Constant High Mains Voltages At Home

[IanB]

When talking about distance from the substation, surely the cable size is also an important factor? If there is going to be a big substation serving many homes over a wider area, maybe the cables are bigger?

Also maybe the cables in typical UK installations are copper? Whereas my service in the USA is delivered on aluminum cables.

[SeanB]

If it was installed in 1992 it definitely is an aluminium core cable, as the 1970's copper price spikes made aluminium cables the dominant cable in all new installations, and the utilities now are actively swapping out older bare copper lines to ABC runs, with the price they get for the scrap copper covering the cost of the new cable and installation.

[richard.cs]

Your analysis seems to be mistaken.  You are claiming the UK line voltage is higher so the current is smaller, etc, but your math is faulty.  The power to any one home is not three phase, it is one leg to the neutral.  So the current drawn for a given load is the same ~240V as in the US.  The voltage between the phases is irrelevant for the feed from the transformer to the home.  You seem to be talking about the higher voltage lines which are three phase in the US as well.

The cable from a typical US transformer to a home is 240 V split phase (with the transformer feeding only one or two properties). The cable from a typical UK transformer to a home is 400 V 3 phase for the majority of it's length, perhaps 180 m of a 200 m run, and then single phase for the last 20m or so. As we were discussing the losses in the length of the cable all the way back to the transformer it is absolutely relevant to consider that most of its length is three phase when talking about power delivery to its many single phase loads. Consider for example that under typical loading (but not unfortunately in Paul's case) the neutral currents from many properties cancel reasonably well and the neutral current ends up being small compared to the phase currents (20-50% is not atypical, and losses go as the square) so in the context of of analysing the losses the 180m of cable from the substation it might as well be feeding a 400 V delta load. Yes the loads are really star and there is some neutral current, but the dominant losses in the 3 phase section are in the phase conductors.

Put another way, from the perspective of one of the single phase customers, the neutral has minimal voltage drop (because their neutral current is mostly returned by other nearby customers to the other two live conductors) and they only observe voltage drop / power loss on the live pole of the supply. In a three-phase cable with many single phase loads at the far end or distributed along it the losses, to a reasonable approximation, are those you would get with a 3 phase 400 V load.  This is the same reasoning as treating the US setup as 240 V rather than 120 V despite the presence of 120 V loads.

To be honest it was largely a flippant remark about the cables being longer because the effective voltage is higher (though the voltage delivered to the end user is not). A full analysis would of course take account of cable sizes, the fact that the UK setup has a greater number of conductors which is copper that could otherwise be used to reduce losses in a split phase setup, etc.

As Sean says for a 90s install it will probably be Aluminium. I think they were using WaveCon by then so 3 triangular aluminium cores surrounded by a wave-wound (not spiralled) bundle of copper wires that form the CNE conductor. 240 mm² is a common size. The branch to the house will probably be concentric with a 35 mm² aluminium live surrounded by 25 mm² of copper strands. This has the nice feature that if you put a spade through it the aluminium tends to burn back up the middle until the arc is extinguished, if you're lucky before the substation fuse blows.

[SeanB]

I remember in primary school, around 1975, they were doing a whole load of cable work right outside the school, and there were lots of little sawn off sections of the aluminium cable there. 4 triangularish conductors, with a coloured PVC sheath, and then that in turn surrounded with a black plastic sheath, and then the steel wire armouring wires around that, and the final protective sheath, with the cable being run along the street in a newly dug trench from the transformer, to a new block of flats being built in the cul de sac at the end of the road. Took some of those little pieces home with me.

Now only ABC cable, as it is faster and cheaper to install, though the quality of the work often leaves a lot to be desired, though it is still stolen on a regular basis.

I am at the moment on the end of 200m of cable, though the substation is 60m away, as it runs in a dog leg, and yes, I have had a few mains voltage excursions that reached nearly 300VAC, which was when the thieves broke in to the substation and removed the neutral and earthing from there, and cut back the armour on the cables, century old, so just unwind the steel strap to expose the paper, and cut the neutral off of the cable for 10m. The load was reasonably constant, so the excursions on the mains were small, till the coffee shop turned on their oven to bake the morning load, and this dragged the one phase down to 160VAC, and the others rose up in sympathy. I had a UPS high voltage warning, measured the mains, and turned off all power at the board, then went to the meter room, measured the phases all over the place, and turned off the building, then phoned the metro to notify them of loss of neutral.

