grid monitoring vs actual usage

What would you guys vew as acceptable for data differences for monitoring grid use?

Looking at last month (November 2022) the ge reports say I used 680.13kwh from the grid. However my smart meter says 739.25kwh. 8-9% difference?

I would check but all the GE systems seem to be down for me at the moment...perfect timing.

judgepd I'm glad you've just put that though - all down here for approx. 30mins now...

Confirmed here http://status.givenergy.cloud/incident/294287

Uh oh, someones not paid the electric bill 😆

Maybe their batteries suddenly dropped to 0% with no warning 😅

judgepd meow 🐱

Right, proper answers now for November:

Octopus meter: 411.2 kWh
GE report: 250.68 kWh + myenergi at 142.6 kWh (my zappi isn't 'seen' by my GE system) 393.28 kWh

393.28/411.2 = 96%.

I wonder if they would be further out if GE was recording the zappi use too.

THALL for last month Octopus charged me for 123.1kWh and GE reports have me at 118.6kWh so 96.3%. However back in summer Octopus charged me for 3.4kWh during July and GE have me at 0.61kWh. So massively different percentage wise.

I totally forgot to check myenergi

That says grid import 751.2 so over what my smart meter says, but a hell of a lot closer than the GE system. I wonder if I need to move the clamp a bit.

I might need to see what monitoring options are available on the solis inverters then.

THALL IMHO If you're trying to reconcile the Givenergy metering with the energy supply meter, you're making a rod for your own back. Both the grid supply meter and the EM115s are certified for energy billing, but they're different. If you could persuade your energy supplier to provide a dual register Modbus energy supply meter that Givenergy could interface to, you'd be in absolute alignment. We've got two inverters and therefore two EM115s with CTs physically located next to each other on the incoming supply cable. The EM115s provide different readings for import and export in the portal.

Moving your EM115 CT so that it recognises load to the Zappi would reduce errors (aka differences), but won't eradicate them. I use the Givenergy portal/app and the myEnergi app for information. Neither agree with each other and they don't agree with my energy supplier meter. I respectfully suggest that you don't waste time attempting to reconcile the energy consumption figures from the different systems.

Tim I wouldn't say expecting a paid for system to use as single point of monitoring was making a rod for my own back. A small difference is fine, the tiny difference recorded by the myenergi system vs the actual smart meter is fine. But I collect this data and share it with others monthly to compare setups, generation amounts, usage etc etc. I simply should not have to waste my time having to check different systems, the system I have paid for should work better. If the myenergi system can monitor with 2% or less then the givenergy system should be able to do it the same.

Expectation, 1 point of monitoring for someone who wants the data.

Also, if the system can't accurately monitor grid usage, then it doesn't know how much to discharge to cover that grid usage. Pretty fundamental to the system really.

judgepd the primary purpose of the EM115 installed with the Givenergy systems is to control inverter output to the home to minimise grid import and export. It does this by driving the inverter so that the value measured by the CT is zero; not a hard target value to measure. The fact that it also provides integrator readings for import and export is a bonus which Givenergy are using in their portal.

Even if Givenergy bought into a metering system that was orders of magnitude more accurate than the energy supply meter, although you would get very accurate import/export and consumption values, you’d still be paying Octopus or whoever your energy supplier is, based on the integrator values in their energy supply meter. But don’t let me get in the way of you trying to square a round hole.

Tim The data sheet for my AC3 says PV monitoring and Grid monitoring by CT / meter, I opened this post asking what people would class as an acceptable tollerance for the GE system vs the actual smart meter. If you were buying a system that lists features you wanted would a 10% plus or minus variance be acceptable to you? Especially if that variance is not documented or mentioned in the sale?

From your other posts I assume this is not a feature you care about, but I and others might want these features and might expect it to be "better" than a 10% variance. Posting the question on here would provide me with data from others to get an idea if my setup was preforming worse than others and I could then escalate it to my installer and to GivEnergy.

THALL the thing is that if you were to send your EM115 and CT to a certification lab for testing, you might find that it has a 0.01% error and an accuracy of 99.99% across the range of 0 to 100A import and export at 240V +6%/-10%; you’d have a near-perfect means of accurately measuring your grid import and export and some really good data on system performance.

With your EM115?meter certificate, you’d then ask your energy supplier to match that level of accuracy and they’d decline saying that the meter they installed for you is compliant with UK metering regulations and is sufficiently accurate for them to bill you according to their meter.

Both the energy supplier meter and the EM115 are certified for metering but they both have errors and will never agree. I already mentioned that we have two EM115s clipped next to each other and they provide different readings. It is irritating when something as regular as voltage, current and power, which absolutely can be measured absolutely, is measured using different metering and different technology and results in different measurements. Which one is right? Who knows?

The only way to align the energy supply meter readings and the Givenergy system is for them both to use the same meter. I agree it would be good if the two systems could measure and report the same values but they don’t.

THALL My view is a 2-5% difference is to be expected. More than that and the figures added up over time lose their value for many purposes.

