Off peak times inaccurate.

Swipe

Stilson said:

My electricity supplier is British Gas. I am on Economy 7.

On this tariff, off-peak periods are from 1am to 8am during British summer time. But the off-peak indicator lights on the night storage heaters are staying on until 9am.

So there obviously is some kind of timing fault. Could anyone tell me what it could be?

Thanks.

Nice problem to have

dunstonh

So there obviously is some kind of timing fault. Could anyone tell me what it could be?

Drift can happen.   Our two previous non-smart digital meters drifted almost monthly.  It got to the point where my off peak rates were late morning to early evening.  That was great.  We just checked the meter time once a month and adjusted the timers to match the meter time.    Now we are on a smart meter that syncs time with a timeserver, that is unlikely to happen.  

Part of me wishes I still had the drifting meter.

Scot_39

https://www.britishgas.co.uk/energy/guides/what-is-economy-7-meter.html#:~:text=If your meter was installed,30am, depending on your meter.

States 7 hrs between 12-8:30 if fitted by BG, 10 - 9 am if fitted by others.

And if gmt - the OPs 9am is within both specified ranges.

Doesn't sadly mention gmt or dst/bst though - but many older systems didnt adjust for dst - staying gmt.

And 3 regions according to one online table have the DNO window ending at 8 (NW and Mersey, W Midlands, or even 830am SScotland - again normally expressed as gmt.  So 9am dst is perfectly valid accurate setting.


The fact the op says 9am - I assume exactly - and hadnt noticed the error before - suggests more of a preset configuration not a drift or fault.

But then as the phrase goes - even a stopped clock is right twice a day.

Chris_b2z

dunstonh said:

Now we are on a smart meter that syncs time with a timeserver, that is unlikely to happen. 

You must have a SMETS3 meter.

mmmmikey

Chris_b2z said:

BarelySentientAI said:

There's a command to sync with UTC provided by the comms network, and if the required change is more than 10 seconds it records it as an out-of-spec event in the log.

I have no idea how often they are configured to run this command.

Thank you. I would expect this synchronisation command to be sent daily especially as smart meters are intended to replace RTS timing. I don't believe that happens.

My smart meter has been 7 hours behind UTC for the past 3 years.

Hi @Chris_b2z - out of interest, when you say your smart meter has been 7 hours behind, what exactly do you mean...? 

i.e. Is it the display on the meter that is showing as being 7 hours behind? If so, does this always stay 7 hours behind before and after clock changes? And is it always exactly 7 hours behind (give or take a minute) or does it change over time? 

If you look at your data using your energy company's app or one of the third party apps, is that also 7 hours behind, or do you get correctly billed and it's "just" the clock on the meter that displays the wrong time?

Sounds like an unusual fault to have with a smart meter, it's certainly not something I can recall seing other posts about. Clock drift on older meters with their own clocks is fairly common, but not for smart meters that are properly set up and communicating.

Thanks, Mike

Chris_b2z

mmmmikey said:

Hi @Chris_b2z - out of interest, when you say your smart meter has been 7 hours behind, what exactly do you mean...? 

i.e. Is it the display on the meter that is showing as being 7 hours behind? If so, does this always stay 7 hours behind before and after clock changes? And is it always exactly 7 hours behind (give or take a minute) or does it change over time? 

If you look at your data using your energy company's app or one of the third party apps, is that also 7 hours behind, or do you get correctly billed and it's "just" the clock on the meter that displays the wrong time?

Sounds like an unusual fault to have with a smart meter, it's certainly not something I can recall seing other posts about. Clock drift on older meters with their own clocks is fairly common, but not for smart meters that are properly set up and communicating.

Thanks, Mike

Thanks @mmmmikey for taking an interest in this issue.

I discovered the 7-hour offset from UTC when I was on a Scottish Power E7 tariff and noticed the half hourly electricity charts suddenly went out of sync compared to actual usage. I can only assume there was a remote update on the date this changed.
My IHD had never worked so I figured out how to navigate through the menus on the meter. This allowed me to check the actual meter time and also the times when the rate changed between peak and off-peak. It confirmed that the RTC was over 7 hours out of sync.

I'm not alone. Two other members posted similar experience on these MSE forums last year so I'm fairly confident that the clock resync does not work. One thing in common is that we all have EDMI brand SMETS2 meters.

Fortunately, I'm tech savvy enough to figure stuff like this out but it would be easy for other consumers to be caught out and overcharged.

Ildhund

Chris_b2z said:

My smart meter has been 7 hours behind UTC for the past 3 years.

