That’s an excellent review @hydrosam and well worth a read. It also confirms some thoughts I’ve had for a while which is that the main knowledge base of much of this new tech are the users themselves. 
I can understand that companies/installers want to move on and not spend time without additional funding but it also means that any training initially dedicated to the installation is diluted as users often cannot get further information or dialogue after a while. I believe that same to be true for Solar PV, Battery Storage and other systems so these reviews are key to inform others in the initial phase of adoption 

Thanks for sharing your thoughts on HP sizing, very interesting.  If I put in my 2 pence I am def leaning toward bigger is actually better, simply because rather than your view of average efficiency, personally i'm more interested in whether my HP will be still be efficient/work when it's really cold, could be considered a critical issue. That being the case, an oversized HP will not have to work as hard in colder conditions and so will both work down to lower temps and higher efficiency. This is because whilst the Spec  for different kw sized HP might have the same min operating temp, an oversized one will be required to deliver less heat and so can drop the flow temps lower, resulting in less Defrosts and a higher COP, a bit like small and large car engines, one is more efficient in low demand cond (Town), the other is more efficient in high demand cond (Motorway).

Therefore a Smaller HP will not be able to deliver enough heat at low flow temps in colder weather, so will have to raise the flow temp, increasing Defrost cycles and reducing COP.

It's also worth mentioning comparing different HP models could  be like comparing Apples and Oranges,  as newer models might just have an improved design and so their  improved efficientcy isn't a function of their size.   I’d suggest to anyone looking to improve their HP efficiency to run through their settings and start from the absolute best case which Heatgeek suggest is the lowest Flow temp you can get away with, set a high flow rate and reasonable target temp 16/17 and just leave it running 24/7 and monitor the target temp. If you can achieve the desired target temp within a few hours then you have prob found the best COP for your system, if its not delivering enough heat or taking too long or you want a higher target temp then put the flow temp up a few degrees and go again.

Obviously the performance will change as the outside temp cond vary but it should be poss to find the best settings for you home/HP/ best COP this way.

It's also interesting that HydroSam mentioned that cycling isn't good, which I completely agree with, but I would also say it depends on when this is happening?.  I have found with the lowest flow temp setting cycling is minimised, and certainly when starting from scratch never occurs, but as the house gets upto temp it should be normal for the HP to come on less and less frequently. This isn't a sign of  a poor system rather it's that the HP heat demand is reducing and that as that happens work the HP has to do drops, flow and return temps converge and COP is at its peak. Then how frequent the HP comes on will be determined by the heat loss factor of the building, which I completely agree is the most important variable and prob differs the most between any two properties. I think HydroSam has done a great job of showing that improving the industry standards/methods for properly assessing both the heat demand (property size and use) as well as the heat loss (U values) are crucial in getting a individual HP system to achieve the manufacturers highest COP performance.

@hydrosam a cracking piece of work. The UK heating installer industry are known as “boiler slingers” for good reason and over sizing is a symptom of poor training largely. Translate your findings across the UK housing stock you will find massive energy savings so clearly Gov should concentrate efforts to improve design.

@hydrosam this is now in the public Smart home category and converted into a blog article - such good insight, thank you for sharing. One thing to finish this off would be a landscape photo of your current heat pump. Feel free to send to me directly or upload as a comment and I’ll add this as a ‘featured’ image.

@Tim_OVO the beast in all its glory

It’s on the side of the house by the vegi patch and we never hear it. It’s also not as high up as it may appear. There is a footpath just below it. 

A short comment on the point above about wanting the pump big enough to work when it’s really cold from @OrphM62. It’s definitely a consideration but in my case the heat pump never put out more than 5kw of heat, even when it was -6C outside in early Jan. I think COP dropped to about 2.5, it’ll be interesting to compare with @jason.lewis who I think has a very similar heat demand to my house if I remember correctly when we compared stats a year or so ago. 

But if you were to operate with a more traditional on/off heating schedule for a few hours a day you’d want a bigger output because you are constantly reheating the fabric of the property every time. It’s a minefield of scenarios to consider. 

@hydrosam when I get a moment I will have a go at HTC calcs to see how our home 4.3kw as calculated by the survey compares. 

I’ve been listening to a series of podcasts from Nathan Gambling called BetaTalk. On the 100th episode he had on Guy Newy (from Energy Systems Catapult) who’s had a heat pump installed and had his heat loss calculation done by Veritherm https://veritherm.co.uk 

They put sensors and heaters all over your house and measure the actual heat loss overnight (with nobody in it). *edit* also mentioned in energy-stats !! But neither gave cost to do this.

