In my quest of getting more accurate estimation of power for systems without metering, I’ve been looking at datasheets a lot recently and it dawned on me that they provide the numbers to calculate Carnot COP (i.e. T_flow and T_ambient), which can then be compared against the manufacturer’s own numbers. So, I did exactly that for a Mitsubishi Ecodan 11.2 kW:
Can see that (on paper) these fit quite closely to a heatpump factor of 0.5, so we should expect any decent installation to get close to that. Here’s how the Ecodans on HPM compare, ordered by size, all R32 except for 1:
I reckon you ought to exclude all the Ground Source systems from this analysis. The issue is that T_ambient for a GSHP should be the incoming Brine temperature from the ground loops, not the Ambient Air temperature, since it’s the Brine that’s acting as the heat source.
I suppose it depends on what GSHP installations have used as their feed for ‘heatpump_outsideT’ but - because of how that is named - I suspect most people (like me) will be feeding that with the ambient air temperature, which is useful to see on a graph but not what you’re looking for for this analysis, for Ground Source systems.
[My excuse for not simply choosing the incoming brine temperature as ‘heatpump_outsideT’ is that the brine temperature reading is only meaningful when the brine pump is running, which is only when the compressor is running, which in my case is only about 20% of the time. My brine temperature sensors must be located somewhere that gets quite warm, since when the compressor is off they tend to ‘float’ up to about 30C, whereas the real brine temperature is currently around 9C. What I ought to do is apply some logic to only report the brine temperature when the brine pump is running…]
Would not a better analogue for the outside air temperature be the outside ground temperature, at the same depth as and a location that’s representative of the collector, but sufficiently distant from the collector so as not to be influenced by it?
I disagree, there must surely be value in including ground source (and water) as separate tables so that it’s possible to compare all three.
For an ASHP, the fluid directly heating the evaporator is the ambient air whereas for a GSHP it’s the brine (actually glycol, but everyone seems to call it ‘brine’) coming from the ground loops. There’s effectively another heat exchanger between the actual ground and the brine which we want to exclude from the Carnot calculation, so the like-for-like comparison is to use the brine temperature (or incoming water temperature for water source). The guidance notes at COP comparator – John Cantor Heat Pumps confirm that.
Ideally yes; it would indeed be valuable to include ground (and water) source systems. I should have added the caveat to exclude them “…until we have reliable source temperature data feeds for the ground-source installations" which is very much not the case at the moment.
