Friends, this week I remembered that I am an idiot.
You may have been aware of this for sometime, and indeed I have known it too, but amidst the buzz of day-to-day activities I forgot. Realisation of my status as a level-10 idiot was triggered by events in my ongoing quest to improve the efficiency (seasonally-averaged COP or SCOP) of my heat pump.
Over the last two winters, the SCOP of my heat pump has been around 3.5, which is a really excellent result – much better than I hoped for when I had the heat pump installed. But the data wizards at Open Energy Monitor have established a dashboard that allows one to monitor dozens of heat pumps across the UK – and many of those installations with the same make and model of heat pump as mine, are achieving SCOPs of 4 and above!
Click on Image for a larger version. Graph showing the performance of the heat pump through the heating seasons of 2021/22 and 2022/23. The red data (read against the left-hand axis) shows the weekly-averaged COP and the red dotted line shows the annually-averaged SCOP. The blue data (read against the right-hand axis) shows the average amount of heat pumped per day in kWh/day. Averaged weekly, 60 kWh/day corresponds to 2.5 kW continuous power.
And friends, I began to wonder what might be causing my system to underperform. Being an experimental physicist, I thought I would try an experiment. This led me to hire HeatGeek elite heat pump ninja Syzmon Czaban to remove the Low Loss Header in my heat pump installation. You can read about that adventure here. But early data indicates that this has made very little difference!
And then I remembered the Golden Rule of Experimental Physics (Do it quick. Then, do it right) and its corollary which states that “Two weeks in the laboratory can save a whole afternoon in the library“. In other words, thinking about things first is harder than just messing about, but it can save a lot of time.
And after thinking about things, I realised something very very obvious: that the SCOP figure is not always a good indicator of heat pump performance, and that my SCOP envy was consequently pointless. Doh! Allow me to explain.
A model of heat pump performance
Making a model of how a heat pump works is tricky. So I made the simplest possible model with the fewest possible assumptions. And I then considered how the same heat pump would perform in three different dwellings with three different levels of ‘fabric efficiency’ i.e. insulation. I then embedded my model in a spreadsheet which you can download here. Most of the parameters are changeable so you can see how the insights gained might apply to your installation and in the description below I have included my default assumptions in [square brackets].
Click on image for a larger version. Screenshots from the tastefully coloured spreadsheet.
Firstly, for each dwelling I worked out how much heating it required. This assumed:
- There was a background level of space heating due to human heating and electrical dissipation [300 W or 7.2 kWh/day] and that the balance of the heating was supplied by the heat pump. Dwellings with a larger heat transfer coefficient (HTC) require more heat to raise them 1 °C above the outside temperature. [100, 200, 300 W/°C or 2.4, 4.8, 7.2 kWh/day/°C]
- Hot water was suppled by the heat pump [3 kWh/day]
- The total amount of heat supplied was the sum of the two figures above. In summer this would be mostly water heating and in winter this would be mostly space heating.
Secondly, I assumed that each dwelling had the same heat pump installation and that the COP was determined by the indoor and outdoor temperature only. In other words, the COP was not affected by whether the heat pump had to pump just a little heat or lots of heat. I assumed the COP declines linearly as the outside temperature falls, with the COP for the DHW operation being lower because it heated water to a higher temperature.
Click on graph for a larger version. The graph shows how COP is assumed to vary with outdoor temperature for both DHW and space heating. Indoor temperature is assumed to be 20 °C.
Finally I estimated how much electrical energy the heat pump would use by adding three terms together.
- The electrical energy used for space heating is the amount of heat supplied each day divided by the COP for space heating.
- The electrical energy used for DHW is the amount of hot water supplied each day divided by the COP for DHW.
- The background electrical consumption by the computer that operates the heat pump. [20 W or 0.48 kWh/day]
Then I estimated the overall COP for each dwelling by dividing the total amount of heat supplied for DHW and space heating by the total amount of electricity used. The results are shown below.
Click on graph for a larger version. The graph shows how overall COP varies with the amount of heat pumped each day for three properties with good, average and poor insulation. Notice that except in very mild weather (when little heat is delivered), the COP for the poorly insulated home is better than the COP for the well-insulated home.