When I queried GE about the differences, they stated that the DC to AC and AC to DC conversions can lead to losses of up to 15% in each direction. I have only had my system fitted for about 3 weeks but I will collect daily records for a couple of months and then analyse them. More in January . . . . .

GeoffEvans The issue is rather that there are at least two "meters" measuring the same values: the energy supplier's smart meter, and Givenergy's EM115 used for inverter control (as well as your MCS meter). They are all certified for billing purposes but register different values for import/export over time. The specification for certification is quite tight and so you wonder how they can all clock up different values. But they do and (IMO) it's a wattless activity trying to work out which is "right" and whose should be changed.

GeoffEvans 15? So for data collection (for sharing) that makes it pretty pointless. The myenergi setup which also uses CT clamps is much better. I know CT clamps are not 100% accurate. I have found a add on module for my 2x solis inverters to give me data access. £42 each....

I'll just use the myenergi monitoring for now.

Seems to me that the industry needs to get together and have a system whereby the electricity company meter 'publishes' its data to any system that wants it, such as solar inverters, battery inverters, EV chargers etc. It already 'publishes' its data to the In House Display. Perhaps SMETS3?

And similarly PV inverters, battery inverters, EV chargers etc 'publish' their data to other devices.

Would save lots of CT clamps and discrepancies between different sources of the 'same' data.

Baychattan Yeh, right! Not much chance of that. The 'industry' screwed up with SMETS1 and SMETS2 is only marginally better. They have been paid huge amounts by the government for a roll out of badly designed devices that even the distribution companies have problems with. Has anyone tried to use their energy monitor any further than about 20m from the meter or from a detached garage? Crap. It beggers belief that they didn't choose to provide a simple, open, wifi based interface. Dinosuars is the 21st century...

Baychattan We have a SMETS dongle that connects to the energy meter using zigbee (same as the IHD). It polls the meter every 6 seconds and sends the energy supplier meter data to a cloud service. The data is published over MQTT every 6 seconds, so we can subscribe to our own meter topic (using a userid and password), giving near-instantaneous meter-read values. The near-instantaneous data is available locally over wifi in the dongle-manufacturer's app. The app provides DCC data for the last 13 months summarised monthly, weekly, daily, hourly, half-hourly and even minutes. The DCC data is available via their app and MQTT service and an API without having to buy their dongle; you download their Bright app, create an account, enter your MPAN (Import and export, works with gas too) and you're good to go.

If the Givenergy plant would speak to the dongle locally over wifi, or if Givenergy would replace the EM115s with their own zigbee-modbus or zigbee-wifi device, they could have energy-supplier meter data every 6 seconds as well, although they'd probably want to keep EM115s as backup just in case there was a zigbee or wifi failure.

The GE grid use readings are next to useless and should be ignored
My last 7 days GE daily figures varied from 4% to 14% less than Smart Meter's
in January I bought in 322kWh but GE reckoned it was 300kWh
This makes me wonder whether I should trust any of the other figures in the reports.
"The more you measure the more you worry"

Tim I see octopus are rolling out a similar bit of hardware… https://octopus.energy/octopus-home-mini-faq/
Might be easier to keep on top of the differences if you can monitor both.

naltsta I've got an octopus mini, just a shame you can't pull the data in to home assistant yet.

judgepd you could look into Glow from Hildebrand, the people behind the bright app. The device connects directly to your meter through Zigbee and has an mqtt feed that you can use in HA

Cdent looks like those with the Bright app, or it could be their dongle(?) are being invited to participate in their “Delight” trial where you are entered into a prize draw for participating. It doesn’t say that you actually have to reduce use at certain times, just opt in to participate in the session. I guess this means that you aren’t tied to a specific supplier to participate. We’ve declined anyway as our consumption is all off-peak battery charging while we wait for longer sunnier days.

Tim Just be careful because the Bright/Hildebrand/Glow demand-shifting has smallprint which says it will forcibly opt-out from any other schemes - which I think means you can't do the Octopus one as well.

I'd rather have guaranteed Octopus revenue than enter into a lottery with presumably quite poor odds...

Riscy national grid rules, you can only be in one scheme.

Thought it used to be per session, but think it's changed going forwards to be one full stop.

Tim delight is just another one of the many Demand Flexibility Schemes, like Savings Sessions from Octopus. I'd rather go with Octopus as I get something for saving energy, rather than a competition where I could (but knowing my luck, won't!) win something.

Cdent I agree that Octopus, where you actually get £credit to your account for matching their criteria, is a good option. But if you don't have Octopus as supplier, it's a tad more difficult!

Tim that's true, I've been invited to another service with Equiwatt, they offer points which you can buy things from their catalogue like smart plugs and various gift cards. They're supplier agnostic and could be worth it. They claim to offer the highest incentives too

Tim I’ve notice my GE usage is quite variable compared to the usage billing log from Scottish Power. But my billing period starts in the middle of the month so no proper alignment. But £45.00 for all my SP grid electricity( exc Standing Charges) over the period beginning of April to mid Sept 2023 compares favourably with about £240 for the same period 2022. However SP were given 800 kWh free . Next year hope to put that into EV charging.