OK so here are some observations to ponder, in no particular order, and with no particular agenda (other than to try to explain the observations)
If we believe the excel regression analysis figures then the observed difference, whilst it may be wrong, is, so far as I can tell, statistically significant.
The average reduction appears to be about 7kWh/day as previously stated by @cathoderay
The Ambient during the setback period was generally higher and less variable than during the control period (average 10C vs average 5.5C
The house seems to cool by about 2C for about 9hrs at an IAT around 19 and ambient around 10. So the reduction in heat lost from the house accounts for a saving in energy supplied to the house of about 4% (1/10*9/24) only, about 1.2kW => something else is going on
In the hour immediately after setback (when the recovery is presumably happening) the COP is actually rather high even though its 4am in the morning. Whats going on here? It then drops, creeps back up during the day (as one would expect) then drops again (see plot below). Is this significant?
The circulation pump(s) is (are?) (presumably) off for 6 hrs. Assuming three are 2 and they are both 50W pump thats only about 0.5kWh, so thats not accounting for the discrepancy
We know (because @cathoderay has mentioned it) that this particular heat pump in this particular setup has a tendency to cycle at modest temperatures. I cant be sure whether its cycling or not on the days of interest (or on the control days, or more to point the control days when the temp is mild). Is this relevant?
On average, whilst setback was operational, the ambient temperature was (apparently) warmer when the heat pump was switched off (which was between 9pm and 3am) than when it was switched on. Admittedly by only 0.33C, but you would have expected it to be the other way round by several degrees
Basically I'm currently struggling to explain the observations which cant, so far as I can see, be explained by savings in heat lost from the house. However there must be an explanation which is consistent with the laws of physics.
4kW peak of solar PV since 2011; EV and a 1930s house which has been partially renovated to improve its efficiency. 7kW Vaillant heat pump.
Can you please confirm your WC curve settings so that we are testing like for like.
It varies, because of my auto-adaption script, which provides the recovery boost. However, much of the time it is 56 @ -4 / 34 @ 15. When the actual IAT falls to one to two degrees below the desired IAT the script increases both end points by 1 degree (to 57 /35), when the actual IAT falls to two to three degrees below the desired IAT the script increases both end points by 2 degrees (to 58 /36), and when the actual IAT falls to three or more degrees below the desired IAT the script increases both end points by 3 degrees (to 59 /37). It is not complicated, so do not complicate it. The reverse also happens when the actual IAT is above the desired IAT. The script runs once an hour (at half past the hour) and sets the endpoints using the above logic.
Midea 14kW (for now...) ASHP heating both building and DHW
@jamespa - I am digesting your latest comment but while I am doing that, there is a question I have been meaning to ask for a while, the answer to which may help our thinking. I think I know the answer, but I would be interested to know what yours is! The question is this: you said a while back that longer setbacks (days/weeks) are a no-brainer ie they obviously save energy, and I am sure we all agree on that. Now, as the setbacks get shorter, at some point they change from a no-brainer to become a brain-teaser, ie we are no longer sure they save energy. What is that point, and why does the change from no-brainer to brain-teaser happen?
Midea 14kW (for now...) ASHP heating both building and DHW
Methodology note (only read if this interests you!): how to get the regression coefficient CIs in LibreOffice Calc
This needs a work around. You have to create a new second independent variable, the square of the first variable, and do multiple linear (not polynomial) regression on that. For this specific example, ambient vs heating energy in, your data columns will be ambient, ambient squared (filled with the square of the ambient) for the independent (x) variables and htg_kWh_in for the dependent (y) variable. You will solve the general equation y = a(x) + b(x^2) + constant (intercept), just as the scatter plot plus polynomial trend line does, but more importantly you get the extra stats including CIs. In Libreoffice Calc, got to Data > Statistics > Regression and fill in the boxes as required (select both ambient and ambient squared as the independent variables, and use linear as the regression type, the quadratic element has been factored in by adding the ambient squared column) and click OK. If all goes well you will get this output (scatter plot on left, regression stats on right):
Midea 14kW (for now...) ASHP heating both building and DHW
@jamespa - I am digesting your latest comment but while I am doing that, there is a question I have been meaning to ask for a while, the answer to which may help our thinking. I think I know the answer, but I would be interested to know what yours is! The question is this: you said a while back that longer setbacks (days/weeks) are a no-brainer ie they obviously save energy, and I am sure we all agree on that. Now, as the setbacks get shorter, at some point they change from a no-brainer to become a brain-teaser, ie we are no longer sure they save energy. What is that point, and why does the change from no-brainer to brain-teaser happen?
I don't know is the simple answer. In my own mind I was thinking of two (practical) regimes, a 24 hour cycle and a 2 week holiday. But of course people do weekends away, long weekends away and anything in between. Somewhere in this continuum, where probably depends on the thermal mass of the house (its also a no brainer for a house with zero thermal mass!).
Incidentally the algorithm for polynomial regression you set out above works also in excel (2003).
4kW peak of solar PV since 2011; EV and a 1930s house which has been partially renovated to improve its efficiency. 7kW Vaillant heat pump.
