Domestic consumer unit rating with PV and battery storage.

Having read the COP on  Electrical Energy Storage Systems and  completed the IET course on the same subject I had a query regarding the rating of domestic consumer units and switch gear which I addressed by email to NICEIC technical. I also came across an older discussion on this forum but am still no closer to a definitive answer. I've included my findings and would welcome constructive input. 

Post by GKenyon in previous thread

Because an EESS charges the battery as well as as discharging it, you will need to check the rating of the CU is not exceeded. For example, if the CU is rated for 100 A, and there's a 100 A service fuse, and a 16 A output battery storage system - by feeding 16 A in at one end through an OCPD, because that OCPD gets hot it contributes to the total heat load in the CU - therefore the CU should be rated for 116 A.

My question to NICEIC.

Hello
Please can you help with the following.
Domestic installations with PV and/or battery storage.
551.7.2 Where the generating set is connected to either the main consumer unit or via a separate consumer unit via Henley blocks the rating of the consumer units shall be protected by a OCPD InA≥In+Ig(s).
Where In = 100A DNO fuse and Ig(s) = 16A MCB or 2 x 16A MCB's which would be 116A or 132A, what inspection code should be given on an EICR where a standard domestic consumer unit is fitted which has a rating of 100A. Can any allowance be given on connected load being less than 100A or as the regulation relates to the rated current of the assembly and is a "shall" requirement does the load have no influence on the code assigned. 
Answer from Certsure

The Certsure Technical Helpline provides general information and guidance for compliance with the British Standard BS 7671, the Requirements for Electrical Installations, and matters concerning electrical safety within electrical installations designed, constructed, inspected, and tested to BS 7671. Without detailed knowledge of your installation, we cannot offer advice specific to your installation and can only generically provide comments based on the information you have provided.

The intent of the regulation is to ensure that the assembly is not overloaded with the additional generating set, as the main fuse may not protect the assembly if for example the internal busbar is pulling 116A.

Regulation 536.4.202 states: see regulation

From the viewpoint of an EICR, we would be looking for evidence that the assembly is being overloaded, such as burning, distorting and the likes.

The above regulation allows for diversity to be taken into account, so we can exercise our engineering judgement in declaring whether or not the assembly is suitably protected.

We trust that we have answered your current question; however if you require any further information or clarification, then please do not hesitate to contact us either by e-mail to helpline@certsure.com or by telephone on 0333 015 6628

I've read 536.4.202 and am interested on your views on the last paragraph with the shall requirement and how this ties in with the answer given by certsure. 536.4.3.2 is also relevant but has not been mentioned in the reply.

Thanks for your time.

  • If the unit can charge the battery using grid power, then in my view no special consideration is needed. The battery charger is simply another load like say a couple of fan heaters. A 100 amp service fuse will limit the sustained total current to 100 amps.

    If the equipment can supply the load by use of the solar power AND grid power at the same time, then it seems to me that the total load could be 116 amps and that the consumer unit should be rated at 116 amps or more.

    If the equipment can supply the load from solar power AND the battery AND from grid power, then it seems to me that the total load could reach 132 amps and that the consumer unit should therefore be rated at 132 amps or more.

    In practice I doubt that it matters much for several reasons. Firstly the overload resulting from use of solar AND/OR battery power as well as grid power will be fairly brief as the sun will go in or the battery run down. No one worries about a 100 amp consumer unit loaded short term to 116 amps, which the 100 amp cut out fuse will pass for hours.

    Also, if the backfeeding MCBs are installed at the end of the busbar distant from the main switch, then nothing will be overloaded in any case. Neither the main switch, nor any part of the busbar can be loaded to more than 100 amps, so why worry.

    However my view that it probably does not much matter counts for very little. A STRICT reading of regulations and guidance thereon would seem to require a consumer unit rated at more than 100 amps as described above.

  • If the equipment can supply the load from solar power AND the battery AND from grid power, then it seems to me that the total load could reach 132 amps and that the consumer unit should therefore be rated at 132 amps or more.

