Struggling to get good COP from Daikin 9kW

Hi Matt recently had 9kw pump installed and am struggling to get cop above 3 though not using much electricity. I suspect the pump is over sized even though I had three surveys all about 8kw. Was wondering about your thoughts on adding a large radiator in a sun lounge with a lot of glass that we tend not to use in the cold months. I’m thinking it would enable us to use the sun lounge more and also possibly make the pump more efficient and so heat the room with very little extra cost! Have enough room to put a large k3 say 4000w radiator, total radiator output at the moment is 26000w. House is very comfortable and I’m very pleased with the heat pump and costs at the moment but feel I should have a better cop closer to the design.

Hi John,
Perhaps you can tell us more about your system configuration.

What weather curve are you running and what temperature control (Madoka thermostat, or leaving water temperature etc.)? Have you experimented with changing the settings?

How are you monitoring the system? Are you reading values from the Daikin MMI, or do you have third party monitoring like openenergymonitor?

Welcome to the forum, John. I’ve split this out into a new topic as the previous one was quite long. There’s probably worth reading through it to see what advice has been given by others.

What and type property do you have?

This may suggests that the heat pump is running fine but the estimation of heat produced is off.

Do you have some daily figures for electrical consumption compared to gas consumption before?

Hi Tim my average gas consumption per day in December was 65kw for hot water and heating and I’m using on average at the moment for heating and hot water 15-16kw which is interesting because points to a cop of 4 ! Can’t remember December being that cold making the gas use that much more than it would be now!? Maybe then then MMI isn’t very accurate.?
The house is a mid 80s house 125m2 good loft insulation, old double glazing and cavity wall insulation put in 2000

House 125m2 good loft insulation and cavity walls done about 20 years ago in a mid 80s built house, so insulation ok but not brilliant. CB heating design 45c @ -2 currently using WDC of 40c @-3 and 28c@18c I’ve set a schedule of 23c from 7am-9.30 pm then 9.30pm-7am 18c with madoka control. At the moment pump is not coming on at night time and is only dropping 1-2 c over night! Data access is through MMI, the energy usage seems pretty accurate 82 kw for last week heating only but have no way of knowing if heat out put is correct! I find it really strange that whatever hot water usage we have its always 2-3 kw a day giving a cop of 1.5 - 2.5 . I’m also running modulation of 1 but finding what ever I change not making a lot of difference to cop which is for last week ranging from 2.7 - 3.7 average being 3 heating only. I’m reading a lot about the 9kw pump being a beast and maybe over sized for my house! I have a 20m2 fun lounge with atrium tagged onto the lounge which isn’t heated at the moment and would be fairly easy to put a big radiator in! By doing this I’m wondering if it would make the pump fit the hose better and we would be able to use the sun lounge on colder days with out costing a lot in heating costs. I’m reading people who seem to get decent cops from the 9kw unit have a large radiator output!
Appreciate any thoughts.

1 Like

Looking at other similar properties on HeatpumpMonitor.org, they’ve consumed around 500-600 kWh of electricity over the last 30 days (so 16-20 kWh per day), and closer to 700 kWh for January (23 kWh/day). Around 2000-3000 kWh of heat, or 66 - 100 kWh per day. This seems in line with the average consumption figures you quoted, so it’s likely that you’re getting reasonable performance from the heat pump after all.

The heat output is estimated by measuring the temperature of the flow and return pipes, and sometimes the sensors are not accurate. May be worth seeing if you can find them, and double-checking that they are securely fastened.

Adding a radiator may help the system run at a lower temperature and cycle less. Other users may be able to advise on how to optimise the controls.

I’m also struggling to get a good COP from the 9kW model. Our house has a heatloss of about 6kW at -2.
What I’ve found is that the COP is as expected below about 5degC. But much above that temperature the heatpump won’t run continuously (as it’s lowest compressor power is about 900W so heat output is around 3.6kW) and so we get significant losses. The only way around this is to deliberately increase heatloss by opening windows!
Best reported COP I’ve seen from the daikin MMI is 4.0.

I think your logic is correct on the extra radiator - it’s “use it or lose it”.

The way I see it I’ve got nothing to loose just turn it on and off as needs be! Could turn out as a useful tool to improve cop. Do you think that the MMI figures are not very accurate then? Had one day reading 3.8 but mainly now around 3. My point being the house is really comfortable and using on average for last week of 13 kw a day heating only! Hot water is always 2 or 3 kw and always 1.5-2.5 cop. Was hoping someone might have compared MMI data other monitor data.

Hi John,

My MMI data matches my MID monitoring pretty closely.

Based on my system, the MMI data is correct.

Your radiators are a reasonable capacity and you should get a reasonable COP at a flow temperature of 36c to 40c.

There is no point going lower than 36c when it is warmer outside as you save no electricity and just reduce the heat output making the COP worse.

Although that is OK if you value comfort over COP.

COP isn’t everything, you will use the same amount of electricity at a COP of 3 as you will at 4 at times.