What this calculation shows is that for identical heat pump installations, a heat pump operating in a well-insulated home will have COP values which are systematically less than a heat pump operating in a poorly-insulated home.
What this means is that over a year, the seasonally-averaged COP (SCOP) will inevitably be larger (better) for a house which requires more heating.
I think this result is both surprising and obvious. It’s counter-intuitive because in general a well-insulated home is easier to heat (with a heat pump or in any other way). But it is also quite obvious because the overall COP is the average of the DHW COP and space-heating COPs, and the more space heating that is performed, the better the expected COP.
Click on graph for a larger version. This is the same graph as shown at the start of the article but with additional labels. It shows how COP for a heat pump is assumed to vary with outdoor temperature for both DHW and space heating. In Summer the average COP is due mainly to domestic hot water heating. As one journeys into winter, the average COP is influenced more by the space heating COP.
Model versus Reality
All this modelling is very well. But does it correspond to reality? In fact the assumptions I have made are pretty reasonable.
Since my heat pump was installed in August 2021, I have each week recorded the amount of heat it has pumped and the amount of electricity it has consumed. From this I have calculated the weekly-averaged COP. The results for the last three seasons are shown in the figure below.
Click on graph for a larger version. The graph showing weekly-averaged data for my heat pump installation over the last three heating seasons. The graph shows the overall COP versus the amount of heat pumped on average each week. Also shown as a dashed green line is a fit to the 2022/23 data which suggest the maximum achievable COP (on warm days) is about 4.9.
During the coldest days last December 2022, I also evaluated the COP daily (rather than weekly). This was the overall COP including both DHW and space-heating. I noticed that the COP fell roughly linearly with external temperature. Based on this data I think the model describes the basic behaviour of my installation.
Click on graphs for a larger version. Graphs showing data from the cold weather in December 2022. The left-hand graph shows how the daily COP varied as a function of the average external temperature. It fits modestly well to a straight line falling from about ~5 to about ~2 at -10 °C. The right-hand graph shows how the average amount of heat pumped each day varied with the average external temperature.
So I conclude from this that it is reasonable to assume that the COP varies linearly with the external temperature.
COP Envy: A discussion
So if the model matches the reality, then my envy of those installations with superior SCOP results is probably misplaced. The most likely cause of my ‘low’ SCOP (and 3.5 is not really low) is that the heat pump simply doesn’t need to do very much space heating: this makes it super-cheap to operate. But it means that in the bulk of the heating season – the times which contribute most to the SCOP – the pump overhead and the amount energy used for DHW was probably the reason that the SCOP did not reach 4.0.
This analysis does not rule out the idea that the installation could be improved. For example, there could be a flow impedance that makes the hydraulic pump work harder than it should. Or my assumption that the COP does not depend on the amount of heat delivered could be flawed because of the way the heat pump cycles on and off at low load. But despite these possibilities, this calculation makes me suspect that the installation probably cannot be improved by much – unless I turn down the temperature to which I heat my hot water – or leave the windows open!
Protons for Breakfast 
November 1, 2023 at 10:14 am |
Michael,
There are only a few people who are level 10 idiots. It is the stuff of dreams. Thank you for what you do, we less grounded fools are genuinely grateful for your comments. Travel well and thank you.
Nick
November 1, 2023 at 10:17 am |
Ian, thank you. Your comment made me smile. But I still feel like an idiot 🙂
Maybe that’s just the way things are.
Have a good day yourself: M
November 1, 2023 at 10:24 am |
Interesting , one thought. I know you have little faith in the valiant recorded figures nor do I. But the yield figures available from the wall unit with the volume knob does give working figure for DHW and heating separately and the split between energy yield (total energy delivered is this yield figure plus the electrical energy used. ) although you have found these to be in accurate i would expect the error to be similar for heating and DHW. so even if the actual figures are not right it can indicate the relative energy uses between DWH and space heating.
