Maximum demand on EICR

The other way of tackling this problem is to use actual maximum demand data for the installation - get the customer to obtain their half-hourly  import metering data from their supplier, then you can use that to determine the actual highest average demand over a half-hour period, which might be a reasonable representation of the demand in thermal terms.

Potential weakness of that approach is it doesn't tell you the maximum instantaneous demand, or MD over shorter periods, or what the maximum demand might be under a different occupier (more people in household, different habits/routines affecting diversity etc). 

But at least it is based on real, accurate data - settlement metering is accurate to within a couple of percent. Only fly-in-the-ointment is where there's solar/battery installed which then needs some more difficult analysis adjusting for solar/battery energy flows.

I'm surprised there's not a more formal, defined process for how Maximum Demand should be determined because with it being inherently probabilistic, rather than purely deterministic, it opens up the space for significant differences in how different people might determine the value, the electrical equivalent of 'how long is a piece of string?'.

TurboGen said:

The other way of tackling this problem is to use actual maximum demand data for the installation - get the customer to obtain their half-hourly  import metering data from their supplier, then you can use that to determine the actual highest average demand over a half-hour period, which might be a reasonable representation of the demand in thermal terms.

Interesting idea, but would that not require a smart meter?

Moreover, presumably, it is the electrician's job to get the info.

TurboGen said:

I'm surprised there's not a more formal, defined process for how Maximum Demand should be determined

In which case, the term is pretty meaningless.

AJJewsbury said:

30A per 1000ft² (or 32A per 100m²)

TurboGen said:

what the maximum demand might be under a different occupier

That formula puts us well beyond the rating of the DNO's fuse, but there are only two of us, so most rooms are unused most of the time.

Of course that could change markedly when we sell/leave feet-first.

Chris Pearson said:

Interesting idea, but would that not require a smart meter?

Moreover, presumably, it is the electrician's job to get the info.

Yes but most homes already have smart meters and ultimately all homes will have them, so I don't see that as a problem in the longer term.

I don't know what it's like with all suppliers, but some already have a function for the customer to download their half-hourly data, so it's a simple job to then just find the maximum value, or rank them to find the top X half-hour periods.

Either way, now that smart metering is available and half-hourly data from them is increasingly available, it seems only logical that the Wiring Regulations ought to reflect this to make more accurate, data-driven assessments of MD, 

TurboGen said:

Yes but most homes already have smart meters

It's still only at 70 % in the UK I believe. Technically "most" but still around 30 % that don't.

TurboGen said:

ultimately all homes will have them, so I don't see that as a problem in the longer term.

That will almost certainly be the case, because even though there's no legal requirement (yet), the fact is that meters have a finite life, and the options for a non-SMART meter, even for companies that have some older technology that's still in date, will vanish in the future.

TurboGen said:

the Wiring Regulations ought to reflect this to make more accurate, data-driven assessments of MD

BS 7671 does not specify what means is used to identify 'maximum demand', so I don't see a need to change BS 7671.

I would, however, bring out the following points:

1. If the data is not available over a sufficient period, covering when peaks in usage are anticipated at different times of year, then that needs to be taken into account by the designer.
   
   
2. Smart meter consumption is recorded as an average over a time period, which could be half an hour (half-hour values), or could be as little as 10 minutes (10-minute values). These values are likely to be adequate to cite the 'maximum demand' required by the distributor for their stated 'maximum demand', but they are not necessarily adequate to rate a circuit, specify a protective device, or rate a consumer unit, etc, according to BS 7671.
   
   
3. Some meters only record 'direct feeding' power, i.e. that supplied by the grid to the installation. And even if the meter records both 'direct feeding' and 'reverse feeding' power, it won't tell you whether there are loads in the installation that are consuming when the installation is reverse feeding.  If there are local generation sources (PV, battery, V2G) then meter data alone (regardless of where the meters are) won't necessarily permit you to see where power is going in the installations ... i.e. if there are certain 'hot spots' of thermal load ... but see also second point above.
   
   
4. Existing data is great, and better than an estimation of usage, but if you are modifying the installation, or tuning EEMS functionality, you still need to calculate the delta of what the new loads/generation or settings will do ... taking into account seasonal usage also !

So, in summary, I think:

(a) BS 7671 has appropriate requirements in its current form;

(b) the main smart meter might be able to be used to provide the 'maximum demand' for the distributor (taking into account any changes being made to the installation, see point 4 above); and

(c) meter data alone can't be used to rate circuits and equipment in the installation according to BS 7671 ... there's more than one 'maximum demand' for an installation ... or at least different parts/components forming it.

That feels like a statement that "maximum demand" is too poorly defined to be of any meaningful use.  Not only is it not defined temporally (1msec, 1sec, 1min, 10min 30min, 1hr, 2hr, 5hr?) bit it is also not defined positionally in an installation.  How is anyone expected to enter a figure on any document? 

When I was having a discussion with the DNO about MD for my own property I had the ability to provide a graph of maximum RMS current vs. duration which seemed to be a lot more meaningful but very few people will have that granularity of metering.

Just to add, note that some DNOs explicitly permit a short term demand of up to 100A on installations fitted with an 80A fuse - who knows what an electrician is expected to write on an EICR in that case...

Just to add, note that some DNOs explicitly permit a short term demand of up to 100A on installations fitted with an 80A fuse - who knows what an electrician is expected to write on an EICR in that case...

Given the fusing curves, they don't really have a choice, if you think the company fuse is the only current limiting object they have at their disposal. 

Actually its not a bad fit to reality in terms of overheating meter tails and things either - the cable does not instantly combust at a 20% overload, rather it slowly heats up over several minutes eventually reaching  a new equilibrium temperature, that is a bit above further above ambient, than that intended for meeting the full lifespan prediction in an ambient of 30C. That time constant is very long for transformers and buried cables and shorter for things with less thermal mass.

If the ambient is below 30, or the overload ceases before it even reaches that state, it may not be even stressed in terms of life reduction.

It is after all quite possible to wire a shower in 2.5mm2 cable and see it last for many years, despite the horrendous overload, on paper,  until someone visits who takes more than ten mins at a time in the shower ;-) 

Regards Mike.