@cathoderay thank you, that is a very helpful reply.

shark infested water...

1. manufacturers can mislead with model sizes
2. installers serve themselves first and not your needs
3. MCS nudges heat loss estimates too high with overstated ACH guidelines

no wonder I'm finding this forum and associated YouTube so rewarding!

@matwin - a good summary, though in fairness I should add there are also a few friendly dolphins out there as well. The problem, as many have discovered, is finding them, because it is not unknown for sharks to dress up as dolphins.

This forum is a great resource. It has both accounts of individual heat pump 'journeys' with lessons learnt, and a lot of more generalised practical expertise. It is also remarkably good natured, largely due to the tone set by our most esteemed leader, Chairman Mars. We also follow the 'only dumb question is the unasked question' maxim.  

Posted by: @matwin

↑

Heating Degree Days

Heat degree days are not always that straight foreward. There are websites that will automatically calculate. But you have to be careful the weather station data is actually representative of your house, plus you need to set the base temperature, this is the temperature you actually stop needing to heat the house. 

Things that affect the above

Local micro climate for house or weather station in a micro climate

House airtightness and insulation levels of house. Well insulated and airtight, this drive down the outside temperature at which heating is needed. Very poor insulation the opposite. Either end of the scale gives very different HDD output numbers - how do you know what is the correct number

House inside temperature also changes the outcome

Posted by: @matwin

↑

MWh

To be representative, the heating system operation needs to be similar to the heat pump, heat pump design (to get grant money) needs house to be at given temperatures, if you operate the house at 16 to 18 then gas consumption will not be representative.

So there is a possibility of a calculation that is very miss leading, house heated differently, miscalculated HDD, equals rubbish in, rubbish out.

A heat loss calculation, using the best known data isn't generally that far out, add in a bunch of assumptions, again rubbish in rubbish out.

A simple whole house heat loss calculation is fine for sizing heat pump. Room by room is required for emitter sizing. Emitter sizing is required for a balanced heating system.

Posted by: @cathoderay

↑

heat pump puts out less than it should according to its badge rating, but at the same time the spreadsheet based heat loss was an over-estimate of the actual heat loss

Not saying you are wrong, but what temperature should they pick, different countries have different minimum design temperatures. Even in this country it varies, your design temperature is most likely -3, while mine is -9.  So do they sell me a different badge, to one they sell to you? The designers job is to to match a heat pump to the location and house. They have to read the performance chart for each and every install if they are designing correctly. So really doesn't make a jot if difference if they add a number or just call them small, medium, large and extra large, or well insulated, average or leaks like a bucket house.

My view who cares what number they feel they should add to a heat pump, you still need to go back to base data and pick a heat pump suited to the application.

Posted by: @Anonymous

↑

what temperature should they pick, different countries have different minimum design temperatures. Even in this country it varies,

My installer used the Freedom heat pump spreadsheet, which until recently was widely available and widely used. It is highly MCS focused (for grant applications) with design outside air temp (OAT) and room temps in effect set by MCS, who if I remember correctly get them from somewhere else, possibly a crystal ball. This particular spreadsheet also has output data for a number of heat pumps, and allows you to match heat loss to the heat pump output. What it didn't say on the tin (it is instead on hidden sheets which you have to unhide) is that the Midea heat pump outputs were guesstimates.

Another source of spreadsheet error not to be overlooked is typos. For a while my largest room with three external walls (the kitchen) had an external wall length of 2 metres. It is actually 12 metres.

I remain convinced that if you can do an empirical heat loss, it always trumps a spreadsheet. Spreadsheets are in my view, as many will know, a form of whatiffery (what if the ACHs are 3, what if the OAT is -2 etc etc), and as you say, the output is totally dependent on the input. Get something in the input wrong, be it a typo or a wrong what if assumption, and the output will be wrong. An empirical (ie measured) heat loss on the other hand if done properly involves few or even no assumptions. Done properly here means steady state room temps plus accurate enough outside air temp and energy delivered to the house data. Given steady state room temp, the energy delivered to the house is the heat loss at each outside air temperature. Plot energy delivered against OAT and then read off the heat loss at your design OAT.         

