Hi

We are looking to install ASHP for our Edwardian detached home. Had a few heat surveys which tend to recommend a 12kW ASHP. However, rule of thumb from Dr Modesta and our annual 22MWh gas consumption indicates only 5kW heat loss. (Our home is well insulated, EPC = B). So my question is...What is the disadvantages of oversizing and would it be smart to compromise at say 8kW ASHP?

Thank you for any ideas and to Mars for this excellent resource.

Rules of thumb are generally pretty rubbish for heat pump sizing and your gas usage patterns may or may not be a good indication.

Common theme seems to be you need 12kW. But the average house in the UK actually only needs sub 6kW.

I would ask to see the heat loss calculation and review it and pass comments back. Number of air changes per hour can be just daft for example, taking no notice of fabric improvements also. Do you have evidence of fabric improvements - without they may have just ignored etc.

General disadvantage of oversizing

Heat pump cycling 

Large heat pumps need very high flow rates, so end up with big pipes, or a buffer (not good for efficiency), bigger radiators than needed etc. So everything gets bigger than needed, you also get a bigger bill than needed, to pay for all that bigger stuff.

The correct size is max heat loss plus an allowance for DHW heating. So if your heat loss was 5kW, or 120kWh a day being able to deliver 120kWh in say 22 hrs. So 5.5kW being able to be delivered at your outside design temperature. 

In addition to the above oversizing means you are likely to be running the compressor at an inefficient point on it's modulation curve for much of the time, reducing COP.

My personal view, which differs a bit from @johnmo, is that gas readings, properly interpreted, can (but won't always) provide the only solid upper bound on heat loss available, so are well worth factoring in in addition to surveys, possibly only as a sense check but possibly as the prime source (depending on circumstances such as how the house is currently heated)

However I agree rules of thumb aren't sufficient, you need a bit more.   Do you have smart meter readings, they can tell you a lot more.  What's your pattern of heating and how old is your boiler, that can help interpret the has readings more intelligently.

I would also do what @johnmo suggests regarding the surveys.

Once you have that data put it all together to get the best data available.

Feel free to ask for more advice/post some of your data.

@johnmo thanks @johnmo that is great info - I will be discussing this with our best prospect installer.

I just needed some ideas to make it a discussion instead of a lecture!

So far I have just too much conflicting information:

1. Dr Modesta of the National Physics Lab in Teddington recommends using annual gas consumption (20MW) required to heat the house divided by Heating Degree Days at the local weather station - gives our heat loss 5kW

2. The "B" EPC estimates 18.5MWh to heat the house and water and the two surveyors are broadly the same.

3. Surveyors have estimated 10kW heat loss. But in the surveys, the ACHs seem unreasonably high (3 in bathrooms, loo & utility, 2 in kitchen, hall & landing, 1.5 in living & dining rooms and 1.0 in the bedrooms). I would expect the average ACH for our well draught-proofed house, especially in winter when we are careful with opening doors etc, to be under 1 and not the approx. 2 that is being used to exaggerate the ASHP sizing.

So what is right - the real world historical heating MW data and a world leading scientist or a detailed fabric heat loss estimation multiplied by MCS nonsensical ACHs?

Is there a heat geek approach to ACH that might sort this out?

many thanks

@jamespa thanks @jamespa. I really appreciate your thoughts on this.

Especially the benefits of not oversizing!

Our boiler is a condensing Worcester almost 10 years old and it does fine. We do have gas meter readings per month but not anything smarter. Annual consumption is a steady 22MWh. 

Another thing that bothers me is that if a well installed ASHP can deliver SCOP of 4, why cant a 5kW heat pump deliver 20kW?

Posted by: @matwin

↑

if a well installed ASHP can deliver SCOP of 4, why cant a 5kW heat pump deliver 20kW?

I love it !

That's an excellent question for the Welcome topic!

The maths works the other way around from what you suppose.
The kW figure for the heat-pump is its output.

Thus a 12kW heat-pump would draw 3kW of electricity if it was operating with a SCOP of 4.0
... but 4kW input if the SCOP was only 3.0.

@transparent thanks. Fighting confusion one step at a time. Do you have a view on ACH exaggeration?

Hang on. Let me just have a look through my library of views....

It's Friday afternoon, and my more serious views have already been taken this week.

I could offer you a distant view of Bodmin Moor; that's an Agricultural Cornish Horizon.
Or, at the smaller end of the scale, I had a fleeting view of an Accelerated Collided Hadron... but it lasted barely a femtosecond.

Would you like to clarify which particular ACH you're interested in,
and just how exaggerated you think it shouldn't be?

@matwin, welcome to the forums.

Expanding on transparent's reply, it's worth saying that a 5kW heat pump can deliver 5kW of heat... that’s its rated capacity. If its SCOP is 4, that just means that over the course of a heating season, it’s using about 1.25kW of electricity to produce that 5kW of heat. But it’s still only delivering 5kW. It can’t suddenly become a 20kW machine. Hope that helps.

@editor thanks Mars. Your channel is making a big difference!

@matwin really pleased to hear that… thank you for your positive feedback.

Feel free to start a new, dedicated topic for your journey.

Posted by: @matwin

↑

@transparent thanks. Fighting confusion one step at a time. Do you have a view on ACH exaggeration 

I don't have much concrete evidence but

• The only way to explain the discrepancy between my calculated and actual heat loss is to assume ach =0.5 instead of 2-3
• Several of the better installers have gone on record as saying that they believe ach is overestimated and/or that when they have measured ach and it's never more than 1
• There is clearly an epidemic of loss overestimation generally.

One thing I don't yet understand is the basis of the figures.  I believe that ach tests for building regs are done by pressuring the house and it's this figure that corresponds to the figures used for loss estimates. 

However we don't generally pressurise houses, so the actual figure is due to natural infiltration or (usually but not always) very localised forced ventilation amounting to nothing like the assumed figure when averaged out over a whole house. 

My strong suspicion, for which I have some very weak circumstantial evidence, is that MCS used the only figures that they could find (and attribute to another body thus avoiding liability) not anything that has a scientific basis for the purpose in question.

If you have consumption per month (particularly if it's for more than a year) you could try plotting that vs degree days or average temperature, which you can download from various websites.  This may give you a bit more info, although unless you have several years of data probably not (but it will still increase confidence levels so very likely worth doing).