I’ve replaced the old bargraph chart tool that was accessible from the top of the system list with something a bit more capable. I wanted to make it easier to generate over sizing factor vs SPF and mean flow temperature vs SPF charts. Not that I’m really managing to pull out insights from this yet 
Click on the chart icon (top-right) next to stats time period:
A somewhat fun one, The top end of the Vaillants is higher than the Ecodan’s but theres a much wider range of overlap…
Is there any correlation between over-sizing and SPF?? It’s very weak!
The data suggests higher correlation between over-sizing and lower SPF for the Daikins vs the Vaillants, which would confirm my priors… but we probably don’t have enough systems for statistically significant results…
Plotting by Refrigerant and mapped by Rating is interesting too
Only really seeing R290 used in the smaller units (blue/purple) while the larger R32 units (green/yellow) don’t manage to reach higher COPs - perhaps due to oversizing?
Nice!
Adding 3 further Daikin’s to the oversizing chart (by adding measured heat loss entry). Drops R2 from 0.88 (n=4) down to 0.35 (n=7). Which goes to show that’s it’s not really possible to say much with such a low sample size…
One of the things that’s very clear from trying to make these comparisons is that you cant distill heat pump performance down to 2 or 3 variables. This does make sense when you think of the large number of variables that can influence performance.
The question is what is the minimum number of factors/variables required to predict heat pump performance and is it possible to develop a model that can reproduce the wide spread of results that we see.
