Current transformers

Also worth considering that, without a resistor or short at the current transformer, the voltage will not be maintained in the SELV range.

You can buy current transformers that are guaranteed SELV, but they are very expensive.

For other current transformers, these need to be treated as FELV at least, if not LV, and preferably go in an enclosure. Similarly, insulated wires not in containment or sheath wouldn't meet BS 7671 requirements.

Only mentioning this, as current transformer installation is ramping up with PV inverters and EV charging equipment ... but not always installed according to BS 7671 requirements because the actual potential voltage range is not considered.

DShutt said:

Ideally we would have rolled out a simple standard interface giving basic electrical data (voltage, current, power, etc.) as part of the smart meter roll-out, completely avoiding the need for clip on CTs (a great application for cheap plastic fibre optics, since only uni-directional communication over relatively short distances is required).

For energy management, it's important to be able to look at import and export current as well as voltage.

There are alternative approaches, where an energy monitoring unit (that has CTs and VTs built in) connects in-line with the tails, and communicates via CANbus/Modus back to the energy management system (e.g. inverter).

DShutt said:

Consumer CTs intended to clip round meter tails with whatever protection their suppliers deem appropriate have been out there for over 20 years (at least since the first home energy monitors appeared in the 1990s) and the fitment rate is taking off rapidly with the proliferation of EVSE and PV. 

Agreed ... it's those I'm talking about, not all are installed correctly (and as discussed above, perhaps some are not installed in accordance with BS 7671).

gkenyon said:

The device you linked to has an audio jack on it, so only one fault (failed resistor) is necessary.

Fair point. Of course wire-ended versions are available too. Interestingly Victron seem to have switched from plugs to hard-wired in their recent models - they might have had similar thoughts.

   - Andy.

as part of the smart meter roll-out, completely avoiding the need for clip on CTs

Except that sometimes what's meant to be measured doesn't necessarily pass completely and individually through the supplier's meter - e.g. total generation for PV or consumption for specific load (heat pump or EV). Mind you extending the smart meter system to PV generation meters might not have been a bad idea...

    - Andy.

DShutt said:

Ideally we would have rolled out a simple standard interface giving basic electrical data (voltage, current, power, etc.) as part of the smart meter roll-out, completely avoiding the need for clip on CTs (a great application for cheap plastic fibre optics, since only uni-directional communication over relatively short distances is required).

That's very close to what the French did, their 'Linky' smart meter, used for the French smart meter roll-out has a transmit-only serial interface, TIC (Tele-Information Client), albeit wired rather than fibre optic, providing the various meter measurements along with a range of smart control signals that can be used to automate different loads, for switching things like EV chargers, space heating, water heating inline with time of use tariffs. The meter measurements can be used by smart EV charge points to dynamically adjust the charging rate, to avoid overloading the incoming supply.

Further, the French make the TIC interface available to the end consumer, so when the electrician is installing new loads, like an EV charge point they can connect up the TIC comms to the charge-point as part of the installation, so it is controllable and can work with TOU tariffs from day 1.

I take my hat off to the French engineers that designed it, it's a very well thought through, logical solution to the problem. As for the British smart metering system.....

AJJewsbury said:

Except that sometimes what's meant to be measured doesn't necessarily pass completely and individually through the supplier's meter - e.g. total generation for PV or consumption for specific load (heat pump or EV). Mind you extending the smart meter system to PV generation meters might not have been a bad idea...

And not all supplier's meters are 'bidirectional', i.e. don't measure 'export' (or 'feed-in', 'reverse feed' or whatever term is in vogue).

I understand SMETS2 and later are supposed to though, so it's coming.

But yes, load control and some energy management system functions need to know what current is on certain circuits, or groups of circuits, rather than the whole installation.

I haven't looked at the specs for the French system but from what you have said it sounds very sensible.

I did read the the SMETS2 spec and have concluded that it is a massive missed opportunity.  It was clearly written based on a very narrow set of requirements, purely around metering.  For example, there is no requirement for a SMETS2 meter to provide details of the current measured at all (only power and voltage but not not necessarily corresponding instantaneous values are available).  If we are going to install a digital current meter in every home, not providing any access to that data is quite frankly ridiculous.

