Heat Pump Sizing?

matt_drummer

Reed_Richards said:

I can't help thinking that the reason @matt_drummer goes not use Weather Compensation is that his starting point (30 C) is so low that there isn't room to go any lower.  I use WC very successfully with my heat pump but my starting point is about 44 C flow at -3 C outside.  Matt will be achieving better efficiencies than I get; I didn't have the space for radiators that would work at 30 C at outside temperatures less than 12 C. 

Yes, that is partly why.

I got very lucky or I am extremely skillful, probably the former, but I just get the right amount of heat at a fixed flow temperature that happens to be at the sweet spot of my Daikin's efficiency.

I would program a weather curve if I could but I need a negative curve and you can't do that.

I want slightly higher flow temperatures as it gets warmer and i think that is because the compressor likes a certain amount of load. 

The Daikin is not very happy below 31c when it is warmer but I have had it running at 26/27c. It is just less efficient that low than it is slightly higher.

In the warmer weather this Daikin is not quite as good as a Vaillant in terms of efficiency but it excels when it gets below 5c outside and I think it matches/beats pretty much everything.

This is based on hours and hours of observing the behaviour of my heat pump forced on me with the grossly oversized initial unit.

It all started from trying to minimise the electrical consumption of my 9kW Daikin.

So when I turn the heating on it just runs continuously only stopping to do dhw.

I think my radiators are so big that they act as the weather curve although I am not sure I can explain it easily. They are probably too big in some respects as they cause issues when trying to run a fixed dT between flow and return.

I have had a software update to the heat pump itself and it should mean I can run it even more efficiently, particularly in the early stages of a heating cycle.

I may try handing over control to the Madoka room control again.

That has never worked very well for me in the past as the hysteresis on room temperature is 2c which I find too much.

The Madoka is supposed to modulate the flow temperature to maintain the requested room temperature but it is difficult when it is already running so low. The new settings I have available may make it work better.

My setup is unique and unusual and I am not saying that 30c flow at negative outside is the target for everybody.

What I do know is that running at 55c plus won't be good.

FreeBear

michaels said: Difficult to add rads in our house downstairs as solid floors and the rooms being designed for the rads in certain positions.

If you can't fit larger radiators or add more, then the only other option is to cut the heat losses. But fitting wall/floor insulation and changing windows gets very expensive.

The alternative to a conventional radiator is a fan assisted unit. In a kitchen, a plinth heater takes up zero wall space, but the biggest I've seen is a 2.5kW @ ΔT50. Smiths, Myson, and others do wall mounted models. Downside is they are expensive, but may be comparable to designer radiators (and not as pretty).

Reed_Richards

.Here is part of the spec. for a Thermix plinth heater:

Build-in thermostat doesn’t allow the fan to blow the cold air until the water temperature in the central heating system reaches above 38°C.

That's not very heat pump compatible because when it's warmer outside then with Weather Compensation you might well want your maximum water temperature to be less than 38 C.  It wouldn't work at all for @matt_drummer with his 30 C water temperature.  But I think a lot of plinth heaters work this way so you would need to track-down one that doesn't.

FreeBear

Reed_Richards said:

.Here is part of the spec. for a Thermix plinth heater:

Build-in thermostat doesn’t allow the fan to blow the cold air until the water temperature in the central heating system reaches above 38°C.

That's not very heat pump compatible because when it's warmer outside then with Weather Compensation you might well want your maximum water temperature to be less than 38 C.  It wouldn't work at all for @matt_drummer with his 30 C water temperature.  But I think a lot of plinth heaters work this way so you would need to track-down one that doesn't.

Got a Smiths plinth heater here. Swapped out the bimetalic switch for one rated at a lower temperature a few years back. Now experimenting with lower flow temperatures, and may well replace that switch with a wifi enabled smart (Sonoff Basic). Then use the home automation system to control the fan.

NedS

matt_drummer said:

Reed_Richards said:

I can't help thinking that the reason @matt_drummer goes not use Weather Compensation is that his starting point (30 C) is so low that there isn't room to go any lower.  I use WC very successfully with my heat pump but my starting point is about 44 C flow at -3 C outside.  Matt will be achieving better efficiencies than I get; I didn't have the space for radiators that would work at 30 C at outside temperatures less than 12 C. 

Yes, that is partly why.

I got very lucky or I am extremely skillful, probably the former, but I just get the right amount of heat at a fixed flow temperature that happens to be at the sweet spot of my Daikin's efficiency.

I would program a weather curve if I could but I need a negative curve and you can't do that.

I want slightly higher flow temperatures as it gets warmer and i think that is because the compressor likes a certain amount of load. 

The Daikin is not very happy below 31c when it is warmer but I have had it running at 26/27c. It is just less efficient that low than it is slightly higher.

Yes, you are correct - heat pump efficiency varies with compressor load, and most modern ASHP compressors will be most efficient when running in the 30-60% range, particularly in milder conditions. Efficiency tends to fall off a cliff below 25% which is why most manufacturers limit the minimum amount their hardware will modulate to around 25%. Yes, it could modulate lower, but you'd get much lower COP and may benefit by running the compressor slightly harder. This is also why you will not see many manufacturers publish data for the minimum input/outputs, as COP values at minimum will not be great, and poor published COPs do not help sell heat pumps.

Your detailed observations fit perfectly with the above.

michaels

So I need a heat pump that can actually output 10kw at -3.

I will probably be running a radiator temp at 45C at -3

I use about 18k of heat and 8kwh of hot water a year.

How would I figure out what SCOP I might expect?!