[gnuarm]

When talking about distance from the substation, surely the cable size is also an important factor? If there is going to be a big substation serving many homes over a wider area, maybe the cables are bigger?

Also maybe the cables in typical UK installations are copper? Whereas my service in the USA is delivered on aluminum cables.

Maybe they do things differently in the UK, but in the US a substation feeds many, many homes, like many blocks of homes.  The substation outputs a high voltage which is dropped to 240 V by a distribution transformer to feed the homes. 

There's nothing wrong with aluminum wire.  They make the cross section larger to account for the higher resistivity resulting in the same resistance and the same losses.  They don't use it in homes anymore because it was not installed properly in the 70's causing fires.  They can use it safely, but like any other technology it has to be done correctly.

Actually, I've been told in the UK older installations were done with aluminum sleeves for the earth ground.  Something wasn't done correctly and the rather thin aluminum sleeve corroded away leaving entire blocks with no safety ground.  Expensive to correct.

[nctnico]

Actually, I've been told in the UK older installations were done with aluminum sleeves for the earth ground.  Something wasn't done correctly and the rather thin aluminum sleeve corroded away leaving entire blocks with no safety ground.  Expensive to correct.

Chances are that the homes have their own grounding anyway. At least that is standard in NL.

[richard.cs]

Maybe they do things differently in the UK, but in the US a substation feeds many, many homes, like many blocks of homes.  The substation outputs a high voltage which is dropped to 240 V by a distribution transformer to feed the homes.

Ah, it's a terminology issue, I understand now. In the UK we would call what you refer to as a "substation" a "primary substation", in the UK this usually outputs 11 kV line to line. These cover quite wide areas, a mid sized town might have one only, a city of a million people might have half a dozen. The substations Ian and I were referring to are more akin to your distribution transformers, we would normally call a ground-mounted distribution transformer and its associated switchgear, etc. a "substation". Our distribution transformers often serve hundreds of customers within a few hundred metres.

Actually, I've been told in the UK older installations were done with aluminum sleeves for the earth ground.  Something wasn't done correctly and the rather thin aluminum sleeve corroded away leaving entire blocks with no safety ground.  Expensive to correct.

There was a short period between the demise of PILC cables and the rise of polymeric ones when cables with aluminium sheaths were attempted (cheaper than lead I guess?). It turned out to be a rather bad idea.

[gnuarm]

Maybe they do things differently in the UK, but in the US a substation feeds many, many homes, like many blocks of homes.  The substation outputs a high voltage which is dropped to 240 V by a distribution transformer to feed the homes.

Ah, it's a terminology issue, I understand now. In the UK we would call what you refer to as a "substation" a "primary substation", in the UK this usually outputs 11 kV line to line. These cover quite wide areas, a mid sized town might have one only, a city of a million people might have half a dozen. The substations Ian and I were referring to are more akin to your distribution transformers, we would normally call a ground-mounted distribution transformer and its associated switchgear, etc. a "substation". Our distribution transformers often serve hundreds of customers within a few hundred metres.

Those numbers don't seem to add up.  Hundreds of customers within a few hundred meters means people are practically standing on top of one another.  If all these customers are fed in a single run the current in the first leg would be huge! 

Actually, I've been told in the UK older installations were done with aluminum sleeves for the earth ground.  Something wasn't done correctly and the rather thin aluminum sleeve corroded away leaving entire blocks with no safety ground.  Expensive to correct.

There was a short period between the demise of PILC cables and the rise of polymeric ones when cables with aluminium sheaths were attempted (cheaper than lead I guess?). It turned out to be a rather bad idea.

Lead???  I've never heard of using lead in electric cables.  I guess there's a first time for everything.

I was interested in ham antennas at one time and found there is a type of tubing with plastic inside and out, but aluminum in the sandwich to act as an oxygen barrier.  This prevents rust formation and other oxidation in heating systems.  If you protect the aluminum from everything else it is not a bad component in a cable or pipe, but on the outside it is a very bad idea. 