Having said that i though you might be interested to hear my reported results from Vaillant. I have no independent heat meter, so have to infer from what Vaillant report. Our scop (working figure ) to September was 5.4 heating and 4.2 DHW! not a fantastically well insulated building but outside the data sheet by quite a bit! i have worked hard to get the performance up but this figure suggest my machine really over estimates the heat delivered. it says it delivered 12,500kwh for the year. our gas use for the previous year was 8,900khw. so with 90% boiler efficiency ~8,000kwh delivered. Although our comfort level has increase I’m would be surprised if the heating demand is up 56% . Based on our heat pump electricity use of 2600kwh and the previous heat demand that is a scop of about 3.1. so with our better comfort and cold December ’22 the real figure lies between these two widely different figures! in future I will just focus on bring the input down and forget about the COP.
November 1, 2023 at 12:28 pm |
Joe, Good Afternoon.
Thank you. As you say, I have largely disregarded the Vaillant data but since it’s easily available on the app, I might have a look at it again. As you say, it would be interesting to see if the ratio of DHW/Space Heating was more accurate than either number individually.
It is frustrating that the Vaillant data is not better, but the main thing to remember is that you are comfortable! Chasing the numbers is just our hobby!
Warm – literally – best wishes.
Michael
November 1, 2023 at 10:25 am |
Michael,
You are a good man Your brother’s not too bad as well
. It has been one of the more fortunate occurrences of my life to have been infested by DePodestas.They are fine fellows.
November 1, 2023 at 10:55 am |
Thanks Michael – all really great intel which I am squirrelling away for the future when I get a heat pump, and passing it on to my son who is probably nearer to the deed.
I love the truism that two weeks in the lab can save an afternoon in the library – same in chemistry of course. Less fun though, and no accidental break throughs!
November 1, 2023 at 11:56 am |
Graham,
Good luck to you and your son with your endeavours.
And Yes, The Golden Rule and it’s corollary are *so* true it’s almost painful to remember of times I have broken these rules and regretted it!
All the best
Michael
November 1, 2023 at 11:02 am |
That’s a good perspective view! Yes indeed, if the same heat pump is used in a well insulated property running at the same weather compensated flow temperature and the same inside and outside temperatures as a poorly insulated property, then the well insulated property’s heat pump is more likely to be cycling – so it has less chance to settle to efficient running spending more of its time ramping up and the quiescent load is also going to be more significant. So, of course the poorly insulated home is more likely to have a higher COP with longer or even continuous running cycles with the quiescent load becoming a smaller relative component.
In this situation however, the better insulated home should be able to lower the weather compensated maximum flow temperature until the same more continuous running cycle periods are achieved as the poorly insulated home and at that point the better insulated home should achieve a better COP.
There is a caveat to this though in that a heat pump should be sized to the heating demand of the property because its output will only modulate within a certain range and if a lower heat output is required below the minimum it will achieve this by cycling. So your poorly insulated home example should have a larger heat pump than the well insulated home and smaller heat pumps trend to have a higher COP.
So anyway, you could try reducing your maximum flow temperature to achieve a better COP. I did this with mine on the coldest days last year – reducing the upper temperature so the heat pump runs continuously yet still keeps the house at the desired temperature.
November 1, 2023 at 11:55 am |
John, that is an interesting point, but it is not one that I investigated in the article. In the article I assumed that the heat pump operated identically in all the dwellings and that the COP depend ended only on the temperature difference. So I am assuming that the flow temperature was the same in all dwellings at the same outside temperature. Implicitly this assumed that the poorly-insulated dwelling had much bigger radiators.
The point that you are making is different. You are saying that because of the cycling effect, a 5 kW capacity pump delivering 1 kW is less efficient than a 5 kW capacity pump delivering 5 kW. I guess that is true – but I don’t know the size of the effect and so I have ignored it in this analysis.
All the best
Michael
November 1, 2023 at 11:13 am |
Ultimately, what matters to us is how much it costs to be comfortable on a monthly/yearly basis. You can get incredibly efficient systems if they run 24×7 at optimum capacity, but a measure of efficiency itself leaves out the number of units consumed/spent. COP/SCOP by itself is a vanity measure. Number of kWh in a year (or even kWh/year/m2) are far better measures of successful heating system design imo.
