there is voltage drop question. It would seam that reading the voltage drop is a good ideas to check to see if it is within local standards? Here in the USA branch circuits to stay within 3% and feeders5% per NEC. But, what if the voltage drop is too high after the installation is complete well this would mean correcting the wire size to compensate a little late for that?

Hence why the correct cable size for the load should be determined before installing it. It wouldn't be much fun to run 200 ft. of #14 Romex to a 15A load and then realize after testing that your 120V at the panel is down to less than 105V at the far end!

We enter readings onto paper work with little or no thought to what the reading actually mean. In the UK 0.94Ω is likely the limit due to volt drop with a ring final yet for years we took 1.44Ω now 1.37Ω as the limit. I am unaware of USA limits with 120 volt it must be a problem which makes UK problems look rather mild.

The smallest branch circuits used are 15A on #14 conductors (just over 2 sq. mm). Bare copper of 14 AWG has a resistance of 2.57 ohms per 1000 ft., so if designing for a full load of 15A with 3% maximum voltage drop that limits the branch circuit length to about 46 ft. for a 120V circuit.
Something wrong with how you came up with this answer

I remember when they came out with ELCB's back when I was an apprentice years and years ago in Ireland
Do they still use them in the UK ? I know RCD's replaced them

Do you mean the old voltage-operated ELCB? They went out of use in the early 1980's in the U.K., but are still around in older installations in many rural areas. Current-operated ELCB's were also employed in some cases, which were just forerunners of today's RCD but with a much lower sensitivity (often up to 500mA).

Why not read line to neutral and line to earth impedance. Approx 95% of the fault current goes back on the neutral from what I understand?

It depends on the fault. Short line to earth at an outlet and the whole of the fault current will travel back on the earth, not the neutral (until reaching the point at which the earthing system connects with the supply neutral in the case of TN-C-S).

Then there is voltage drop question. It would seam that reading the voltage drop is a good ideas to check to see if it is within local standards? Here in the USA branch circuits to stay within 3% and feeders5% per NEC. But, what if the voltage drop is too high after the installation is complete well this would mean correcting the wire size to compensate a little late for that?

Hence why the correct cable size for the load should be determined before installing it. It wouldn't be much fun to run 200 ft. of #14 Romex to a 15A load and then realize after testing that your 120V at the panel is down to less than 105V at the far end!

In the UK power is allowed 5% lights 3% volt drop from the supply head. The supply has to be no more than +10 or -6% volt drop so taking both to limits with a 230 volt supply 205.39 volt to 253 volt at the socket.

Where taking the supply from a board which already has a volt drop one has to measure the loop impedance to ensure the volt drop is not exceeded. With our ring final system that means no more than 106 meters of 2.5mm² cable. When I first tried to work this out I could not get the correct results. On enquiry at a IET meeting I found we don't consider the load to be 32A in centre of ring but at 20A and the remaining 12A is considered as even spread so we work it out at 26A not 32A. They also take into account that the two cables are rated at 42A but drawing 26A so correction factors are used. So instead of using 18mV/A/M we use 16.5mV/A/M so the line - neutral impedance incomer of typical 0.35Ω would go to 0.94Ω at the midway point. However the impedance required to trip a B type MCB at 32A was 1.44Ω this has been reduced a little with last amendment to allow for volt drop, but to comply with volt drop we should not reach the limit of the MCB anyway.

Rating factor for grouping Cg and Rating factor for ambient temperature Ca and Installation Method all have an impact on the final result and without having a computer or PDA and a java script or excel program it is near impossible to work out the exact limits on site. So we use rule of thumb so we know on a rewire never to use more than one drum (100M) of cable on a ring.
We basically apply the same rules here in the U.S.




With 230 volt it's not too bad but with 55-0-55 or with three phase 63-0-63 volt systems the volt drop is a real problem glad I don't live in USA. Our 110 volt system used in building sites is suppose to be safer with call it reduced low voltage but in real terms the amount of fires caused with the lower voltage means it's not really any safer than using 230 volt specially since with 230 volt we use RCD protection.

We have some in-line RCD's for 110 volt but not really a success. Rules say no more than 50 volts to earth but only 55 to start with. We never have a neutral with our 110 volt just to lines and an earth.

In UK neutral is considered as live so we call phase wire line I think you call it the hot wire?

In the UK power is allowed 5% lights 3% volt drop from the supply head. The supply has to be no more than +10 or -6% volt drop so taking both to limits with a 230 volt supply 205.39 volt to 253 volt at the socket.

Where taking the supply from a board which already has a volt drop one has to measure the loop impedance to ensure the volt drop is not exceeded. With our ring final system that means no more than 106 meters of 2.5mm² cable. When I first tried to work this out I could not get the correct results. On enquiry at a IET meeting I found we don't consider the load to be 32A in centre of ring but at 20A and the remaining 12A is considered as even spread so we work it out at 26A not 32A. They also take into account that the two cables are rated at 42A but drawing 26A so correction factors are used. So instead of using 18mV/A/M we use 16.5mV/A/M so the line - neutral impedance incomer of typical 0.35Ω would go to 0.94Ω at the midway point. However the impedance required to trip a B type MCB at 32A was 1.44Ω this has been reduced a little with last amendment to allow for volt drop, but to comply with volt drop we should not reach the limit of the MCB anyway.