Just don’t feel that something is wrong, it isn’t.

You have a minimum electrical input of 900w or so, you can’t get below this.

Adding a radiator and heating another area is a good idea, most of the time it will cost you nothing or next to nothing.

Quoting a daily electricity use doesn’t mean much, from the amount you are using you cannot be running it all day and night.

Running in shorter periods may sound Ok but the first hour or so will probably be at the worst COP, you could find your efficiency drops.

You really need some monitoring to get the best out of your heat pump, this one particularly.

I can only recommend heat and electricity meters although you can extract data directly from the heat pump.

Proper monitoring is worth every penny and will help you get the best from your heat pump.

I have found some settings that work for me but the first thing you need to do to improve COP is stop it going below about 36c flow temperature, you are giving away efficiency if that is what you are after.

1 Like

@JohnC

Madoka
Since you are using Madoka thermostat control, you might want to adjust the settings on the Madoka menu to get more accurate readings from it. I’ve seen quite a lot of self-heating from the blue LEDs and backlight.

Flow temperatures
I agree with Matt that you probably don’t want to go lower than mid-30s for the flow temperature. So I’d suggest you make your weather curve much shallower, 40C-35C perhaps.As Matt has noted on many occasions, the larger Daikin units (9-16kW) don’t have a huge range of operation for our normal sized houses, and we are running them towards their minimum output.

Monitoring
I don’t get any meaningful values from the MMI as my electricity usage shows up as basically 0 or 1kWh per day. (No one has seemed interested in looking into this, either Octopus or Daikin).

I’ve gone with ESPAltherma monitoring. Stephen has written a great post explaining how to do it here:

You need to be comfortable opening up the outdoor unit and probably want some experience with Home Assistant. It gives a lot of insight into how the unit is operating, but does rely on the accuracy of Daikin’s sensors. There’s debate on how accurate they are…

And of course, since you are on this forum, you could look at getting the full monty heat pump monitoring from

Continuous running or switching the unit off?
Again, a topic of some debate. While I think we all learn that running a heat pump continuously is most efficient, that requires the heat pump to be able to match all of the ranges of heat demand for your property. It’s likely that the Daikin 9kW can’t go low enough for medium sized UK houses, so continuous running is impractical on warmer days. On the cold days, you are probably better running nearly 24 hours a day, at a medium flow temperature (40C?) than asking the heat pump to ramp up the temperature in the morning with 50C flow - that will just defrost a lot.

Switching the heat pump off, either via thermostat or schedule, is certainly reasonable. It is very efficient not to run the heat pump :slight_smile: I have been meaning to measure the ‘startup’ cost of the first hour or so until it reaches stable operation (after a previous heating cycle), but it depends a lot on ambient conditions.

I would suggest that if you are using thermostat control, have a look at the backlight settings mentioned above. I saw some horrible behaviour where the Madoka thermostat temp was reached, the blue light would go off and the measured temperature dropped and after a few minutes the heating came back on! Frequent on-off-on cycles for the heat pump are not good for efficiency, but if you are running it for a couple of hours at a time, it should be pretty good. (From looking at some of the 4-8kW units on heatpumpmonitor.org, it looks like they can be cycled a lot and still give decent efficiency! Grr.)

There’s a long thread covering some of these topics:

2 Likes

Back up heater
One more thing, consider turning off the back up heater:

The wisdom here is that we don’t typically need it in the UK. If you find your system is unable to heat your home or hot water during a cold snap, you would re-enable it. But it seems to come on more often than it should in normal operation, and is just a 3kW electric heater with a COP of 1.0.

There might be an argument that it helps the system get up to temperature more quickly, but most people here who are getting good efficiency have the BUH disabled.

Radiator balancing
Try to balance your radiators. Note that the ones nearest to the heat pump in the circuit may require the lock shield valve almost fully closed. Preventing too much hot return water getting back to the heat pump should help prevent cycling, as well as directing the heat to where it is more needed.

TRVs, open-loop and auto bypass valve
The perceived wisdom is that you want to open all your TRVs and give the heat pump as much water volume as possible to use. If radiators valves shut down, other parts of the system have to emit more heat to compensate and heat pump has less water volume to work with.

If you have an automatic bypass valve (ABV), you may want to consider closing it. (Take a picture of the initial setting, so you can return it to that value if you need to.) I’m definitely not an expert on this, but as I understand it, ABV is not required if you have some (or all) TRVs always fully open. I think the theory is that an incorrectly setup ABV will again let some hot return water back to the heat pump, and cause more short cycles.

Daikin MMI / installer settings
Increase around 0C
This makes the pump increase the water temperature around 0C to compensate for melting snow and ice on your building. You probably want to switch this off, as it just makes the heat pump work harder at cold temperatures, probably increasing defrosting.

Overshoot
Set this to 4C, to give your heat pump more leeway to have water temps above the set point.