What heat pump would people recommend?

Currently my electricity costs approx 8.5p a unit and gas about 5.5p per unit.  If I got a scop of 3 then it would save about 2.5p (5.5 - 8.5/3) x 26000 kwh = £650pa so if I paid 7.5k for a heat pump rather than 2.5k for a new gas boiler it would take 8 years to break even in cash terms.

Seems like 10kw or above is not ideal with most ranges as they only have two actual sizes and 10+ puts you into the bottom version of the bigger size so peak output capped 'by software' but minimum modulation relatively high

stripling

Octopus suggested their biggest unit, 11.2kw but is that the right size
I currently have a modulating boiler with weather comp and a smart meter so I know that the boiler runs at a steady 8.2kw (gas input) to maintain the house temp at 20c when the outdoor temp is 0 degrees (flow temp low 50s C).  It uses a fair bit less overnight when we reduce the temp to 16C but I suspect we might want a steady 10kw if it got to say -5 and we wanted an indoor temp of 22C.

@michaels
I'm having the exact opposite problem - Octopus are proposing their lowest size and only 2 radiator upgrades - I think it is not enough. I am replacing a dying, inefficient boiler (85%) in a leaky house. My annual gas use was approx 16277Kwh for heating and hot water.

You could try doing your own heat loss survey on Heat Punk and there's also Michael de Podesta.
This is a good place to start 

I had 2 surveys from 2 different companies and I agree with your perception - both of them designed by tick box not for me and my use of the house. The MCS surveys are ridiculous and full of presuppositions but if you read the MCS guidance the surveyors are supposed to consider variations. It seems none of them do..

Reed_Richards

The MCS calculation uses data from your nearest weather station to work out a temperature that is exceeded 99.6% on the time.  So if your heat pump is exactly right-sized that means on an average of 35 hours per year it won't be powerful enough to keep your house at the 21 C interior temperature that the calculations assume.  In my case that 99.6%-of-the-time temperature was -3.7 C (I'm up north).  

FreeBear

stripling said:

Octopus suggested their biggest unit, 11.2kw but is that the right size
I currently have a modulating boiler with weather comp and a smart meter so I know that the boiler runs at a steady 8.2kw (gas input) to maintain the house temp at 20c when the outdoor temp is 0 degrees (flow temp low 50s C).  It uses a fair bit less overnight when we reduce the temp to 16C but I suspect we might want a steady 10kw if it got to say -5 and we wanted an indoor temp of 22C.

@michaels
I'm having the exact opposite problem - Octopus are proposing their lowest size and only 2 radiator upgrades - I think it is not enough. I am replacing a dying, inefficient boiler (85%) in a leaky house. My annual gas use was approx 16277Kwh for heating and hot water.

You could try doing your own heat loss survey on Heat Punk and there's also Michael de Podesta.
This is a good place to start 

I had 2 surveys from 2 different companies and I agree with your perception - both of them designed by tick box not for me and my use of the house. The MCS surveys are ridiculous and full of presuppositions but if you read the MCS guidance the surveyors are supposed to consider variations. It seems none of them do..

Came across this the other day -> https://github.com/TrystanLea/heatlossjs (found it after following your link) - It uses u-values of each main component of the house along with orientation to calculate the heat loss. A lot more detailed than your average radiator sizing calculator. Maybe a bit more detailed than the MCS spreadsheet.
When I get a spare moment, I'm going to have a play with it.

michaels

FreeBear said:

stripling said:

Octopus suggested their biggest unit, 11.2kw but is that the right size
I currently have a modulating boiler with weather comp and a smart meter so I know that the boiler runs at a steady 8.2kw (gas input) to maintain the house temp at 20c when the outdoor temp is 0 degrees (flow temp low 50s C).  It uses a fair bit less overnight when we reduce the temp to 16C but I suspect we might want a steady 10kw if it got to say -5 and we wanted an indoor temp of 22C.

@michaels
I'm having the exact opposite problem - Octopus are proposing their lowest size and only 2 radiator upgrades - I think it is not enough. I am replacing a dying, inefficient boiler (85%) in a leaky house. My annual gas use was approx 16277Kwh for heating and hot water.

You could try doing your own heat loss survey on Heat Punk and there's also Michael de Podesta.
This is a good place to start 

I had 2 surveys from 2 different companies and I agree with your perception - both of them designed by tick box not for me and my use of the house. The MCS surveys are ridiculous and full of presuppositions but if you read the MCS guidance the surveyors are supposed to consider variations. It seems none of them do..

Came across this the other day -> https://github.com/TrystanLea/heatlossjs (found it after following your link) - It uses u-values of each main component of the house along with orientation to calculate the heat loss. A lot more detailed than your average radiator sizing calculator. Maybe a bit more detailed than the MCS spreadsheet.
When I get a spare moment, I'm going to have a play with it.

Thanks, will take a look.

I am not to worried about the overall heat loss calc I have from Octopus as it ties in with the heat input to maintain temp based on gas usage although individual rooms may be off.

I think I can work out the minimum flow rate I can sensibly aspire to given the positioning and size of rads and I know my annual heat usage and hot water usage but it is then deciding which heat pump is going to be most efficient in delivering that output. 

Being at the 10kw level seems to out me into the 'large' size where they have 2 sizes but that then also means I am at the bottom of 'large' so have least modulation and thus most likelihood that I will be having inefficient cycling when it is not cold enough to need maximum output.

I also need a heat pump that can deliver my required output continuously not only for 45 minutes per hour with the other 15 spent in a defrost cycle.