November 1, 2023 at 11:49 am |
Paul,
I agree with you.
COP/SCOP are useful indicators, but without knowing the amount of heat pumped, it is – as I have discovered – an indicator not worthy of envy.
As you say, kWh/year/m^2 is a good indicator, but one also needs to know geographical location and internal temperature. A house with an internal temperature of 22 °C will (in the UK) consume about 20% more than a house kept at 20 °C.
All the best
Michael
November 1, 2023 at 12:40 pm
You are of course right! “Degree days” are one common further refinement to understanding this metric like-for-like with other properties.
November 1, 2023 at 12:54 pm
Indeed.
November 1, 2023 at 11:38 am |
Michael, I do appreciate the content, and the humour with which you deliver your musings.
November 1, 2023 at 11:44 am |
Craig: Thank you: Humour is a weak antidote to despair ;-). All the best: M
November 1, 2023 at 12:02 pm |
Michael, thank you! This is a really valuable insight. Now that you’ve pointed it out, I can see that the SCOP is a function of the intrinsic properties of the heat pump, intrinsic properties of the house, how it’s installed and how it’s used. It’s the house part that messes up comparison. I wonder if there’s a way to factor that out, perhaps dividing SCOP by some measure of the house’s heat loss rate. I don’t know the answer, but I do know you’ve moved our thinking forward significantly.
November 1, 2023 at 12:32 pm |
Duncan, Good afternoon,
Yes, SCOP is a global measure, and a useful one, but not the only one that matters. As another correspondent here pointed out, it should really be quoted alongside : kWh pumped/year and possibly kWh pumped/year per square metre, and possibly internal temperature.
I do love heat pumps, but as one learns to love something (or someone) over time, one comes to appreciate their fascinating mysteries.
Best wishes
Michael
November 1, 2023 at 1:23 pm |
Cop will start to rise if the temperature goes even lower. The hardest conditions for h.p. are from 0 to -5 to -7 Celsius. You are hitting frequent defrost cycles. Search for defrost algorithms of your machine in service manual. When – 5 or more air is becoming more dry and less ice is accumulated… Or other staf i dont understand. I know that my air to air daikin worked best that 1 week at -20 in the nights and -10 to -15 during the days, or my prayers kept it working or physics and good engineering .
November 1, 2023 at 1:58 pm |
Dear Est Mo, Good Afternoon.
If you are experiencing weeks at these low temperatures then I guess you are not writing from the UK.
You are right, the de-frost cycles impact heat pump performance most in the temperatures around °C – I would estimate ±5 °C around 0 °C.
With an Air-to-Air heat pump it’s hard to know the COP because it’s tricky to measure thermal output, but it may well have been that your Daikin unit was operating with a COP between 1 and 2. As you say – excellent engineering keeps these units working at the time they are under most stress.
All the best
Michael
November 1, 2023 at 3:24 pm |
Fantastic analysis, as ever.
I think the summary might be: your heat pump is too big, even though it’s the smallest Arotherm you could fit. That your install only runs continuously when heat demand is 1.8 kW and upwards is very interesting; I’d love to know the same figure for the smallest Ecodans and Samsungs.
I bet your CoP would be higher, and your kWh electricity consumed lower, if you set the DHW cycle to early afternoon when outdoor temperature is usually highest. I know that could reduce your rate of return on your battery, but what’s more important?
November 1, 2023 at 3:46 pm |
Zarch AKA Mick Wall went down the rabbit hole trying to establish the minimum output of Vaillant heat pumps here: https://community.openenergymonitor.org/t/vaillant-arotherm-owners-thread/21891
It is also not helped that i have found MSC required heating calcs to be a significant overestimate. In my case at -2c the heat pump was running at about 50% to maintain our comfort. it feels measurement may be a good tool in the future rather than calculation leading to oversized systems.
November 1, 2023 at 5:00 pm
Joe. Those are good points, but that is not what the article is about.
