It seems someone else stumbled upon the same issue earlier this year too here, albeit it's not clear if it affected their billing.

The starting point of that discussion on Camelot-Forum is valid.
But alas, the correspondents have failed to appreciate that it's an integral function of the Smart Meter design which means that the Randomised Offset to the Tariff Switch Table exists.

The delay they then talk about being implemented by Octopus is the one related to remote switch on/off of an EV charger.
Their discussion has muddled the two offsets.

We had similar discussions on the OVO Forum 4-5 years ago. That preceded the random-offset required under the EV Charging Regulations (2022), and was based purely on incorrect bills. That analysis work by Simon1D and I continued after we both ceased to be OVO customers. Simon created mathematical modelling tools which were used to produce some of the probability histograms which I posted here on this topic earlier.

We have used the SMETS2 Specification as the basis of that modelling, which means it's theoretical. We don't (yet) know what the several authorised SM Manufacturers set as the default limits for the Randomised Offset Limit. Ofgem have not enforced the spec requirement that this register be set randomly between 1-1799.

Nor do we yet know if DCC and/or any Energy Suppliers change that Offset Limit after a meter has been installed.

All we can state for certain is that the Randomised Offset cannot be zero.
No SMETS2 meter can start its half-hour tariff periods coinciding with UTC because that would generate a "divide by zero" error within the firmware!
The smallest Offset possible is 1-second.
However it would be reckless for DCC or a Supplier to configure such a small offset because that would result in demand surges on the national grid.

The absence of enforcement by Ofgem and the SEC means that it's difficult to build future smart technology which uses SMETS2 features.

It's noticeable that neither DESNZ Policy nor the Labour Party Clean Energy Pledge (March 2024) include technology based on Smart Meters within the tool-set for future electricity supply and governance. The politicians' gaze is instead focused on which sources of zero-carbon generation they favour.

The widespread lack of understanding/knowledge of energy issues is constraining what we can achieve.

I received the outcome of the complaint with the Energy Ombudsman today.

The investigator agreed there were billing discrepancies on the account that were in EDF's favour, and agreed with the point that it will be affecting other customers on EDF tariffs. EDF confirmed that the off-peak period can start a few minutes either side of the stated tariff time. EDF state this will cease to be an issue when they move to half-hourly billing of customers.

The investigator has required EDF to reimburse me with £150 as settlement, and has requested that EDF look into other customers on their time of use tariffs to proactively remedy billing issues similar to mine as to avoid future complaints. Whether EDF will comply with the request...I am rather unconvinced.

Posted by: @korwraith

↑

has requested that EDF look into other customers on their time of use tariffs to proactively remedy billing issues similar to mine as to avoid future complaints. Whether EDF will comply with the request...I am rather unconvinced.

I think we all know the answer to that. £150 multiplied by however many million customers they have is never going to happen. But well done for seeing it through!

Posted by: @korwraith

↑

EDF state this will cease to be an issue when they move to half-hourly billing of customers.

https://www.edfenergy.com/energywise/improving-customer-products-services implies that EDF are moving onto the same billing platform as Octopus Energy, so hopefully that means the Octopus workaround will be applied.

From a technical viewpoint I don't see how 'moving to HH billing' will make the bills correct.

All Energy Suppliers, including Octopus, need to make the adjustment for the Randomised Offset to the Tariff Table.
At this stage it still isn't known if the SMETS Command to read that Offset is one of those which can be sent via DCC.

Posted by: @editor

↑

£150 multiplied by however many million customers they have is never going to happen.

The usual level of awards made by the Ombudsman is £50 - £70.
So an award of £150 suggests they believe this to be serious and that they have taken into account the depth of research which went into making this complaint.

Do I understand correct @korwraith that the £150 is a full and final settlement, or is it an award in addition to EDF also having to reimburse you for the amount they over-charged you?

I disagree with @editor on this point.
EDF and all other Energy Suppliers will have to pay back any amounts which they have over-charged customers.
They have no right to hang onto money to which they have no legal claim!

Whether they will also be required to pay interest and compensation for their errors is another matter.
Ofgem will probably try to argue that they should not, because they're worried about more Suppliers suffering financial failure.

But that argument could quickly spiral out of control.

a: what about Energy Suppliers who have previously been fined £-millions for incorrect billing?
If Ofgem now protects their rivals, won't they want to be reimbursed for their historic fines?

b: what about consumers that have been 'tipped over the edge' by this relatively small repetitive over-charging?
There may be some for whom it was sufficient that they were forced onto pre-payment meters.

This story has only just started...

@transparent Alas, the settlement is not in any direct addition to overcharging reimbursement. It probably splits along the lines of a £120 settlement and a £30 overcharging reimbursement.

