Snug Octopus (Economy 7) Problems

Qyburn

Looking at the original issue, where conventional dumb storage heaters are meter controlled, they will virtually always come on and thus draw full load the moment the circuit comes on. Hence the randomised delay. But they'll switch off when they're "full" at varying times afterwards depending on their capacity, settings and the inside temperature. So what I'm getting at is that unless the NSHs are seriously undersized they mostly won't be charging during that last half hour. 

Chrysalis

Does this only affect E7 or other TOU tariffs as well, its a bit alarming, as although meter controlled devices will still come on at the right time, I do turn on/off devices the minute the clock changes at the time boundary.  If the offset is only set at the time of the tariff setup (not randomised each day), then should the customer be told the exact offset so they know when its safe to use heavier devices?

Phones4Chris

Qyburn said:

Looking at the original issue, where conventional dumb storage heaters are meter controlled, they will virtually always come on and thus draw full load the moment the circuit comes on. Hence the randomised delay. But they'll switch off when they're "full" at varying times afterwards depending on their capacity, settings and the inside temperature. So what I'm getting at is that unless the NSHs are seriously undersized they mostly won't be charging during that last half hour. 

Whilst the switch off when "full" at varying times is most likely the case for the overnight off-peak block except perhaps in extremely cold weather conditions - that's what's happening with mine - (whether split timing or not), I'm not sure that it's likely to be the case for the afternoon boost as most of our older NSHs will have discharged a fairly good amount of their capacity and so are likely to charge for the whole of the boost period.

Phones4Chris

@Chrysalis I'm fairly sure that most suppliers don't know the "offset" for individual customers. It's determined by the DNO and whilst it is fixed at the same time each day for the overnight night rate block, it depends on how it's determined. It might be based on "Load Block" group (ie. ESEC blocks - the rota for Disconnection when electricity is in short supply) but it might be determined down at Sub-Station level. I'll see what I can find out - unless there's someone working in the industry on here that knows the answer.
What may happen with the afternoon boost, as I get the impression that the timeslot can vary on Snug on different days, I can only guess, but one might expect the "same offset" to be automatically incorporated in the ALCS programming - it would be nice to discover exactly how it's done.

Other TOU tariffs are a different kettle of fish!

(Thinking more as I've been typing, IF it's determined by "Load Block" group, then data ought to be readily available to suppliers as both the "Load Block" (it's on the bill) and the Region are known. Even if the "offset" for a given Block may vary from region to region (although the ESEC timings do not) information should still be readily available from DNOs!).

Scot_39

Qyburn said:

Looking at the original issue, where conventional dumb storage heaters are meter controlled, they will virtually always come on and thus draw full load the moment the circuit comes on. Hence the randomised delay. But they'll switch off when they're "full" at varying times afterwards depending on their capacity, settings and the inside temperature. So what I'm getting at is that unless the NSHs are seriously undersized they mostly won't be charging during that last half hour. 

Except in the afternoon slot - the 3pm-4pm - the OP is most concerned about.

Where many in winter will be after the heaters - especially older lossy ones - have been releasing overnight charged heat for 8.5 hours plus (06:30 - 3pm) between charges (and where the OP's complaint lies - about paying for his 10min at peak rate). 

As that is half the nominal 17 hr discharge period - on block e7 - and the hottest part of it of course for a lossy style heater without any dynamic air vain / flow type control - chances are many could be capable of taking up half the nights charge by that period - so likely on cold days far more than the full hour.

I'd really like to see the kWh graphs rather than just the costs in £0.xx p.

The OPs time graphs suggest their heaters had given up early overnight - the last 2 green time slots are background type level - but may have been far closer in fact had they not started 1 hr (they show a large power usage reading - I assume Octopus app reflects the actual day - so that last green bar is 23:30 to 23:59:59 etc - and that is use genuinely prior to midnight) early - but later at the FAQs specified supposed 0030 start time too.  As shows an albeit tailing off non background type load for 12 half hour slots - albeit with potentially one or more heaters giving up before the 12 slots - and probably the last one early on during the final use slot. 

Although the OP appears to have 14 green time slots - the spec is only 6 hours - already 1 hr shorter than e7 - overnight.

Although I have often wondered - apart from switch on from cold - how many NSH even on max woulld charge the full 7 hours continuously - or even the 6.

On the reddit thread link above - 2 of the users have for instance suggested they set their time to 2.5 hours of the 6 - but then that appeared to be split on their 1/2 hour use charts - into 5 charge periods. 

And in the original FAQ - there was reference to the ability to set a target "home hot" time - which may or may not have gone to avoid many specifying the end of the 06:30 - perhaps reducing Octopus's ability to optimise at cheapest rates.

QrizB

Until @worlestone checks their bill and answers this question:

QrizB said:

worlestone said:

But it's not, for us it's 50 minutes off peak which then rolls into 10 minutes at peak, so any saving is gone. An hour is 60 minutes after all, not 50

Do Octopus charge for Snug based on the 30-minute metering values (like they do Agile or Go) or do they use two registers in the meter (like they would for E7)?

If the latter, it makes no difference. You'll get the full 60 minutes at the cheaper rate.

... we won't know if this is a real problem or simply an artefact of how Octopus's app displays the 30-minute data.

Scot_39

I am not sure how often Octopus are actually going to tweak users timings - but there's lots of stuff about having the right - e.g. for DSR type savings in the SNUG terms.

See the specific smart billing rules for smart tariffs - and the special conditions sections for snug (2.14.x) therein - which are currently date stamped just last Monday on 13th Jan 2025.

https://octopus.energy/policies/smart-tariffs-terms-and-condition/

As per other Octopus smart - the default on loss (temporary or more permanent ?)  of comms isn't block time of use - but "typical" consumption (for profile class 1 - so single rate metering) variable.  Section 1.5  But there have been forum posts here regarding this in more detail when problems occurred.

