Home  5  Books  5  GBEzine  5  News  5  HelpDesk  5  Register  5  GreenBuilding.co.uk
Not signed in (Sign In)

Categories



Green Building Bible, Fourth Edition
Green Building Bible, fourth edition (both books)
These two books are the perfect starting place to help you get to grips with one of the most vitally important aspects of our society - our homes and living environment.

PLEASE NOTE: A download link for Volume 1 will be sent to you by email and Volume 2 will be sent to you by post as a book.
Buy individually or both books together. Delivery is free!


powered by Surfing Waves




Vanilla 1.0.3 is a product of Lussumo. More Information: Documentation, Community Support.

Welcome to new Forum Visitors
 Join the forum now and benefit from discussions with thousands of other green building fans and discounts on Green Building Press publications: Apply now.




Heating and cooling: How to size a heat pump

Bottom of Page

1 to 25 of 25

  1.  
  1.  
    When planning a heat pump as a replacement heat source for central heating and DHW how many hours should be the target daily run time for the heat pump ?

    Some thoughts

    Heat load for the coldest month (kWh) / 30 = CH power demand for a day

    CH power demand for the day / (how many hours ? ) = size of heat pump for CH.

    DHW is fairly constant throughout the year so a daily consumption is easy find so –

    DHW power demand for the day (kWh) / (how many hours ? ) = size of heat pump for DHW

    But is the CH load on top of the DHW load or do you target the hours run so that e.g. CH needs 10 hours run time and DHW needs 5 hours run time = 15 hours run time a day so that the heat pump needs to be sized only for the greater of either load.

    Playing with the target run times for each of the CH and DHW will alter the planning size of the heat pump.

    But what should the target run times be ?? And how much spare should capacity be built in ?

    Or have I got it all wrong?
  2.  
    My recent experience was that the installers like to use the MCS heat pump calculator spreadsheet
    https://mcscertified.com/mcs-launch-new-improved-heat-pump-calculator/

    This does a basic room-by-room heat load calc for keeping the rooms at quite a warm temperature, on the coldest night of the year, without incidental gains or any other heating sources. It adds on the DHW load, and an optional 20% uplift for intermittent running.

    Then the installer picks the next biggest model of heat pump from their accredited brand (typically the choice is around 5kW, 9kW, 12kW or 16kW). In my case this was much bigger than my existing oil boiler.

    Then they send the house EPC off to the heat pump manufacturer who confirms yes/no.

    I had expected some logic like you did, but no... they have to follow these rules in order to issue the MCS certificate, which is needed for deemed planning permission and for any funding.

    We disagreed, and the poject is now on hold...!
  3.  
    No MCS ticket needed and no external funding available (sigh)

    I am planning to move from the existing wood burning stove to heat pump so no gas bills to gauge kWh used.

    Room calcs are theoretical and prone to error when dealing with old buildings that have variable wall structures (i.e.my 2 houses linked on one combined system)

    However I have heat meters (kWh readout) on the output of the TS to each CH system so that I have a reasonably accurate measure of the CH load for each house. DHW is done on a night rate meter so direct kWh reading is easy.

    Which is where my thoughts above came from.
    • CommentAuthorRobL
    • CommentTimeNov 16th 2022
     
    Personally, I used our peak daily gas use (our smart meter knew this), and divided by 24. That has both dhw and heating in it. Then went for something 30% bigger, to avoid disappointment (colder weather than last year, and I would like a bit of headroom for warming the house up from cold).

    Ours heats a 200l dhw tank at night for 4 hours in cheap rate, and is allowed to heat the house at all other times whenever it likes.

    You could restrict the HP to shorter times; you will need a bigger one, and it will have to run hotter, with a slightly lower cop. This bigger HP could likely run just as efficiently, if you increased the on duration and crucially backed off the weather compensation. Monoblok HP usually can modulate down by 3:1 or so I think, hence it’s ‘ok’ to oversize them, unlike older on/off units.
    • CommentAuthorbhommels
    • CommentTimeNov 17th 2022
     
    2 rules of thumb calculations, copied from a good source, and actually 2) is more or less equal to RobL's calc above:
    1- Take annual gas consumption in kWh and divide by 3,000. This will be approximately the size of the heat pump required in kW. If on oil divide litres used in year by 300.