Posted by: @cathoderay

↑

design outside a

ir temp (OAT) and room temps in effect set by MCS, who if I remember correctly get them from somewhere else, possibly a crystal ball

Room temps are set temperature for different rooms apply to all buildings. OAT is a simple equation to cover 99.7% of likely minimum outside temperature. So ignore the lowest temperature for 2 days.

Posted by: @cathoderay

↑

An empirical (ie measured) heat loss on the other hand if done properly involves few or even no assumptions. Done properly here means steady state room temps plus accurate enough outside air temp and energy delivered to the house data

Perfectly acceptable way to do it, but to do it properly, you and your family get to go into a hotel for several days, while data collection is completed. Several thousands of pounds later, you get a nice spreadsheet of data showing heat loss over a range of temperature. And an even more expensive install when you factor in this added cost. So you need to be realistic.

A spreadsheet model is perfect acceptable, as long as the person completing know what they are doing. And have suitable data to input.

Posted by: @cathoderay

↑

Midea heat pump outputs were guesstimates.

Midea like every company that manufacturer, publish full data for their heat pumps. Midea chart every flow temp and outside temp on a table for each model.

Posted by: @matwin

↑

shark infested water...

It's not really that bad if you just take it methodically, using the knowledge on this forum, and have confidence in your own ability to think logically, which you have already demonstrated by asking the very first question you posted.

Interrogate the detail of the survey calculations (or do the calculation yourself) to find out what assumptions have been made that are wrong or maybe wrong.  Correct the wrong ones and be aware of the scale of the maybe-wrong ones.

Collect whatever data you have on actual consumption, plot vs degree days or average temperature (if you have sufficient data), use it either to sense check the spreadsheet calculations or make a reasoned estimate for the maybe-wrong parameters, applying some common sense as to what the error margins are depending on your situation.  By putting all the data available together you should hopefully be able to get some confidence in the figures.

Using the combination of all the data (measured and calculated) available select a heat pump (or candidate heat pumps) based on performance tables and other considerations (physical, aesthetic, noise, price, control capabilities, quality of tec support, available installers).  

Likely the above narrows it down to a fairly small set of installers/models.  The gaps between models in any one manufacturer line are large so nothing has to be super accurate, for example Vaillant in reality do 12, 7, 5 only (headline figures), the others are simply software limited versions of the above.  Other manufacturers are similar.

The penalties for mild over or under sizing aren't disastrous, mild under sizing means you will need to use supplementary heating (or the immersion) on a few more days of the year when it's very cold all day for several days.  Mild oversizing may have a tiny performance penalty but is much to be preferred over mild undersizing.  Gross oversizing is to be avoided because the machine will always be operating at the inefficient end of the compressor performance curve and always cycling.  Gross underdizing also means that the machine will be operating at the opposite inefficient end of the performance curve and of course won't heat your house some of the time.

System design anyway matters more than choice of heat pump, the key factors to watch for, because it helps differentiate installers, being to design for lowest ft reasonably possible and do not include buffer/llh/phe between heat pump and emitters (two port volumiser is OK).

@jamespa thanks yet again James. Very clear and usable insights.

As a complete newcomer to SHPs I have to say that there is room for your advice and simplification - written up as a "Key steps to sourcing a succesful heat pump" for us novices.

Mars, a one or two pager that holds the attention, minimises the PhD physics and empowers the consumer to deal with the variable competencies/helpfulness of installers!

Sorry to be so blunt but if the revolution is to get going it needs to drive quality standards up from the paying end - ie us!

and thanks everyone else for the tips and explanations so far

Posted by: @matwin

↑

…

As a complete newcomer to SHPs I have to say that there is room for your advice and simplification - written up as a "Key steps to sourcing a succesful heat pump" for us novices.

Mars, a one or two pager that holds the attention, minimises the PhD physics and empowers the consumer to deal with the variable competencies/helpfulness of installers!

…

With 20:20 hindsight, we probably ought to have provided you with a link to this article before.

https://renewableheatinghub.co.uk/the-abcs-of-ashps-a-jargon-free-introduction-to-heat-pump-basics/
Is that what you mean?

