@technogeek

If your hot water cylinder has an immersion heater, you could try the same test using direct electric at a known COP of 1.

I can't remember if Samsung are one of those, but I know that some heat pump manufacturers report thermal energy produced as the quantity absorbed from the ambient air, to which should be added the quantity of electrical energy used, thereby giving the total thermal energy output. In your case it would therefore be:-

1.1 kWh + 2.2 kWh = 3.3 kWh Total.

A further problem with single short duration tests is the accuracy with which the measurements are being made and the results reported.

@abernyte I did think of the oil boiler efficiency however it is a well maintained  5 year old Grant Vortex Blue flame and very efficient, so ruled 35% out. Even very old yellow flame boilers had efficiencies of 60%+ for space heating, would not hazard a guess for DHW generation on those models.

Reading manufactures data the boiler is 59.1% efficient for DHW and 93% for space heating 🙂

@derek-m yes I do have an immersion heater however it is positioned at the top of the tank and would not be heating the same volume of water sadly. In addition the immersion heats to 60C which is not the same temp range of 38C to 55C so sadly is not the same work load for both systems.

I suspect you could be correct regards how manufactures report thermal energy as all the figures add up. My oil boiler has used a litre a day for hot water since day one, currently at 63p a litre. The HP does the same with approx 2.2 Kwh a day on an average 20p / Kwh, so regardless of efficiency the HP is proving a 1/3 cheaper, so happy days!

Cheers for the suggestion 🙂

In relation to Samsung heat pumps in particular, Glyn Hudson has indicated that the COP is calculated differently than usual:

Heat Output = Energy Consumed + Energy Generated

COP = (Energy Consumed + Energy Generated) / Energy Consumed

I haven't seen this discussed anywhere else but he is comparing the values from the Samsung controller to a class 1 electricity meter and a class 2 heat meter. 

https://community.openenergymonitor.org/t/samsung-air-source-heat-pump-installation-with-monitoring/21254/10?

@redzer_irl From my own research over the last few weeks I would be inclined to agree with Glyn Hudson.

I think for me, I incorrectly interpreted the energy generated to be the output of the system not what was generated from the ambient air. After thinking about it, I would assume the manufactures would have called the screen "Output Power" or "Total Output Power" to represent the heat power being fed into the radiators / DHW tank, if that is what that screen represented, however I believe now it is the heat generated from the ambient air which added to the power input gives the correct COP and power output. The power output was cross referenced with my metered oil boiler and similar heat load in the DHW tank, to find the heat energy required to heat the body of water to 55C.

Thank you for the link, I will have a proper look 🙂

Posted by: @technogeek

↑

Hi All,

 

I have had my ASHP (Samsung HT Quiet) fitted about 2 months ago and slowly getting to know its functionality. Currently it is used to just heat the hot water in my old cylinder (170 litre solar tank with 2 coils, upper coil is used by the ASHP) which it manages to get from 36C to 55C in approx 30mins on average. The COP is working out to be about 2 for this. The hotwater cylinder is getting upgraded in October to a modern tank more compatibile with ASHP's.

I am wondering what COP other owners are getting for DHW generation to see if mine is OK or bad and if the new tank upgrade is going to make an improvement?

TechnoGeek

so one key thing that I have noticed is that you need to wait for water that you are heating to be actually cold. I have temperature probes at 4 locations in my cylinder, in the following locations down from the top : 1: well above coil: 2: same height as the input of the coil , 3: halfway between 2 and 4, 4: very near the bottom, level with the output of the coil. the coil covers roughly the lower 2/3's of the cylinder.

The heat pumps thermostat for the hot water goes in the "middle of coil" probe location, at point 3. If I simply let the hot water come on automatically when stat 3 goes below the necessary value, the COP is rubbish , 2 or below.

This is because there is still hot water above it that the coil goes through. The reading of stat2 could still be 50C when stat3 is 20C. So if you run it like this, the heat input is actually negative for several minutes:  as the HP's LWT ramps up , its still lower than the water surrounding the upper part of the coil (stat2) for quite a long time, thus heat input is negative. the water around stat2 actually gets cooled because of this, and the water returning to the heat pump warmed by the cylinder. or at best, you move a bit of heat from the upper part of the cylinder to the lower, but don't actually input any energy to it.

On the other hand, if I wait for stat 2 (and by implication stats 3 and 4) to go below 30, then I get a much better behaviour, because all the water around the coil is cooler than the water coming from the HP. So the heat pump can thus provide input energy to the DHW almost from the first minute.  COP of around 3 as per below from today (nice warm OAT of 17)

this basically requires dual thermostat capability, which heat pumps don't have. I am trying figure out a way to automate it.