The podcast is definitely up our street, it even tops the Heat Geeks !

Great analysis and explanation. Very useful and thought provoking. 

Be interesting to hear any experience of what the maximum output of say a 6kw heat pump actually is at say - 5 ambient temperature 

My understanding is that the maximum output depends on the ambient temperature to some extent. The colder it gets the lower the maximum output. 

For those Dakin models at what ambient temperature is that 6kw rating specified and how much does is vary based ambient temperature? Is the variation so small that it doesn't need considering when sizing the pump at very close to the building heat loss? 

@Jeffus your understanding is the same as mine. Output depends on ambient temperature, and also the running/leaving water temperature the pump is trying to achieve. Also the refrigerant used has an impact. I’ve looked at a few different manufacturers and it’s difficult to find the data, not many publish it. 

I know the Valliant heat pumps output higher than their named output, for example their 5kw unit can output 6kw at -2C outside/

After a bit of trawling I found some info on my pump. Daikin seem to name their heat pumps based on Ambient air temperature 7°C and leaving water temperature (LWT) 35°C. The output table for my 11kw is:

If you can keep the LWT low then the output doesn’t drop off significantly until -2C, whereas for higher LWT, which is how systems tend operate under colder conditions, the output drops. For info my LWT is set to 40C at -5C outside, so the theoretical max output is around 10.5kw under those conditions.

@hydrosam thanks, that is good to know the information is at least available, albeit there isn't a standard way of providing it which is a shame. 

Have you seen anything about how much the defrost cycles in practice limit the maximum heat output over and above this? Are there calculations made to allow for this?

Thanks again for sharing your practical experience and knowledge on the forum. It will be really useful when asking installers questions i am sure. 

@Jeffus the defrost cycles don’t bother me much at all. They tend to last about 5mins and don’t happen very often now. 

The challenge is that you will be more defrost cycles under certain conditions. 
1. It’s cold and damp, 2. The LWT is at the higher end.

So when my system was first setup with a fixed LWT of 50C, the heat pump was defrosting a couple of times an hour even when it was just mildly cold outside (e.g., below 7C), meaning the system struggled to get to temperature and the house struggled to warm.

Now I run a much lower LWT, defrost cycles are less common and not noticeable within the house to us, it maybe occurs 2-3 times a day when it’s below 0C. The data will be available within the high resolution data collected on our trial but we don’t have access to that data so it’s difficult to give definitive evidence on this. I can only judge based on the electrical consumption data I have from my battery from which I can tell when the heat pump cycles. 

Do you have any photos of ice on your heat pump @hydrosam ? It would be useful to compare to mine, which is soon to be tested for a low refrigerant level.

@juliamc surprisingly I don’t have any photos. It was basic like a light frost. Only once was it frozen solid, like the inside of my unthawed freezer. And this was after a fault cause by the first cowboy installers. 

Looking at the electricity profile, whilst the actual Defrost only takes 5mins the whole process of the HP stopping, defrosting and then getting back up to working conditions is about 15mins and uses some energy.

The LWT setting has a big effect, dependant on outside humidity and temp. It would be great to have a graph plotting the three variables inorder to avoid defrosts/ improve COP, but I just monitored my HP and if I saw it frosting up, I just lowered my LWT by a few degrees and frosting was far less frequent, it does seem as though a few degrees can make a big difference.

Any photos @OrphM62 ? 

First Pic is HP frosted up, Second is after defrost

Date was Morning 17th Dec 2022, Oxfordshire, overnight temp -3 degrees.

Overnight temp the day before was 16th Dec Temp was -5 degrees.

So the early morn temp when the photo was taken was prob between the two.

At the time I still had Heat batteries so the flow temp LWT was 45 degrees, this frosting was happening every 45mins.

This Feb after my Heat batts were removed and on a night when the outside temps were the same -3, I found the HP frosted within 15 mins at 38 degree LWT but after I lowered it to 35 frosting was much reduced and at 33 didn't occur within the few hours time frame I monitored it.

I should also mention that I believe the frosting in Feb occurred at a lower LWT than in Dec because whilst the outside temps were roughly the same the humidity on that day in Dec was lower, after a number of cold nights in a row with no snow, whereas in feb we had alternate warmer and cooler days with some rain.

Is that similar to what you see @juliamc 

@Jeffus I have only just seen this question - almost exactly a year ago. Yes, it confirmed the fact that there was something seriously wrong with mine. It had been suggested that the refrigerant was low when I mentioned the bands of ice (on another forum). Eventually I had the refrigerant removed weighed and recharged. Spot the difference !!