Took just over an hour for both. Same fitter, pretty chatty chappie.]]For some reason didn't like the fuses that prepended the existing meter, but worked with them.
All up and running now - with the obligatory 2 week wait for gas to start reading (I have no idea !).
One thing which I was mildly surprised at was that contrary to something I picked up in my youth, meters are not required to be changed every 10 years.
I am also under no illusion that Octopus will just throw the old meters away without any testing. Which slightly saddens me as I could have been arranging my own 50% discount for years without their knowing.
Since they (dumb and smart) are a measuring device, there is a period of time for which their calibration is valid. That was always (AFAIK) 10 years. In the past, energy suppliers haven't usually bothered to replace an out of date meter which doesn't seem to be causing any problems, but that was pragmatism.
It's likely that the suppliers have some idea of calibration drift, ie they know the calibration can drift (according to some statistical model) by X% per year. They have certain standards that they must meet, eg the meter must be accurate to within 0.5%, and so if the calibration drift is 0.05%pa then they meet that standard by recalibrating every 10 years.
However it's possible that mechanical and electronic meters have different calibration stability: eg the spinning metal disc loses calibration as its bearings get sticky, while the electronic meter's current sense resistor may age. The manufacturer's datasheet may have some specification for this.
So the figure of 10 years isn't set in stone, it'll depend on the numbers. Additionally the supplier may load the measurement in the customer's favour
- like the greengrocer's scales are allowed to report things underweight but never over. If you're prepared to accept the customer underpaying for their usage then you can buy yourself a lot of time without needing to recalibrate. You don't get the full revenue from the customer but you save money swapping meters.
Long before smart meters the industry had a target of replacing meters every 10 years, although this target was seldom met. I had a friend who worked for BG and was responsible for arranging appointments for these meter replacements. What BG tended to do was to blitz one area at a time and bring in contractors to support the local BG staff. I was in one house 20+ years before I had an old meter replaced under this scheme.
Basically, the meters are allowed to over-read up to 2.5% and under-read down to 3.5%.
I was trying to work out how an electromechanical meter could "drift" into over-read over time, but couldn't see it. As you've mentioned, sticky bearings might affect calibration, but wouldn't that always lead to under-read? Fully electronic meters are a different kettle of fish and, I guess, are equally as likely to go over or under.
On a related matter, do domestic electromechanical meters care about power factor or things such as SMPS? Or do they "integrate" accurately and always give a true power consumption over time no matter what is using the power?
Yes, a couple of years ago, BG insisted that my meter needed to be replaced as it was out of calibration, and offered me the chance to 'upgrade' to a 'smart' meter. I refused the offer, and suddenly the calibration issue disappeared.
A couple of years ago Octopus started nagging me to have my GAS meter replaced as its accuracy couldn't be guaranteed.
The gas meter is "only" 40 years old.
Of course they said it would HAVE to be a smart meter, and it was up to me whether I wanted both done. A gas meter without its companion would bot be very smart.
I ignored the letters and they stopped bothering me.
The digital meters use a current shunt for measuring current flow. The shunt would be made of something stable, like manganin or Zeronel. If you used a piece of copper, the thermal coefficient of copper is very high, and it would form a lousy basis for building a meter. You could measure the instantaneous temperature of the copper and compensate for the error, but it just would not be the same.
The meter uses a couple ADC to convert from analog to digital. The converters run at 500KHz. The maths are done in real time. While a mechanical meter used by consumers measures kilowatts (real component), the digital meter can measure both real and reactive power. The power company may only bill on real power, so the reactive power is ignored for billing purposes.
The electromechanical meter is a motor, and it likely has limits as to what it can do with harmonic consumption. The digital meter is not limited in the same way.
This is an example of a current shunt. It has four screws. The large screws are for the current connections. In the sixth picture down, are the small screws for voltage-developed measurement. The selection of the shunt, is a compromise between elevated operating temperature, and the developed voltage. 100mV full scale, is a good number for the converter to use (as going much lower, attempts to apply "gain" to the voltage can throw off the metrology.
formatting link
The shunt could be thrown off, by corrosion over the years, and I don't know if there is any observed changes like that, in meters of this type. Whereas the bearings on the motor-based meter, will eventually be a concern (the meter back home stopped rotating, after enough years had passed and no "preemptive" replacement had been done by the supplier).
If the real-time reporting system is working on the digital meter, it is useful for network surveillance, such as starting a truck roll as soon as a power failure occurs in a neighbourhood. What is weird here, is at first, that scheme was working (I no longer needed to phone, to get the power restored -- service time as short as 20 minutes). The last time power went off, *I* had to phone to get service. None of my neighbours having used their smartphones to report the outage. It was still some time after I called, before the truck showed up. And the two crew had to clear foliage (tree trimming) in the dark, near where the tree had shorted the line. A crew just came out a few days ago, to do tree trimming on that property, and clear away more of the line. The more things change, the more they remain the same.
Yes, I wasn't aware until the guy was putting his tools away that the gas meter is relaying through the electricity meter.
I can just about see some logic there. However adding the stability of a wireless connection into the mix seems to be extra complexity for very little advantages. I appreciate I may not be typical but my meters are
40ft apart with 4 solid brick walls between them.
I look forward with interest as to whether the gas meter ever starts reading.
My SMETS1 meters were fitted a decade ago, they work in the same fashion
Gas--->Elec-+->Supplier | +->Display
They started talking immediately, I can see no reason why SMETS2 shouldn't be able to do that. Of course, the Supplier bit stopped working as soon as I changed supplier.
At the Electricity Board where I worked, the electro mechanical meters were brought back to the workshop and recalibrated before being re-issued. They only got thrown away if they could not be recalibrated properly.
Meters no longer report directly to the supplier, they report to DCC, who make the readings available to the supplier, so presumably mixed supply is possible?
Are you with Octopus? If so there will be a web display for both electricity and gas, also available on a mobile phone via their app and a monthly download in a CVS format. You can also request a free Octopus mini to give near real time (30/60 second update) electricity readings on you smartphone.
They work this way as, in general, gas meters don't have an electricity supply nearby, so they rely on a 10-year battery which is enough to power a low power transmitter to send the reading to the electricity meter every 30 minutes, but not enough to power a cellphone (or other) transmitter to send the readings to the supplier.
But I wonder how they manage to do this: "Meter readings: the DCC does not have access to individual meter readings. These are encrypted and securely transported to energy providers and distribution networks."
So does each meter encrypt its readings, before sending them to the DCC, which then sends them on to the appropriate suppliers, so that the suppliers can un-encrypt the readings?
Strange.
I also note this: "When will the smart meter roll-out be complete?
The roll-out is forecast to be completed by 2025. Last year (2022), a total of 6.3 million smart meters were connected to our network."
So this data was gathered in 2023. And they haven't succeeded in persuading me to have 'smart' meters installed in three possible service locations, in two buildings. I have not yet seen why such a meter would benefit ME, I have no solar, no Electric Vehicles, and certainly no Heat Pumps. They would not work in my old, draughty house without re-piping and replacing the whole heating system, and I doubt that I would get a sufficiently large grant to cover it. Apart from which, it would be a waste of money.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.