What would a clean-sheet set of wiring regulations look like today?

Standardisation of wiring colours would be a useful step. Green/Yellow for ground/earth is almost there but the rest of the colours are scattered in function between the different countries. If I remove a light fitting in my Swiss apartment I may find a black wire and a yellow wire. The black is phase, the yellow is neutral. 

Redefining any of the current colours would merely increase the already huge legacy problem 

DShutt said:

For the consumer, the economics are skewed towards being able to take advantage of cheap electricity whenever it is available.  The larger the charger, the more likely that they will be able to maximise the benefits of the availability of cheap power.  The whole concept of cheap rate electricity is to push consumers to concentrate demand at certain times of day - I suspect there is a conflict here between what suits the transmission / generation system and what suits the distribution system(s).

I think 3kW (let alone 7kW) of charging demand per car is probably enough to handle any glut of electricity. There are around 30 million cars in the UK, on average the vast majority are stationary and thus theoretically available for charging. If we could provide chargers, even low power ones, to cars wherever they go then we will be able to absorb enormous pulses of surplus electricity.

Say 15 million cars are available for charging at 3kW. The electric car system can then soak up around 45GW of overproduction on demand. At 7kW it will be 105GW.

Likewise if we can persuade everyone to have heat store style hot water (be it a cylinder or one of those compact stack of block ones) then we can soak up something like 90GW (at 3kW each) of surplus power, on command (At least in the short term).

If we assume that most electric cars will be plugged in at low demand periods, which may be night or the middle of the day, the amount of electricity each individual car will have to absorb to eliminate gluts will be relatively small.

I actually considered some of this when I did my PhD thesis. I concluded that total decarbonisation, without resort to amazing efficiency assumptions (or even heat pumps!), can probably be achieved in a world with 100A single phase supplies were the predominant domestic supply, although massive reinforcement would be necessary to meet the low voltage demands. Albeit it should be possible to avoid wrecking every street.

Average car mileage is only 7,700 miles per year after all, or about 20 miles per day. That's only about 5kWh at 4 miles per kWh

Even with only 5 hours of charging, a 7kW AC charger will pull down something like 35kWh, or around 140 miles per night.

I am sceptical there are really many people for which this is unacceptable, and in that case by far the lowest capital cost solution would likely be to just tolerate charging at slightly higher rates on a longer duration.

Economy 7 would get you 49kWh after all.

MOT statistics seem to indicate only around 5% of vehicles are driven more than 15,000 miles a year, which still only amounts to 40 miles per day, on average.

Taking your points in turn:

• Yes moving the AC-DC conversion from the car to the charge-point would reduce car manufacturing costs, but it then increases the charge-point cost by a similar, or possibly greater amount, so domestic charge-point cost and public EV charging costs would have to increase to cover this extra infrastructure costs.
• I can't see it being realistic currently because of the restriction on charge-point access, where DC charging infrastructure is still relatively limited and AC charging is pretty much essential to avoid range-anxiety / accessing most public charging. DC rapid charging is also more expensive than AC public charging (p/kWh), reflecting the higher infrastructure cost of the DC rapid charge-points.
• Your experience, I would suggest, reflects that of typical early adopters who normally have their own off-road parking and their own home charging facilities, but going forward EV ownership needs to expand to include motorists without home charging. So where are these drivers likely to charge? Currently it looks likely this will be a mix of road-side charging (lamp-post charge-points, kerb-side charge points), workplace charging, public car parks, shops/leisure venues.
  • Some of these 'use cases' are not suitable for rapid charging, with people rapidly moving their EV once charged, they rely on the vehicle being plugged in and left to charge for long periods while the owner is busy doing something else, at home busy or asleep over-night, or during the day at work. These are not suitable for DC rapid charging, where the vehicle blocks the charge-point for long periods, resulting in poor utilisation of the infrastructure.
• For the tariffs savings, there's a trade off between high power charging at off-peak times to maximise unit cost savings vs. the cost of providing the home with the supply capacity and a suitably powerful charge-point to provide that. And when we get above 7kW per house, needing supply upgrades etc, then there's a question of whether it makes sense, or whether a different tariff design (for example, including the cheap, afternoon off-peak period for some extra hours each day) gives a better result.
• For the DNO domestic unlooping, the DNO recovers those costs as part of its regulated costs from bill payers and the consumer getting the unlooping does not pay the direct costs of the work, so there's no financial incentives on either of these to use technologies like smart dynamic control to maximise the use of the existing supply capacity. 
  • The UK should learn from France and adopt their system of having defined supply capacities, with standing charges based on those charges, that provides financial incentives to use smart / innovative controls and use existing infrastructure more optimally. 
• Re. smart meter control, yes that largely what the French Link smart meter did, had an accessible serial port with real-time metering data and control signals for loads, signals when the off-peak rate was live etc, to support these types of smart controls. The GB smart meter design was much more basic and limited, the old auxiliary load control switches plus a later attempt to graft on some proportional load control capability which didn't deliver much / scale-up.

