These were not available at the time of this trial.
They haven’t been available for long enough to have gone through one heating season either.
Look again come January 2026 for those I think…
…but expect the “built in” data collection to suppress demand for independent data collection… …and the results to be just as absent from the public domain as the rest of the Octopus data. (they already collect plenty; but it’s considered a private asset for their own internal use)
[quote=“TrystanLea, post:37, topic:27445”]
The trial data only provides cumulative energy figures (in Wh) and system temperatures. I’ve made use of this to calculate power values based on the difference in the energy readings every 120s. This works well for populating the Emoncms dashboards.[/quote]
Ah.
[quote=“TrystanLea, post:37, topic:27445”]
I do think visual inspection of the data is key, it looks like this was a missing component of the project so far given some of the examples above. Though perhaps limitations in software visualisation made it difficult to see what we are seeing?[/quote]
For sure.
The analysts probably didn’t get to see the data until it was too late either. By which point the principal contractors had successfully ignored the problems for it to be too late to be worth fixing them.
i.e. Delay long enough and you don’t have to deny or defend. Don’t look too close either. You keep the installation money and it’s no use sending you back because the trial is already over. @SarahH will be here to remind me that I’m a cynic shortly. 
Glyn and I are currently trialing Homely on two Grant ASHP’s as the controls that come with the Grant units is so bad. Given the potential annual saving going from sub 3 to near 4, the application of something like Homely if it does optimise well could pay for itself very quickly!
Data from users who care/understand enough to pay for independent data collection will tell us little about the experience of normal people.
Thanks @Marko_Cosic will update with progress on checking for errors in the data, while there are some issues there’s plenty of good data in there for useful learning and comparison with our HeatpumpMonitor data, this will be really useful I think!
And would let me buy a lot cheaper heatpump if Homely works as well as “Degree Minutes” on a Vaillant. Afterall Homely is cheaper then the “optional” Vaillant display!
(But Homely will not work for rental properties as landlords can’t require tenants to pay for an internet connection and even when the tenants have broadband the wifi password will need resetting each time the tenant changes or they change provider.)
Yep.
There’s no real commercial incentive for a business selling electricity / selling heat pumps on the basis of an “adequate” standard (i.e. not inadequate) in the fine print and “up to 400% COP” in the marketing either.
Not until they run out of people who don’t ask “what does ‘up to’ mean?” anyhow. I can’t see the data being published for these from “normal” people.
Those who care will invest and publish. There will probably be enough of them (initially measuring the “default” Octopus install / not touching it whilst the warranty is in place etc) to see the performance of a default / first year install with Open Energy Monitor. Not until end fo 2025 though I should think.
Absolutely. Huge set of installations there. Props to the Energy Systems Catapult for publishing it. Props to OEM for making it come to life. 
@TrystanLea Thanks for pulling the data together. I’ve found my system on there, the Daikin 11kw in GU9 here. I might see if I can get my head around merging the historic data with my site on Heat Pump Monitoring for a complete record. If you have any tips on this let me know.
I’ve had a scan through this thread and seen the various comments. If I can answer any questions as a trialist do ask them.
From my point of view the EoH project was split across three organisations to deliver, OVO (SE England), EON (Newcastle) and Warmworks (SE Scotland). It was in the OVO part. From what I learnt from the installers and auditors, OVO subbed the work out to three main contractors on a fixed price per install basis. From feedback on the private OVO forum at the time it looks like two of the contractors were reasonabily competent with a background in installing heat pumps, and the final one wasn’t. I got the latter contractor and the initial install was a catastrophe, taking lots of phone calls, complaints and finally audits to resolve.
The initial install was Jan 21, it took about 3 week to get the heat pump running, then after a few more visits by someone else they managed to make the monitoring live in Feb 21, but still had issues with the logging equipment (sensors installed wrong way round, resolution issues, etc) and the installers not understanding the instructions, so it wasn’t logging correctly until June 21. A new installer came and rebuilt my install in Aug 21, and the original installer was struck off the trial and I believe are no longer installing heat pumps (and no longer registered with MCS).
