Hi folks, I’m getting poor COP readings for my DHW run (Altherma 3 8kw monobloc), these are typically 1.6 - 1.9. The 150 Ltr tank is set to 50 degs and runs around 12am for 60 mins, I’m only used about 50 Ltrs per day of the hot water, so it doesn’t drop too much in temp. The install was a retrofit by an independent installer, it’s the Daikin tank. The pipe run is about 8m to the tank and most of the pipe run is 28mm. Peak flow temp is 64.4 and return 61.1
I’ve dropped my data below.
The heat output seems to collapse at the end of some runs, but is okay in others.
Consider dropping the target tank temperature to 45°C. This requires less energy to heat, with higher COP and will avoid the top end of the heatpump’s capacity.
It would be useful to plot the tank temperature, if you have it. Set up another instance if ‘My Heatpump’ app with that as the room temperature feed.
There’s possibly some air trapped in the system, so maybe try bleeding the radiators and automatic air vents a few times.
If you are heating to 50 and not using a lot and it has to reheat from say 48->50 it won’t get great efficiency.
If you aren’t using that much water why not heat it to just 40C? Or think about doing a scheduled heat every other day?
Thanks for the ideas, I’ll drop the target temp to 45 and look to get the tank temp added to the data plot. I have been struggling with air in the system since the buffer tank was removed, a deaerator was added but it hasn’t fixed the air problem.
One other thought… I’m using the ESP Altherma and HomeAssistant link to flag the DHW water cycle, I can see from my data that is sometimes dropping out, which is likely impacting the COP reading. Is there a better way of telling emon that the unit is on a DHW cycle?
Hi Adrian,
It seems odd that the Flow & Return temps are over 60C when heating the tank to 50C. I also have a DHW target temp of 50C and typically the Flow peaks at 55C with the Return around 51C, so you’re showing a whole 10C higher. It will be interesting to see what happens with the DHW target set to 45C instead.
Running for a full hour at roughly 7kW of heat output is a lot of heat to dump into a 150 litre tank. (My DHW run-times are around 20 minutes.) 7kW for an hour is enough to heat 150 litres by 40C (assuming I’ve managed to use one of the online water heating calculators correctly), i.e. taking your whole tank from 10C to 50C (or 20C to 60C).
Making the tank temp visible somewhere would definitely help with understanding what’s going on.
Thanks. I should have added, that it doesn’t normally heat for an hour, I’ve just got an hour set on the schedule, it normally heats in 30-40 mins. I’ll get the tank temp added and report back…
High DHW temps are not your friend! From your first post, it looks like you are at the limit of what the EDLA8 is designed for, and from the Daikin capacity tables the COP is not surprising.
I successively reduced my tank temp down to 34C. This is enough for a deep bath at “quite hot enough”, only running the hot tap. I avoid blending with cold, as that just throws away all the extra energy in pushing the heat up to the higher levels. I also set it to come on as close as possible to when I want to use the water. I suggest trying the same process to find the set-point which gives you just hot enough water from the hot tap at point of use.
Also: its not just a good COP which matters, but the total energy in to achieve sufficient hot water. Decreasing the set-point is a double “win”.
Thanks. I do have an EV so benefit from circa. 8p / kWh, but it’s still nice to maximise efficiency
TLTR, but here goes:
It seems to me that DHW tanks are the trickiest item in a HP circuit to analyse/optimise – fluid properties and temperature differences are changing rapidly with time (so you can’t apply steady state heat balance equations), and it also takes a significant time for the convection pattern to develop (an essential component of heat transfer). Also DHW tanks are prone to temperature layering, which takes time to break up.
Many heat pump controllers do not constrain LWT during DHW heating (unlike space heating where temperature compensation limits LWT), which is a particular problem if the DHW tank coil is undersized – the circulating fluid can get very hot (at a rubbish CoP) before significant heat is transferred to the tank contents, because convection takes a while to get going.
If your DHW demand is modest, or at predictable times, you may be able to programme DHW just once a day (or even two). In this case, one partial answer to energy minimisation may be to heat DHW in two or more short bursts, with a few minutes’ gap between. This (I surmise) allows the convection current to start developing before the LWT has wound itself up too far, and distribute some heat around the tank before the next burst.