Rating factor for grouping Cg and Rating factor for ambient temperature Ca and Installation Method all have an impact on the final result and without having a computer or PDA and a java script or excel program it is near impossible to work out the exact limits on site. So we use rule of thumb so we know on a rewire never to use more than one drum (100M) of cable on a ring.

With 230 volt it's not too bad but with 55-0-55 or with three phase 63-0-63 volt systems the volt drop is a real problem glad I don't live in USA. Our 110 volt system used in building sites is suppose to be safer with call it reduced low voltage but in real terms the amount of fires caused with the lower voltage means it's not really any safer than using 230 volt specially since with 230 volt we use RCD protection.

We have some in-line RCD's for 110 volt but not really a success. Rules say no more than 50 volts to earth but only 55 to start with. We never have a neutral with our 110 volt just to lines and an earth.

In UK neutral is considered as live so we call phase wire line I think you call it the hot wire?

We call our hot wires live and our neutral
Wire the grounded conductor or neutral

 

[Bus bar]

We enter readings onto paper work with little or no thought to what the reading actually mean. In the UK 0.94Ω is likely the limit due to volt drop with a ring final yet for years we took 1.44Ω now 1.37Ω as the limit. I am unaware of USA limits with 120 volt it must be a problem which makes UK problems look rather mild.

The smallest branch circuits used are 15A on #14 conductors (just over 2 sq. mm). Bare copper of 14 AWG has a resistance of 2.57 ohms per 1000 ft., so if designing for a full load of 15A with 3% maximum voltage drop that limits the branch circuit length to about 46 ft. for a 120V circuit.

I remember when they came out with ELCB's back when I was an apprentice years and years ago in Ireland
Do they still use them in the UK ? I know RCD's replaced them

Do you mean the old voltage-operated ELCB? They went out of use in the early 1980's in the U.K., but are still around in older installations in many rural areas. Current-operated ELCB's were also employed in some cases, which were just forerunners of today's RCD but with a much lower sensitivity (often up to 500mA).sounds right !I left Dublin in1981 worked with a English bloke back then he was an electrician working in dub at the time. He knew his stuff

Why not read line to neutral and line to earth impedance. Approx 95% of the fault current goes back on the neutral from what I understand?

It depends on the fault. Short line to earth at an outlet and the whole of the fault current will travel back on the earth, not the neutral (until reaching the point at which the earthing system connects with the supply neutral in the case of TN-C-S).

Then there is voltage drop question. It would seam that reading the voltage drop is a good ideas to check to see if it is within local standards? Here in the USA branch circuits to stay within 3% and feeders5% per NEC. But, what if the voltage drop is too high after the installation is complete well this would mean correcting the wire size to compensate a little late for that?

Hence why the correct cable size for the load should be determined before installing it. It wouldn't be much fun to run 200 ft. of #14 Romex to a 15A load and then realize after testing that your 120V at the panel is down to less than 105V at the far end!

We enter readings onto paper work with little or no thought to what the reading actually mean. In the UK 0.94Ω is likely the limit due to volt drop with a ring final yet for years we took 1.44Ω now 1.37Ω as the limit. I am unaware of USA limits with 120 volt it must be a problem which makes UK problems look rather mild.

The smallest branch circuits used are 15A on #14 conductors (just over 2 sq. mm). Bare copper of 14 AWG has a resistance of 2.57 ohms per 1000 ft., so if designing for a full load of 15A with 3% maximum voltage drop that limits the branch circuit length to about 46 ft. for a 120V circuit.

I remember when they came out with ELCB's back when I was an apprentice years and years ago in Ireland
Do they still use them in the UK ? I know RCD's replaced them

Do you mean the old voltage-operated ELCB? They went out of use in the early 1980's in the U.K., but are still around in older installations in many rural areas. Current-operated ELCB's were also employed in some cases, which were just forerunners of today's RCD but with a much lower sensitivity (often up to 500mA).

Why not read line to neutral and line to earth impedance. Approx 95% of the fault current goes back on the neutral from what I understand?

It depends on the fault. Short line to earth at an outlet and the whole of the fault current will travel back on the earth, not the neutral (until reaching the point at which the earthing system connects with the supply neutral in the case of TN-C-S).

Then there is voltage drop question. It would seam that reading the voltage drop is a good ideas to check to see if it is within local standards? Here in the USA branch circuits to stay within 3% and feeders5% per NEC. But, what if the voltage drop is too high after the installation is complete well this would mean correcting the wire size to compensate a little late for that?

Hence why the correct cable size for the load should be determined before installing it. It wouldn't be much fun to run 200 ft. of #14 Romex to a 15A load and then realize after testing that your 120V at the panel is down to less than 105V at the far end!

 

[Bus bar]

I think AFCI technology will cross the pond before to long maybe it will replace certain testing all together? But don't get sold on the idea as they really only mask the problem that we are dealing with here in the USA