To those who say "just install a separate shunt" then remember that this is a domestic setting with electricians used to working with domestic supplies.  I would be quite happy mounting a din rail Modbus based digital power meter in a suitable enclosure and wiring the necessary 25mm sq cable to it but (a) those are more high current connections which can potentially cause fire if poorly made, (b) making such an assembly is probably outside the (assessed) competence of most domestic electricians, (c) it takes up space and (d) it is extra unnecessary cost.

In the vast majority of cases, what needs to be measured (certainly in the clip on CT case) is the whole house load.

Are there any current green funding initiatives which rely on measuring what is generated rather than what is exported?  Historically, where measurement of generation has been required for subsidy purposes, the requirement has been for a meter where the accuracy / tamper resistance requirements preclude clip on CTs as part of the measurement arrangements.

Clip on CTs are generally used for load / export limiting rather than accurate metering.

SMETS2 meters will cope with export quite happily, which makes it all the more frustrating that the specifiers of those meters omitted the ability to avoid potentially dangerous CTs.

DShutt said:

In the vast majority of cases, what needs to be measured (certainly in the clip on CT case) is the whole house load.

That could be challenging when the power is coming into both ends of the busbar.

Isn't the requirement not to export until the battery is full, and not to import until it is empty. So the inverter balances the voltage so that there is no current in the consumer's tails until there is a surplus or none left respectively. It does not have to be as precise as a meter.

If you do not have a battery, is there any need for this monitoring?

DShutt said:

In the vast majority of cases, what needs to be measured (certainly in the clip on CT case) is the whole house load.

There are other installations than domestic.

DShutt said:

lip on CTs are generally used for load / export limiting rather than accurate metering.

load curtailment can be applied to a circuit ... or an installation.

I agree that in the case of domestic EV charging installations, where the existing demand is large, or the supplier might take some extensive period to uprate the fuse, the quick and easy method used is load curtailment.

Similarly, clip-on current transformers offer a quick and easy method for export limitation, and probably all that's needed for energy management.

DShutt said:

SMETS2 meters will cope with export quite happily, which makes it all the more frustrating that the specifiers of those meters omitted the ability to avoid potentially dangerous CTs.

Yes, the reason for mentioning this was people often ask why we don't use the smart meter for some of this ... why is a separate CT necessary. There are, of course, many reasons for this, including GDPR and information security, in the layers of the smart metering protocols ... and as mentioned, earlier versions of SMETS being 'import only' ... which may be OK for the EV case, but other difficulties (and hence costs) also get in the way.

Chris Pearson said:

Isn't the requirement not to export until the battery is full,

This doesn't cover the complexities of schedule based pricing. There is a competition between the middle-men/women and the users as to who gains the benefit of the overall grid variable pricing. Users with large battery stores want to charge while cheap and feedback while costly, much to the chagrin of the "eco-suppliers" offering loss-leader overnight pricing.

Even with a simple Solar - Battery system (8 panels, 5kWh) and a Gas tariff (cheap 4h on Sundays), I'm being asked to 'flex' the charging regimes .

Maybe all the RCBOs need a second CT for just the supply side (rather than the L-N difference), with cheap consumer units pre-populated with 100% monitored circuits.... Brave New World?

Chris Pearson said:

Isn't the requirement not to export until the battery is full,

This doesn't cover the complexities of schedule based pricing. There is a competition between the middle-men/women and the users as to who gains the benefit of the overall grid variable pricing. Users with large battery stores want to charge while cheap and feedback while costly, much to the chagrin of the "eco-suppliers" offering loss-leader overnight pricing.

Even with a simple Solar - Battery system (8 panels, 5kWh) and a Gas tariff (cheap 4h on Sundays), I'm being asked to 'flex' the charging regimes .

Maybe all the RCBOs need a second CT for just the supply side (rather than the L-N difference), with cheap consumer units pre-populated with 100% monitored circuits.... Brave New World?