In my case I can heat my 150 litre DHW tank from 20degC to 40degC in 2 x 15 minute bursts, with 5 minutes between bursts. Although the total heat required is the same as a 1 x 30 minute burst (obviously), it is delivered at a maximum LWT of 55degC rather than the 60+degC that occurs during a single longer burst, with the consequent improved CoP.
The above observations are for a Samsung HP but I thought it might also be relevant to some Daikin users. Why not give it a try and see if it helps?
Interesting notes on the physics of the matter, @SarahH. I’m intrigued by the 2 stage heating; the Daikin systems do seem to rapidly escalate the HP LWT, so maybe the same approach could work.
Something which I think I have observed, which you could maybe comment on from your ChemEng perspective, is that the temperature of the drawn-off water declines by more than I can account for by the stated cylinder heat loss when I allow it to stand for a while.
So: with my 34C set-point, if I draw off as soon as that is reached, I find the water starts off very hot. 1 bath-full is somewhat less than 1 tank-full (150 litres) so the water is still clearly warm when there is enough water in the bath; it is nowhere near cold supply temperature.
If I leave it for an hour or more, it starts off much cooler. At a 2 hour wait, the bath is either deep and a bit tepid, or shallow. I put this down to convection giving a fairly-well stratified tank while the HP is running, with conduction spreading the heat in the 1-2 hour period, such that the total heat drawn into my bath-full is less. I also suppose the fact that I start with 20C water throughout the tank gives a strong temperature gradient during the heating phase. If this is true, it further illustrates the trickiness of the analysis.
I’d guess that this is caused by the inrush of cold water when a large water volume is drawn off the tank top (e.g. a bath’s worth). It’s likely just simple mixing, rather than convection. Maybe it’s time to switch from a bath to a shower (the lower draw off rate of the latter doesn’t seem to upset temperature layering too much, so retaining a decent hot water temperature for a couple of days 
).
Thanks Sarah - really helpful. I like the idea of running two short heat up cycles to avoid longer runs at 60+ flow temps, I’ll try that.
Let us know how you get on 
.
I did think of trying 3 x 10 minute bursts with 5 minutes between each, but diminishing returns and all that…
I get the best cop for DHW on my daikin EDLA04 with 185l joule cylinder by:
Lowering the target temp (43 is plenty for me).
Letting the tank cool down before scheduling a heat up.
Setting the cylinder to 3rd party, small coil. (My coil is larger but this forces it to keep the output low)
Enabling Quiet mode on the quieter setting.
This gives me a long, slow, efficient cycle
That’s unusually good for the time of year as it was quite warm overnight, heating from 26.6° to 43.3°, using just over 1kWh at at COP of 4.7
Oooo this is a nugget of information I’ve not seen before on how to configure the Althermas to get better COP in DHW cycles. Going to try it out!
So experimenting with these settings I’m not sure that changing the tank to 3rd party, small coil made any difference to my HP performance (I have the 6kW EDLA, so essentially the same unit as your 4kW).
However, setting the quiet/more quiet/most quiet (I’ve settled on “more quiet”), as you mentioned, did makes a noticeable difference to the power consumption during DHW cycle. The good thing (imo), is that this mode only knocks down the top-end output of the unit (and so increases DHW COP) whilst not impacting CH performance. I will keep an eye on when temperatures get colder to see if this stops the unit from being able to return the tank to target temperature (mine is set to 46C, which the unit always overshoots by 1C 
)
Ive had a play with Quiet mode today,
these are on reheat - not shedule
before quiet mode, starting temp of water heating is 20 degrees C
Running the DHW in 2 x bursts has definitely helped with the COP, last night it was ~2.9 in phase one and 2.15 in phase 2. In phase one the flow temp is hitting 51 deg and 60 is phase 2. I’ve also added a plot of the DHW temp rise (I don’t think I can add that to the emon plot)?
Good work, @UB71.
Just for a laugh why not try 3 x 10 minutes (with 5 minutes between each) one day if it’s easy to set up. Might improve your CoP even more. Let us know if there’s any further progress - we might even have a Best Practice on our hands here…