From a technical viewpoint I don't see how 'moving to HH billing' will make the bills correct.

My (simple) viewpoint is that the HH usage data I download from my smart meter via n3rgy has no random offset so matches my (theoretical) off-peak period 1:1. Hence why my Octopus bills were accurate.

I've received a bill from EDF covering my most recent usage from 18th January to 20th May and it compares as follows to my HH data (note that during this period I had already set my devices not to draw for the first 10 minutes of my off-peak window):

Prior to my discovery of the random offset I had my car charger and home battery charging schedules set to activate at 3 minutes past the start of the off-peak period (e.g. 01:03am), upon discovery of the billing issues I pushed that back to 10 minutes past (01:10am) which has resulted in a decrease in overbilling by ~35%. I've now pushed them back to 15 minutes past the hour (01:15am) and will see what that brings it down to come my next billing cycle.

Honestly, if I didn't have a £75 exit fee attached to my EDF contract until October I'd have already switched back to Octopus. But the £150 compensation and comparatively lower overcharging incentivise me to see the contract period out.

I feel this is the sort of situation that the likes of Martin Lewis would be all over (his disparaging views of energy suppliers are well noted) but alas MoneySavingExpert can only fight so many battles and there's a lot of competition for attention.

 From what I can see all the problems so far identified  have occurred with people that were with Octopus Energy and had been on their Go tariff .

For some reason OE changed the timing period of  the R1 & R2 registers to coincide with the Go time period .

And failed to switch it back when the customer either left the company or as in my case they moved me on to their  Eco7 tariff .

( From what I can gather the have no record of this having been done )

Moreover  they cannot or are not in a position to find out when the meter transitions from the one register to the other and as I understand it , is not sent out in the data stream 

It is my belief that the way to discover the offset that is applied to your particular SMETS2 meter is to actually watch and record the time of the change over between the registers.

You will be able to hear the internal contacts switch as well as see the display change to total usage from the one to the other at the time of switchover .

Just how many 5 port meters have been installed ?

 In my opinion the country has made a grave error by no insisting that all new meters were so equipped so we at least  had the option of controlling peak demand at times of need .   

@korwraith

It would appear that you have been incredibly  lucky in  that your  "trustalliancegroup investigator " did NOT follow company policy and make  the standard 50  award in your instance.  

Please keep adjusting the time at which you start drawing 'cheap rate' electricity @korwraith 
It seems that you have a meter where the Randomised Offset to the Tariff table is particularly large, and therefore easier to work out.

One of my fears is that DCC or some of the Energy Suppliers have been setting the Randomised Offset Number low.
That reduces the possible delays, but adversely affects the grid by causing larger demand surges.

@bxman  - the great majority of users on ToU tariffs don't have 5-terminal meters.
So listening to contacts closing isn't an option.

Posted by: @bxman

↑

For some reason OE changed the timing period of  the R1 & R2 registers to coincide with the Go time period .

Do you have a source reference for the function of such registers?

They are not mentioned in the Specification documents for ESMEs.

Are they in the Communications Hubs?

@korwraith  I believe that n3rgy and Octopus both get their meter data from the DCC, so the data should be identical (bear in mind potential 1 hour offsets as the DCC uses UTC time). I have tried to identify my meter offset by comparing household consumption data from my EV charger, but I’ve not been able to reach a conclusion that I’m comfortable with as yet. 

Posted by: @ianmk13

↑

I believe that n3rgy and Octopus both get their meter data from the DCC, so the data should be identical (bear in mind potential 1 hour offsets as the DCC uses UTC time).

Indeed, for any future readers, UTC matches the 'clock' time during winter, my EDF off-peak period is 12am-5am in winter and 1am-6am in summer. Therefore, when looking at my HH data I always use the 12am-5am fields. When the clocks change in March I adjust my EV and battery charger schedules forward an hour but as the clocks also move forward an hour this keeps it in the same 12am-5am UTC offpeak cells (despite the charge schedule now being between 1am and 6am).

Posted by: @bxman

↑

  From what I can see all the problems so far identified  have occurred with people that were with Octopus Energy and had been on their Go tariff .

For some reason OE changed the timing period of  the R1 & R2 registers to coincide with the Go time period .

And failed to switch it back when the customer either left the company or as in my case they moved me on to their  Eco7 tariff .

( From what I can gather the have no record of this having been done )

That's an interesting theory that it only affects previous Octopus Go tariff users. Evidently, it is the case for me. Any non-Octopus customers fancy switching to EDF for theory-testing purposes 😆? I've a £50 refer a friend link for anyone interested 😉