@cathoderay

You would appear to be quite good at producing charts using flawed data, and misinterpreting the results.

The lower of the two charts, which you claim is correct, appears to show that during the setback period the heat pump uses little or no energy (agreed), and that during the operational period it uses a certain quantity of electrical energy (blue trace) to supply the present heat loss and also raise the IAT back to the desired level.

According to your regression theory, if setback had not been initiated, the heat pump would not only uses the same quantity of electrical energy (red trace) during what was the operational period, but also an additional quantity of electrical energy during what would have been the setback period, and in so doing does not cause the IAT to increase above the desired level. I would be interested to know how putting in more energy does not cause the IAT to increase?

The upper chart, which you discount, does indeed show a lower electrical energy consumption when a setback period has not been initiated, thereby spreading the electrical energy consumption over the full 24 hour period, rather than an 18 hour period. Would you not agree that this should be the correct interpretation?

Posted by: @cathoderay

↑

a) the formula predicts that halving the OAT from 10 to 5 degrees all but halves the savings (may or may not be consistent, we suspect setbacks do better in milder conditions,

As a percentage of the total consumption this is correct.  Tha absolute saving for the same setback is the same.

Posted by: @cathoderay

↑

0.25*(18-17)/(18-10) = 0.25*1/8 = 3.1% (compared to 5.5% with full recovery)

ie not fully recovering (to a lower IAT1) makes less of a saving than fully recovering - something isn't right here!

Again you are looking at percentages which can of course be misleading if it's the absolute that matters.  Also in this case the way you have done the calculation is only a 1C reduction during setback which you are comparing with a 2C reduction. Quite obviously a 1C reduction will save less than a 2C reduction.  The calculation is correct.

@derek-m - I am getting rather fed up with your insinuations that I either cook the books and/or am a buffoon, malign or otherwise. Of course I am only human, and so am fallible (but not deliberately so), and am more than happy to have my mistakes pointed out to me in a civilised manner. There are some oddities in the chart, which neither of us has mentioned, me because I don't think they are noteworthy, but as I have repeatedly pointed out, it is a very small sample because your have only supplied predictions for a small number of days. I am more than happy to provide more data, but as we both know, it means a lot of processing, including that needed to generate the charts. 

That said, I honestly can't make head nor tail of your last two paragraphs. The first of these two paragraphs mentions the IAT, which isn't even on the charts (for the reasons I gave earlier, clutter). Perhaps you might translate them into English, preferably civilised English. 

Posted by: @jamespa

↑

The calculation is correct.

Got it, thanks! As I have just said, I am more than happy to have my mistakes pointed out when it is done in a civilised manner.

BTB, don't get me started on relative (gee whizz!) vs absolute (not so impressive after all...) reporting of results. It is one of my biggest gripes in the medical literature and subsequently in the lay press. 

Posted by: @cathoderay

↑

@derek-m - I am getting rather fed up with your insinuations that I either cook the books and/or am a buffoon, malign or otherwise. Of course I am only human, and so am fallible (but not deliberately so), and am more than happy to have my mistakes pointed out to me in a civilised manner. There are some oddities in the chart, which neither of us has mentioned, me because I don't think they are noteworthy, but as I have repeatedly pointed out, it is a very small sample because your have only supplied predictions for a small number of days. I am more than happy to provide more data, but as we both know, it means a lot of processing, including that needed to generate the charts. 

That said, I honestly can't make head nor tail of your last two paragraphs. The first of these two paragraphs mentions the IAT, which isn't even on the charts (for the reasons I gave earlier, clutter). Perhaps you might translate them into English, preferably civilised English. 

I appreciate that you are not happy when someone points out the weakness of your theories, but I don't think that I did so in an abusive manner, and was certainly not in the same class as some of the abusive statements that I have received from you in the past.

If you wish to make claims of electrical energy savings in the region of 20%, then you should expect such claims to be questioned.

You could try doing something constructive and ask Midea to explain why you need to apply a 1.18 correction factor to the V x I calculation.

I supplied predictions for all the days for which raw data was provided, but once again you try to imply that I am the one who has failed to deliver. Unfortunately your data has inherent weaknesses, of which you should now be fully aware, though seem to make little effort to correct.

From the raw data it would appear that your heat pump is using a great deal more electrical energy that should be required, so maybe that is the saving that you make when you switch it off.

