While there are clearly monitoring issues, I think there are enough systems with good data in the dataset to make the overall result a robust one (fortunately for the study authors but unfortunately for the industry). I think it’s within the realm of possibility that correcting low % carnot under reading heat meter issues could raise the average SPF by around 0.1 but the study is also missing the heat lost during defrosts which typically reduces SPF by about 0.1 over the year. I think the overall result would therefore be unlikely to shift that much…
All of my corrections to the data so far have only changed the result by -0.03 SPF which is really in rounding error territory.
I think the overall result reflects what we can see in the data, my working hypothesis is that most of it is down to non-optimal control settings, systems running hotter than they need to (flow temperatures not room temperatures). Other issues like too much zoning, 3rd party controls, system sizing etc will clearly contribute as well and the issues you experienced e.g refrigerant charge etc
I wonder if a large enough group of the participants would be interested in re-enabling their monitoring and using it as part of a peer-to-peer learning and advice process to improve the performance of their systems? much like we do here on this forum?
I have to admit I chucked out my monitoring equipment (particularly as I’d seen so many flashing red lights on it). It’s likely some others did the same. If only we’d been given access to the data as it was being collected, not the point of a ‘blind’ trial I suppose. So glad now I can see what’s happening using my heatpumpmonitor kit .
Was there some factor during the trial that has caused the controls?
Were installers penalised for callbacks for instance? If that was the case you may find unscrupulous actors turning off WC and setting static flows in order to achieve that.
I wasn’t aware of that - my installers returned quite a few times to get the legionella cycle to work (eventually replaced immersion thermostat). They appeared to set it up from the start with a fixed flow temp though.
.