The article assumes that heat pump COP only depends on delta T, the difference between the indoor and outdoor temperature. The article assumes that the scaling of the heat pump output happens perfectly.
The effect you’re talking about, in which heat pump outputs well below maximum output have lower COP is *additional* to the effect I am describing.
November 8, 2023 at 11:14 am |
OK I now admit that on first reading I didn’t understand what you are saying. Second time around, I think I do!
So yeah my comment about heat pump sizing was off beam. However I stand by suggesting doing the DHW cycle early afternoon to increase your sCoP and decrease total energy consumption!
November 8, 2023 at 1:52 pm
I agree. The overall COP could be quite sensitive to small changes in DHW COP. I am investigating this, but the whole processor gathering data is tedious.
All the best
Michael
November 1, 2023 at 4:49 pm |
Any sort of boiler/heater that has to cycle on and off will be less efficient than one tha runs less but for longer when it is running. It may be the other systems have large thermal stores. It may also be the pump location is on a warm south facing wall so the heat pump spends less time in a defrost cycle in cold conditions. It could also be that the other heat pumps are run with underfloor heating where the heat pumps run for long periods at lower temperatures.
November 1, 2023 at 5:04 pm |
David, good afternoon.
How cycling affects COP is unknown, but this calculation assumes that cycling has no effect. It assumes that COP depends only on the difference between the indoor and the outdoor temperatures.
Any ‘cycling effect’ would be in addition to this.
M
November 2, 2023 at 10:11 am |
How hard would it be to seperate the data so you only use the electricity usage and heat output in your calcs from times of day when dhw is not being heated? (I would also ignore the 1st half hour after switching from dhw to heating)
November 2, 2023 at 11:12 am |
Ian, Good Morning,
“How hard would it be?”. Well- it’s not trivial. I am currently looking at the COP for DHW and seeing how it depends on the heating power. On the Vaillant Arotherm plus, one can either heat on full power (about 6 kW) for about 30 minutes, or in ECO mode – which heats at about 3 kW for an hour or so. But the analysis is slow and tedious precisely because it isn’t straightforward to separate DHW and space heating.
I’ll take a look it again and perhaps I will be inspired!
M
November 2, 2023 at 6:40 pm |
I agree with Ian Nicholson — if you’re indeed an “idiot”, then you’re the sort who makes the world a better place. Dostoevksy’s classic novel “The Idiot” comes to mind!
I’m also reminded of standard practice in computer performance systems engineering — where (following Jain’s advice) the competent engineer establishes the goals and define the boundaries of the system *before* we select a performance metric that’s appropriate to this particular project. Before signing-off on the final report, we carefully review the important decisions we had made along the way, notably including our choice of metric. If you want to get a high-level overview of this style of performance engineering, you’ll find some nice lecture slides at https://tu-dresden.de/zih/ressourcen/dateien/lehre/ws1112/lars/vorlesungen/lars_lecture_02_requirements-metrics-techniques.pdf. And … it’s not standard practice among computer system performance engineers to call ourselves “idiots” when we make a mistake!
Hmmm… consider this proposition (copied from Brunst & Mueller’s lecture slides) that “The right metric to measure the performance of an analyst is not the number of analyses performed but the number of analyses that helped the decision makers.” Seems to me quite likely that you’re doing quite well on this metric! (Prince Myshkin move over, there are additional moral goals in the 21st C! 😉
In the vernacular of computer systems performance engineering… I believe you have a (morally-inspired) goal of minimising your domestic consumption of electricity for purposes of heating your house and supplying its hot water. You have quite a few factors you can vary, e.g. tweaking the heatpump by removing its LLH, adjusting the thermostat on your space heater, adjusting the water temperature, taking longer (or shorter) hot showers. Some of these factors affect both the workload and the COP; so if you focus solely on maximising the efficiency metric, then you won’t be reliably minimising the energy consumption of your household. And it’s certainly the case that what you’re describing as “COP envy” makes about as much sense as believing that a car that’s fuel-efficient in city driving must be also fuel-efficient on the motorway. It’s always a case of “horses for courses”, whenever you have a technology (such as a heatpump or a passenger vehicle) whose workload is quite variable and whose efficiency varies significantly with one or more characteristics of the workload.