I tried to make the text as simple as possible, so please explain which part of the two paragraphs that you do not understand.

Posted by: @cathoderay

↑

BTB, don't get me started on relative (gee whizz!) vs absolute (not so impressive after all...) reporting of results. It is one of my biggest gripes in the medical literature and subsequently in the lay press. 

They both have their place but as you say need to used carefully.  It's human nature to present things in a way that grabs attention and/or supports preconceived views..  There are three kinds of lies...

@cathoderay, @derek-m 

I wonder if its worth turning the question round and asking instead, what are the circumstances where setback will use more energy?

The reason I'm thinking that is because actually that's a 'good enough' answer.  We don't really need to know how much might be saved, but we do really need to know if there is a likelihood of loss.

From a purely physics perspective we know that there is always a saving in energy lost from the house and we know how much that saving is.  To reach a position where the saving is negative we therefore somehow have to reduce efficiency for a period, or do something else which negates the known saving in lost energy.

The most likely ones, again purely from a physics point of view are (I think)

1. recovery is too aggressive (reducing COP because its necessary to raise FT higher to get enough out of the emitters)
2. recover happens at a time when its colder than during the setback period (ditto)
3. recovery doesn't happen fully, so the human (or machine) responds by whacking up the temperature or WC curve
4. the house has a very high thermal mass so acts as an integrator.  Effectively setback doesn't occur (in the sense that the house doesnt cool) so the same amount of energy must be delivered in a shorter time, only possible by increasing FT (ie #3)
5. others that I cant yet think of

The first two are more likely if the setback 'ends' in time for breakfast, or is timed for people on a night shift.  

The third is a hidden gotcha, particularly if done shortly after acquiring the heat pump when its not yet properly tuned.  Its more likely to happen for longer set backs ('ready for teatime')

The fourth applies to a specific type of house.  However to achieve this the house needs to be both high thermal mass and well insulated - a typical passivhaus/low energy house build.  Some people 'batch heat' such houses (often at night) to take advantage of ToU tariffs.  For sure this results in a lower COP (and higher energy consumption) than if it were heated 24x7 (assuming adequate modulation depth of course), but the ToU tariff dominates.

Interestingly, if you have fancoils, then it is not necessary to raise the FT to increase the output from the emitters, so (1) can be circumvented.  Makes me think that a couple of fancoils downstairs (which is where you want morning heat I would suggest) and maybe in the bathroom, pretty much guarantees a win from setback, unless its timed for a night shift.

Just some thoughts I might try to quantify some of this over the next week or two.

Posted by: @jamespa

↑

 

So maybe, just maybe, setback and the associated effects on non-optimal systems will, on perhaps a majority of occasions, deliver benefits, and on some/many occasions the benefits may be bigger than one might attribute to setback alone.  That's about as much as I would be prepared to say for now and we definitely need more data

Are you sure about that? Really sure? 😆 

Posted by: @derek-m

↑

I appreciate that you are not happy when someone points out the weakness of your theories

For the umpteenth time, I post my data and ideas so they can be reviewed, successes approved and errors pointed out. If I am not happy with this, I have one very simple solution: stop posting. But I am still posting. 

The problem I have with your comments is that most of the time they just in effect say 'you are wrong' without giving adequate explanations beyond 'you're wrong'. As I said before, I can't make head not tail of your recent comments. Here's the first paragraph in question:

Posted by: @derek-m

↑

According to your regression theory, if setback had not been initiated, the heat pump would not only uses the same quantity of electrical energy (red trace) during what was the operational period, but also an additional quantity of electrical energy during what would have been the setback period, and in so doing does not cause the IAT to increase above the desired level. I would be interested to know how putting in more energy does not cause the IAT to increase?

I don't actually have a regression theory, unless having a regression equation amounts to a theory. I see it more as a summary of observed data: how one observed variable varies with another variable, and as such, I don't add any particular theory, I have just summarised observations. Perhaps thinking I can summarise observations using regression is a theory...

Having re-read the paragraph a few times, I think you are asking why, given the 'extra' predicted energy in during the setback - and this is where I get confused, because there is no actual additional quantity of electrical energy in, meaning it can't affect the IAT value. In effect, your question becomes a non-question. What the red line attempts to show is what the predicted energy in would have been had there not been a setback, but it never actually happens, so it cannot affect let alone raise the IAT. As I have said IAT is not on that chart because it gets too cluttered, but it is clearly visible on many other charts, and what happens is, as expected, the IAT falls during the setback.