Posted by: @majordennisbloodnok

↑

Is that what you mean?

I think @matwin may be after the next stage. The ABC guide covers the basics - jargon and principles - but doesn't go so far as how to select an installer and get a satisfactory system. That information is here on the forum, in many places, but so far as I know hasn't been collected into one practical user guide. For greater depth, there are of course Mars's books.   

Posted by: @Anonymous

↑

Perfectly acceptable way to do it, but to do it properly, you and your family get to go into a hotel for several days, while data collection is completed. Several thousands of pounds later, you get a nice spreadsheet of data showing heat loss over a range of temperature. And an even more expensive install when you factor in this added cost. So you need to be realistic.

It doesn't need to be that complex (the great being the enemy of the good). Also removing the inhabitants removes one of the key variables - human behaviour - and so may actually make the result less accurate. Furthermore, you want as long-ish period of data, with as wide a range of OATs as possible. 

Posted by: @Anonymous

↑

Midea like every company that manufacturer, publish full data for their heat pumps. Midea chart every flow temp and outside temp on a table for each model.

I know that now (only it is not every flow temp, and every OAT, but a selection, albeit at reasonable intervals), but I didn't at heat pump selection time, and it is buried away in their Engineering Data book. Is it really reasonable to expect all prospective heat pump owners to dig deep into not freely available technical manuals? Is it even practical and realistic given you need some technical knowledge to make sense of the tables? I wonder what Mrs Trellis of North Wales would make of it all? Why not instead require manufacturers to publish low OAT performance in their brochures and marketing materials - the stuff the punters actually get to see?

The point about the guesstimates is that Freedom Heat Pumps put them into their spreadsheet, and then hid the fact their were guesses (as it turned out, over-optimistic ones). It is also yet another example of how whatiffery (what if Midea output is X?) can lead you up the garden path.    

Posted by: @cathoderay

↑

Is it really reasonable to expect all prospective heat pump owners to dig deep into not freely available technical manuals?

Why should the owner really need to understand or delve into any data - you are paying a hefty price for the installer to know and select correctly. If that doesn't happen the MCS system is flawed and needs fixing. Which is really a different kettle of fish.

My wife would have zero interest or a clue about heat pump sizing selection, why would she, she was the client, if the house is comfortable zero comment. If it cold I get in the neck. But I designed the system selected the heat source, so I need to get the complaint. 

If your heat pump isn't correct complain, if that gets ignored, go to MCS and complain further. Keep complaining go to the press - our tax money isn't being spent well.

Posted by: @cathoderay

↑

Furthermore, you want as long-ish period of data, with as wide a range of OATs as possible. 

Not really as long as you know steady state internal temperature, once fully heat soaked you can extrapolate for design temperature.

People moving, opening doors windows may sound like it's normal life, but it isn't the design parameters required. So become rubbish. Sorry not the answer you believe is correct. Spreadsheet is fine, with the correct numbers. I did my own from house design data, and is within a few percent at -9. 

Posted by: @Anonymous

↑

Spreadsheet is fine, with the correct numbers. I did my own from house design data, and is within a few percent at -9. 

Yes that's true of course 

Unfortunately installers all too frequently don't use the correct numbers, and it's not always possible to tell, which is why using gas consumption intelligently interpreted (IE not just rule of thumb), at least as a sanity check (or more depending on the available data) is a useful thing to do. 

If you can rationalise data from spreadsheet and measured data then you can increase confidence and eliminate obvious sillies like 16kW survey results for a 7kW house (mine) or 12kW for a likely 5-7 kW house (IE the one in question). 

Surely you can accept that neither is perfect (except possibly with a newbuild where you.know the construction and ach accurately),  together they are better than either alone.

Absolutely no need to go into a hotel if you have a smart meter, just look back a year and plot daily consumption vs either average oat or degree days, whichever you can get hold of.  Make a sensible assumption about gas efficiency depending on age of boiler and operating temp, if necessary a (relatively small) adjustment for part time heating and you are there (including dhw use).  Not perfect but better than having nothing to compare often hopelessly inaccurate surveys with.