[gnuarm]

Actually, I've been told in the UK older installations were done with aluminum sleeves for the earth ground.  Something wasn't done correctly and the rather thin aluminum sleeve corroded away leaving entire blocks with no safety ground.  Expensive to correct.

Chances are that the homes have their own grounding anyway. At least that is standard in NL.

I was discussing grounding in a UK ham group and they made the point that many homes simply can not obtain an adequate ground locally.  Practice is to bond the ground wire to everything metal that could be grounded, like plumbing, so if there is an open protective earth/neutral (PEN) there won't be a voltage differential.  If the ground is poor, everything in the home is at the same high voltage.

This falls apart when power is used outside the home.  I don't know how they manage that situation.  I'm sure I heard about it, but I don't recall the details.  Maybe that would be handled by an RCD/GFCI device since at that point the current in the circuit is not balanced.

[themadhippy]

This falls apart when power is used outside the home.  I don't know how they manage that situation.

By following the relevant  part of the wiring  regulations that don't permit the cpc to be exported outside the equipotential  zone if a TNC-S supply method is used. What you do is convert the outside circuits to TT (use an earth spike) so only live and neutral are taken outside

[richard.cs]

Those numbers don't seem to add up.  Hundreds of customers within a few hundred meters means people are practically standing on top of one another.  If all these customers are fed in a single run the current in the first leg would be huge! 

A few hundred customers within a few hundred metres radius, not all on the same cable. Commonly 5-10 cables might radiate out in different directions, each 100-200 m long in an urban area, longer in a rural area. A 200 m radius is 120,000 m², a typical suburban UK property is maybe 200 m² (land area not building area), 120,000/200 = 600.
 But yes the currents can be fairly high, a 1 MVA transformer is around 1300 A per phase, but that's on the larger end.

Lead???  I've never heard of using lead in electric cables.  I guess there's a first time for everything.

Yup, lead. A good search term is "PILC cable" - Paper Insulated Lead Covered. Lots of photos covering a range of construction types and voltages here:
https://www.spenergynetworks.co.uk/userfiles/file/scottishpower_cables_equipment_metal_theft.pdf
LV PILC on page 18, note the undersized neutral on one of the examples.

Long ago (1930s) we also had lead covered rubber cables like this in domestic use:
https://img1.wsimg.com/isteam/ip/14f7f3aa-bbda-4bcf-bf19-2ab202e8e95c/81a326a6-24ce-4b6d-ae58-460e8af5ce60.jpg/:/rs=w:1280

[bsdphk]

Lead???  I've never heard of using lead in electric cables.  I guess there's a first time for everything.

Before plastic, cables for wet environments were encapsulated in extruded lead.

Here is a great article from 1931 about how they did that for telephone cables: https://archive.org/details/bstj10-3-432/mode/2up

Lead has also been used as conductor, but almost only for very high currents, where the "cable" were more or less cast in place, where it would be needed, for instance the main busbar in the first generation of power-plants.

[coppice]

Lead???  I've never heard of using lead in electric cables.  I guess there's a first time for everything.

Most houses built in the 1930s and 1940s, at least in the UK, used rubber insulated tinned copper wire with an overall lead sheath. The tinning was needed to limit the reaction between copper and rubber. In the 1950s the price of lead skyrocketed, as the nuclear industries took off, and early plastics, like PVC, had been around long enough to see that they were pretty stable long term. The industry moved to plastic insulated and sheathed cable. By the 1960s most of the rubber was rotting, so by the late 1970s most of this lead sheathed cabled had been ripped out and replaced.

[mikeselectricstuff]

Lead???  I've never heard of using lead in electric cables.  I guess there's a first time for everything.

Before plastic, cables for wet environments were encapsulated in extruded lead.

My incoming supply is on lead cables (mid- 1960s)

[nctnico]

Lead???  I've never heard of using lead in electric cables.  I guess there's a first time for everything.

Yup. And only old-timer electricians have the skills to deal with these cables. Soldering a ground wire onto the lead shield using a blowtorch is a bit of an art. One of my relatives used to be an electrician. IIRC they still called him when they needed someone to deal with a lead clad cable after he retired.

