Can anyone explain the algorithm used for calculating the emitter spec and system volume? It’s doing something rather weird!
Firstly, why is Radiator Heat Output zero (or infinite?) while the HP compressor is not running? You can clearly see that the flow and return temps gradually cool down - and the only place that heat is going is into the house via the radiators - they are definitely emitting heat during these periods and it should be calculable how much given the system volume, rate of temp loss and specific heat capacity of water.
Secondly, why is my calculated system volume increasing? I have one radiator which is being controlled by a thermostat and on/off zone valve - which you can clearly see is causing ripples in the flow/return temps every 20 minutes. I would expect the system volume, and emitter spec, to go up and down by the volume/power of that radiator. The Emitter Spec more or less does this - but the System Volume increases in that zig-zag slope. That can’t be right!
The other thing that can’t be correct is that in the nice steady-state period between 05:30 and 06:00, how can the radiator output be higher than the heat-meter measured heat power?
I know these algorithms are trying to deduce this information based on some heuristic assumptions that are clearly incorrect in my case here. It would be nice to know though exactly how they are intended to work? Thanks.
When compressor is not running, the dT between return and flow is zero because no heat is being produced by the heat pump. The app does not show radiator output during this phase as there’s no way to know the dT across the radiators.
The optimal state to calculate emitter spec is during the steady state when power and temperatures are level. The orange line shows the radiator capacity at that mean flow temperature, and so long as the generated heat is below that line, then the heat pump will settle into a stable state. If the radiators cannot emit that heat being generated, then the flow temperature will rise and the heat pump will cycle. This is correct and as expected.
The calculation for system volume is trickier, and works best when the emitter spec is known and the system is heating up from cold. There are some manual steps you can do to get a more accurate reading. For more information about this, see Infer radiator spec and system volume using the MyHeatpump app
When the compressor is not running, but the water pump is, typically dt is still there, your return water is cooler than what you send out. It sort of depends if you want to measure how much you boost the return temp to get lwt (OK, that’s zero) or how much the lwt has lost when it arrives back (likely not zero, unless your house is as hot as the lwt). In steady state the two calcs give the same answer, with no compressor Input they dont .. one says zero, the other says still heating the house at an ever declining rate. Of course, you didn’t want the compressor to stop anyway. 
That depends on where the temp probes are. In my case, I have a longish pipe run (15m each way?) between my airing cupboard, where the heat meter temp probes are, and the outdoor unit.
The flow and return temps measured are thus pretty much the actual radiator temperatures, and when the compressor is not running the return temp is a little lower than the flow temp. It takes about a minute for the return flow to loop around the outdoor unit and come back again - so in this condition, the flow temp is more or less equal to whatever the return temp was a minute previously.
The effects of the long pipe runs and varying temperatures around that loop make it very hard to infer any meaningful data from the flow and return temps in anything other that steady-state conditions, unless you take into account the lengths of the pipe runs and hence the time delay that it takes for water to flow round the loop from the flow temp sensor to the return temp sensor, and also from the return temp sensor through the outdoor loop and back to the flow temp sensor.