    2- Take gas meter readings each day during a cold period of the year. Convert to kWh used each day, divide by 24 and plot this against average outdoor temperature. Where the trend line crossing -2ºC is the heat load at the design temperature.

    (assuming average hot water usage, continuous occupation, some cooking on gas, main rooms heated to approx 21ºC with insignificant heating from wood stoves or other non-measured heat sources)
  6.  
    RobL and bhommels Thanks for that. It would seem that the modern modulating heat pump would be running most of the time and modulating down to match the heat demand - Correct?
    That being the case I will continue collecting the heat loads but in January (our coldest month) I will go to daily readings / 24 then account for the DHW and then add some for the cushion effect.

    RobL - is there a reason that you heat the DHW as a dedicated run time rather than an on demand like the CH? Is this the normal way of running a heat pump when doing CH and DHW or is this particular to your system?
  7.  
    In our case, installers followed the MCS process and/or consulted the HP mfrs, and came out with much larger sizes than I had worked out when I used roughly the methods above. In fairness, they don't make the rules, which are skewed towards ensuring nobody ends up disappointed because their HP was too small. I was willing to accept cooler rooms.

    PiH you could ask some local installers to quote, and see what they recommend? They had a pretty good idea what would be specified for our house just by looking at it.

    If you oversize the heatpump, you also need to oversize the radiators (even more), or have a thermal store, otherwise the HP can never run at its full capacity and will have to modulate even in very cold weather. Then, in mild weather it won't have any modulation left, especially with weather compensation which reduces the radiator output capacity. Put another way, the radiators and HP need to be specified as a combined system.

    Beware also that a "12kW" air source heatpump might only give 10kW when the ambient air temperature gets very low, check the fine print on the datasheet.
    • CommentAuthortony
    • CommentTimeNov 17th 2022
     
    In Canada they design heat pump to cover all but the coldest two weeks and top up with direct electric if needed. Additional capital cost outweighs the cost of the energy needed for the additional heating

    Belt and braces approach costs too much.
    • CommentAuthorRobL
    • CommentTimeNov 18th 2022
     
    Posted By: Peter_in_HungaryRobL and bhommels Thanks for that. It would seem that the modern modulating heat pump would be running most of the time and modulating down to match the heat demand - Correct?
    That being the case I will continue collecting the heat loads but in January (our coldest month) I will go to daily readings / 24 then account for the DHW and then add some for the cushion effect.

    RobL - is there a reason that you heat the DHW as a dedicated run time rather than an on demand like the CH? Is this the normal way of running a heat pump when doing CH and DHW or is this particular to your system?


    Yes, it is normal - most heatpumps installed into wet systems now are mono block ashp that can modulate down to 30% of full power.

    We run ours that way for legacy reasons - I inherited the controls and valves from the existing gas heating system. Having said that, it’s generally more efficient to heat a dhw tank just before you need it. Topping it up all day is relatively energy intensive.
    Usually a HP heats either dhw or radiators, not both simultaneously - again, you could, but that would not be as efficient, as it would likely prevent weather compensation turning down the flow temperature to the rads.
    To get best heatpump efficiency, the flow temperature out should be as low as possible.
  10.  
    This is a really comprehensive write up of the subject, well worth reading through

    https://www.energy-stats.uk/what-size-heat-pump/

    He tried all the different methods he could find to size the heatpump, and got a range of answers, some 2x as big as others..

    He was able to choose a heatpump near the top of that range, but which could modulate down.

    Looking elsewhere at his twitter, it does seem to be running in stop-start mode, suggesting that it is a bit generously sized and not able to modulate right down to match actual need, but Im sure he'll publish his analysis after the winter. Getting very good CoP. https://mobile.twitter.com/Zarch1972/status/1587733561934811136
  11.  
    Posted By: tonyIn Canada they design heat pump to cover all but the coldest two weeks and top up with direct electric if needed. Additional capital cost outweighs the cost of the energy needed for the additional heating

    Belt and braces approach costs too much.