And there is a whole - perhaps new - perhaps just recently updated - section on Snug - sections starting 2.14.x

2.14.4  makes clear reference to half hourly billing - so I assume it is the 48 1/2 hr blocks

2.14.5  covers the OP's 10 min delay

I guess the OPs 11:30pm to 12:30 charging are like the afternoon covered by the "schedule" flexibility rather than the core off peak hours 0030 to 0630 charged at night rate regardless of heater schedule as I read 2.14.1 - and like the afternoon slot - charged at night rate - again as 2.14.4.

I am rather curious as to how 2.14.9 works - as could be taken to imply a reversion to perhaps old block time of use parameters - especially when phrased as if lost comms (in future say after a period of successful operation on snug with their optimised schedules).

Maybe there is some other timing control exchanges or tables in the smets specification to parameterize dynamic ALCS.

Phones4Chris

@Scott_39 Thanks for the above, 2.14.4 and 2.14.5 are obviously the key paragraphs.
2.14.9 really requires further clarification.

Phones4Chris said:

@Chrysalis I'm fairly sure that most suppliers don't know the "offset" for individual customers. It's determined by the DNO and whilst it is fixed at the same time each day for the overnight night rate block, it depends on how it's determined. It might be based on "Load Block" group (ie. ESEC blocks - the rota for Disconnection when electricity is in short supply) but it might be determined down at Sub-Station level. I'll see what I can find out - unless there's someone working in the industry on here that knows the answer.
What may happen with the afternoon boost, as I get the impression that the timeslot can vary on Snug on different days, I can only guess, but one might expect the "same offset" to be automatically incorporated in the ALCS programming - it would be nice to discover exactly how it's done.

Other TOU tariffs are a different kettle of fish!

(Thinking more as I've been typing, IF it's determined by "Load Block" group, then data ought to be readily available to suppliers as both the "Load Block" (it's on the bill) and the Region are known. Even if the "offset" for a given Block may vary from region to region (although the ESEC timings do not) information should still be readily available from DNOs!).

Well it turns out that this is all a bit of a hypothetical idea. It may have applied pre current smart metering (who knows) BUT having checked back through this thread - because something suddenly clicked that I'd read somewhere about this "randomised delay" it seems this reply actually covered it, para.5.7.5.28 of the document is what defines it.  So clearly suppliers do NOT know an individual customer's delay.

Gerry1

Ildhund said:

As I understand it, the variation in switching times is always a delay of up to 30 minutes (so not in each direction). The amount of the delay is randomized, unique and fixed for each meter:

@Ildhund That may be the case for smart meters but not so for conventional meters with RTS unit.  Mine states it's 2230-0030 and 0230-0730 GMT ± 15 minutes.

With smart meters, I wonder if it's to simplify it so that the offset is always a positive value?  It could be a bit like the fudge on supply voltage tolerances where they were changed to match the EU 230V rules but the UK effectively stayed exactly the same with 240V.

A nominal 2230 start time could be redefined to be 2215 with an offset delay between 0 and 30 minutes with an average of 15.

Phones4Chris

Gerry1 said:

Ildhund said:

As I understand it, the variation in switching times is always a delay of up to 30 minutes (so not in each direction). The amount of the delay is randomized, unique and fixed for each meter:

@Ildhund That may be the case for smart meters but not so for conventional meters with RTS unit.  Mine states it's 2230-0030 and 0230-0730 GMT ± 15 minutes.

With smart meters, I wonder if it's to simplify it so that the offset is always a positive value?  It could be a bit like the fudge on supply voltage tolerances where they were changed to match the EU 230V rules but the UK effectively stayed exactly the same with 240V.

A nominal 2230 start time could be redefined to be 2215 with an offset delay between 0 and 30 minutes with an average of 15.

I don't think I've ever come across a case where the offset in non-smart meter situations was SET early (unless a timeswitch was wrong!). I'm guessing that what it says on yours may be to cover situations where it's lost its RTS signal.
As far as Smart meters go, it is always "plus" (which can be zero).

Phones4Chris said:

@Scott_39 Thanks for the above, 2.14.4 and 2.14.5 are obviously the key paragraphs.
2.14.9 really requires further clarification.

Phones4Chris said:

@Chrysalis I'm fairly sure that most suppliers don't know the "offset" for individual customers. It's determined by the DNO and whilst it is fixed at the same time each day for the overnight night rate block, it depends on how it's determined. It might be based on "Load Block" group (ie. ESEC blocks - the rota for Disconnection when electricity is in short supply) but it might be determined down at Sub-Station level. I'll see what I can find out - unless there's someone working in the industry on here that knows the answer.
What may happen with the afternoon boost, as I get the impression that the timeslot can vary on Snug on different days, I can only guess, but one might expect the "same offset" to be automatically incorporated in the ALCS programming - it would be nice to discover exactly how it's done.

Other TOU tariffs are a different kettle of fish!

(Thinking more as I've been typing, IF it's determined by "Load Block" group, then data ought to be readily available to suppliers as both the "Load Block" (it's on the bill) and the Region are known. Even if the "offset" for a given Block may vary from region to region (although the ESEC timings do not) information should still be readily available from DNOs!).

Well it turns out that this is all a bit of a hypothetical idea. It may have applied pre current smart metering (who knows) BUT having checked back through this thread - because something suddenly clicked that I'd read somewhere about this "randomised delay" it seems this reply actually covered it, para.5.7.5.28 of the document is what defines it.  So clearly suppliers do NOT know an individual customer's delay.