I don't know is the simple answer. In my own mind I was thinking of two (practical) regimes, a 24 hour cycle and a 2 week holiday. But of course people do weekends away, long weekends away and anything in between. Somewhere in this continuum, where probably depends on the thermal mass of the house (its also a no brainer for a house with zero thermal mass!).
I said I thought I might have the answer, and I think on one level (though there are others), it is when you impose a return to baseline constraint over a short period cycle, but the bit I can't work out is how short the cycle has to be. The longer periods are also cycles, albeit much extended ones. There becomes a point where the savings always outweigh any extra recovery boost costs, and if my data is correct, then that happens with 24 hour cycle. But then again, a shorter cycles (and shorter setbacks), you get a smaller saving, but also less of a recovery cost, so maybe it is always a no-brainer! O do think this riddle is worth thinking about, as a way of understanding what is happening.
Incidentally the algorithm for polynomial regression you set out above works also in excel (2003).
The potentially confusing bit is you have to do a multiple linear regression to get the polynomial regression with the stats, while the scatter plot polynomial trend line does the polynomial bit itself, but doesn't provide the extra stats. Nonetheless, this is still less painful than trying to do it in R.
Midea 14kW (for now...) ASHP heating both building and DHW
@cathoderay@derek-m Health warning: This post doesn't contain any conclusions, just some more observations to trigger thought, pursuant to the overriding issue that the apparent saving cannot, so far as I can see, be explained by the saving in heat lost from the house.
Further to my comments above here is an interesting plot of the hourly data during the 'control' period (I haven't excluded the DHW periods so the correlation isn't quite as good)
Note that
correlation of energy energy in with OAT is MUCH better than the correlation of heating energy out
the latter appears to have an intercept of 2kWh at 20C whereas you would expect it to be 0 or thereabouts at 20C. There is a lot of uncertainty on this number but its definitely looking like a positive intercept. Its a similar intercept if you plot heating energy out vs (IAT-OAT).
So Im thinking two things:
(1) is perhaps fairly easily explained by WC. If the only control loop is the WC control loop then the system is only marginally responsive to IAT and thus the correlation with IAT will be better than the correlation with OAT assuming that the emitters are oversized (so that they just emit everything thrown at them). Oversized emitters are probably not uncommon, and certainly a case worth considering (@derek-m isn't there an emitter capacity parameter in your model).
(2) suggests the possibility of a vampire load. If there is a vampire load then this could be part or most of the apparent saving through setback, particularly in view of (1)
If the vague speculations in (1) and (2) are correct, then the implication of this and the data may be that 'batch heating' is cost effective in the particular circumstances we have here.
Here is another interesting plot, showing the minute by minute data for the morning of 7th November, the second day of the setback
Here you can see the initial peak after the heat pump is turned on, presumably due to the compensation routine. Thereafter you can see cycling which carries on throughout the day. We did already know that cycling occurs so this just confirms what we already knew, but somehow I have the feeling it may matter.
he longer periods are also cycles, albeit much extended ones. There becomes a point where the savings always outweigh any extra recovery boost costs, and if my data is correct, then that happens with 24 hour cycle. But then again, a shorter cycles (and shorter setbacks), you get a smaller saving, but also less of a recovery cost, so maybe it is always a no-brainer! O do think this riddle is worth thinking about, as a way of understanding what is happening.
In the limit shorter cycles eventually become 'short cycling'!
And yes I agree thinking about it is worthwhile. As I said above we simply cant do enough experiments to cover all (or even a material proportion) of the cases of interest, so theory and modelling will have to supplement experiment if we are to reach conclusions that are anything other than so specific as to be useful only to one person.
I am new to the forum, but have been following this topic closely, as I am sure are many other heat pump owners.
To quote Sherlock Holmes - When you have eliminated the impossible, whatever remains, however improbable, must be the truth.
I think we are all agreed that we can’t rewrite the laws of physics, even my 0-level Physics tell me this, so the appeared discrepancy in energy use after a setback period must has some explanation. All the figures used, can, and have been cross referenced with other sources, indoor temperatures, external temperatures etc. there is one exception the COP.
Just throwing this out, to other people that have more expertise it this than I have. As far as I know none of the domestic heat pumps have an energy out meter, they all use a company produced algorithm, using all the data, temperatures, flow rates, and pressures to calculate an output, this is then used to give a COP figure. What if this is flawed and is exaggeration the energy produced, and therefore the COP by say 25% in the long steady runs (no setback) and under reporting in the higher faster conditions (after a setback). Heat pumps have been designed for Scandinavian conditions; does anybody know if the software has changed for the Uk’s warmer wetter ones?
One of the nagging doubts in my mind is @cathoderay's previous difficulty in reconciling his estimated delivered energy with his previous oil consumption. Doesn't this analysis depend on accurate and consistent measurement of delivered energy? If that were neither accurate nor consistent would any of the conclusions be different?