    Exactly - but part of the reason for that is that pretty much all heatpumps here are also used for A/C in summer. If they were sized for the worst-case winter load, they would be massively oversized for A/C usage. Sizing for 90% of the worst case winter temperatures gives a cheaper unit plus smaller borehole/slinkies (if GHSP) and the saves cost. The extra top-up resistance heating required covers about 20 years of usage, versus the extra cost of sizing to 100% (or more) or the worst-case load. Of course, our electricity here is cheap and is not affected too much by fossil fuel prices either, so that helps too.

    Paul in Montreal.
    • CommentAuthortony
    • CommentTimeJan 3rd 2023
     
    Thanks Paul, I have always appreciated your input.
    • CommentAuthorborpin
    • CommentTimeJan 4th 2023 edited
     
    Calculating the right HP seems to be a bit of a dark art.

    Be very careful of your installer. So many installs are badly executed. HeatGeek has an excellent reputation.

    AIUI, a heat loss survey is vital and is the starting point.

    Posted By: WillInAberdeenHe was able to choose a heatpump near the top of that range, but which could modulate down.
    Yes. An oversize HP doesn't help as it cannot modulate down to where you need.

    Sizing for the 90-95% of the time solution, that needs topping up with direct heating for 5-10% seems to be the way forward.

    I'm looking at a DHW boost solution currently that can be fed warmed water eliminating the need for DHW cycles (I'm still on gas and a Thermal Store so the principle stands). https://www.stiebel-eltron.co.uk/en/products-solutions/dhw/instantaneous_waterheater/compact_instantaneouswaterheater/dce-s-plus/dce-s-10-12-plus.html

    The other consideration is monitoring of the HP. Design this in and make sure it has local access to data.

    The OpenEnergyMonitor community has lots of folk with experience of Heat Pumps.
  14.  
    Not sure I fully understand... (my head perhaps still fuzzy from festivities... happy new year!). Why would such a big direct electric heater (CoP=1) be helpful in a heatpump system, wouldn't it drag down the performance of the heatpump? Lots of people are reporting CoPs of 2-3 just by using their heatpump to heat their DHW, and are able to schedule that consumption for times of day when electricity is cheaper/greener.

    Is your mains connection strong enough to deliver that heater's current requirements, at the same time as running the HP/cooker/WM, or would you need to go to 3 phase?

    Completely agree that the HP shouldn't be oversized, I was bit disappointed that the installers round here had to follow the MCS process which seems to push them towards larger HPs (eg they have to size for very cold days, with very warm inside temperatures, using assumed heat losses, no supplementary heating, etc etc, then bump up to the next bigger model in the range). If actual heating measurements or energy bills are available, they cannot be used.

    That seems to be because of the risk that the customer might come back complaining their HP "isn't big enough", even if any problems actually lie somewhere else in the system. But in doing this, they introduce an unnecessary risk that the system will be oversized and run inefficiently. And all the components like radiators and pipes also have to be oversized to dissipate the heat so the costs go up. Unfortunately MCS certs are needed for all kinds of reasons.


    The user who I linked to above (Zarch, who you might know from OEM) is publishing all his monitoring data at https://heatpumpmonitor.org/ . He hasn't drawn conclusions yet afaik, but it looks to me that his HP was plenty big enough for the cold snap without top-up heating, and perhaps a little large during the milder weather, he seems very pleased with it and getting good CoPs.
    • CommentAuthortony
    • CommentTimeJan 4th 2023
     
    What are good CoPs in real numbers please?
  16.  
    https://heatpumpmonitor.org/ have a look

    Good: 3.5- 4.0 over the year (SCoP)
    Less good: 2.5- 3.5
    • CommentAuthorborpin
    • CommentTimeJan 4th 2023
     
    Posted By: WillInAberdeenWhy would such a big direct electric heater (CoP=1) be helpful in a heatpump system, wouldn't it drag down the performance of the heatpump?
    It is a question of whether the losses from the system (holding hot water that isn't used and cools) and the lower COP of heating to a higher temperature are outweighed by the COP of One for the direct heating. In a household with highly variable DHW requirements (think teenagers), the flexibility of just heating to a lower temp using the HP, might be more efficient overall.

