Hi Ian

3000 kWh per year doesnt sound that high to me. I was using about 50 kWh per day during winter. Obviously the more insulation you can install the better.

I found this guide very useful What Size Heat Pump? - Energy Stats UK (energy-stats.uk)

which gives several different ways of estimating your heat loss

Thanks for that, @bontwoody. Very interesting.

I'm quite comfortable with my room by room heat assessments, which pass several of my 'sanity checks', although, as others have said here, the calculations have the whiff of voodoo around them. There are very few data points relating to very cold days on my graph so I am tempted to think that I should rely on supplementary heat sources (or an extra clothing layer) on the data extremities (i.e. the handful of coldest days) in an attempt to reduce heat pump cycling at the other end of my graph, where the data points are concentrated. I'm not planning on removing gas completely; we have a gas hob in the kitchen and a gas stove in the lounge.

As I mentioned previously, I suspect that the radiators are undersized (they are all type P1 or K1) and I wonder if the system as originally designed (before being hobbled by the replacement boiler) was intended to be run 24/7. There are signs that there was originally an external temperature sensor on the front of the building.

The radiators will be a ball-ache to change. The existing widths don't seem to be a standard size and the downstairs units are plumbed with 10mm copper inset into dot & dab plasterboard walls. I suspect that a big problem will be the kitchen/diner which we renovated last year and fitted a pair of tall 'designer' radiators. Plumbing to both includes runs of 10mm copper and one is now the furthest radiator from the boiler (it was extended with 15mm PEX) and takes a while to get to temperature even now. Maybe I should augment the heating here with an A-A heat pump.

I'm wondering if there is anything to be gained by a detailed analysis of pipe sizes. I haven't done fluid dynamics since the first year of my engineering degree in the mid 70's. Since I've now forgotten it all, it's more black magic to me now. Will this all 'come out in the wash' or do I need to persevere to ensure that the pump is dimensioned to cope.

Posted by: @ianmk13

↑

The annual gas consumption for CH and DHW for my house is around 3000kWh.  I'm told this is high.  The boiler was replaced ten years ago (by BG and before I bought the property) and was left configured as 24kW output. It is not ageing well. During the heating season, I set-back the temperature to 19degC overnight, although in the depths of winter it will take hours for the house to slowly get back to a comfortable temperature in a morning.  The radiators are hot, but perhaps too small, so I suspect that the boiler may be cycling rather than modulating, resulting in high wear and tear.  Since I need to be considering boiler replacement, I have started to look at ASHPs. I am trying to be open-minded despite many aspects presently being non-intuitive to me.

To characterise my existing system, I have plotted my gas consumption (from my smart meter and converted to kWh) against degree days (derived from my weather station records) for the past year. I was interested to see that peak gas consumption was around 260kWh at a degree-day value of 19 (referred to 15.5degC). This gave me confidence to adjust the boiler output setting down to 12kW.  The trend line for my data is defined by y = 12.822x + 13.658. '13.658' would appear to be the kWh energy expended on DHW per day. Is the '12.822' my property's 'heat loss', or something else? I commissioned a (remote) on-line heat loss estimate  which came out at 11kW. My own detailed estimate, which erred on the side of pessimism, was 13.8kW. Being relatively new to this branch of engineering, I'd be interested to hear comments from those with experience.

3MWh/year is pretty tiny, are you sure of this figure?

If this is correct then it implies that your house requires a boiler with an output of roughly 1.5kW only.  So either your house is very small or very well insulated.  Your estimate isn't consistent with this, something is wrong! 

12.8 is kWh per day per degree delta T.  So at eg -2 you expect to require 12.8/24*(20--2) = 11kW (or arguably a bit less), pretty consistent with your estimate but not your annual consumption.

How embarrassing. I dropped a zero.  I must stop ignoring those Specsavers reminders 😉 

Thanks @JamesPa.  It looks as though my pessimistic approach was unnecessary, then.

Posted by: @ianmk13

↑

How embarrassing. I dropped a zero.  I must stop ignoring those Specsavers reminders 😉 

Thanks @JamesPa.  It looks as though my pessimistic approach was unnecessary, then.

OK 30MWh/yr is quite high.  Can you tell us something about your house and water usage pattern?

@ianmk13 Sounds as if you have a lot of the same issues as I had. in the end i changed the 10mm pipes for 15mm and upped several radiators. I ran the gas CH at a low temperature during the winter to see how the house would cope when I switched to a HP. I have also put a A-A heat pump in the kitchen as the radiator gap was small without having a very long run, plus I get cooling in the summer if needed.

I have a small wood burning stove in the lounge to top up should we want to be extra toasty. You can see all I have done at my blog, the address is below.

Posted by: @jamespa

↑

Can you tell us something about your house and water usage pattern?

What would you like to know?

