Posted by: @cathoderay

↑

@judith - thanks. Can you get leaving and returning water temperature at one minute intervals? That would let us see frequency and duration of defrosts on a Vaillant if you can get a chart covering a defrost period.

I’m using different systems to look at similar information but I’m afraid lwt is not one of them.

On 3rd Jan we had mostly -ve temp rising to ~0C midday. The most frequent deicing was in the evening as the temperature dropped again. The longest run without was 7:30 to 9am other than 0-4am with set-back of 18C meaning lower heat demand. So it must have been a fairly dry atmosphere. The total energy usage was 52kWh for ashp alone.

You have to ignore kettle spikes and other appliances since it’s total load on the GivEnergy figure. Even though it’s 5 min intervals de-icing can be clearly seen. 

The sampling rate of 5mins is not quite fast enough to see all the features. The ‘off’ period is ~7mins out of ~40. A real monitoring system would be better, but I’m not that software minded.

https://emoncms.org/app/view?name=Primary&readkey=eec7334e58f2a6927a9ff7b27c8c7e63 this is a same sized heat pump but not geographically very close, it is running at lower flow temperatures than ours at similar temperatures on the same day

@cathoderay you can zoom in as much as you wish on their data

@judith - thank you very much. I take it the dips you can see in the mainly green chart you posted on the right above are defrosts, on the assumption the load power (whatever that is, I do wish people would consistent and meaningful terms! load on its own is ambiguous). Between 1800 and 2100 there are five dips, ie I agree cycling every 40 mins or so. The duration is a little more difficult to know, because we don't know how much of the dip is actual defrost, and how much is recovery from the defrost. Your ~7mins for the defrost itself seems plausible. If we zoom in on some cycles and put one against a scale, about 20% of the cycle period is apparently defrost, 20% of 40 mins = 8mins duration.

From the other chart, it looks like it was about -3 degrees OAT at that time. All in all it looks like Vaillant's claim that defrosts start to become less of a problem below zero degrees is shall we say optimistic. Your defrost behaviour looks rather like mine, still in full swing at -3 degrees OAT.

I'm not a great fan of emoncms.org/app charts myself, visually OK, but labelling can be less than clear. However, I have managed to get (I think, the labelling isn't there!) a 24 hour view for 3rd Jan for this other heat pump and this does have LWT and RWT: 

and we can see clear defrost behaviour (LWT below RWT) and if we zoom in still further:

it looks like the frequency for this Vaillant at ~0 degrees OAT might be less, once every 90 mins, and the duration marginally shorter, but it is still defrosting. That said, it does appear to recover well, the LWT rises fast after the defrost, but at the cost (to be expected) of a spike in 'Electric' (whatever that is, I guess power in as it's measured in watts. I presume the blue shaded area maybe a DHW heating period.

All useful extra information, thanks.

Intrigued by this conversation , I thought I would attach the spec of My Samsung . I assume by the fact it says inverter technology, it has soft start?

Posted by: @alfapat

↑

Intrigued by this conversation , I thought I would attach the spec of My Samsung . I assume by the fact it says inverter technology, it has soft start?

 

@Alfapat: The inverter should provide the capability to ramp up the compressor speed but I notice that there is almost a factor of two difference between the rated current and the peak current. The difference between these numbers for my Arotherm+ is relatively small. Perhaps that's progress.

I'm also interested by the topic of frequency of defrosting as I've only twice seen my heat pump produce a large cloud of water vapour. I have therefore just ordered a pair of Elitech RC-5+ thermometers with loggers to put on the heat pump flow and return pipes. These:

https://www.amazon.co.uk/gp/product/B07P5YTN9G/

Wow @johnr

Either those Elitech temperature loggers are really small
... or possibly transparent, in which case I'll endorse them.

Posted by: @transparent

↑

Wow @johnr

Either those Elitech temperature loggers are really small
... or possibly transparent, in which case I'll endorse them.

@transparent Amazon, for some reason, isn't cooperating with me today and 2 of the 3 links I have posted aren't working properly. Search Amazon for "Elitech RC-5+". They claim to be able to store up to 32,000 data points.

Posted by: @transparent

↑

I'd also been keen to attract here some people who might later play a part in rebuilding Ukraine.

Hi @transparent, that sounds interesting. I have some contacts in Ukraine and am closely following daily life there as well as the news. Maybe another topic sometime.

NEWS UPDATE:

The team of moderators has been reviewing our thoughts on Heat Pump starting current.
We've reached a point where we need wider input from the main Forum.

1: Heat Pump technology has progressed over the past 5 years.
The implementation of

• soft-start electronic motor control
• integral inverters
• modulating water pumps

has greatly reduced the power required when the heat-pump starts.

So when you comment here on the Forum, it matters as to whether you have an older heat pump without such features.

2: The use of thermostatic controls (TRVs and room-stats) has risen in significance as a cause of heat pump inefficiency.

Such practices do reduce the COP, and there may be more than one reason why this is so.

We are looking at the issue of hysteresis, where a room (or zone) falls below the set-temperature and the heat pump then starts to drive it above the set-point.
There will be different opinions on this and how it affects the ASHP flow temperature and velocity.
Your own system design/layout matters, and needs to be mentioned when posting here.

In brief, a thermostat applies a 'digital' on/off signal to the heat pump,
whereas the Weather Compensation curve, based on the OAT, is a slow-acting analogue approach.

3: Manufacturers' specifications stipulate the need for a mains supply cable (and fuse) which can carry a current much larger than is required by the heat pump alone.

The mains power input to the heat-pump might also need to operate:

• an immersion heater in the DHW tank to allow anti-legionella processes, typically 3kW (12.5A)
• a heater to defrost an external ASHP

Please provide this information about your particular heat pump when posting here (if you know it).

 Let's remember that the issue which concerns the DNOs is numerous ASHPs in an area simultaneously entering a defrost cycle.

Solutions are needed to overcome high-demand surges,
and DNOs are not permitted to control or own any generation or storage assets.

That constrains them in ways which don't apply to us!

@transparent to give context to my results of slow switch on current. Our Vaillant is inverter controlled, the system is open and the overall system is under WC control, so nothing to trigger a fast switching transient. 
I recognise that this screenshot of a switch on (posted earlier) with 1minute sampling does not have the necessary speed of monitoring to prove the lack of fast transitions but it is a good indicator. 
I’m not sure what speed of monitoring would answer your question, any suggestions? Any product for example? It is obvious there could be as yet unseen fast edges within that.

I've done a quick analysis of data collected by a pair of Elitech RC5+ temperature loggers for the period 20 - 22 January to look for defrosting cycles on my 7kW Arothern+ heat pump. The loggers' probes were fitted to the heat pump flow and return pipes close to my buffer tank. I found four defrosting cycles on the morning of 21st January when the outside temperature was around zero C according to my weather station in the garden https://www.wunderground.com/dashboard/pws/ISWINDON65/ . The heap pump's OAT sensor is fitted to the north side of the house and usually reads slightly high (unless there's an NE wind).

I found four defrosting cycles which lasted about 8 minutes each (some variation depending on the water temperature prior to the defrosting). My 5 minute electricity consumption data shows power spikes after the defrosting but typically up to about 2kW. The Vaillant app shows that 7am to 8am was the peak heat generation time with 5.4kWh plus the preceding and succeeding hours producing 5.1 and 5.0kWh.

PS: The Elitech sensors claim an accuracy of +/- 0.5C. The absolute accuracy isn't critical as the clue for the defrosting is the flow temperature dropping below the return temperature is heat is extracted from the buffer tank.