    Although you wouldn't have such an arrangement in the UK on a simple domestic under G83/G98 as there's the 16A/phase limit on embedded generation. If you went down the G59/G99 route you could have more but then you'd have all sorts of extra current sensors and relays that could limit the totals anyway.

    If the equipment can supply the load by use of the solar power AND grid power at the same time, then it seems to me that the total load could be 116 amps and that the consumer unit should be rated at 116 amps or more.

    Agreed. In fact there was a suggestion for CUs to be rated at 116A for that very reason several years ago -  New concept in consumer units? (see small PDF attachment to the original post) - but real CU manufacturers don't seem to have taken up the idea.

    Also, if the backfeeding MCBs are installed at the end of the busbar distant from the main switch, then nothing will be overloaded in any case. Neither the main switch, nor any part of the busbar can be loaded to more than 100 amps, so why worry.

    The was my original argument too. There is another argument however - since the SSEG current can pass through twice as many MCBs as grid current (i.e. the one for the generator circuit as well one for whatever load it supplies) and MCB thermal elements naturally produce some heat when loaded, then the overall amount of heat produced by the MCBs is higher than if there was just 100A coming from the grid - so the overall assembly could in theory be subject to overheating even though no individual section of it carries over 100A. My gut feeling however is that the extra heat would be minimal - and other arrangements - i.e. the Irish approach of using a DP MCB for in the incomer - would produce far more heating effect in an otherwise similar arrangement without apparently suffering any problems.

    551.7.2 is still in need of an overhaul to my mind - originally it only addressed final circuits, now switchgear has been added, but distribution circuits are still conspicuous by their absence (although maybe 551.7.1(i) is meant to be a catch all). 551.7.2 is at the same time over-zealous demanding the entire circuit meets Iz ≥ In + Iz even though in many topologies (e.g. broadgage's example of feeding a conductor from opposite ends) overloading of parts is impossible anyway.

       - Andy.

  • You are perpetuating a myth Broadgage. A 100A fuse will never blow with 116 amps passing, and the protection current is considerably more than 100A. I regularly come across "electricians" who do not understand circuit ratings at all, and assume that fuses fail at their nominal current rating. They also assume that minor cable overloads will immediately cause melting or a fire, and even that ring finals are inherently "unsafe". I do not understand why this should be, but taking fig 3A1 in BS7671 as an example, a 5A BS88-3 fuse will carry 8A for an indeterminate period, as you see the graph does not continue because there are simply too many unknowns to the conditions. Given a 20C ambient temperature I would expect it to never blow, as the power dissipation would not raise the fusebox temperature significantly. You will notice that a 100A fuse would carry at least 150A for a similarly long period, and in a cool environment for a significant period of time, probably at least 10,000 seconds or 3 hours. The 100A rated CU could therefore be subjected to 150A, but this is unlikely to cause a failure except possibly if the main switch is expected to break this current. Given long enough the CU temperature could rise significantly, and most domestic CUs are probably not really rated for the high ambient temperatures of some switch-rooms with inadequate ventilation.

    To the OP, I agree with Andy and Mike, this is not a real or particular problem, and given normal load diversity it will not occur except possibly for insignificant periods. Domestic CUs with ratings over 100A are not available, and the switch to commercial or industrial switchgear with higher ratings is significantly expensive. Essentially you are looking for a problem where none exists, unless there is some very particular and strange domestic installation. Electric heating along with a car charger could change this, but such in one CU is probably unknown, as electric heating is usually off peak storage on cost grounds.

  • Domestic CUs with ratings over 100A are not available

    They might need to be soon, if we move to all electric heating and cooking as well as having to address EV loads.

    and given normal load diversity it will not occur except possibly for insignificant periods.

    Again, what is "normal load diversity"? Many of the assumptions we have traditionally made regarding what a "domestic load" looks like are not valid with all loads now being installed.

    For example, it has become commonplace for EV charging points to be set up for load curtailment at the rating of the distributor's cutout (or, an assumption made that this is 60 A). It could well be, that, with heat pumps, there are now long periods when the demand is way above what we were used to, for a number of hours in a row, and with other appliances on, well, who knows ?

