For this test I set a WC curve to target 90C flow temp at 0C, however the heat pump only managed 46C. I’ve always been concerned with designing systems to run much higher than 45C, in my experience looking at monitoring data high temperature heat pumps struggle to reach 50C, let alone 60C+ in cold weather. For a system to reach high temperatures, it would need to be significantly oversized, resulting in a very inefficient system all year round.
As far as I understand heat pumps are lab tested using standard European climate data, which is colder and drier than UK, ASHPs suffer more from the effect of defrost in UK, this is not a problem, it just needs to be taken into account during system design.
Installers have a tough job avoiding over-sizing and under-sizing a heatpump. Better data is needed to inform this design, IMO. Hopefully this is something we can help with using data from https://heatpumpmonitor.org/ and real-world capacity testing like this.
i have been shocked by the efficiency of most heat pumps on heat pump monitoring when defrost cycles are often( in certain weather conditions that we are currently having , this is certainly under discussed and often forgot about in a lot of reporting.
even though this is my first winter with a heat pump and i new about defrost cycles i was unaware of impact it has on efficiency and costs of running.
NW of England i do seem to be suffering a lot at the moment (my system Emoncms - app view
i have been thinking about using the emersion heater for the hot water cycles as hot water cycles seem to kick off some kind of chain reaction where the defrost cycles are becoming more often as the heat pump works harder , thoughts ?
wonder if the manufacturers took a different approach after a defrost cycle this would improve thing ( instead of trying to quickly recover the flow temps with working the heatpump harder they should just increase the load gradually recovering the flow temp slower)
ive setup some automations to turn off the HP hot water cycles when it below 0 degrees C outside and use my emersion instead to see if this improves the defrost cycles . this is just a experiment but if anyone has any thoughts would be appreciated.
the idea behind this is the HP works harder for the hot water cycle which when its cold causes defrost cycling , and once these start the hp works harder to recover the flow temp , so hopefully this will reduce the defrost cycles and keep the heating efficiency higher at the expense of 100% efficiency of the emersion heater
Great experiment. I might try this with my 10 kW unit this weekend. To better compare systems with in different humidity scanarios it would make sense to also allow humidity as an input feed to the heatpump app and show related stats like dew point.
you can see here how the 50 degree flow temps (hot water cycle) triggered the defrost cycling at 8 pm last night and in the early hours of this morning caused the amount of times to increase
This would be a good idea if your HP was struggling, but looking at your data it looks your HP is working fine, it’s not even 100% compressor and looks like it’s maintaining indoor temp with no issues.
I don’t see the need for this in your system. Why get a COP of 1 when you could get a COP of 2+ for DHW?
it is just a experiment , but my thinking is if i can reduce my defrost cycles for heating i will be hitting cops of over 3 probably over 4 for heating , it seems most systems once they start defrost cycles they require better outside conditions to come out of it , if i can stop or reduce the defrost cycles it improves the heating efficiency at the expense of a cop of 1 just to heat the water .
but again its just a experiment if we dont try things we cant improve
the other advantage of this which is financial is if i heat hot water at a cop of 1 at off peak to a higher temp i am more likely to get through the peak period costs on battery power at off peak
That’s indeed concerning but highly dependent on where the system is installed. To offer a different perspective, heating systems in Germany will typically be dimensioned to an outdoor design temperature around -10°C (for me in Berlin it’s actually -12.4°C). This is a requirement as per a DIN standard and whether this makes sense is an entirely different conversation - we only get these temperatures on a few days per year, if at all.
Nevertheless, if your heat pump is chosen to supply 10 kW at -10°C then you don’t really care about the “defrost dip” around freezing temperatures because your heat loss will still be quite far from its design point.
Manufacturers should provide power graphs for different humidity levels so people can judge actual power output in their environment.
Looks like tonight will be perfect for a full-power test .
