I bought a Heat Pump

Reed_Richards

Most modern heat pumps can do Weather Compensation; that's modulating output to respond to outside temperature.  That's not quite the same thing as matching demand (which they might be able to do also) but it does require that the heat pump can modulate down to quite a small fraction of its peak output.  It looks as if the the nominal 8.5 kW Mitsubishi heat pump @QrizB chose as an example can modulate from 9.1 kW down to 3.2 kW of output.

The other thing is that any heat pump or boiler that cannot modulate down to near zero will turn on and off more frequently when it's warm out and less frequently when outside temperatures are cold.  Heat pumps use much more electricity in the latter case so that's the time you worry about them using more electricity than they need to, also the time you worry about the heat pump keeping your house warm enough.  I would happily sacrifice some economy of operation at mild outdoor temperatures if it gave me better economy at cold outdoor temperatures.

There is an awful lot of myth and misinformation surrounding heat pumps and I think the statements that @Cardew found may be just another example thereof.  Perhaps they stem from the days before inverter-driven heat pumps?     

shinytop

Reed_Richards said:

Wow, thanks @QrizB, that's great research.  Lets look at the figures now for WT at lower temperatures. AT at 2 C.

WT = 35 C

• Max output 8.5 kW, COP 3.15 - that's a better COP than the best case for WT = 45 C

WT=40 C

• Max output 8.5 kW, COP 2.89 - that's a worse COP than the best case (Mid) for WT = 45 C although marginally better than Min and better than Nominal

So if a Mitsubishi heat pump starts from cold and works really hard ("Max") to get the WT up to 45 C it will be working at a better COP for as long as the WT is less than about 35 C but if it continues at Max output until it gets the WT right up to the target 45 C it will be operating at a worse COP between about 35 C and 45 C (assuming that it can operate at Mid output once it gets to 45 C).  Of course it could "decelerate" as it approaches 45 C to achieve better economy at the expense of a slower response. 

The Ecodan data book is what I've been using to build the model of my setup and why I'm saying what I do.  There is a slightly later version here.

I don't think it works as above.  The databook numbers are steady state measurements made under strict conditions as set out in EN 14511. I can't find a free copy (even if I wanted one!) but there is an article about the standards here. The numbers in the databook are the COP at the stated outlet temperatures and assume a delta T of 5K.  You can't therefore take the Mitsubishi figures and apply them to a circuit where the inlet is 20C and the target is, say, 40C or anything in between.  That would be complete guesswork. It's like looking at the WLTP mpg numbers for your car for 50mph and trying to use them alone to work out how much petrol you use to get from 0-60.  It can't be done. 

I'll try and dig out some figures that show exactly how much energy my ASHP uses to heat up from a low flow temperature.  

Reed_Richards

Here's a typical early-morning power consumption profile in my house.  My heating controller (a Drayton "Wiser") is tasked with bringing my house up to temperature by 7 am ("Comfort Mode").  It knows it is dealing with a heat pump but (I think) treats it in the same way as an oil boiler so does not allow cycles shorter than at 15 minute intervals.  The first cycle always goes to a higher peak power.  Possibly after the first three cycles it decides the house is not warming fast enough so then the power is maintained.  At 6:30 my hot water heating comes back on and there is a big peak as the tank temperature is brought back up to 50 C.  I have no idea what caused that peak just before 6:30.

There is nothing going on here that is specific to heat pumps and that Mr & Mrs Average couldn't do.  The only "techy" thing I have done is to change the weather compensation line from the one preset by the installer.  I did that because I came to see that  it was unnecessarily conservative at colder outdoor temperatures so the Leaving Water Temperature was higher than it needed to be (implying reduced economy of operation).  My weather compensation is probably still too conservative but I don't want to change things too often. 

shinytop

Reed_Richards said:

Here's a typical early-morning power consumption profile in my house.  My heating controller (a Drayton "Wiser") is tasked with bringing my house up to temperature by 7 am ("Comfort Mode").  It knows it is dealing with a heat pump but (I think) treats it in the same way as an oil boiler so does not allow cycles shorter than at 15 minute intervals.  The first cycle always goes to a higher peak power.  Possibly after the first three cycles it decides the house is not warming fast enough so then the power is maintained.  At 6:30 my hot water heating comes back on and there is a big peak as the tank temperature is brought back up to 50 C.  I have no idea what caused that peak just before 6:30.