  • taking fig 3A1 in BS7671

    Be wary of those graphs when looking for non-operating times - as with any device there's a significant band of acceptable operating times for a given current and the BS 7671 charts only give the slowest edge of that range. For most BS 7671 purposes (e.g. ADS or fault protection) where you want to make sure that the fuse definitely does open within a given time for a given current that's just what you want, but always be aware that it might operate significantly faster in many cases.

    It's the same with MCBs - generally for small overloads we can only say that somewhere between 1.13x and 1.45x In they'll open within an hour.

      - Andy.

  • and there's a 100 A service fuse

    From the perspective of EICR, do you know the service fuse is actually 100 A?

    So, from a coding perspective, you'd need to be careful. EICRs ought not to record assumptions as facts.

    the rating of the consumer units shall be protected by a OCPD InA≥In+Ig(s)

    "Which consumer units?" is a good question.

    Regulation 551.7.2 addresses only the assembly through which the generator (inverter or storage system) connects. It takes no account of the diversity of the loads connected to that consumer unit, only the protective device at its input, or upstream of it, along with the output rating(s) of the generator(s) connecting through it.

    Regards the other CU (if you have two and split the tails via suitable connection blocks) this needs to be protected according to instructions from the manufacturer (which may or may not take into account upstream protective devices, and therefore you may or may not need to take into account the contribution of the generator(s) upstream of this other (loads only) CU.

    I think Regulation 536.4.202 is the correct reference for the "other" regulation covering this, statement regarding overload protection being the last para of the Regulation:

    However, overload protection shall not solely be based on the use of diversity factors of the downstream circuits. To achieve overload protection of RCCBs or switches, the rated current of the OCPD shall be selected according to the manufacturer’s instructions.

    So, clear as mud really ... a lot hinges on the manufacturer's instructions (apologies to   for adding that) and their applied assembly ratings, according to BS 7671.

    Sadly, I know that doesn't make it an easy job to assess for EICR (it's not a design job; however, things are always difficult when BS 7671 changes design premises like this) and therefore the only thing you could go off is evidence of overload or clear disparity between the distributor's cutout and the rating of an assembly.

  • Perhaps there will be a need in the future for more than one domestic consumer unit, rather than assume higher rated units will be available. The current ranges are quite economical to purchase, so why increase cost by going outside the norm? We used to have at least two consumer units years ago, one for the 24 hour supply and the second for off peak storage heaters and water heating. Could we do that again?

    Z.

  • Perhaps there will be a need in the future for more than one domestic consumer unit, rather than assume higher rated units will be available.

    That would work, but only if you install another overcurrent protective device (switch with integral fuse probably - see note) before new CU, so the sum of the ratings of the generators plus the nominal rating of the overcurrent protective device, is less than the current rating of the CU connecting the generation.

    Note: The overcurrent protective device prior to the CU connecting the generators would have to be stand-alone, in its own enclosure, and not in a CU either!

  • A far nicer solution, as the DNO  will have to re-cable the street anyway, if the predicted EV and heat pump loads come to pass, will be to go continental, not perhaps in the sense of coffee and croissants, but to make 3 phase supplies to the house more common.  3 phase boards with a decent current rating are already available.

    In the meanwhile the load cutting car chargers and so on are a good stop-gap and the same idea may extend to other loads so the heating goes off during peaks of cooking or something.

    For now the risk is very much on paper. A CU that is really good for 100A all day in all possible domestic installation situations, probably does not exist right now, and if it did then it would probably also be good for 110A in all but the warmest.

    Mike,

  • A far nicer solution, as the DNO  will have to re-cable the street anyway, if the predicted EV and heat pump loads come to pass, will be to go continental, not perhaps in the sense of coffee and croissants, but to make 3 phase supplies to the house more common.  3 phase boards with a decent current rating are already available.

    I wouldn't disagree with that, but perhaps a consideration is that it would be recommended to use three-phase inverters in domestic situations ... or over-rate the Neutral [in the supply main and upstream of the connection points of inverters] by roughly the sum of all inverters on the phase with the largest current output from inverters. The reason being, if the inverters are exporting on a phase with little load, and the other two phases have a lot of load, consider the vector sum of the Neutral currents