What is the logic for solar sub boards avoiding overload of main board bus bar

Does it not depend to some extent upon where the solar is connected?

On the naive assumption that your 100 A fuse will blow at 101 A, If you have a 4 kW solar array connected upstream of the DB, you could potentially draw 117 A through your main switch and proximal busbar. However, we all know that you could draw 117 A from the grid, at least in the short term (if you can afford it). Presumably manufacturers take this into consideration, so your nominal 100 A board will cope with more.

What the board cannot do is draw the slight overload from the grid plus the solar output, so the board's rating needs to be observed.

If you feed the solar in at the distal end of the busbar, only the nominal 100 A goes through the mainswitch and somewhere in between there is no flow in the busbar, but as Alan says, there remains the question of heat dissipation in the board as a whole.

I suppose that in this instance, diversity is your friend, but I would be very wary of knowingly allowing a DB to be overloaded - it might catch on fire.

If you are going to have two separate CUs then for this to work, you need at least three fuses, one of which will be the DNO fuse, arranged in a Y.

If we assume that the DNO supply comes in at the bottom of the Y then you will have a (in this case) 100A fuse there.  You will also need a 100A fuse from the centre of the Y to the main CU (say at the top left) and you will need a fuse, rated at a maximum of (100-max solar current) for the second CU going to the top right of the Y.  The wiring between the fuses will need to be rated at the sum of the DNO fuse and the solar inverter rating.

Basically, each CU needs to be on its own fuse (just install a standard switch fuse before each CU).

Note that the wiring between the fuses will include the service head to meter and meter onwards tails, there may be those who would argue that these need to be protected, in which case you need a fourth 100A fuse in series with the DNO one, I would argue that the DNO fuse is sufficient to protect these tails (assuming that the other conditions applicable for the use of the DNO fuse to protect tails are met).

If you go over a potential flow of more than 100A through a CU then it is your job to do the engineering to confirm that it is acceptable - unless you happen to have insider access to the design arguments that the manufacturer uses to underpin a 100A CU rating, you are unlikely to be able to do that in a way that is considered "competent".

DShutt said:

If you are going to have two separate CUs then for this to work, you need at least three fuses, one of which will be the DNO fuse, arranged in a Y.

If we assume that the DNO supply comes in at the bottom of the Y then you will have a (in this case) 100A fuse there.  You will also need a 100A fuse from the centre of the Y to the main CU (say at the top left) and you will need a fuse, rated at a maximum of (100-max solar current) for the second CU going to the top right of the Y.  The wiring between the fuses will need to be rated at the sum of the DNO fuse and the solar inverter rating.

Basically, each CU needs to be on its own fuse (just install a standard switch fuse before each CU).

Note that the wiring between the fuses will include the service head to meter and meter onwards tails, there may be those who would argue that these need to be protected, in which case you need a fourth 100A fuse in series with the DNO one, I would argue that the DNO fuse is sufficient to protect these tails (assuming that the other conditions applicable for the use of the DNO fuse to protect tails are met).

If you go over a potential flow of more than 100A through a CU then it is your job to do the engineering to confirm that it is acceptable - unless you happen to have insider access to the design arguments that the manufacturer uses to underpin a 100A CU rating, you are unlikely to be able to do that in a way that is considered "competent".

In my opinion this suggestion goes way beyond anything that would be considered as a reasonable level of complexity for a single home installation.

To avoid this I would certainly be going down the route of using a CU manufacturer that states a 125A load is acceptable even if I am not a fan of their products. If the solar is connected to the main CU the tails and main switch will not see an increased load beyond what the 100A main fuse lets through.

I am probably more interested in the debate about does current guidance make sense; to me from the perspective of max current through busbar and switch connection to the main board is more logical. But thermally it may not be.

I agree the point of having a separate CU is to allow for overload protection for the original CU - I reckon you don't necessarily need a fuse for every CU though - just the ones liable to overload (e.g. where the sum of the outgoing devices exceeds the overall rating) - so a dedicated PV CU with say a 16A device in it (or a PV & EV one with say a 16A and 32A or two 32A) can't draw (or supply) excess of that CUs rating, so the fuse can be omitted,

   - Andy.

Just found this.

https://electrical.theiet.org/wiring-matters/years/2024/103-november-2024/how-does-the-installation-of-microgeneration-affect-the-rated-current-of-a-consumer-unit/

Will need to read it and think about it more later.

I was going to add that the Irish seem to have a policy of using a DP MCB as their incomer - which does nicely provide overload protection without needing a separate switchfuse ... I'm not sure if does anything for discrimination (selectivity) through ... it would be a bit of poor result to have the "main switch" trip out for almost any fault in the installation (as discrimination between MCBs is almost non-existant, regardless of rating).

   - Andy.

Alan B said:

To avoid this I would certainly be going down the route of using a CU manufacturer that states a 125A load is acceptable even if I am not a fan of their products.

125 A may become the new normal, but I am bound to wonder where all the leccy will be coming from.

According to Ofgem, the typical annual consumption of a medium user is 2700 kWh/year = 7.4 kWh/day. Make the sums easy and assume that is consumed over 7.4 hours and we end up with an average consumption of 1 kW.

The figures allow about 4 times as much for gas, so if that is replaced by electricity, we still get only 5 kW or 22 A.

Alan B said:

Just found this.

Well worth reading. :-)

The figures allow about 4 times as much for gas, so if that is replaced by electricity, we still get only 5 kW

Hopefully we won't be installing too much resistive electric heating - with heat pumps the COP should be at least 2 even in the worst case conditions (higher temp for tap water and low outside temps) so the overall energy figures for gas replacement shouldn't be even that bad.

Instantaneously (or even over a few hours that would mean thermal stability for smaller conductors) things might not be quite so nice through - my heat pump tries to concentrate it's running to off peak hours (early hours of the morning and early afternoon) - EVs (not covered by the gas figure of course) will likely draw significantly more than 5kW (more like 7.2lW min -maybe double that for 2-car households?) and again will tend be concentrated to the start of off-peak periods.

On the other hand high load households aren't a new phenomenon - we've had all electric storage heaters + immersions since probably the 1950s - so we should be able to cope with adding a modest amount of local generation into the mix.

   - Andy.  

Where the lecky is coming from will be a question for the DNO.

Before someone askes where's it going:-

A fairly big heat pump

A substantial aircon/heating device supplementing the ASHP which isn't big enough for a fairly large poorly insulated house.
Additional immersion to supplement the ASHP

Double oven, large electric hob with diversity
Normal white goods with some diversiy

Add a bit more for normal stuff like lights, TV's, computers atc and I am at 80A. Only take someone to plug in a hot tub or something and I could be approaching 100A. Maybe a bit over cautious.

AJJewsbury said:

so the overall energy figures for gas replacement shouldn't be even that bad

Quite so, but even with worst case, the DB doesn't get overloaded.

AJJewsbury said:

EVs (not covered by the gas figure of course) will likely draw significantly more than 5kW (more like 7.2lW min -maybe double that for 2-car households?)

EVs are certainly not the panacea that the Government seems to think.

12,000 miles seems to be a reasonable annual mileage and at 4 mi/kWh (please forgive the non-SI units) that is 3000 kWh/year, so your medium consumption household has doubled its leccy.

If say, two-thirds of the mileage is commuting, and assuming a slightly onerous 240 day working year, that is 50 miles/day = 12.5 kWh or a couple of hours on a 7.2 kW EVCP. So your 2-car household could swap over, or charge for 4 hours on alternate days.

However, I do take the point about a 2-car EV household.