Put another way, perhaps you are asking why something that didn't happen didn't happen. If so, the answer is in the question, it didn't happen because it didn't happen! Somehow, it seems you may have mixed up a hypothetical situation (what would happen without the setback) with the actual situation (a setback). Or perhaps something else, I don't know, as I say, I can't understand the question.  

It does occur to me that maybe you have dismissed the basic idea, comparing observed vs expected values, and so struggle to grasp the idea that the expected value, in the context of actual use of the method, as opposed to testing the predictions, as we are doing here, is never a real value that actually happened, it is always a predicted but never realised value that would have happened had something else happened, rather than what actually happened (now I'm becoming incomprehensible).  

Or perhaps you are just over-complicating things. The basic idea behind observed vs expected could not be simpler, take the observed value, and compare it to the predicted value. That said there are of course very legitimate questions over how the prediction is made, and how accurate it is, and that is exactly what I am getting at with this chart: where we do have both, I can test the actual observed value against the predicted value, visually, by assessing how close the red line is to the blue line.

The second paragraph:

Posted by: @derek-m

↑

The upper chart, which you discount, does indeed show a lower electrical energy consumption when a setback period has not been initiated, thereby spreading the electrical energy consumption over the full 24 hour period, rather than an 18 hour period. Would you not agree that this should be the correct interpretation?

I don't 'discount' it, I merely observe the upper chart red line is further from the blue line. You then state this upper chart "does indeed show a lower electrical energy consumption when a setback period has not been initiated" [emphasis added]- but I don't see this - and so am confused because it is something I can't see. In both charts, the non setback period (left hand two thirds of the charts) shows both prediction methods get fairly close to the actual values most of the time; perhaps your spreadsheet method is a little spikier? But overall, they really don't do too badly. The spreadsheet predictions only become noticeably low during the setback 'on' periods, suggesting there is something the spreadsheet model fails to take into account during those periods. The reason this matters is those too low values will in turn lower the 24 hour predicted totals, which in turn will have the effect of lowering the predicted savings.

The whole (and only) point of the charts is to compare the predictive ability of the two prediction methods, to see which gets closer to the actual values when both observed and predicted values are available, with the OAT added for context. During the setbacks, the OAT does rise (not unexpected, but how much of that is heat pump off and how much is natural variation I do not know), and both methods show a fall in predicted values. But we have no idea of how they compare the the actual values, because they don't exist (or perhaps more accurately they are zero) during the setback. That leaves the recovery period and the remainder of the 'on' period, where we can compare the actual and predicted values. The regression method makes no attempt to spread anything out over the full 24 hour period, it just predicts what the energy in would be for a given hour at that the current OAT for that hour. It also misses the clearly visible recovery boosts, because it doesn't know about recovery boosts (the regression was deliberately done only on 24/7 running, no setback data).     

Posted by: @derek-m

↑

You could try doing something constructive and ask Midea to explain why you need to apply a 1.18 correction factor to the V x I calculation.

I supplied predictions for all the days for which raw data was provided, but once again you try to imply that I am the one who has failed to deliver. Unfortunately your data has inherent weaknesses, of which you should now be fully aware, though seem to make little effort to correct.

There is no Midea UK, just Freedom Heat Pumps, and as you know they will not usually talk to end users. It is true I haven't asked Midea in China, but I am not sure where to start (Midea is huge), or how far I will get. Nonetheless, I will give it a try, thanks for the suggestion.

I'm not trying to imply you failed to deliver, because the very next thing I said was I am more than happy to provide more data, ie I know why you have not supplied more data - if anyone has failed to deliver the data, it is me. I will post the whole December 2023 data shortly, which has the full recent set back period in it, and can post January's data when it becomes available. 

I am not sure what I am supposed to have not corrected. Most of it is raw data, just numbers made available by the Midea wired controller, inherent weaknesses and all, and the one calculated value which clearly does need a correction (energy in, because it doesn't match the external kWh meter) already has a long established correction. Maybe I will get a modbus enabled external kWh meter, maybe I won't, if I do, then I hope we can agree its data is as close to gold standard as is possible. The OAT I have left as it is, because it represents what actually goes into the heat pump, rather than some other OAT from air that never goes anywhere near the heat pump.   