    Also remember, all HP performance is seriously impacted by the fabric of the building.

    Posted By: WillInAberdeenusing assumed heat losses
    Assumptions make an ass out of u and me and are the mother of all foul ups.

    You need a proper heat loss survey as the starting point, anything else is just guesswork.
    • CommentAuthorRobL
    • CommentTimeJan 4th 2023
     
    There's an article linked below on different ways a heatpump is installed giving rise to different SCOP values. It also gives a graph of predicted versus actual SCOP, piccy below. Sadly, the average seems firmly in the "Less good" bracket - I think largely down to overcomplex installation leading to high heatpump flow temperature.

    I think we need legislation to force some sort of performance guarantee which works - I expect this will persuade companies to install systems with higher resulting performance. At the moment, the imperative is to install systems that don't require a call-back, so run the rads hot and add on a whizzy "smart" controller like it was a dino-gas boiler.

    I think that the user/installer shouldn't need to choose the weather compensation curve - surely this day and age this can be automated? Start on a high (hot, inefficient) weather comp slope, back off over time.
    The heatpump should have the COP clearly displayed at all times. It should include externalities; pumps, heaters etc.

    https://renewableheatinghub.co.uk/how-to-correctly-install-heat-pumps-so-that-they-work-properly-and-efficiently
      Capture.JPG
  19.  
    Agreed! But beware that data was from the older R410A era of heatpumps, the current R32/R290 machines are rather better than those were.

    Borpin, here is the MCS's heat loss survey, it's a very detailed room-by-room calculation and takes hours of the HP designer's time to complete it. But if you dig into the Excel, it is full of assumptions about as-built U values and air permeability and winter temperatures, which cannot be verified during the survey.
    https://mcscertified.com/wp-content/uploads/2021/04/MCS-Heat-Pump-Calculator-Version-1.10-locked.xlsm

    In our case, it calculated a HP about 30-40% bigger than I wanted, based on my data of the house's heat load using the existing heating system.

    Trystan Lea (he off of OEM) found his real-world heat consumption was 45% less than the survey his HP had been designed from, his investigation why is here:
    https://trystanlea.org.uk/heatpump-oneyear
    Also has some explanation on that page how he heats DHW with a CoP of 3.5 - just heating it from cold feed temperature directly up to shower temperature, rather than heating it to 60⁰+ and then blending it back down with cold water.

    It's becoming clearer that the fabric of the building doesn't prevent good ASHP performances, people in poorly insulated stone houses are getting good CoPs. What matters more is the radiator temperature, so if the house is old and leaks lots of heat, then it will benefit from bigger radiators to deliver all that heat. Or, if that's not possible/economic, then the trade-off is poorer CoP.
    • CommentAuthorborpin
    • CommentTimeJan 4th 2023
     
    Posted By: WillInAberdeenpeople in poorly insulated stone houses are getting good CoPs
    Of course you can get good COP, but does it keep the house warm. Just because you get a lower COP, that doesn't mean it actually can do the job.

    Posted By: WillInAberdeenAlso has some explanation on that page how he heats DHW with a CoP of 3.5 - just heating it from cold feed temperature directly up to shower temperature, rather than heating it to 60⁰+ and then blending it back down with cold water.
    Thing is, there are lots of other factors that must be considered. Is the shower close to the HP outlet? how quickly does it heat, what do you do for other DHW, what happens when the incoming temp drops? how hot do you like your shower? What sort of HP will be fitted?

    Just because it works there, doesn't mean it works everywhere. HP installations are very sensitive to other forces so you could even end up with 2 identical installs, in identical buildings and one is considered satisfactory by the owner and the other not. Boiler based heating / DHW is far less sensitive to outside factors and can adapt more easily (for a cost) to the occupants requirements.