I really could not justify running any longer at full power because it gets truly tropical. Time between two defrosts (there was one just after I cranked it up) was around 2 hours at roughly -0.6°C and 85% RH. Average power over the cycle + defrost is around 10.6 kW, which is approximately as expected from the datasheet. Take this with a grain of salt as I really could not run this any longer. My bathroom got up to 28 °C…for science!
Thanks for your sacrifice! You must have some large emitters in your bathroom!
What model is this?
Averaging the output data over several defrosts, ideally 3 or 4 is key to getting a true average heat output figure. The instantaneous heat output between the defrosts will be much higher and should be hitting the quoted Vaillant figures. It’s the defrosts which pull down the average output.
Here’s another data point, this a 3-phase 10kW Vaillant Arotherm+ running flat out at 1C outside, it’s averaging 10kW between defrosts, datasheet says 12.4kW.
It looks like Vaillant units will give their badge output i.e 5kW from a 5kW unit and 10kW from a 10kW unit but no more in worst case taking defrost into account.
It’s the one with the strongest fans under the radiators. In hindsight, I should have just turned them all off.
True, but looking at the examples here in the thread it appears that the duty cycle seems to stay approximately constant with outdoor temperature. What pulls the power down is the frequency of defrosts. Under that assumption, the average over one heating + defrost cycle should already give a good idea of what the true power output is.
Hi Glyn, these test-data are excellent. I wish i’d knew them BEFORE choosing my ASHP (Vaillant 55/6). probably I should have choosen the 75/6.
I live in Romania (-9C design temp), so much colder climate than in the UK, but I see the HP struggles somewhat when +80% humidity, even if it does only 1 defrost every 45min. (no buffer, no zones, ~ 140 liters system volume, 860l/h flow)
Any instrument manufacturer has a challenge of defining the conditions for a specification (yes I have been a technical manager in an instrumentation company). So to define the conditions which are repeatable is best. That is without cycling. We don’t have this HP monitoring system but can see the cycles on the mains power monitoring that we do have. On the 21st Nov which was our coldest day (-1C) the household peak draw was consistently 3.6kW of which our 7kW Vaillant was 3.0kW max. This happened to be a full app monitoring hour with no de-icing. As the de-icing cycles kicked in the mean HP load went down to 2.3kW, with cycles at every 30-45mins. We can’t see the effect as clearly as you guys can but the output power was the same at max throughout the day (lwt ~45C).
In the next cold snap I would be interested in a repeat (Glyn) but with the heat curve set to 50or 55C.
There is just a possibility that setting it as the complete maximum temperature takes the HP over the maximum power point. The thermodynamics will determine that.
Just for your amusement the app said the max thermal power output was 9.1kW (a classic overestimation due to thermistor inaccuracy.)
Those Vaillant thermistors are indeed highly inaccurate, but your 7kW Vaillant unit can easily supply 10 kW of heat at A0W35 and a bit more than 8 kW at A0W55, so your 9.1 kW peak is totally within the expected output.
It is possible it is 9kW but since the Czech data sheet shows 9.0 at full speed at -2C for 45 &40C lwt I was assuming the app might be on the optimistic side by 5-10%. Without a true heat measurement it can only be an educated guess.
Agree, it’s a challenge! However, currently the Vaillant datasheet does not mention under what test conditions the datasheet output figures were derived. The conditions the unit I tested are very typical for the UK. They just need to add a bit of context info to make the system designs aware that real-world output taking into effect defrosts will be lower than stated. I feel bad for installers who have correctly sized systems based on the manufacturer’s data, and are now out-of-pocket having to upgrade units for customers when it’s not really their fault.
I don’t think this will make a difference since compressor was running 100% and the flow temperature was only reaching a max of 46C between defrosts. If the compressor speed reduces, the max output will also reduce.
Yes I agree with both your points. I had missed your information that at 100% compressor power the temperature only reached 46C.
In this regard Vaillant is better than most manufacturers such that the unit will reach its nominal badge power under de-frosting conditions.
We all need to disable the legionnaires cycle under a cold snap, since none of us would like the cold house while it tried and failed to get above 60C.
.
