Calculating COP of your heat pump without extra sensors

[DoItDreckly]

I have a NIBE heatpump and in this country they don't seem to want to sell the flow meter which would allow calculation of the COP.  You can install a flow meter in the central heating loop but unfortunately with the all in one NIBE you would not be able to see the DHW flow figures as the shuttle valve and pipes are internal.  I wanted to see if the heat pump is matching the performance as advertised.

I ended up adding an ESP8266 based device which had an rs485 connection to the heat pump.  This is essentially emulating the NIBE modbus part you can buy for circa £600 !#!@.  I made the equivalent for about £20.  NIBE seems to have a policy for charging too much for a lot of accessories and the online tracking software.

I am running a node-red server which frequently polls the ESP8266 and saves the data in an influxdb.  It calculates the COP as part of the process.

On my NIBE F1255 I can disable the compressor and use only the internal resistance heating.  The process is as follows.

1. The immersion works in 1 kWh increments.  It is worth having a high power setting for the process.  
2. Put the pump for the central heating loop in manual control.  I also set a demand for 30 degrees in the house.
3. Start with whatever the lowest pump % setting you have and then for each 5% setting record the difference between temperature out and in on the central heating loop.  It is worth waiting for about a minute for the temperature to stabilise after each pump setting before taking the reading.
4. Then switch the heating to the DHW and repeat the process.

You should have a nice graph of delta T vs pump %.

Now the process works on the idea that DT for a volume of water is given by DT = power in / (volume  * 1.06ish)

So for each of the readings you have you can get a value of V = immersion power / DT.  You can ignore the 1.06 as you don't need to know the V exactly.  Just a proportional value for use later.

So in the future you can set the heatpump to work as normal.  Using the modbus, read in the power input into the compressor;The pump % setting for the hot water loop; The temperature in and out.
Interpolate the values from your graph using the % pump setting.  Call this value V

then power into the hot water = DT * V
COP is power into hot water / power into compressor.

The accuracy is obviously dependant on the accuracy of the temperature readings and the resistance in the hot water loop not changing.  I have the radiators balanced so the number of radiators on the circuit isn't changing.  If you have thermostats opening and closing all the time then your numbers will not be so accurate.

Things I have noticed with my COP.  With 3º OAT I am getting about 4.5 COP at the moment. In April last year when the borehole had cooled I was getting 4.0 COP. I haven't had a chance to look at SCOP as I still need to gather a years worth of data.  I shall update with some more figures during the winter if people are interested.

The DHW COP really falls after 50ºC.  If I luxury boost to 60C then the COP moves from 2.5 to finish at about 2.0.

I have attached the node red flow which might be a useful starter for 10.  I never coded in JS before and don't code much anyhows so there is not a lot of documentation.  It might help somewhat.

The basis for my ESP8266 code. This will be need to be modified to use the Wifi.

https://github.com/openhab/openhab-addons/blob/2.5.x/bundles/org.openhab.binding.nibeheatpump/contrib/NibeGW/Arduino/NibeGW/NibeGW.ino

[DoItDreckly]

Oh, the attached node red flow includes the solar edge stuff which works with the modbus over IP on the Solar Edge allowing the data from that to be saved.  It should work with any Sunspec based inverters.  I'm not sure how you'd turn that on with any of the new inverters they sell as they don't have a control panel. You might need to phone them.

[djs63]

Thank you! Will study your report closely. We have had a gshp for 10 years and I am impressed that it works fine but unsure about its COP

[DoItDreckly]

Thank you! Will study your report closely. We have had a gshp for 10 years and I am impressed that it works fine but unsure about its COP

Glad to hear it’s working well for you. Breakdowns are my biggest fear as the prices for most parts seem to start close to £600.

[marshman]

Interesting.  Had my heatpump for coming up to 4 years, trouble free so far.  Just started  the season for DHW heating - not enough PV to keep the water hot at this time of year. Not cold enough yet to need it for heating, but give it another weeks and it will be on.