As trialists we were advised to leave the systems as configured by the installers. As my first installer didn’t know what they were doing I did learn about the system myself and tried to improve the basics like setting the system up with weather dependant operation rather than a fixed leaving water temperature, and then some other basic improvements, but I didn’t have access to the data to see what was really going on. I resorted to logging readings off the meter in a spreadsheet daily to take the complaints back to OVO. The second installer did improve things but they were drafted in to ‘fix’ the issues but not start from scratch. They didn’t stick around as they came in from another part of the country. Servicing was handed over to Daikin.
From the 2.8 SPF on the trial, I’ve now moved to about 3.2 since having access to the data and being able to optimise the setup with the help of people on this forum. But I think that’s about as good as it gets with a grossly oversized Daikin heat pump (11kw heat pump vs 4.5kw heat loss).
Assuming the 4.5kw is a heatlose calculated with standard methods (eg assuming standard airtightless erc) then 11kw would be expected for a condensing gas system boiler with a normal modulation range.
I keep hitting this issue when thinking about rental properties, as a tenant who works all day and then often go out with mates will get a lower running cost by working an oversized heatpump hard when they want heat, the COP will be significantly lower then 24/7 heating but that not important. But the next tenant may be in the property all day, so wish to get good COP with 24/7 heating. The modulation range of heatpumps seem too low for rental properties unless the properties are restricted to old people. (The EPC rules removes the possibility of using A2A.)
I expect that unless you can force them to replace the heatpump with a correctly sized heatpump for free, you would get a greater saving from home batteries than a smaller heatpump. But I wonder what a large volumizer would do on Octopus Agile?
@Ringi 4.5kw heat loss is the actual calculated using the data from the metering. The estimated heat loss by the ‘assessor’ using MCS standards was over 9kw.
The modulating range of this particular Daikin model of heat pump is a big restriction. Its minimum output for steady operation is around 4kw but that’s been covered in great detail on plenty of other threads.
I can understand the challenges of trying to accommodate different tenants and their use of the property, and different levels of knowledge or desire to operate the heat pump effectively. A system like Homely seems like the best option to help optimise running costs for the less in-the-know.
My running costs - I have a 18kWh battery+Octo Intelligent Go, so fair pretty well. My average unit price for this November was 9.6p/kWh.
Great to see what my system was doing during those 2.5 years when we had no access to the data. I made notes of major events and dates, eg stuck 3 port valve, major refrigerant leak, discovering the flat WD curve 
and making changes to it, and thankfully having the refrigerant recharged. I’ll make a fuller report in due course, but although my SPF was 2.47 over the trial I’m glad to say that it’s now 3.7.
My trial data user ID is 529 and I’m on the main site as the Daikin LT Split in Fetcham, Surrey.
Julia, there is this document, which was published with the trial data.
In it they suggest that users were contacted and offered optimisation advice, is that what you experienced?
Looking fowawrd to reading your fuller report!
Hmm, interesting. Thanks !
I’ll read that carefully and see if anything sounds familiar…
Thanks @JuliaC
That’s a great result, well done!
For the OVO part of the trial we had a self-help group through a private forum they setup for us and we helped each other.
Audits took place on some systems, mine was audited several times through the first six months and the system repaired, but there was no professional advice offered on how to optimise our own systems. At least that’s my experience.
Thanks for sharing your experiences @HydroSam and @JuliaC. We need heat pumps that are more forgiving & consistent in terms of modulation control etc + self optimising curves. Realistic heat loss calculation and design (plus good data on defrost max outputs) and then lots of good learning resources both for customers and installers on how to get the most out of these systems The second part of this is probably happening slowly what with social media discussion, our learning here, videos such as HeatGeek’s content and others. Together with better guidance and standards from MCS and industry bodies such as CIBSE, hopefully things will start to move in the right direction!
I’ve just started reading this rather verbose document (re optimisation). It seems to imply that with the benefit of the monitoring that took place, the consumer would be able to make improvements to the running of the system. It’s possible that my installer made some tweaks to the flow temperature, but without telling me what he had done.
That was one of the most frustrating parts of the project; we knew the data was being collected but we were never permitted to see it. On my Daikin I had to make regular readings from the control panel (on the indoor unit at the far end of my loft) to calculate the COP.
New URL is live: https://eoh.heatpumpmonitor.org