In the UK context — where (in my vague understanding) heatpumps are routinely used as retrofits for fossil-fueled boilers which supply hot water for space heating *and* for showers — the COP is especially problematic as a one-size-fits-all efficiency metric. Googling, I see there’s also a reliance in the UK on an SPF — which I’d characterise as a combined metric (on the space-heating portion of the workload, multiplied by its COP). But if you don’t also use a combined metric (maybe call it HWF?) for the hot-water-supply portion of the heatpump’s workload, then I can’t see how the UK could draft adequate energy-efficiency regulations for its heatpumps — unless perchance the houses are generally so poorly insulated that the total energy used to power heatpumps (across the nation!) is controlled mostly by the SPF (i.e. the HWF isn’t very important because the total energy used for hot water supply is so much smaller than the total energy used for space heating).
All to say that your “mistake” is great food for thought!
November 2, 2023 at 10:16 pm |
Good Evening, And thank you for your kind words.
The general point made in the slides – that ‘performance evaluation is an art’ – is a very interesting idea, and I think I agree with it. And it is *very* important.
Because of the way improved performance is chased by seeking to improve ‘metrics’, if the metrics are not well-aligned with the real performance requirements, the nominal ‘improvements’ can drive the actual performance further and further away from optimal. This is a corollary of Goodhart’s Law.
Your comment has now changed my perspective. I still think I am an idiot: or at least I feel like one so much of the time I just feel very comfortable with the epithet. However, I think you have set this issue of heat pump performance in its proper context, and it definitely needs a multi-dimensional performance metric. I will set my dumb brain on to the task of thinking what might be smart.
Warm best wishes:
Michael
July 1, 2024 at 9:40 am |
I am pretty sure your analysis is wrong. The COP falls as the output increases. Not vice versa.
A well insulated house requires less heat pump output, there is less entropy, and a higher COP.
July 1, 2024 at 9:53 am |
Roderick, Good Morning,
And I am pretty sure my analysis is right! The key figure is this one:
The graph shows that at low outputs, the COP is poor because the DHW and static electricity consumption are relatively large. The average COP rises as more space heating is required and then at large loads (i.e. when it’s cold outside) the COP falls. If you insulate a home, the amount of time spent at high COP operation falls and the SCOP does too. The data on my own system back that up. You can find the latest data in this article.
Of course the well-insulated house will have much lower bills, but the SCOP might not be so good.
The article includes a spreadsheet and if you disagree with me, I invite you to find the error in my reasoning.
Best wishes
Michael
July 2, 2024 at 10:32 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 10:42 am |
[…] 12 months, de Podesta wrote a weblog submit entitled “COP Envy Is Pointless,” by which he defined that when a home is extraordinarily nicely insulated—as his is—the warmth […]
July 2, 2024 at 10:47 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 10:47 am |
[…] yr, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” during which he defined that when a home is extraordinarily nicely insulated—as his is—the […]
July 2, 2024 at 10:50 am |
[…] yr, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” by which he defined that when a home is extraordinarily properly insulated—as his is—the warmth […]
July 2, 2024 at 10:50 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:01 am |
[…] año pasado, de Podesta escribió Una entrada de blog titulada “La envidia de la policía no tiene sentido” En este artículo, explica que cuando una casa está muy bien aislada (como la suya), el SCOP de la […]
July 2, 2024 at 11:04 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:08 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:08 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:09 am |
[…] yr, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” through which he defined that when a home is extraordinarily properly insulated—as his is—the […]
July 2, 2024 at 11:15 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:16 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:17 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:22 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:27 am |
[…] 12 months, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” wherein he defined that when a home is extraordinarily properly insulated—as his is—the warmth […]