There is nothing going on here that is specific to heat pumps and that Mr & Mrs Average couldn't do.  The only "techy" thing I have done is to change the weather compensation line from the one preset by the installer.  I did that because I came to see that  it was unnecessarily conservative at colder outdoor temperatures so the Leaving Water Temperature was higher than it needed to be (implying reduced economy of operation).  My weather compensation is probably still too conservative but I don't want to change things too often. 

Interesting.  I wonder why it cycles to start with.  Is it the ASHP or the Drayton?  The ASHP will get the flow up to temperature quite fast (mine takes a 30-40 mins to get from 20 to 45).  If the room is still cold, wouldn't it be better if the ASHPjust kept running until it had warmed up? 

With HW, I always see 2 peaks; one to heat the water then one to reheat the heating circuit that has cooled down in the meantime.  

Reed_Richards

I have just been watching the power consumed by my house (which is as near as I can get to the power consumed by my heat pump).   The pump or pumps seem to consume about 150 W of power and the unit pump has to run for 3 minutes before the compressor comes on.  The compressor starts at about 1 kW of power drawn and stays there for a couple of minutes whilst the unit monitors temperatures.  It then goes up to 2 kW and then drifts up by about another 250 W over 10 minutes or so.  Then the compressor goes off.       

Reed_Richards

shinytop said:
Interesting.  I wonder why it cycles to start with.  Is it the ASHP or the Drayton?  The ASHP will get the flow up to temperature quite fast (mine takes a 30-40 mins to get from 20 to 45).  If the room is still cold, wouldn't it be better if the ASHPjust kept running until it had warmed up? 

I have a night time set-back to 17.5 degrees.  I want the house (strictly the living room) to be at 18.5 between 07:00 and  09:30.  The easiest way to do this is to set my Drayton controller to "Comfort Mode" which causes it to try to reach the required temperature by the time that temperature is required (i.e.18.5 C by 07:00).  The Drayton does not know the outside temperature but it saw how fast the house cooled overnight and has some access to weather reports for my area.  I suspect that it turns the heating on in plenty of time (04:00) then runs through some cycles to see how hard it needs to work.  But I have no easy way to tell for certain whether what happens is due to my ASHP or the Drayton controller.  

shinytop

This is what my ASHP does on a cold start.  This is letting the thermostat control things. You can see the energy used as the flow is heated up and also the COP.  A slight caveat on the COP.  This is the ASHP's reported COP; it's not a properly calibrated number as you would get from proper metering.  I don't know exactly how Mitsubishi measures/calculates.  I'll do this again when I get my MMSP up and running. 


On the LH scale, 50 corresponds to 3kW consumed.

    

Reed_Richards

Funny, (as you know) I would have expected that when the water was sufficiently cold the improvement that gave in the COP would outweigh any loss of COP from working the heat pump hard.  But manifestly that does not happen, almost from the outset the COP is worse.  The COP seems to show a clear negative correlation with energy which, if anything, is stronger than the positive correlation with flow temperature. 

shinytop

Reed_Richards said:

Funny, (as you know) I would have expected that when the water was sufficiently cold the improvement that gave in the COP would outweigh any loss of COP from working the heat pump hard.  But manifestly that does not happen, almost from the outset the COP is worse.  The COP seems to show a clear negative correlation with energy which, if anything, is stronger than the positive correlation with flow temperature. 

Well it's one heat pump out of many out there.  And I'm just a little suspicious of how good that negative correlation is.  As I said, I don't know exactly how the energy delivered is measured/calculated.  

Reed_Richards

After the first year of operation of my heat pump I was able to report that I had used 6930 kWh of electricity.  Now, about seven weeks later my consumption for 28/1/21 to 28/1/22 is 6026 kWh.  This means that I used far less electricity over the last seven weeks than in the corresponding period a year previously (about 900 kWh less).  This is very gratifying but I'm not sure how to account for it.  The possibilities I can think of are:

1. Commissioning and, perhaps, "bedding-in" of the heat pump caused it to use more power than in normal operation.
2. For the first few weeks of installation I kept the house about a degree warmer than normal (I cannot remember exactly how long for).
3. The corresponding period a year ago may have been much colder (I cannot remember).
4. I am now using more "aggressive" (but perfectly functional) weather compensation settings.

I continue to operate my heat pump as if it were a boiler, using a night-time set-back and using the "smart" feature of my third-party controller to try to ensure that the target temperature is reached at the start of the target period (so-called "comfort mode").

If anyone else keeps track of their monthly energy use for heating and hot water I would be interested to know if they have also seen a drop in the second half of December and January, compared to last year.