Posted by: @jamespa

↑

I wonder if its worth turning the question round and asking instead, what are the circumstances where setback will use more energy?

The reason I'm thinking that is because actually that's a 'good enough' answer.  We don't really need to know how much might be saved, but we do really need to know if there is a likelihood of loss.

Interesting approach, will give it some thought too. At the same time, I don't think we ever answered the 'when does the no-brainer become a brain teaser' question: when, and more importantly why, do longer definitely make savings setbacks become short enough to mean that the saving is no longer 'beyond reasonable doubt'? It may also throw light on the matter because of what happens at that point.

@derek-m (and anyone else who wants it) - here is my December 2023 hour and minute data in csv format:

@cathoderay

For the umpteenth time, I post my data and ideas so they can be reviewed, successes approved and errors pointed out. If I am not happy with this, I have one very simple solution: stop posting. But I am still posting. 

The problem I have with your comments is that most of the time they just in effect say 'you are wrong' without giving adequate explanations beyond 'you're wrong'. As I said before, I can't make head not tail of your recent comments. Here's the first paragraph in question:

So please explain why in the past have I received so much verbal abuse from you when reviewing your data and ideas.

For the umpteenth time, I advised you of the inherent problems with your system and the raw data produced from it. I think that it may have been over one year ago when I identified how the PHE would have a detrimental effect upon overall efficiency, and more recently the 'cold well' effect, again affecting efficiency, and also affecting the OAT sensor reading upon which your predictions are totally based.

You are obviously struggling to understand some of my explanations, so in future I will try to keep them as simple as possible.

Posted by: @jamespa

↑

@cathoderay, @derek-m 

I wonder if its worth turning the question round and asking instead, what are the circumstances where setback will use more energy?

The reason I'm thinking that is because actually that's a 'good enough' answer.  We don't really need to know how much might be saved, but we do really need to know if there is a likelihood of loss.

From a purely physics perspective we know that there is always a saving in energy lost from the house and we know how much that saving is.  To reach a position where the saving is negative we therefore somehow have to reduce efficiency for a period, or do something else which negates the known saving in lost energy.

The most likely ones, again purely from a physics point of view are (I think)

1. recovery is too aggressive (reducing COP because its necessary to raise FT higher to get enough out of the emitters)
2. recover happens at a time when its colder than during the setback period (ditto)
3. recovery doesn't happen fully, so the human (or machine) responds by whacking up the temperature or WC curve
4. the house has a very high thermal mass so acts as an integrator.  Effectively setback doesn't occur (in the sense that the house doesnt cool) so the same amount of energy must be delivered in a shorter time, only possible by increasing FT (ie #3)
5. others that I cant yet think of

The first two are more likely if the setback 'ends' in time for breakfast, or is timed for people on a night shift.  

The third is a hidden gotcha, particularly if done shortly after acquiring the heat pump when its not yet properly tuned.  Its more likely to happen for longer set backs ('ready for teatime')

The fourth applies to a specific type of house.  However to achieve this the house needs to be both high thermal mass and well insulated - a typical passivhaus/low energy house build.  Some people 'batch heat' such houses (often at night) to take advantage of ToU tariffs.  For sure this results in a lower COP (and higher energy consumption) than if it were heated 24x7 (assuming adequate modulation depth of course), but the ToU tariff dominates.

Interestingly, if you have fancoils, then it is not necessary to raise the FT to increase the output from the emitters, so (1) can be circumvented.  Makes me think that a couple of fancoils downstairs (which is where you want morning heat I would suggest) and maybe in the bathroom, pretty much guarantees a win from setback, unless its timed for a night shift. 

Just some thoughts I might try to quantify some of this over the next week or two.

Item 1. is by far the biggest factor IMO.  This is from when I let the Ecodan algorithm do its thing unfettered.  It's hard to factor in defrosts but this uses the same method as previously where I thought savings were around 20%. I've included LWT on this one; it's up to 55 deg and delivered power is over 13kW at times.

The other point that hasn't been mentioned is that many people (I am included) have a setback to enhance comfort because they don't like the house to be 21 deg at night.