Interest Ali.
I have only seen this on mine at the start of a space heating or dhw cycle. The same on @KnightPhoenix
So, fault or design design feature?
1 Like
Just been looking at his again, could it be some sort of airlock in the system that is fooling the kramstrup meters again?. When the System restarts it is cleared (as the pump starts at max flow rate and purges anything) and the issue seems to happen when the higher flow temperature is required (such as for DHW cycles or space heating in colder days).
The Carnot heat on @alibloke’s graph above seems to show we are still getting the output…
Have you tried checking the aav’s or radiators for trapped air?
I have a Sontex heat meter.
Carnot heat output on @alibloke is way low
And this is yours on 20 November 2024, Carnot heat is way too low?
Hi Matt,
I’m still working through purging “air in the system” errors, but I’m not thinking that today’s trace is connected with that…
First time I’ve seen this dip in Flow temp and CoP during my DHW cycle. 14:15 to 14:25
“It came out of nowhere” and cleared up about 10 mins later. Curious.
Daikin 8kW. Sontex heat meter.
What has happened to all the extra electricity consumed in those 10 mins? Does it just warm up the outdoor chassis for some deliberate antifreeze reason?
Hi David,
I had Octopus engineers here yesterday.
It was a good visit but we made no progress other than agreeing it appears to be a `design feature’ rather than a fault that can be rectified.
These units have been around fr a year or so now and maybe `cracks’ in the design are showing up.
I am no expert but I tend to agree with @johncantor that liquid refrigerant is finding its way into the compressor.
I only see this at the start of a space heating/dhw cycle but @alibloke has the same as you.
I can only assume that the liquid refrigerant drops the flow temperature and the electricity is consumed vaporising the refrigerant for no heating gain.
I am sure @johncantor can correct me and/or explain better.
It is an annoyance at best, not the end of the world, but a concern if it gets worse or occurs more often.
I am at a dead end with it for the time being.
Just something else to add to my growing list of dissatisfaction with owning a Daikin heat pump.
That is a dramatic drop off isnt it. As you say, the flow temperature drops (or rises less), and of course this shows up as lower output. As you say, instead of the evaporating happening outside, some of it is happeneing inside the compressor. If it all evaporated there, the COP would be 1! That is a long time for this to be happening… 10 mins. I guess the discharge temperature will be very low, and normally this would ‘signal’ for the expansion valve to close. Are we seeing the water flow-rate reducing when it sees the dt recucing?? There is of course a lag before a flooded evaporator corrects itself, and many compressors would be signalled to rev-up if discarge temp is dangerously low, in order to get out of trouble.
I need to spend a bit of time looking at the graphs… excuse my forgetfulness, but I struggle to remember whos system is who’s.
I could believe that as they ‘tune’ systems to be more energy-efficient, and like a performance car, it may cough and splutter occasionally. Are we seeing a cough? If they change the algorythm to fix this, they might reduce the net averge COP, but I really have no idea. I do have a hunch that the normal sampling rate for feed-back control is too slow. maybe it has been fine for air con. Maybe they all need to sample more often… Its only a hunch though. Maybe for the maunfacturers point of view, its not a big enough or frequent enough problem?
2 Likes
Hi John,
Mine is here, in Ipswich
I see no change in flow rate.
The COP did get pretty low on mine, less than 2, and I think I saw the same on others.
Perhaps my biggest concern is that this wasn’t a feature that was apparent until recently.
That makes me feel that it is something that has developed over time, maybe a wear issue?
I don’t know.