[gnuarm]

This falls apart when power is used outside the home.  I don't know how they manage that situation.

By following the relevant  part of the wiring  regulations that don't permit the cpc to be exported outside the equipotential  zone if a TNC-S supply method is used. What you do is convert the outside circuits to TT (use an earth spike) so only live and neutral are taken outside

So when you are using a hand power tool you add an earth spike to your extension cord? 

That was how the conversation went with a description of all manner of messy, complicated and potentially unworkable approaches.  It also doesn't protect against any number of innocuous situations where the "equipotential zone" is violated.  Off the top of my head someone might have a clothesline strung from the house to a metal pole that can be reached from a window.  These are not so rare.  If the line has metal in it and the plastic cover degrades, there is a potential of a ground point that is reached through the window while in contact with the laundry equipment. 

Obviously these things are rare or they would have been factored into the standard.  I've just never understood why it is so important to minimize the cost of running proper ground wires and sharing the neutral.  That simply isn't done in the US.  I can see the aluminum ground wire in the cable that runs to my house.  It stands out clearly against the insulated current carrying wires. 

[gnuarm]

Lead???  I've never heard of using lead in electric cables.  I guess there's a first time for everything.

Before plastic, cables for wet environments were encapsulated in extruded lead.

My incoming supply is on lead cables (mid- 1960s) (Attachment Link)

Looks like that cable has paint on it.  I hope it isn't leaded paint.  They might have to rip that out to get rid of the paint! 

[coppice]

Lead???  I've never heard of using lead in electric cables.  I guess there's a first time for everything.

Before plastic, cables for wet environments were encapsulated in extruded lead.

My incoming supply is on lead cables (mid- 1960s) (Attachment Link)

Most houses in the UK are like that. The insulation in those underground cables didn't degrade like the rubber insulation used for the lead sheathed wiring inside the house, so there has never been a need to replace the underground stuff. 100 year old cables are basically as good as new, and only need replacing if a higher rating is needed.

[richard.cs]

So when you are using a hand power tool you add an earth spike to your extension cord?

We sidestep the problem by having product standards that mandate equipment for outdoor use (power tools, lawnmowers, etc.) is Class II double insulated, not requiring an earth. This falls over a bit for non-domestic things like welders, and for things like electric cars where they should probably have been Class II but aren't.

Obviously these things are rare or they would have been factored into the standard.  I've just never understood why it is so important to minimize the cost of running proper ground wires and sharing the neutral.  That simply isn't done in the US.  I can see the aluminum ground wire in the cable that runs to my house.  It stands out clearly against the insulated current carrying wires. 

Mostly, it's a cheap-arse 1970s bodge that was supposed to save money, and one which is looking increasing dubious in the modern world. That said, it has some advantages; open-circuit PEN conductors are very noticeable and therefore quickly repaired in comparison to rarely-use earth conductors, the low fault impedance gives fast disconnection during a fault, etc.

If we were designing a system from scratch, I would favour an impedance-earthed transformer star point, say about an Ohm. It would limit earth fault current to a few hundred amps, so earth could be distributed cheaply on smallish (say 25 mm² copper cabling, but such a current is plenty large enough to reliably trip an RCD(GFCI)-like device without needing carefully balanced mechanics with low reliability like lower value RCDs have. That said if we were designing from scratch I'd push the distribution voltage up as well. Single phase domestic supplies are also clearly a UK mistake.

Actually, electrical deaths are remarkably low, especially shock deaths (as opposed to fire, etc.) so it doesn't actually make sense to spend a load of money to reduce shock risk when investing that same amount in road safety, anti smoking campaigns or healthcare would save more lives. In fact I suspect that in the big picture many electrical safety things are barely justified.

[mikeselectricstuff]

We sidestep the problem by having product standards that mandate equipment for outdoor use (power tools, lawnmowers, etc.) is Class II double insulated, not requiring an earth. This falls over a bit for non-domestic things like welders, and for things like electric cars where they should probably have been Class II but aren't.