@Iancalderbank

Your heatpump is likely to have a flow-return temperature of around 5degrees.  So under most dhw circumstances, the tank will have been part used - it will have stratified water, hot at the top and cold mains below.  Thus usually the heatpump will push heat in and simultaneously destratify the tank - taking heat from the top, moving it down, bringing the whole tank up in temperature.  I don’t think this is inefficient.

The best way to heat a dhw tank with a heatpump will be just enough heat, just before you need it.  High heatpump flow temp is inefficient, also high standing tank loss causes inefficiency.  
Ours comes on overnight, showers tend to be morning ones.

Posted by: @iancalderbank

↑

Posted by: @technogeek

↑

Hi All,

 

I have had my ASHP (Samsung HT Quiet) fitted about 2 months ago and slowly getting to know its functionality. Currently it is used to just heat the hot water in my old cylinder (170 litre solar tank with 2 coils, upper coil is used by the ASHP) which it manages to get from 36C to 55C in approx 30mins on average. The COP is working out to be about 2 for this. The hotwater cylinder is getting upgraded in October to a modern tank more compatibile with ASHP's.

I am wondering what COP other owners are getting for DHW generation to see if mine is OK or bad and if the new tank upgrade is going to make an improvement?

TechnoGeek

so one key thing that I have noticed is that you need to wait for water that you are heating to be actually cold. I have temperature probes at 4 locations in my cylinder, in the following locations down from the top : 1: well above coil: 2: same height as the input of the coil , 3: halfway between 2 and 4, 4: very near the bottom, level with the output of the coil. the coil covers roughly the lower 2/3's of the cylinder.

The heat pumps thermostat for the hot water goes in the "middle of coil" probe location, at point 3. If I simply let the hot water come on automatically when stat 3 goes below the necessary value, the COP is rubbish , 2 or below.

This is because there is still hot water above it that the coil goes through. The reading of stat2 could still be 50C when stat3 is 20C. So if you run it like this, the heat input is actually negative for several minutes:  as the HP's LWT ramps up , its still lower than the water surrounding the upper part of the coil (stat2) for quite a long time, thus heat input is negative. the water around stat2 actually gets cooled because of this, and the water returning to the heat pump warmed by the cylinder. or at best, you move a bit of heat from the upper part of the cylinder to the lower, but don't actually input any energy to it.

On the other hand, if I wait for stat 2 (and by implication stats 3 and 4) to go below 30, then I get a much better behaviour, because all the water around the coil is cooler than the water coming from the HP. So the heat pump can thus provide input energy to the DHW almost from the first minute.  COP of around 3 as per below from today (nice warm OAT of 17)

this basically requires dual thermostat capability, which heat pumps don't have. I am trying figure out a way to automate it.

 

Hi Ian,

Did you manage to automate the process?

I can suggest a solution if you wish.

Posted by: @derek-m

↑

Did you manage to automate the process?

yep. The key was to get the modbus control of the heatpump up and running. Its a a simple logic loop running in NodeRed on top of HA, with HA calls to get the sensor values and write to the numbers containing modbus values. did not take long to figure out.

starting state:

• DHW Status on heat pump controller set to OFF

This is same effect as pressing the power off button for DHW on the controller - it will never run DHW in this state. Needed so that it doesn't run the DHW prematurely.

every minute:

• check temperature of cylinder stat2 
• If >= 35, do nothing and exit.
• If < 35:
  • Set DHW status on heat pump controller to ON , via modbus write command. You will see the screen on the controller go from Grey to Active (blue) at this time.

we know that stat3 at the middle must be 35 or cooler so no need to check that one. The heatpumps DHW stat is at that point in the cylinder. Therefore, as soon as DHW status is set to ON, the Heat pump , provided it has an appropriate setpoint, will go "hello, the cylinder is too cold and I'd better run a DHW cycle", so will go into DHW mode and reheat the cylinder. When the heat pump detects that water is hot enough according to its setpoint it will go out of DHW mode and back into CH mode. 

Note that the DHW status value is still set ON at this point.

• • 80 minutes after setting DHW status to ON, set it back to OFF, to get back to the starting state. 80 minutes being assumed more than enough for a full reheat

tested this last night with the entire 300L tank,at the start the  top was at 35, bottom 22. got pretty efficient behaviour to get the cylinder up to 55C , COP of 3.0 . With a lower target temp, COP could have been better. You can see a clear tracking of input power to LWT over the run, meaning that instantaneous COP drops from about 4.0 at the start to just over 2 at the end. So perhaps 3.5 would be achievable if stopping sooner, for say a 50C target temp instead of 55.

Although it is of course still quite warm OAT. note the times - sad heating geek that I am sat , I and watched it until later than I should have!

there could of course also be time of day inputs to this loop to schedule it and also the option to tweak the DHW setpoint as well.