I do like the analogy to a performance engine though 
Sorry seeing this thread so late, but it’s definitely EEV problem. Your Daikin heat pump is not the only one. I have Atlantic Alfea with just the same problem. My guess is a lot of other brands have the same problem. It’s definitely a design issue for me. Getting the exact superheat from the heatpump is the most difficult moment in refrigerant cycle. For giving the correct value for the EEV you need evaporation pressure and evaporation temperature. Difference from both of this give superheat. So you need two sensors - pressure sensor for the evaporation temperature and temperature sensor. Most of the producers have only one temperature sensor at the evaporator and the control logic is dependent only from compressor rpm and evaporation temprature. And its just logic - not dependent from real data. A bug in the logic will give wrong EEV setting and flooded evaporator. You need also the evaporation pressure to exactly control superheat. Most of the heat pumps miss that sensor. I’ve seen Vaillant have second thermistor at suction before the compressor. Other brands don’t have even this. I monitor the EEV for my heatpump. When it floods the evaporator the setting of the EEV is two times greater than a normal superheat. I guess it’s an error in algorithm. This is a screenshot of this type of I say “cold start”…
In red is discharge temperature, green is suction line temperature. Even though suction temperature looks normal the EEV setting is wrong and we have flooded evaporator. At this case it think it was 250 when it should be around 90. Heatpumps have a lot of defences for preventing flooded compressor so my guess is that no liquid reaches the compression chamber but at wrong EEV settings we have flooded evaporator working with low efficiency.
4 Likes
Ah yes, it might be Davids one that has a varible flowrate/variable pump. Your one does a lot of oscillation, and I really dont know why. Some bits are fairly level, and for no apparent reason, the input seems to vary. I guess it gets into ‘hunting’ where it overreacts, and it could easilt stray in and out of these oscillating conditions. How many other Daikins are doing this?
1 Like
Hi John,
This is my second Daikin, I had a 9kW replaced with an 8kW.
My 9kW was very stable in flow temperature although I always noticed that the flow temperature decayed over time when running for long periods.
The hunting is seen on all of the 4/6/8kW models from what I have observed.
I don’t know enough to explain it.
What I think I know is that Daikin heat pumps focus on controlling the return temperature as opposed to the flow. You set a flow temperature and a dT, so say 35c flow with a dT of 5c it tries to keep the return at 30c. It will vary the flow rate to do this but I have always found that quite erratic, it never seems to really get to grips with it so the flow rate swings pretty wildly from one extreme to the other.
It is one of the reasons I run with a dT of 10c in the radiator setting, I never get it so it keeps the flow rate constant at the lowest level possible.
Just to add
Mine has done it this morning on space heating. For the first time it wasn’t at the beginning of a cycle, it was about 7 minutes in.
COP was down to 1.3 at around 3c outside, flow temperature was around 29c and it dropped to 26c consuming about 1.8kW to produce 2.25kW of heat.
Virtually a COP of 1 as @johncantor suggested.
I have been doing a limited amount of checking back to when we all first got these heat pumps last winter.
I cannot find any instances of this behavior at all even in identical weather conditions and heat pump settings.
One 8kW was producing 8,000W of heat for 3,300W of electricity at -5c in January shortly after installation. Today, the same heat pump was consuming 3,700W of electricity and only producing 5,200W of heat at +3c outside, quite a difference!
Everything suggests that this is now showing up as the heat pumps have worn in, it just wasn’t there at the beginning.
I cannot see any particular set of circumstances that cause the heat pumps to behave this way, it just seems to happen randomly.
So, before inverters, all HPs were controlled by the return sensor. This makes sense. If sensing flowtemp, it would tend to start, then turn itself off, and be very sensitive to flowrate. Inverters are different, and sense flow because they can ‘throttle off’ as the flow temperure rise, so there can be confision. I dont know why Daikin dont do as everyone else, and simply go on the flow temperature. Just looking at yours. the DT is quite wide, especially since you operate at such nice low temperatures. I think I saw return temperatures of 23. I would think a dt of 4 might have been better here. Yes, I see thismornings oscillations. what happens if you reduce the dt setting?
1 Like
Hi John,
I run with Daikin emitter settings as radiators which gives a fixed dT of 10c, which I never get.
What it does do is stabilise the flow rate.
Otherwise the flow rate is all over the place and I lose efficiency.
I also found that I needed a different dT for different flow temperatures and outside temperatures.
With a dT of 3c or 4c I will run out of flow rate, it won’t go fast enough, although perhaps I didn’t test it enough.
I arrived at this setting because it just worked and gave the best efficiency.
I am also fed up with heat pumps and really couldn’t be bothered to fiddle any more.
I am not sure any of my settings cause the issue, I see it on dhw too and the others that are seeing this are not running like me.
Hi John,
Have a look at my data, maybe 10 March 2024 and look at the flow rate and heat output.
