Last night was a good test whether your ASHP was specified correctly to your house heat loss.

I’m at 52 degrees North, Central England, OAT levelled off last night at -4C. My Cosy6 is working hard and is managing to keep the house at 20C. 
Octopus said the install should cope down to -6C and wanted to fit the Daikin 8kW unit, but that was too big to fit the space I had and would use more electickery to run it.

COP for December was 4.2, January, being colder, COP has dropped to 3.7

Posted by: @agentgeorge

↑

COP for December was 4.2, January, being colder, COP has dropped to 3.7

You are doing well in January.  Dec COP with my Vaillant was 4.2 which is also a typical SCOP for me.  Jan so far 3.2.  I suspect the fact I run in 'low noise' mode, which limits output, at night (probably unnecessarily) isn't helping because it means the heat pump has to work harder during the day.  

My Cosy 9 is managing to keep pace with the cold weather with downstairs rooms at around 22.0 - 22.3c.  Downstairs hall, which is mostly an internal room, is at 23c and my office, at 22.2c with its TRV set at 4.  I've had to set the TRV in the kitchen to 3.5 in order to keep that space at around 22.2c.  Upstairs most rooms are around 22.2 - 22.5c except the bathrooms which are around 23.2c.  The landing, a lost cause in terms of heat control, is around 23.5c but its radiator never comes on and it is purely heated by air from the rest of the house.  There are still two cold rooms due to undersized radiators: the entry hall will typically be around 18.2c but will drop into the 17s and the downstairs loo hovers around 19c: it is the room with the Primary Pod with a set temp of 19c so it "hovers" because of hysteresis.  System is still thermostatically controlled.

COP for December is 3.90 and January to date is 2.93.  Today so far is 2.53 including hot water with an overnight low of -4.3c and 5 hours with a temp below 3.0c.  

The Cosy is definitely working hard as it has consumed 10kWh out of my battery in 3.5 hours.  Which begs a question: is 10/3.5 =2.857kWh an indication of hourly heat loss at -3c to -4c?

It looks like it will be an expensive day.  Usually this cold the sky is clear and solar can add around 7.5kWh of power during the day.  Its cloudy now and just started snowing so I'm not going to see that little boost.

Posted by: @andrewj

↑

The Cosy is definitely working hard as it has consumed 10kWh out of my battery in 3.5 hours.  Which begs a question: is 10/3.5 =2.857kWh an indication of hourly heat loss at -3c to -4c?

If you consume 10kWh in 3.5 hours at a COP of 2.5 then the 'calculated heat loss' is 10/3.5*2.5=7.1kW which quite plausible. 

However 3 Hours is probably too short to get an accurate read out because of the lag in house response etc.  The past few days are, though, a great time to get a measured figure for house loss, although of course its only as good as the sensors in the heat pump.  These can be notoriously inaccurate, but the same can be said of heat loss surveys which can be out by a factor of 2 or more!

@jamespa thanks James.  That’s an interesting figure: just before Christmas, Octopus visited to do another heat loss survey and determined the original 8.5kWh figure was over stated by around 1kWh and heat loss is actually 7.5kWh at -3.2c

Isn’t there too much averaging if calculating over a longer period - wouldn’t we want worse case scenario, or at least design temp verification? Taking last week’s figures: 227kWh/(7*24)*3.11=4.2kWh

Posted by: @andrewj

↑

Isn’t there too much averaging if calculating over a longer period - wouldn’t we want worse case scenario, or at least design temp verification? Taking last week’s figures: 227kWh/(7*24)*3.11=4.2kWh

 

In this case you need to use average OAT and energy supplied to the house (as opposed to energy consumed) to do the calculation.  You then get the loss at the OAT corresponding to the average.  The problem with a 3 hr window is many houses take 12 hours or more fully to respond so it may well not be representative unless the preceding conditions were very constant.  

Unfortunately although its conceptually very simple, practically its not so simple because the outdoor conditions keep changing, unless of course you have a house in a lab or you get one of these days when the temp is stable for 24hrs, which does occasionally happen.

The more robust way is to plot heat supplied to the house vs average OAT like this (the plot used degree days rather than average OAT but the concept is similar).  You can see from the scatter that taking even a single day in isolation can lead to a considerable error.  This was from my gas boiler and is basically how I sized my heat pump.  Installers were then told that they had to make their GIGO calculations fit the facts otherwise I would look elsewhere (I didn't quite put it that bluntly but this was the message).

My Cosy6 heat output was 116kWh yesterday, giving an average of 4.8kW/h. Temperature here dipped to -5C last night, coldest so far.

COP has dropped from 4.5 last month, average temp 6C, to 3.7 this month, average temp 0C.

@jamespa Thanks - I learn a lot more about heating systems asking simple questions than I ever thought I would need.  Seems to come with the ASHP territory!

Posted by: @andrewj

↑

@jamespa Thanks - I learn a lot more about heating systems asking simple questions than I ever thought I would need.  Seems to come with the ASHP territory!

It does but it almost all applies to boiler systems and I wish I had understood when I had a boiler as much as I understand now.  Just before I replaced it (after I stared learning about ASHPs) I was operating my boiler in a way that was as similar to an ASHP as the hardware would allow, which improved comfort and reduced cost relative to the way we are 'told' to operate.

I did not hear back from the tech team. (Assume they are busy).

But I have managed to solve my cosy noise issue by about 95%. I can now run it at night.

I removed the excess flashing around the case exhaust area, which makes that whole area smooth and this seemed to stop the fan booming noise and moved the resonance frequency down a little, which has made the structure noise way less noticeable.

Obviously octopus can't expect customers to do this 👍

Personally i think this needs a software patch where the fan driver and inverter avoid resonant frequencies.

Posted by: @agentgeorge

↑

Just had my annual service by Octopus. The engineer replaced the pcb in side the Cosy6 without me prompting him.

I asked what the new PCB did, all he new is the firmware on the board is different and there is a change to the earth wire.

I am just wondering whether you have noticed anything different following that PCB exchange during your annual Cosy service?

@swwils Have you got pictures of what you did?