As it was just doing the DHW, heating the 300L tank roughly once every 2 days or so I decided to measure the COP.  I know the tank size, 300L, I know the start temp and the end temp so can calc the temp increase and I have a separate electric meter just for the heatpump so can measure the energy used. As the PV diverter to the immersion is still working the tank starts at the same temp throughout, usually around 35 deg C. I run the heatpump and switch it off at 50 deg C - plenty hot enough for us - and note the electricity consumption. (Diverter obviously not doing anything whilst heatpump is running!)

A few calculations - actually I cheated and used and online calculator - and I get a COP of around 3.2 fairly consistently, which I am happy with for DHW. The incoming brine temp is at 16 deg C at the moment - this gradually drops through out the heating season to around 7 deg C at the end of Feb.

Roger

[DoItDreckly]

 The incoming brine temp is at 16 deg C at the moment - this gradually drops through out the heating season to around 7 deg C at the end of Feb.

Roger

Wow, those are nice temperatures for the brine. Your Scop must be pretty high.  Up here in Scotland the granite never gets above 9 degrees. . So the borehole starts at about 8 degrees going into the winter. With -10 we hit about 2-3  degrees from the brine.  It seems to be determined by the extraction rate. So as we get back to 0 degrees air temp it will get back to about 4-5 for the brine. Thankfully we have only had brief periods each year where the return brine drops below freezing.

[nowty]

When I installed my GSHP I calculated its COP by running the output into a builders bucket filled with 35 litres of water. I measured the temp in the builders bucket every minute and monitored the current draw so I could calculate both the input and output power. From that I could build up a table for COP vs output temp.

I had a thread about it a couple of years back, https://www.navitron.org.uk/forum/index.php/topic,29762.0.html

[marshman]

 The incoming brine temp is at 16 deg C at the moment - this gradually drops through out the heating season to around 7 deg C at the end of Feb.

Roger

Wow, those are nice temperatures for the brine. Your Scop must be pretty high.  Up here in Scotland the granite never gets above 9 degrees. . So the borehole starts at about 8 degrees going into the winter. With -10 we hit about 2-3  degrees from the brine.  It seems to be determined by the extraction rate. So as we get back to 0 degrees air temp it will get back to about 4-5 for the brine. Thankfully we have only had brief periods each year where the return brine drops below freezing.

My ground loops are really way over sized because at the time we didn't know what the ground was like. The area is mainly shingle but with "fingers" of blue clay and the water table is only 4 or 5 feet down. The house is built on shingle but the field to the rear where the majority of the pipework is happens to be is nice wet blue clay which we didn't find out until the trenching machine got out there.  The field is totally open so gets the full summer sun to warm up - and we do get a lot of sun I get around 3700kWh from a 3.15kW pk PV system. When we had the "Beast from the East" a couple of years ago outside temps here were down to -4 deg C or so (that is cold for us!!) and the brine still didn't dip below 7 deg C.  In fact I have just looked back and that is the lowest it went.  If the pump runs for a when a hotwater cycle coincides with a heat cycle and it is cold outside it can run for the best part of an hour and then the brine will drop to around 6 deg C, but by the start of the next cycle 30 mins later it is back up to 7 or 8 deg C.  Not sure of the SCOP, I just know that the system works, keeps the house warm and snug 24/7 and is incredibly cheap to run. (240 sqm detached house, around 1350 kWh per year for heating and DHW)

The picture shows the ground loops being "installed". To the right is the "shingle" the house is built on, round to the left is the clay area.  Looked an absolute mess but now you would never know the ground had been disturbed.




Roger

[Philip R]

Just seen the circ pumps. Are they supposed to be up ended like that?
Most of them are supposed to lie shaft horizontal as the bearings are not designed for axial thrust. Also they will always have air in the pump inlet .
Philip R

[nowty]

Just seen the circ pumps. Are they supposed to be up ended like that?
Most of them are supposed to lie shaft horizontal as the bearings are not designed for axial thrust. Also they will always have air in the pump inlet .
Philip R

I presume your talking about the pumps in my picture ?

They are an efficient shaftless design with ceramic bearings and can be installed that way. There is just a ring that rotates inside.

They also automatically detect air and go into an intermittent mode to expel it.