I suspect the long term outcome will depend on what happens with power conversion technology and costs, if 7kW DC charge point equipment falls in cost/size/cooling requirements, to the point that a DC charge-point is competitive we may see more DC charge points appear. If V2G takes off, that might accelerate the uptake in the domestic space.

I suspect it's likely going to be at 10 years or more until DC charging starts squeezing AC charging out.

Going back to the original question about rewriting the Wiring Regulations for today, I think a good point coming out of this discussion on EVs is that in our electrified homes and businesses, there will be more high power loads like AC / DC for EV charging, heat-pumps, water heating / showers with a total maximum power demand far beyond anything that can reasonably or economically met by simply increasing the supply capacity.

So the topic of load diversity and the use of smart / dynamic control of loads becomes much more important to ensure that installation design is economical and efficient.

DShutt said:

off peak tariffs encourage the use of electricity at certain times, it is likely to result in a degree of discussion if the message turns into get "cheap electricity between midnight and 05:00 but don't use too much".

Or rather spread out the usage rather more evenly over time ... e.g. my supplier offers an EV tariff between 0000 and 0500 but the Heat Pump tariff runs from 0500-0900 (and then 1300-1600).

With the advent of smart meters it should be possible to smooth things further - e.g. one group of customers get 2300-0400 another gets 0000-0500 and another 0100-0600 say. Loads will tend to be higher when off-peak starts then decline as less than empty to start with batteries become full (or charging rates reduce above the 80% capacity mark). All sorts of possibilities to protect looped supplies or overall distribution capacity.

  - Andy.

TurboGen said:

So the topic of load diversity and the use of smart / dynamic control of loads becomes much more important to ensure that installation design is economical and efficient

And safe ... where are the system standards for this technology, that ensures the load curtailment provisions don't get into an unsafe condition (i.e. cause overloads) when things go wrong ... particularly long-term small overloads are potentially the most damaging (larger overloads can be covered by operation of overcurrent protective devices)? There is a functional safety element to this.

As always seems to happen, whenever something gets discussed here, eFixx has a new video on it: https://www.youtube.com/watch?v=Fl3kpGxYvcs

But you do need a stack of up to 5 off 5kWh battery packs, and also a decent amount of power to the grid. At the end of the charge, the batteries will be flat, so no running the house off them until they charge again.

The whole lot looks really expensive when compared with a standard 7kW EVSE.

See also https://www.sigenergy.com/en/products/dc-charger

Yes quite and that's something which you want covered in the 'clean sheet' Wiring Regulations because these smart / dynamic load controls will be or already are, an integrated part of many different types of consumer or commercial loads each with their own respective standards, so it needs some overall / wrap-around requirements which presumably could be part of the Wiring Regulations which is bringing those products together through the fixed wiring installation.

Thinking about a future electrified home, it's not unreasonable to expect it may have:

7.4kW EV charge point = 32A

9kW (thermal) heat-pump - typically 4.6kW = 20A electrical (worst case cold-weather / poor CoP conditions)

Hot water cylinder back-up/legionella heater 3kW = 13A

Home BESS 6kW = 26A

Electric cooker = 20A

Sockets = 32+ A

That totals up to 143A before you add any additional socket circuits, lighting, electric shower, possibly an air-conditioning unit etc. 

Now in practice there will be some fortuitous diversity, but given the growth in time of use tariffs incentivising off-peak consumption which result in loads being synchronised to operate during those periods, it's not unreasonable to think there may be periods where the EV is charging at full power, the heat-pump is running flat out to provide space heating, the legionella cycle may be running on the hot water cylinder, the home BESS may be charging up and appliances are running.

It strikes me that smart / dynamic load controls are inevitable if the future, electrified home is to be feasible and so the Wiring Regs needs a good framework for handling this topic going forward.

My understanding is that the National Grid DNO operation (is that Western Power Distribution?) has started replacing 100A cutout fuses with 80A fuses, because the traditional "peaky" loads are being replaced with more continuous loads such as EV chargers. The report I read suggested that they are concerned with the true rating of their service cables under continuous load.

I agree that some sort of standardised, non proprietary, control apparatus is required. I doubt the signalling channel needs to be particularly fast, a 16 bit 'half precision' float every second or two from a meter giving current on the inlet cable is probably sufficient to avoid a slow overload.

I know I keep going on about it, but the recent DALI-2 standard could easily do it! However that would require cables. It could probably be done with something like power line signalling but that would require some method to prevent the signal interfering with other houses on the same feeder. Wireless could also do it, but I would worry about the integrity of available low cost wireless formats, and an accidental misconfiguration or dropped signal could get very expensive for a householder if it blows the DNO fuse.