Modulation
I’m unclear how effective this is on the 9kW and larger units. It seems like a great idea, but having the thermostat often changing the leaving water setpoint seems to cause lots of changes in power and people seem to get better efficiency with this reduced, or switched off. It also seems to affect the Madoka thermostat control, allowing the room temp to go up to a1.5C above target, although I haven’t seen this confirmed in writing.

It depends on how responsive a system you want, but having the heat pump boost the target water temperature by 5C on a cold morning will probably lead to more defrost cycles. Slow and steady wins the efficiency race :smile:

Hey @JohnC ,

As you have seen a few of us here tackling this unit, I am someone who had a recent swap.

With regards to house size comparison 1976 4 bed detached build, 110m2, 600mm loft insulation, cavity wall with thermal beads, pre1997 double glazing. Our heat loss is circa 5.4kw@-2 (original survey was 7.3kW as they forgot the loft insulation). Had the 9kw fitted and a key indicator for me was the COP increasing below 5°c.

If you have not already try putting your postcode in here and see what the heat geek guesstimate recommends, Upgrades - Heat Geek, you can also run your own heat loss survey with heatpunk free online.

If the figure is below 7/8kW then you would be best placed to follow up with the installer about a swap (best thing I did), but otherwise some more radiators or opening some windows to allow for heat to “escape” may help. @matt-drummer is the expert here as he has virtually tamed the 9kW beast.

Also remember two points

  1. Standard MCS heat loss surveys overestimate heat loss by up to 30% (usually due to air changes being overstated or insulation discarded).
  2. You will only ever need the full power for 10/12 days a year on average, any other time of year the unit will be oversized (hence you need to be VERY close with the heat loss survey for it to be as efficient as possible).

Personally I would rather have a pump that is slightly undersized and for the couple of days a year (and optimized for the majority of it Spring/Autumn) have a small fan heater or jumpers for if / when needed. In reality it is more likely that your house would just be a little cooler and not actually cold.

2 Likes

Thanks Zak interesting heat geek came back with similar heat loss of over 8 kw but I’m sure that includes sun lounge which I don’t heat. Will definitely put a large radiator in there and use it more in the winter now. Will be interesting to see if the cop is affected!

Thanks Jon will give your suggestions a try. I’ve had to set TRV to 3 (5 fully open) upstairs just gets to hot! Not sure how to balance the radiators, should this have been done by the installers? Will investigate how to do this. Thinking I might turn modulation off. Modoka is definitely 1-1.5 c higher than it is or the other thermostat I bought to move around rooms is wrong!
Overshoot I haven’t even looked at, set to 4 and see what happens.

Thanks Matt I’ve been going the other way reducing temperatures. Will change to 45@-3 and 30@ 18 then go from there! Definitely going to add a big Radiator in the sun lounge as I’m sure the pump can handle it! You’re right I’m running the pump from 7am -9.30pm at 23c then reducing to 18c between . The house in this milder weather is only loosing 2-3 c so in affect is turning off. It only seems to draw more power 2-3 kw for the first 1/2 hour before dropping to 900w. Gradually over the day temperature builds up so warmer in the evening and appears not to cycle much.

Balancing radiators
Should the installers have done it? Maybe, but probably not. My installer just roughly made sure every radiator was heating up…

There’s a big thread here:

The Heat Geek article a few posts down is probably a good place to start. I’ve used a little infrared thermometer and a two probe thermometer to try and balance the radiators, with varying success.

You don’t need to aim for perfection, but if your radiators are wildly unbalanced then it’s undoing some of the purpose of the radiator replacements and might be leading to too much warm water returning to the heat pump, and more frequent cycling.

One thing that took me a little while to get used to: if you have a cool radiator (kitchen, in our case) you have to restrict the flow on the radiators “before” it in the piping run, to send more heat to the one you want. It’s a slow process, make small changes and leave it a bit to see if it improves.

1 Like

I see most people here are targeting dt5 on every radiator.

While this is a good start if you have accurate heat loss figures per room you can mathematically work out the DT per room based on the exact output of the radiators and the heatloss. Most radiators will be slightly oversized and by differing amounts compared to the rooms exact heatloss, so while a standard DT of 5 is a good start it can be improved upon (for the truly COP driven)

While this takes a bit more effort and does require some time in excel it should be easy enough for most people to do.

Feel free to take a look at my spreadsheet attached for my current system and the upgrades (new radiators) (due tomorrow)

It also helps to show system volume and makes it easy to work out both the lowest system design temperature with specific radiators and the lowest temperature the system can realistically run at.

Zak-W_SystemOverview.xlsx (16.4 KB)

Most of the formula is courtesy of Paul Spence from the Heat Pumps UK Facebook group, I just popped it into a pretty spreadsheet.

Negative figures show your radiator is undersized, any DT over around 10 is not going to be too realistic, but you can adjust room temperatures to suit your personal requirements (we use two bedrooms as offices so they are kept at 20°c/21°c

Also another point for the Testo Bluetooth clamps, they are great!!!

Any questions feel free to ask.

5 Likes