    You design for a 20°C temperature. What happens if you are ill and actually need a higher average temperature? A boiler could do that (for a cost), a HP less so. The margins are very fine.
  21.  
    Posted By: WillInAberdeenBorpin, here is the MCS's heat loss survey, it's a very detailed room-by-room calculation and takes hours of the HP designer's time to complete it. But if you dig into the Excel, it is full of assumptions about as-built U values and air permeability and winter temperatures, which cannot be verified during the survey.
    https://mcscertified.com/wp-content/uploads/2021/04/MCS-Heat-Pump-Calculator-Version-1.10-locked.xlsm" rel="nofollow" >https://mcscertified.com/wp-content/uploads/2021/04/MCS-Heat-Pump-Calculator-Version-1.10-locked.xlsm

    In our case, it calculated a HP about 30-40% bigger than I wanted, based on my data of the house's heat load using the existing heating system.


    So it seems that the MCS's heat loss survey is about as useful as the typical EPC !

    Back to the original question that I asked, all the above is good info.
    I am logging the kWh used for both houses that are heated by the system along with the run times (daily duration) of the thermostat controlled CH pumps and I have reduced the flow temps to 45 deg.C.

    Unfortunately (sort of) we have an unusually warm start to 2023 with day time/night time temps of 15/6 but we are forecast some -7/-2 later in the month which is what I want to get a target heat demand.

    Reading through the above it seems that it is generally accepted that heat pumps run most of the time at a low output to get a good COP. Does this shorten the elapsed time to failure. After all if the machine is designed for X hours life then wouldn't running it twice as long / day reduce the years in service by half? against which a better COP over its service life. Is there a balance somewhere between daily run time / COP achieved / years in service.

    I like the idea of running the DHW at 45 deg. and not mixing down as a way of getting a better COP (and reducing standing losses) with presumably monthly (?) 60 deg. boost to keep the legionella happy.
    • CommentAuthorborpin
    • CommentTimeJan 4th 2023
     
    Posted By: Peter_in_Hungarywith presumably monthly (?) 60 deg. boost to keep the legionella happy.
    Only required if you are storing the water to be used. If you supply main water to a coil in the tank that heats the water, then there is no (well, virtually zero) risk so no cycle required.
  23.  
    Posted By: borpin
    Posted By: Peter_in_Hungarywith presumably monthly (?) 60 deg. boost to keep the legionella happy.
    Only required if you are storing the water to be used. If you supply main water to a coil in the tank that heats the water, then there is no (well, virtually zero) risk so no cycle required.

    My DHW is an indirect tank of 300 lts (serving 2 houses) that is heated either by an immersion in the summer or by a coil from the TS - When moving to a heat pump I would hope to use the same tank, substituting the TS input for the HP input - so anti legionella cycles presumably would be needed. (although the DHW is not mains water it is bore hole water as we are off the water grid)
    • CommentAuthorRobL
    • CommentTimeJan 5th 2023 edited
     
    Posted By: WillInAberdeenAgreed! But beware that data was from the older R410A era of heatpumps, the current R32/R290 machines are rather better than those were.


    R290(propane) was used in heatpumps back in 1922 apparently, touted back then as a non toxic safety refrigerant! They moved away from R290 to ‘freons’ as you need to take a lot more care brazing a system together when flammable gasses are involved. Now R290 is promoted as ‘more efficient’. I think the industry has twisted back and forth, and is now coming back to propane and other simple gasses, as the f gasses all have high GWP so are gradually being outlawed. R290 is higher COP, but tbh it’s a small difference - piccy below stolen from this paper (worth looking at)
    https://pubs.acs.org/doi/10.1021/acs.jced.0c00338#

    We heat our dhw to 46C by the heatpump, and the tank immersion gets it over 60 once a week straight after its got to 46 by heatpump. 46C is accepted by all in our family :bigsmile: initially it was 41C, which was deemed not ok by Mrs Robl
      C2AF7A74-CDFE-4110-AC99-54E7DCA061D0.jpeg
    • CommentAuthorborpin
    • CommentTimeJan 5th 2023
     
    Posted By: Peter_in_HungaryMy DHW is an indirect tank of 300 lts (serving 2 houses) that is heated either by an immersion in the summer or by a coil from the TS
    If the water stored is used, then yes, you need the legionella cycle.

1 to 25 of 25

  1.  
Add your comments

    Username Password
  • Format comments as
 
Back to Discussions Top of Page
   
The Ecobuilding Buzz
Site Map    |   Home    |   View Cart    |   Pressroom   |   Business   |   Links   
Logout    

© Green Building Press