The property is an unusual design : split level, L-shaped, 2 storey; not a rectangular box. Insulated cavity wall, outside skin mainly brick, but some wood cladding. Glazing predominantly on the eastern side. Double glazing throughout, but 1990's standard apart from 3 sets of bi-folds we had installed a couple of years ago. I used my detailed knowledge of the construction to use the standard heat-loss spreadsheets to arrive at my figure of 13.8kW.  I suspect the boiler isn't working in condensing mode as it's normal to see plumes of steam from the boiler.

There are 2 of us in the house and we have showers rather than use the bath. I forgot to mention that the figure of 13.658kW will include use of the gas hob in the kitchen.  The gas stove in the lounge wasn't used last winter (the ultrasonic remote upsets the cat !)

@bontwoody

I don't relish the idea of having the first floor up and chopping the plasterboard walls to replace all the 10mm pipes. It will be hard enough to move the 10mm pipes to fit replacement radiators (both the pitch and distance to wall are wrong). I was tempted to reduce the flow temperature but thought it best to ask the plumber to do it when he services the boiler.  We have similar ideas regarding supplementary heating (and cooling).

@ianmk13 The pipework replacement was easier for me as I live in a bungalow. Your house does sound better insulated than mine so maybe you could get away with the 10mm pipework. I think John Cantor talks about that on his webiste somewhere but it may mean running at a higher temperature and hence lower COPs. However if your heat loss is low enough the cost may not be prohibitive.

We are slightly confusing two things here, the total system capacity (~12kW?) and the emitter capacity (linked to pipe size).  They are separate but linked.

Pipes to individual emitters must have sufficient diameter to take the demand from that emitter (which must be physically large enough to emit the required power at the temp difference between it and the room - note for rads this is non linear - the power emitted varies as (emitter temp-room temp)^1.3).  Where several pipes join in a manifold or through branches, the feed to the branch must have sufficient capacity to accommodate the total, culminating in the pipes at the output of the heat pump ('primarie's) which must be able to accommodate the total system capacity.

The key parameters are:

1.Flow temp, which defines emitter temp, which defines how much power is lost by the emitter to the room (strictly its the average temp of the emitter, its colder at the output end than at the input end).  Typically 35-55C for ASHP, 55-75 for boiler.  

&

2.Emitter delta T (temp loss across emitter or between flow and return) which is a result of the above and the flow rate through the emitter.  Typically 5C-7C for ASHP, 20C for boilers.  The principal reason ASHPs are designed for a lower emitter deltaT is that it means that the average emitter temp is nearer to the flow temp, meaning that the flow temp can be lower for a given emitted power, and ASHPs are more efficient at lower flow temps

So because the delta T is less for an ashp, you need larger pipes, because the heat transferred from the pipes to the emitters) is flow velocity*cross sectional area*deltaT*volumteric heat capacity, and velocity is limited because of pipe erosion and noise.  So a 22mm pipe which is good for perhaps 30kW at a deltaT of 20C, is good for only 6-8kW at a deltaT of 5C.  

I think a 10mm pipe is good for about 2.5kW at a deltaT of 20 and 1.5m/s (which is at the upper end), so thats only 625W at a deltaT of 5, insufficient for many rooms in a typical house.  You could run at a deltaT of 7 maybe, upping it a bit, but its still not going to get to 1kW+ which is likely to be needed if your total demand is 12kW.  Realistically many rads need 15mm.  But you also have to think about manifolds/junctions and the sizing of the pipes to them.

For a 12kW total demand you will need, I think, 28mm from the pump to the point it first splits.  if it splits about equally (eg upstairs/downstairs) then 22mm each should be good.  Then 15mm to the emitters (but 10mm will be sufficient for small/well insulated rooms).  this is rough and ready though, you need to do the calcs on room demand etc.

Hope that gives you some high level pointers, in the end there is however no substitute for doing the spreadsheet!

This chart from Heat Geek shows the capacity of different pipe sizes. 10mm microbore is 1.15kW/s at a DT of 5. My heat pump is on the last day of install today  (9kW Panasonic) and I have 28mm put in to the split then 22mm primaries and 10mm drops to all the rads. System has been designed for 35C flow temp, open loop and a DT of 5 and the designer was happy to keep all the existing pipe work. It just depends on your heat loss results- mine was about 55W/m2.

Posted by: @dougmlancs

↑

 

-- Attachment is not available --

This chart from Heat Geek shows the capacity of different pipe sizes. 10mm microbore is 1.15kW/s at a DT of 5. My heat pump is on the last day of install today  (9kW Panasonic) and I have 28mm put in to the split then 22mm primaries and 10mm drops to all the rads. System has been designed for 35C flow temp, open loop and a DT of 5 and the designer was happy to keep all the existing pipe work. It just depends on your heat loss results- mine was about 55W/m2.

kW/s, that's not a real unit (yes I know it's what heat geek says).  You mean kW.

The heat geek figures are a bit better than I remembered, 10mm microbore must have a thin wall.  Good enough for many cases then so long as it feeds a single rad.