Would be nice if EVs were class II, but would be pretty tricky to make an 11 or 22kW onboard charger that passes EMC without a ground connection!
I suppose a double-insulated barrier between the mains and DC side, and also between the AC and body might be possible, but it's too late now.

[richard.cs]

We sidestep the problem by having product standards that mandate equipment for outdoor use (power tools, lawnmowers, etc.) is Class II double insulated, not requiring an earth. This falls over a bit for non-domestic things like welders, and for things like electric cars where they should probably have been Class II but aren't.

Would be nice if EVs were class II, but would be pretty tricky to make an 11 or 22kW onboard charger that passes EMC without a ground connection!
I suppose a double-insulated barrier between the mains and DC side, and also between the AC and body might be possible, but it's too late now.

A functional earth for EMC-only, not connected to the chassis would be fine. Isolation mains to battery is likely to add a lot of cost and weight, easier to let the battery wiring flap around at mains live and double insulate that. The main problem there is that you can't really manage to meet double insulation in the motor windings as it adds significant volume, so they would need to be isolated during charging with a contactor that achieves a suitable gap. This then precludes Zoe-like charging where the main inverter is used as the charger with the motor as the inductors.

I can see why cars are Class I, but that doesn't mean I have to like it. I can also see arguments that PE should be safe to bolt to random bits of outdoor metal and therefore that the fault lies with TNC-S not the cars.

[Monkeh]

That simply isn't done in the US.  I can see the aluminum ground wire in the cable that runs to my house.  It stands out clearly against the insulated current carrying wires.

That is your neutral.

[Alti]

If we were designing a system from scratch, I would favour an impedance-earthed transformer star point, say about an Ohm. It would limit earth fault current to a few hundred amps, so earth could be distributed cheaply on smallish (say 25 mm² copper cabling,(..)

I'd say that for residential, TN-S earthing is rare as hen's teeth and is the only one that distributes PE. Second popular is TT but does not distribute PE at all. So the applicability of this invention would have been quite narrow. Don't you think that a failure of single RCD could put whole neighbourhood in danger with this 1 ?

[bsdphk]

If we were designing a system from scratch, I would favour an impedance-earthed transformer star point, say about an Ohm. It would limit earth fault current to a few hundred amps, [...]

That is called a "Petersen Coil" or "Arc Supression Coil" and have been in widespread use in Europe for 104 years.

[richard.cs]

If we were designing a system from scratch, I would favour an impedance-earthed transformer star point, say about an Ohm. It would limit earth fault current to a few hundred amps, so earth could be distributed cheaply on smallish (say 25 mm² copper cabling,(..)

Don't you think that a failure of single RCD could put whole neighbourhood in danger with this 1 ?

I don't. How I see it working is that the supplier provides each customer a ~10 A or so rated, simple, robust, reliable differential current trip in the same assembly as they currently provide a fuse, and a second backup unit at the substation. Householders are free to and encouraged to have lower rated* RCDs to avoid nuisance tripping of their whole supply. If the device fails to open a higher rated and slower one at the substation does preventing the PE being too far from true earth for an excessive time. i.e. if the downstream device fails are larger area is off supply, but the system is still safe. It's not all that different to the current setup really.

Of course it'll never happen, too much installed infrastructure exists, and nor do I claim to have "invented" it.

*And faster, proper coordination with RCDs requires time delays.

That is called a "Petersen Coil" or "Arc Supression Coil" and have been in widespread use in Europe for 104 years.

I've not come across that naming, but yes I know it exists. I wasn't aware of anywhere that currently uses it for LV? In the UK we use such a setup on our HV distribution, all HV loads are delta connected but the source is a star winding with impedance earthing. Any imbalance current on the phases is a fault which is quickly and easily detected, but limited in energy by the impedance in the star point. This is different to American practice that tends to distribute HV Neutral (or HV CNE? I'm not sure?) with pole pigs normally connected between it and a single HV phase.

Actually reading this was an interesting read, Petersen coil's are not quite the same thing, instead of simply adding impedance (R or L) to control current to a few KA for damage reduction, they actually try to make them resonant to limit the current to just a few amps so the system can stay live. Interesting stuff. https://www.westernpower.co.uk/downloads/4102