July 2, 2024 at 11:28 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:32 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:34 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:35 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:35 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:36 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:39 am |
[…] metais rašė de Podesta dienoraščio įrašas pavadinimu „COP Envy Is Beprasm“ kuriame jis paaiškino, kad kai namas bus itin gerai izoliuotas – kaip ir jis – šilumos […]
July 2, 2024 at 11:45 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 11:49 am |
[…] yr, de Podesta wrote a weblog submit entitled “COP Envy Is Pointless,” during which he defined that when a home is extraordinarily properly insulated—as his is—the […]
July 2, 2024 at 12:26 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 12:28 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 1:04 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 1:27 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 6:59 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 7:23 pm |
[…] year, Podesta wrote A blog post titled “Police envy is pointless” In this article, he explains that when a house is very well insulated (like his), the SCOP of the […]
July 2, 2024 at 7:38 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 8:42 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 9:37 pm |
[…] yr, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” by which he defined that when a home is extraordinarily properly insulated—as his is—the warmth […]
July 2, 2024 at 9:42 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 2, 2024 at 9:47 pm |
[…] 드 포데스타는 현재 여러분과 같습니다. “COP Envy Is Pointless”라는 제목의 블로그 게시물 집이 잘 저온되어 있을 때(그의 집처럼) 열펌프의 SCOP은 실제로 저온에 […]
July 2, 2024 at 10:24 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 3, 2024 at 1:19 am |
[…] ano passado, de Podesta escreveu uma postagem de blog intitulada “A inveja da COP não faz sentido”, no qual ele explicou que quando uma casa é extremamente bem isolada — como a dele — o SCOP da […]
July 3, 2024 at 6:30 am |
[…] 12 months, de Podesta wrote a weblog submit entitled “COP Envy Is Pointless,” during which he defined that when a home is extraordinarily nicely insulated—as his is—the […]
July 3, 2024 at 2:00 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 4, 2024 at 7:07 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 4, 2024 at 1:16 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 7, 2024 at 12:19 am |
Well done for writing this contrarian article. It’s overdue.
I found your youtube videos on heat pump sizing from gas boiler consumption and HDD data extremely helpful.
However, in those videos you do not mention the electricity consumed to power both the existing boiler and the central heating circulator pump. For this reason it’s not true comparison to include the power of the ASHP circulator pump in efficiency calculations.
Armed with an estimated Heat Transfer Coefficient (HTC) for my house and a design temperature from HDD most installers just will simply not install the smallest heat pump to match that figure. They really want some margin for underestimation.
The current ASHP technology lacks a wide range of modulation, particularly at moderate ambient temperatures. This makes sizing the heat pump with the objective of making nice SCOP figures over a wider temperature range a challenge.
The data on OEM shows why this affects the overall system COP.
The ideal SCOP system only runs the pump and compressor simulataneously and almost continuously.
At moderate ambient temperatures with weather compensation the system temperature is lower allowing the heating to be incredibly efficient with a high instantaneous COP.
However, the efficiency is lost because once the central heating water is at temperature the compressor stops leaving only the circulation pump(s) running.
The sawtooth pattern of this water temperature modulation is clearly visible in the OEM charts.
In a system with a buffer the saw tooth has longer time constants. The RMS/average temperature of the water is lower than the leaving flow temperature of the heat pump.
The argument goes that the additional pump and slightly higher temperature necessary to run a head/buffer decreases efficiency.
Although there is some truth in this the reality is apparent lower COP is a result of oversizing because this prevents the heat pump from running continuously at more ambient temperatures.
The trade-off is at low ambient temperatures and running the ASHP at 100% (which makes more noise).
Ideally at the design temperature the ASHP should run at approx 75% max output for that temperature.
AFAIK the manufacturers prefer the sawtooth water temperature modulation over the continuous running approach.
