System has been up and running for two weeks now. There is a difference between the daily import figures as given by the inverter and Octopus. The inverter is giving a figure of around 85% to 95% of what Octopus says. This is on a range of between 1 and 10 kWh per day.
It would help if you stated what user interface you are taking the readings from (Web or App), and what type of grid meter you have connected to the GivEnergy inverter.
Commonly used GivEnergy meters are models EM115 and EM418 as shown on pages 28 and 30 of the online brochure.
https://www.givenergy.co.uk/pdf/PRODUCT%20BROCHURE_FINAL%20(ONLINE).pdf
The EM115 meter has a current transformer (CT) wired to it, and the CT needs to be clipped around the correct cable, which should be the live cable in or out of the utility meter that reports back to your supplier (Octopus).
Have you tried taking import readings from the meter displays, say 24 hours apart, then compare the values?
At least then you will know if it is the metering itself, or what is being reported on the GivEnergy cloud.
Thank you Simon C
I am taking readings from the web portal as supplied by the daily report.
The meter is an Owen Brothers OB115-Mod.
I will try taking some readings direct from the meter displays as you suggest.
In my experience, readings from CT clamps are not very accurate. My figures are similar (today my Glow Stick - which is reading the smart meter - reports 13.12, the GE portal says 12.3) The PV readings have a similar discrepancy (my PV gen meter is 9.5 higher than last night, while the GE portal says 9.1)
The readings from your smart meter will be the accurate ones, and should match what Octopus report.
Thanks andrew_bond
Accuracy of CT clamps was my explanation - I was just curious. I would prioritise the smart meter as that is what I am charged on!
I've never seen the figures match between GE and Octopus, which I put down to inverter sampling. The same applies for both import and export, so I tend to take the figures from Octopus for those values, and only rely on GE for generation, battery charge and battery discharge.
The same issue will mean that your consumption figures on the GE portal aren't precise, but since Octopus have no way of tracking what you're generating vs importing there's not really a better figure to use. I personally take my generation, battery charge and discharge from GE, then the import and export from Octopus, and generate a slightly more realistic consumption figure using inputs - outputs (so Generation + Import + Discharge - Charge - Export)
I have had a Giv Battery system since August 2023, but the SMETS2 Meter hasn't been communicating with Octopus since 31st July 2023. Octopus have only this week replaced the meter, so this is my first opportunity to compare Giv with Octopus.
Just over a couple of days I am seeing whole periods when Giv reports absolutely no usage (I have a ventilation and heat recovery system running 24hrs - so the house should NEVER be using no energy). The discrepancy over a 30 minute period is as high as 0.649KWh, which over 24hrs is 8%. I am expecting an argument with Octopus when they finally bill me for 7 months of electricity and was hoping that keeping daily readings from the Giv system would stand me in good dtead when they inevitably try to charge me for total usage, with no consideration for peak and off-peak usage and peak export. But if the Giv readings are this far out, I'm not sure how I will be able to argue anything with them.
lozzer1962 As I think was mentioned above, it matters which GE numbers you're using. One set, the energy flow I think, is based on extrapolating sampled instantaneous power readings. (And I think the performance report may be using these ?) The meter readings are a better bet: AFAIK, they accumulate on the inverter itself, and it's the totals that are reported to the cloud. (So missed values don't matter that much.) And on my system, are a better match to the Octopus numbers.
ah - I notice that the missing readings are 2am to 6am - does your router go to sleep overnight, by any chance?
On the portal, look at the 'inverter data' section (square grid icon ?)
lozzer1962 Agree with DD’s comments, the total import meter readings are the set to use. I find mine are a pretty close match to what Octopus says, with a decimal of a kW a day.
If your meter wasn’t communicating then the meter readings shouldn’t be lost. The meter holds them locally for 18 months I believe.
Worth looking at https://www.n3rgy.com/ to see what meter data the DNO has
geoffreycoan The meter holds them locally for 18 months I believe.
13 months for import 30-minute data; 3 months for export 30-minute data.
Also its not the DNO, they have nothing to do with the process, its DCC who act as a proxy between the meter and the supplier (or other authorised users - like n3rgy). I know you know that but others will get confused.
Hi all
Thank you all for your help. The meter was replaced on Thursday, so all the 1/2 hour readings have gone for good. They didn't give me an option to get the data before it was removed.
You are correct DD, I hadn't accessed the Giv Meter readings before, previously relying on the dashboard daily reports to produce the following...
It's not as easy to work out peak and off-peak import and export from the raw data, but only takes me a minute or two once a day to add the above readings, which I have been doing since 8th August 2023.
However, if, as it turns out, those readings are inaccurate, I guess that it's back to the drawing board.
If Octopus bill me for something around what I estimate I won't argue.
I am on the Octopus Flex Import/Export Tariff and Giv does charge the battery off-peak. It's my usage Peak v Offpeak that Octopus won't be aware of if all the 1/2hr data is lost now that the meter has been replaced.
No worries. I don't suppose it'll make a massive amount of difference to the final bill. The current credit on account is way more than I have calculated I owe them, even allowing for a 10% discrepancy in mine and their readings.
Thanks all for your comments.
Now we have an Octopus mini we can review more live consumption data but also can compare Give Energy’s totals for a day (yesterday 17th September).
So GE say we used 21.28kWh and Octo say 23.46kWh. That is a significant difference (10%).
I’m looking at daily totals on energy graph tab on the portal, so is that the best data?
Obviously we’ll end up paying Octopus’s figure. But the 2 figures are not just a decimal place apart.
Judith I’m looking at daily totals on energy graph tab on the portal, so is that the best data?
Givenergy seem to have two different measurements available. Eg if I look at the energy graph, it says I imported 7.89kWh yesterday, but looking at the meter data on the portal, it was 8.0kWh. (Octopus say it was 8.15kWh. And I also have a car chargepoint that monitors these things, and FWIW it reckons it was 8.4kWh.)
I think the energy data may be accumulated by the portal, whereas the meter data probably accumulates on the inverter itself and it's the total that's reported, so that's probably more accurate.
GivEnergy has to rely on the CT clamp, which can never be as accurate as the smartmeter. But perhaps worth checking that it's securely attached, and if possible not to close to other cables.
Normally about 400w difference on my import between Octopus (higher) and Givenergy.
I think it's mainly around the readings of the CT clamp where the difference will occur.
I always forget the meter data tab!
BUT now I’ve looked it says 21.3kWh imported yesterday so that is simply the same GE data but rounded to 1decimal point.
Curiouser and curiouser?
I don’t want to derail the thread on discrepancies since I’d like to understand it better.
(But I was investigating since it was higher than our normal Sept usage, our monitors are not attached at the mo, but a DIY project meant 5kWh of power tool usage.)