DALI in the house for everything has some simplicity to it, and makes it all strictly plug and play with no worry of accidentally linking to something next door.

I think all this reiterates that all new supplies should be at least two phase, and probably three phase since I doubt it costs much more.
That itself has interesting implications - 400V single or three phase would be available for large appliances.
1.5mm T&E can supply an awful lot more at 400V than 230V!

EIreland said:

I agree that some sort of standardised, non proprietary, control apparatus is required. I doubt the signalling channel needs to be particularly fast, a 16 bit 'half precision' float every second or two from a meter giving current on the inlet cable is probably sufficient to avoid a slow overload.

I agree, to avoid the current situation of duplication of household load monitoring devices - a CT for the BESS, possibly an extra CT for a stand-alone solar inverter, CT for the EV charge-point, a CT for the heat-pump, some of these have remote DIN rail meters, so the amount of duplicated equipment at the meter/CU becomes significant. But also it needs some standardisation to handle how to coordinate the sharing of available capacity between loads.

That said, I think when you get into these detailed load monitoring design, communication standards, arrangements for capacity sharing and prioritisation between loads, this is a taking you toward a separate, technical standard to cover this aspect as it's beyond the scope of the Wiring Regulations.

For the Wiring Regulations, I was thinking something more high-level around how these smart controls should be used when designing an installation and determining maximum demand - what allowances need to be made for minimum loads from these different load types, how much turn-down / load reduction is acceptable, some form of framework for design.

Also re. Graham's point on the safety topic, what approach the Wiring Regs take to providing additional overload protection for multiple dynamic loads. There's conceivably quite a number of different ways this could be achieved, each with various pros and con, so having some standardised approaches within the Regs is likely to ease and simplify this.

Whilst I agree that such systems are somewhat beyond the scope, I think the best way to avoid a zoo of incompatible equipment, with all the problems that causes for the householder, would be to specific a communications standard, if only a physical layer, to be used.

If we go with a DALI compatible physical layer, as an example, the wiring regs could reasonably impose a pair of thin data cores into cables in the same way they impose an earth conductor. It is a bit beyond the traditional view of BS7671, but if we can redesign wiring regulations (and their related regulations like the socket regs), we could impose such things if we wish.

I think smart controls using a single unified spec an excellent way to rationalise the current arrangements for lighting and the like that cause such issues in the field. If we have digital control of lighting just written into the specification, as an example, new installations never have to worry about the pain of two or three way switching and the confusion it can cause.

Such things are often referred to in the context of confusing DIYers, but during recent work in my house an apparent professional electrician managed to rewire the lighting to put a dead short between a switched live and a neutral, so the breaker immediately tripped (complete with a spark visible through the gaps in the casing!) whenever a certain switch was closed.

Just think of all the hours of fault chasing that could be eliminated if lighting was always "wire up like colours"? That could meaningfully ease the burden of the current electrician shortage and thus meaningfully advance the goal of all wiring regulations - effective, affordable and safe electrical installations for all.
We now live in the era of smart technology, I think we should fully embrace it to gain its benefits. If we don't specify smart controls we will have serious problems with lock in, and the inevitable attempts to turn literally everything into an "app".

Also we could have the remotely switched sockets that the Americans always sing the praises of! Without the problems of their American implementation.

Whilst I agree that such systems are somewhat beyond the scope, I think the best way to avoid a zoo of incompatible equipment, with all the problems that causes for the householder, would be to specific a communications standard, if only a physical layer, to be used.

If we go with a DALI compatible physical layer, as an example, the wiring regs could reasonably impose a pair of thin data cores into cables in the same way they impose an earth conductor. It is a bit beyond the traditional view of BS7671, but if we can redesign wiring regulations (and their related regulations like the socket regs), we could impose such things if we wish.

I think smart controls using a single unified spec an excellent way to rationalise the current arrangements for lighting and the like that cause such issues in the field. If we have digital control of lighting just written into the specification, as an example, new installations never have to worry about the pain of two or three way switching and the confusion it can cause.

Such things are often referred to in the context of confusing DIYers, but during recent work in my house an apparent professional electrician managed to rewire the lighting to put a dead short between a switched live and a neutral, so the breaker immediately tripped (complete with a spark visible through the gaps in the casing!) whenever a certain switch was closed.

Just think of all the hours of fault chasing that could be eliminated if lighting was always "wire up like colours"? That could meaningfully ease the burden of the current electrician shortage and thus meaningfully advance the goal of all wiring regulations - effective, affordable and safe electrical installations for all.
We now live in the era of smart technology, I think we should fully embrace it to gain its benefits. If we don't specify smart controls we will have serious problems with lock in, and the inevitable attempts to turn literally everything into an "app".

Also we could have the remotely switched sockets that the Americans always sing the praises of! Without the problems of their American implementation.