Yeah that’s good to hear about the upstairs and downstairs, I agree radiator balancing will help and I’m gonna sort that next week.
It’s just slightly short and sitting around 17.5 (desired temp 18) in living room, but like I say need to get some of that heat from rads upstairs to downstairs.
I would like to lower the heat curve though as it’s 0.75 but would like to get it close to 0.6.
Yeah we are comfortable with the desired temps and setback, don’t think we would like it any warmer, probably 19 tops.
I think I’ll have a play with HC and setpoint. Is it worth scrapping the setback and just keeping temperature at one temp day and night?
it is interesting to see what other temperatures have
Check this excellent post & the article I linked there as well. House Thermal Inertia and Roomstat Setback (some cautionary notes)
I’m not running with setback because my effective setback timeframe would be something like 4 hours to get temperatures back up where I want themwhen I do. The savings for me would be minuscule if existent at all.
So experimented with the HC last night, lowered to 0.7 (from 0.75) but instead of having a setback to 17.5 degrees from 2130 until 0700, I just left it on constant 18 degrees and see what happened.
Living room (where the OEM thermometer is) was 17.9 at 2200 hrs but by time we woke up this morning it was 17.5. However upstairs was 19.1 degrees when we woke up so a degree and half difference which is indicating that my HC can be lowered but got to get my radiators balanced first. I know there will be a bit of that heat rises but I’m sure I can get closer between upstairs and downstairs and even lower my HC to 0.65 or 0.6 once done.
Defrosts overnight was 7 defrosts since midnight up until now with overnight temp of 3.0 degrees here.
https://emoncms.org/app/view?name=MyHeatpump&readkey=8f21605274780a0df694c37d97b64bd1
Don’t forget that you probably occupied the living room in the evening and not overnight. 1 person is approximately 100W of heat and that can make quite a difference, any electrical device (TV,…) also produces heat. In the evening, my living room is typically at 22-23°C, so around 1.5°C above what it would be without any occupancy.
Another new article I’ve written over on Energy Stats
Use Noise Reduction mode to get a half-way house between Eco and Normal
Thanks to @dfeist for inspiring this article and helping with the proof reading! 
I am finally seeing some temperatures where my 10 kW unit has a chance to run continuously (currently pushing -10°C)…but it’s not. Here’s what’s happening (left y = temperature, right y = compressor rps, all values off ebus, so the heap pump is using exactly these values for its computations):
The first visible cycle (until 16:30) is rather typical for me: Flow temperature keeps creeping further and further away from the target even though compressor speed is at its minimum (30 rps), EnergyIntegral (not shown) increases and the compressor is stopped once it reaches 0°min. I’m using a comp. start. heat value of -100°min if that matters.
Starting at around 18:30, the compressor starts working above its minimum speed and the flow temperature more closely adheres the target - but with an offset. Consequently, the EnergyIntegral still slowly builds up and I’m still seeing cycling. Granted, a 3-hour cycle length is OK but I still don’t understand what’s going on. Why is my compressor currently at 39 rps, with flow temperature roughly 1°C above target? I understand that the underlying algorithm still needs to decrease the integral term and thus has to overshoot the target for some time, but why doesn’t the compressor speed decrease once I’m getting close to 0°min? Is there some physical limitation that requires higher minimum rps at lower temperatures?
Can someone with a more appropriately sized pump please post an EnergyIntegral curve for a long steady-state run so I can compare?
Edit: It seems this is normal - minimum power draw & heat output starts to increase for minimum speed below certain temperatures. I’m not complaining about a COP of 3 at -10°C, but it would have been nice to at least once see this baby run steady for a night 
.
This is great. I have always been thinking the same thing. Eco too slow, the other modes not good enough in the efficiency department. I was thinking about using noise reduction for it too, but never got around to proper testing.
Very, very nice information!
Something strange happened with our hot water run this morning, it looks like the flow temp reached nearly 70c and the water in the tank reached 8c over the target (target of 55c, so 63c). Charging offset is set to 10K.
This is yesterday for me Andre on my 5kW unit at heat curve 0.5, active room temp mod, target 20C indoors (apart from small 19.5C setback 22:00 to 01:00)
Avg outside temp of 1.4C
This is the matching period on emoncms
https://emoncms.org/energystatsuk?mode=power&start=1739715120&end=1739796480
Apart from the hot water run, constant running.
Re: Something strange happened with our hot water run this morning, it looks like the flow temp reached nearly 70c and the water in the tank reached 8c over the target (target of 55c, so 63c). Charging offset is set to 10K.
Is it the legionella cycle? That seems the most likely explanation.
Only you know when your Legionella cycle is timed for surely
Legionella cycle is disabled
Fascinating read.
When we had ours installed at the end of Apr’24, I spent weeks fiddling about with all the different settings (offset, hysteresis, NR, ECO etc) trying to get the best CoP and then optimum balance. Knowing how all the settings interacted would have saved me a load of time.
In the end, turns out it was a fruitless exercise as, virtually no matter what I do, our metered water costs more than the electricity used by the HP.
I have a feeling that with many DWH tanks the heat transfer capacity of the coil reduces as more of the coil is within the warm zone of the tank. This will clearly be different for a PHE based DHW tank.
If I am correct then measurements of DHW COP need to be done on each type of tank to work out best control methods.
That’s my understanding
Correct, inactive has no room influence.
Full breakdown of the modes here:
I’ll try and post mine mate, where do I access this graph and I’ll post mine for you
