Octopus Tracker

Spoonie_Turtle

Am I being really dense?  I'm really not understanding why a power cut should affect billing later in the month.  If they can't retrieve a particular day's data can't they try again? 
Or does it bork something in the half-hourly data for the period of the power cut instead of just coming back with 0 for the time you had no power?

Telegraph_Sam

It should make no difference IMHO whether you were denied power between x and y am/pm, or voluntarily went into deep hibernation in this period. Whether daily or monthly charging the effect on the average should be the same.

mmmmikey

I would expect all the data needed to bill you correctly is still in the meter in NVRAM like the settings, i.e. no battery backup needed  Otherwise every time you had a power cut (which could be a blip for a few seconds overnight) all the history would be lost and these meters store a considerable amount of data.

So my guess is the billing compromises / estimates are more likely due to not having the systems in place to cope with interruptions due to power cuts. So they probably treat it like missing data in the way they would if there was a comms issue for example, and I've always believed that on Agile even if one 30 minute slot is missing the whole day gets billed at SVR. I presume this uses the index readings at the start and end of the day rather than the half hourly data.

It sounds like it should be something that could be handled better in future, but to be balanced about this it's worth remembering that these are still sold as "beta" tariffs (I think). Octopus make it clear that although they'll be fair we shouldn't expect perfection.

masonic

Spoonie_Turtle said:

Am I being really dense?  I'm really not understanding why a power cut should affect billing later in the month.  If they can't retrieve a particular day's data can't they try again? 
Or does it bork something in the half-hourly data for the period of the power cut instead of just coming back with 0 for the time you had no power?

If the meter itself is powered off, or in the process of powering up, it simply won't have the data, so this type of missing data cannot be manually retrieved.

Presumably a sudden loss of power would result in some HH consumption not being captured either side of the power loss, as well as any full HH periods of zero usage during the outage. That's why I was surprised to learn electricity meters aren't fitted with a battery backup.

Telegraph_Sam

Distinguish between 2 different scenarios
1. Local power cut. This would be when the presence of a backup battery or 
not becomes relevant.
2. Consumer decides for whatever reason to switch every appliance, light bulb etc off. Meter is working but there's no consumption to record.
Event 2 would presumbly be recorded as a zero consumption day in the invoice. This should reduce the average charge for the period.
Event 1 ??

Qyburn

Telegraph_Sam said:

It still seems to me to amount to a double whammy if you were being charged for kWh's that you never used.

That's what I first thought, but the overall total matches the meter readings. So the over estimate on the two power cut days is cancelled out by underestimates elsewhere.

[Deleted User]

masonic said:

Spoonie_Turtle said:

Am I being really dense?  I'm really not understanding why a power cut should affect billing later in the month.  If they can't retrieve a particular day's data can't they try again? 
Or does it bork something in the half-hourly data for the period of the power cut instead of just coming back with 0 for the time you had no power?

If the meter itself is powered off, or in the process of powering up, it simply won't have the data, so this type of missing data cannot be manually retrieved.

Presumably a sudden loss of power would result in some HH consumption not being captured either side of the power loss, as well as any full HH periods of zero usage during the outage. That's why I was surprised to learn electricity meters aren't fitted with a battery backup.

Smart meters store 13 months’ worth of index readings and 30 minute usage data. This data isn’t lost during a power cut. If, say, a power cut occurred at 5am the meter would have the midnight index readings in store. If power came back 3 days later the meter would respond the following day with the index reading for midnight. Clearly, the meter will not be recording usage during the time that there is no Grid power. I really do not see what the issue is.

Qyburn

Spoonie_Turtle said:

Am I being really dense?  I'm really not understanding why a power cut should affect billing later in the month.  

That must just be Octopus' decision. They could perfectly well have billed every day including the affected days on their actual consumption, and actual day price.    

Spoonie_Turtle said: 

Or does it bork something in the half-hourly data for the period of the power cut instead of just coming back with 0 for the time you had no power?

I don't know what the supplier actually received, but Octopus data export show those time slots missing rather than zero. 

MultiFuelBurner

We accepted that our last strange 8 day billed length was to do with our smets 1 meters being replaced with a smets2 meter so was ok with them averaging our usage and cost out as there may well be 30 minute data appearing to be missing due to the power being off for that meter change.

I have just noticed that we have had some money go out of our account today so just waiting for the bill to appear to see what they have done this month. How long does it take for the bill to appear for everyone else on tracker. Seems strange it's money out first then bill.

Telegraph_Sam

I'd be suspicious. I believe any D/D is legally obliged to give (10?) days notice. In fairness to Octopus that is what I have experienced. That said I am still waiting for the latest bill + notification to arrive.