That's not at all normal. Smart meters contain a clock that measures time by the frequency of the supply, which DNOs work very hard to keep as close to 50Hz as they can. If there's a power outage, the meter's internal battery takes over to keep the clock on time, but it should synchronize again as soon as power is restored.

Do you have a working IHD? Most of them show the current time somewhere, retrieved from the meter. 

Perhaps you mean something else, like heating circuits are switching on and off at the wrong times?

BarelySentientAI

Ildhund said:

Chris_b2z said:

My smart meter has been 7 hours behind UTC for the past 3 years.

That's not at all normal. Smart meters contain a clock that measures time by the frequency of the supply, which DNOs work very hard to keep as close to 50Hz as they can. If there's a power outage, the meter's internal battery takes over to keep the clock on time, but it should synchronize again as soon as power is restored.

Do you have a working IHD? Most of them show the current time somewhere, retrieved from the meter. 

Perhaps you mean something else, like heating circuits are switching on and off at the wrong times?

DNOs can do very little about system frequency, that's a system operator concern (so National Grid ESO here).

Serbia & Kosovo had another falling out a few years ago that meant Europe ended up 6 minutes slow because of exactly this mechanism though.  They rely on SwissGrid to fix the clocks, which is kind of nice.

Chris_b2z

Ildhund said:

That's not at all normal. Smart meters contain a clock that measures time by the frequency of the supply, which DNOs work very hard to keep as close to 50Hz as they can. If there's a power outage, the meter's internal battery takes over to keep the clock on time, but it should synchronize again as soon as power is restored.

Do you have a working IHD? Most of them show the current time somewhere, retrieved from the meter. 

Perhaps you mean something else, like heating circuits are switching on and off at the wrong times?

I agree that it's not normal. An analog Nokia phone managed to synchronise clock back in the 80's but the latest state-of-art digital 21st century 'smart' device is not capable. It does seem strange.

As mentioned previously, my IHD has never worked ... and nor did the replacement that SP sent. I can assure you that the clock in my smart meter is 7 hours out of sync and neither DCC or the supplier seem to be aware.

We should be questioning whether software testing is adequate with smart meters. This is a defect that can result in incorrect billing.

mmmmikey

Chris_b2z said:

Ildhund said:

That's not at all normal. Smart meters contain a clock that measures time by the frequency of the supply, which DNOs work very hard to keep as close to 50Hz as they can. If there's a power outage, the meter's internal battery takes over to keep the clock on time, but it should synchronize again as soon as power is restored.

Do you have a working IHD? Most of them show the current time somewhere, retrieved from the meter. 

Perhaps you mean something else, like heating circuits are switching on and off at the wrong times?

I agree that it's not normal. An analog Nokia phone managed to synchronise clock back in the 80's but the latest state-of-art digital 21st century 'smart' device is not capable. It does seem strange.

As mentioned previously, my IHD has never worked ... and nor did the replacement that SP sent. I can assure you that the clock in my smart meter is 7 hours out of sync and neither DCC or the supplier seem to be aware.

We should be questioning whether software testing is adequate with smart meters. This is a defect that can result in incorrect billing.

Hi @Chris_b2z - your meter appears to be faulty and/or wrongly configured and as you have pointed out this isn't a fault unique to you, although it doesn't appear to be widespread with the vast majority of smart meters working as they should. If I'm understanding your posts correctly, there's no real evidence to suggest that there is a general synchronisation issue or that all meters have the same issue. It seems more likely that the sychronisation fault lies within your meter and a small number of others with the same issue.

If your meter has been consistently out by the same amount (7 hours) for a lengthy period that would point to an offset being incorrectly set up somewhere rather than a badly calibrated clock drifting over a period of time. It's possible that it has been set up to work in a different time zone or there is an internal fault (hardware or software) in converting the internal digital representation of time into a human readable representation like UTC, BST or whatever. It's quite possible the synchronisation is working to keep the clock consistently 7 hours away from GMT/BST but the problem lies elsewhere within your meter. 

You say your supplier is not aware that your meter data is 7 hours out of sync - presumably you have reported this to them so surely by virtue of that they must be aware? When you reported this as a fault what did they say? If your individual meter is sending out properly formatted data and "my clock is fine thanks" messages to your supplier via DCC then it's quite possible that they would be unaware of any issue and they would be dependent on the customer spotting it and reporting it. 

Obviously not an ideal state of affairs, but although smart meters are infinitely better at alerting energy suppliers of faults they are not fool proof. Smart meters don't have to be 100% perfect to be better than the older meters they replace.