It’s not at all stable and I don’t understand why these heat pumps won’t settle at a steady output.
All of these 4/6/8 kW models are the same, the larger units do not behave like this, at least mine did not.
I am starting to get the feeling that these smaller units have been designed to a tighter budget and some savings may have been made in the design!
Yes, the variation in performance is odd… So, do heat pumps wear out? They shouldnt. In the old days of piston compressors, reed valves did wear, but rotaries and scrolls dont have/need them, and if you cut an old compressor open, it should be as clinically clean as new, unlike a car engine. The oil will be clear. What else can go wrong? Slow loss of refrigerant? if heatechangers were clogged etc, it would be a constant drop, but it looks like your Carnot COP is good at times… Is it? can you see recent examples of good-as-new performance. It does seem more likely like a refrigerant control issue. sensor not sensing true readings. It really can be quite tricky to get to the bottom of some of these problems. With an Ecodan, you can read various conditions like various temperatures, sometimes refrigernt pressures. Does Daikin let you see any for diagnosis?
1 Like
Hi John,
I too very much doubt that it will be a problem with wear on a compressor, I think they are pretty much bullet proof.
But what it may be is a sensor or other part that just doesn’t hold up well over time and starts to have intermittent problems.
I am not familiar with these components so I can’t really judge what could be happening.
I think it is clear that this is not supposed to happen, whatever it is that is actually happening, and that it was not there at the beginning.
I would think that a loss of refrigerant would result in a consistent loss of performance, but maybe not? Maybe it is just enough to be a problem at certain times and conditions only?
I think my Carnot COP is good all the time that I am space heating apart from these incidents and the dhw is not so good, it wasn’t good over the summer but I didn’t realise it until these recent incidents.
1 Like
Hi @johncantor, Hi @matt-drummer, Hi @alibloke, Here are a couple more instances from today where some “phantom electricity usage” occurred. I’ve included a trace for flowrate which remains constant through the 15 mins of abnormal output. See how for these 15 mins the simulated carnot trace is so much higher than the heat meter readings.
This one when the system is waking up on a chilly 4C morning after being idle for 45 mins
And this one after the pump had been idle for 90 mins in 7C outside temp.
Could there be a trigger connected with these idle periods in lowish temperature?
My system is Lindfield 8kW Daikin, but please ignore the rather severe air errors during DHW which I’m trying to purge with limited success. But that’s for another thread
1 Like
Hi David,
This is my one today, caught me by surprise, I was awake but driving to an airport so couldn’t do anything about it.
I haven’t been able to link this behaviour to any particular set of circumstances, mine had not been idle here. It had defrosted but I haven’t seen it after a defrost before and thought I was safe if it was running.
1 Like
This is another 8kW this morning.
Looks awful and we didn’t see anything like this when they were installed last winter.
It is worrying me a little I have to say.
1 Like
This is quite dramatic isnt it? Its not at all good. When I first saw this type of thing on my 6kW Ecodan, I fitted a sensor of the discharge pipe from the compressor. This would at least confirm that liquid refrigeant is entering the compressor. Mine was. The next question is … what is the expansion valve doing at that time. I had an idea to look at the Expansion valve (LEV or EEV). I got as far as very carefully putting something like 2kohm (I think) resistors on the LEV signals, straight onto LEDs. This gives opening and closing pulse flashes. Next stage would have been an EmonTH with optical pulse sensor to count pulses. To know it the valve is opening or closing, a 2nd optical is needed… e.g. If ( LED2=on 10ms after LED1 pulse ; LEV opening; else LEV closing. It would be nice to see what a LEV was doing… but what will it tell us?? I would expect the LEV to be closing quite rapidly when the discharge temperature is so low (when heat output drops). However, maybe the sensors are oblivious to conditions, and the valve is opening up, making it worse. I dont actually have the knowhow to be able to configure an EmonTH with 2 optics, and the ability to read the state LED2 10ms after sensing LED1. How easy would this be to achieve? Can LEVs stick?? I dont know. However, if this were the case, we should see closing pulses with no response. Its still a thing I would like to complete. I’m sure something would be learnt
[Discussion about monitoring the Expansion Valve continues here:
Heat Pump Expansion Valve behaviour ]
1 Like