As far as I know none of the domestic heat pumps have an energy out meter, they all use a company produced algorithm, using all the data, temperatures, flow rates, and pressures to calculate an output, this is then used to give a COP figure. What if this is flawed and is exaggeration the energy produced, and therefore the COP by say 25% in the long steady runs (no setback) and under reporting in the higher faster conditions (after a setback). Heat pumps have been designed for Scandinavian conditions; does anybody know if the software has changed for the Uk’s warmer wetter ones?
This has been a consistent concern of mine (which I have called compressor-gate, the problem is the manufacturers are marking their own homework). They could also tweak the energy in, and 'adjust' the COP that way - or even do both! However, the Midea total lifetime energy in figures are not that far from the figures taken from my external 3rd party heat pump only kWh meter, so it looks as though Midea don't cook the total energy in figure. I don't have full independent energy out monitoring (though some do, but it is costly and invasive) but I do have an inline analogue flow meter, and its readings are close to the Midea wired controller flow rates, which just leaves the LWT/RWT delta t. These do at least appear credible on the minute data plots, but perhaps rather more importantly, we use delta t in the equation, meaning even if both LWT and RWT values are cooked, but in the same direction, then the delta t may still be near enough correct. On the balance of probabilities, I am prepared to give Midea the benefit of the doubt, and say that the raw data I use to calculate the energy in and and out is 'good enough'. That said, my calculated energy in (amps x volts) is always less than that shown on the external kWh meter, by a factor of about 1.18 on average (need to multiply the calculated value by 1.18 to get the metered value). I have put this down to the amps readings perhaps not including things like one or both circulating pumps, and furthermore have not lost sleep about it because, with the 1.18 correction applied, I get an energy in value that matches the meter reading well enough.
All that said, one of the key points about the analyses I have presented here is that I only look at energy in as the variable of interest. I do this for two reasons, one is simplicity, and the other it is the one that matters for saving energy/costs, you get charged for energy in, not energy out (though of course the two are connected)!
One of the nagging doubts in my mind is @cathoderay's previous difficulty in reconciling his estimated delivered energy with his previous oil consumption. Doesn't this analysis depend on accurate and consistent measurement of delivered energy? If that were neither accurate nor consistent would any of the conclusions be different?
I think we cracked that one, the explanation being that I was running a very different heating regime, only on (and fast and furious when on) for 12 out of every 24 hours vs continuous running, and the fact I was keeping the house at design IAT 24 hours a day with the heat pump vs two significant but recoverable setbacks (the oil boiler could manage the recovery in minutes not hours) with the oil boiler explained the apparent discrepancy.
Also note my 'delivered energy' with the heat pump is subject to independent verification, by way of the external 3rd party kWh meter. The amps x volts figure is an underestimate, but a consistent enough under-estimate to mean the correcting it with the 1.18 factor gives me energy in values that are close to the external meter values.
Edit: corrected some pure twaddle I wrote about amps when I meant LWT/RWT delta t
If we're getting R-squareds of 0.9 with just the OAT vs energy consumed, then we're probably not going to be able to discern a 5-10% difference in energy consumption that hypothetically might result from a setback. You need more modelling and more data.
I am new to the forum, but have been following this topic closely, as I am sure are many other heat pump owners.
To quote Sherlock Holmes - When you have eliminated the impossible, whatever remains, however improbable, must be the truth.
I think we are all agreed that we can’t rewrite the laws of physics, even my 0-level Physics tell me this, so the appeared discrepancy in energy use after a setback period must has some explanation. All the figures used, can, and have been cross referenced with other sources, indoor temperatures, external temperatures etc. there is one exception the COP.
Just throwing this out, to other people that have more expertise it this than I have. As far as I know none of the domestic heat pumps have an energy out meter, they all use a company produced algorithm, using all the data, temperatures, flow rates, and pressures to calculate an output, this is then used to give a COP figure. What if this is flawed and is exaggeration the energy produced, and therefore the COP by say 25% in the long steady runs (no setback) and under reporting in the higher faster conditions (after a setback). Heat pumps have been designed for Scandinavian conditions; does anybody know if the software has changed for the Uk’s warmer wetter ones?
Thank you for your input. You are indeed correct, the Laws of Physics are rather strict and don't like someone messing with them.
You are also correct about the COP, or efficiency, having a role to play in the drama. Whilst I freely admit that OAT is one of the major actors on stage so to speak, there are quite a number of supporting cast.
To maintain a constant IAT, the Thermal Energy Output from the heat pump needs to match the heat loss of the building, this being very much set by the weather conditions, which have other factors besides OAT.
Whist the Electrical Energy Input is itself dictated to a great extent by the Thermal Energy Output, the efficiency of the system varies with the heating demand, and even the overall design of the system, the size and type of heat emitters, the thermal mass of the building, the overall heat loss.
All of the above combine to make interesting reading for yourself and other forum members, interesting discussion for the active participants, but as yet no definitive answers.
Thinking about installing a heat pump but unsure where to start? Already have one but it’s not performing as expected? Or are you locked in a frustrating dispute with an installer or manufacturer? We’re here to help.
