What was the reason to not use the samsung display/controller itself? Can you relocate that to your living room? That may provide the most significant improvement over using the 3rd party thermostat.
It may not be included by the samsung calculation, im not sure, but it should be included in wider boundary calculations as you say. It is included on mine, so is all other heat pump related consumption e.g the central heating pump, diverter valve and indoor controller of my unit. I also include any top up from the immersion heater in my monthly/annual summary calculations.
Interesting yes questionable, but I’m actually surprised that it would over-estimate the COP. It turns out that for Glyn’s it actually under-estimates the COP significantly due to taking pipe surface temperature measurements of flow and return to calculate heat output. Did the heat pump come installed with a separate electricity meter from which you could read kwh readings? If not I would suggest installing a class 1 SDM120 modbus output electricity meter if you get the chance to do so in future.
Thanks I will take a closer look at this later.
Changing the topic a bit. I decided that if I was to really make sense of your setup and @thedawnbefore’s system which also has a heat exchanger, I needed to get my head around heat exchanger maths. The effectiveness NTU method is a particularly useful way to calculate temperature drops etc across a given heat exchanger. There’s a nice python library that can be used for heat exchanger calculations here that implements this method https://github.com/CalebBell/ht.
I’ve ported the calculation for a counterflow heat exchanger from this library into a little javascript web tool calculator that combines a heat pump, heat exchanger and radiator system. It calculates the system temperatures required to give a certain heat output from the radiator system (with all parameters adjustable). Here it is: https://openenergymonitor.org/heatexchanger/2.html
The code for this is all here: https://github.com/TrystanLea/heatexchanger
A couple of interesting results:
-
The COP penalty from introducing the heat exchanger is not as large as I would have thought, it appears to be around a 0.1-0.2 drop in COP. Maybe at most a 6% drop. This doesn’t take into account additional pumping electricity consumption from requiring two pumps but assuming low energy pumps the penalty for an additional 20W of pumping power is only a further 1% drop in COP, so maybe not that significant.
-
Varying the secondary flow rate on the heat exchanger has less effect on COP than I would have thought , which is probably the point! the heat exchanger allows for a greater range of conditions on the heat emitter side.
-
The variation of expected specific heat values (3800-4200) for the primary and secondary fluids have relatively little impact on the COP.
-
If you drive the same radiator system directly rather than through the heat exchanger at the same flow temperatures used to obtain a given heat output with the heat exchanger you can get ~20% more heat output (mean water temperature is higher). You can alternatively get a 20% boost in heat output by increasing the flow temperature and decreasing the COP by 0.1-0.2 as above.
Caveats are of course that this is theoretical steady state calculation. It doest take into account the buffer or cycling or anything that dynamically varies like that but hopefully an informative theoretical background on which to expand the science experiment! 