July 7, 2024 at 10:16 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 8, 2024 at 4:09 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 9, 2024 at 11:01 pm |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 16, 2024 at 10:30 am |
[…] year, de Podesta wrote a blog post entitled “COP Envy Is Pointless,” in which he explained that when a house is extremely well insulated—as his is—the heat pump’s […]
July 16, 2024 at 1:17 pm |
[…] yıl de Podesta şunları yazdı: “COP Kıskançlığı Anlamsız” başlıklı bir blog yazısı bir evin son derece iyi yalıtılmış olması durumunda (kendi evi gibi) ısı pompasının […]
October 23, 2024 at 11:01 am |
#1 A nice approach, and once you jump to the final conclusion it is indeed obvious. It all hinged on the ratio of hot water to central heating usage. The COP is determined by ambient temperature and the design water flow temperature. So if radiators are resized to suit a heatpump (they should be) then the optimum operating conditions can be met, whereas the hot water store temperatures will always require higher, less efficient heat pump operation.
#2The focus on COP/SCOP is completely fixating the heating industry at the moment mainly because electricity cost is three time that of gas and they need to show a saving, but as this article proves it is just one metric in a bigger picture. The fixation also ignores (the COP figures don’t include) the heat taken from the home to defrost the heat pump in our cool (5°C) humid winters. This heat lost to defrosting can be a significant at times. A worried heat pump commissioning engineer confided he had seen systems not delivering the expected appliance output to the home because at times they spent 20% of the runtime extracting heat from the home and dumping it back into the garden. Beware installers under sizing the system, they assume a 6kW heat pump (at its correct rated temperatures) will always deliver 6kW. In reality during defrost it will take heat back out, then need to recover it again, before the house can return to heating again.
#3I suspect retrofitting heat pumps in our marine climate is going to be an ongoing challenge if the assumption is that ‘heat pumps work in (dry cold) Scandinavia so they must work here’ when our current poorly insulated, leaky housing stock is also included in the issues to be solved.
#4Not by choice I have a 6kW electric boiler in a wet central heating system (yes, a terrible COP of just 1.0) so you would think I would be ideal and desperate for a heat pump? However when I looked at the costs I took a different approach. I have a fully zoned heating control that controls the time and actual temperature of each room separately using the existing radiators. It can (and does) operate down to a heat input of around 10% of just one of the radiators output (say about 100Watts) Because it reduces the heated area and the time it means the USAGE is very low. In my case the summer usage (household and hot water) is 197kW.hr per month and 634kW.hr per month in the winter. I would love a heat pump, they are a great idea, but reducing my £104 difference per month to a third (with a COP of 3) and saving £400 in electricity a year would struggle to pay for itself (don’t get me started on the ‘free’ £7500 installation grant – we are all paying for this in the end!)
#5I suspect the same financial or upheaval arguments will be true for a lot of gas boilers installations. They should move away from gas – but maybe not to a heat pump? Electric boilers are very small and cheap and they don’t need all new radiators and a high efficiency cylinder or the £7500 gift from all of us. Doh! like I said…. bigger picture.
October 23, 2024 at 2:08 pm |
Andrew, Good Afternoon. I have updated your comment to include paragraph numbers so I can refer to them.
#1: Yes, exactly. The more space heating an installation does the better the SCOP.
#2″The focus on SCOP/COP is understandable. It is a good general measure of a good installation. If COP is high then there are probably no major errors, but if COP is low it tells one *something* is wrong, but not precisely what.
I have not found a problem with de-icing cycles. It is perhaps a function of the placement of the outdoor unit. If there is an element of re-cycling of the air, I think the de-icing cycle could well spiral.
#3. YEs it will be an ongoing issue. I don’t think the climate is an issue, I think the technology is. ASHP with air-to-water technology are a good replacement for a boiler in many houses. But in smaller houses and flats – particularly ones without existing hydronic installations – air-to-air systems make much more sense. I understand these are teh systems popular in Scandinavia. They have fantastic COPs for space heating because they only have to heat the working fluid to maybe 25 °C to deliver heat to a dwelling at 20 °C.
#4 I have meant to write about dwellings such as yours – I have a couple of other examples I have encountered – which are already electric (so low emissions) but where a heat pump doesn’t make sense. I’ll get around to writing about that one day.
#5 Regarding the avoiding a heat pump, I cam across a house nearby which had a marginally-sized heat pump and a supplementary electric heater which might be needed for a couple of days a year.
Anyway, thanks for your considered comments.
Best wishes
Michael