Dear all,
Because I often have a much lower performance on my heat pump than specified by the manufacturer and received feedback so far has not been helpful to pinpoint causes I decided to just buy a copy of the industrial standard used to test heatpumps in Europe. The EN14511 (I purchased part 2, 3 and 4).
Internet research and my own heat pump documentation pointed me to a number of different standards (Eurovent for example), many of these standards are not usable for testing at home.
The only one that did turn out to be usable (in my search) is the EN14511. This one defines how heat pumps are tested at various load points. So this is the only one that gives you single load points that you can try to replicate and benchmark against.
As far as I could find all heat pumps supplied in Europe should be benchmarked according to EN14511-2 and EN14511-3 even if their construction design code is different.
EN14511-2 is about testing conditions
EN14511-3 is about testing methods
EN14511-4 is about requirements on heat pumps themselves.
When purchasing these standards I was hoping for 2 things:
-To get clarity on the testing conditions and requirements on testing instrumentation accuracy (which I did)
-To get clarity on requirements on the accuracy of instrumentation installed on heat pumps as they are delivered to end users (My interpretation is that there are none).
Disclaimer: If you get any questions after this post: These standards are copyrighted so I am not able to share them or really parts of their contents here. What I am sharing is my interpretation of some core information I found that you could use to approach these conditions as good as possible when doing your own tests. I am not liable for any misinterpretation of these standards.
Preparation: What information to look for about your heat pump
For heat pumps tables similar to the one below are published with their technical information. It states for a number of combinations of ambient temperature and leaving water temperature the following data:
-Supplied heat at high load
-Consumed electricity at high load
If you divide supplied heat by consumed electricity you get the COP at that load point.
Supplied values by Samsung for example are both āpeakā values and āintegratedā values. Peak are values without defrost and integrated are values with defrost cycles included.
Note on SCOP vs COP: SCOP is a value determined by combining a lot of the tests from the table below with specific weighing factors and not workable to benchmark in a real field installation. So if you want to benchmark anything verifyable you do so based on the tables below or similar info for the specific heat pump you have.
Power consumption and supplied heat of a Samsung AE050RXYDEG Heat Pump (Samsung EHS Technical Data Book)
Outside air temperature and humidity (relevant mainly for defrosts)
EN14511 specifies a number of outdoor temperatures, wet bulb and dry bulb. The differences between the two are such that they correspond to a relative humidity of approximately 93%. My interpretation is that any relative humidity below 93% should probably lead to lower amounts of defrosts and therefore lead to a higher COP-as-measured than published performance figures indicate.
That said I am not sure if lower relative humidity could lead to some ice building up and staying there but not building up thick enough to trigger a defrost.
Testing time and stability of operating conditions
EN14511 specifies the testing procedure with how to start circulation pumps, equipment in the testing chamber and so on. Temperatures have to end up being stable within a few tenths of a degree for a few minutes before the benchmark can start.
If you end up doing a test that includes 1 or more defrosts you always benchmark from end of defrost to end of defrost. It is best to make sure you have more than 1 defrost in your benchmark if they occur at all. Benchmark test time is more than 1 hour in the standard so make sure you take a similar time frame (or longer) for your own benchmark. EN14511 states that if a defrost occurs at the end of the testing time the full defrost cycle has to be included (in that case the actual test time will be longer than the standard testing time).
Then during testing it is permitted to change water outlet setpoints to keep the actual water outlet within boundaries required by the standard.
If all temperatures remain sufficiently stable for the entire test the test is deemed a steady state test. If deviations occur outside these boundaries (for example due to a defrost), the test is deemed a transient test. āIntegrated valuesā for the Samsung performance data for example are all āDynamic testsā So having 100% stable temperatures is not a requirement to say anything sensible about the data you measure.
Testing measurement accuracy
EN14511 specifies a number of sensor accuracies which I am not going to cite all but the following are most important for air-water heat pumps:
-Air temperature: within a few tenths of a degree
-Water temperature: Within 0.15 C
-Water temperature increase: Within 0.15 C
-Consumed electric power: Within 1%
-Water flow rate: Within 1%
Below is my interpretation of the required accuracy during the EN14511 tests compared to my intepretation of the accuracy I have been able to find for the instrumentation on my Samsung heat pump. I presumed full water flow at all part loads (which is the case for my setup as my pump is not PWM controlled). Then plotted both accuracies as a function of temperature increase. These line of Samsung also includes inaccuracy due to rounding off the measured values to 0.1C.
In my view the measurement requirements during the EN14511 tests are not exotic compared to the instrumentation that came with the heat pump. Power consumption accuracy is also comparable to fairly standard 3phase modbus power meter on sale for about ā¬100,-.
Thatās it. If you need to find more I recommend to purchase these standards. They can be bought -among others- on www.nen.nl
.
