ASHP manufacturer choice and sizing

Hardware Heatpump
1

Hi all, building work is just about to start on our eco retrofit, which should leave us with a low energy home (somewhere in between EnerPhit and AECB in terms of fabric heat loss and airtightness). We’ve contacted a couple of local MCS certified ASHP installers and they estimate our coldest day heating demand will be somewhere between 6 and 7kwh. One installer has recommended sizing up to a 9kw unit which we are loathe to do as we understand the MCS calcs often over-estimate the demand for low energy homes. Another is suggesting 7kw will be sufficient. We are also being offered ASHPs from a range of different manufacturers (i.e. LG, Acond etc), all using R290.

A couple of questions:

  • are we taking a risk by installing a 7kw pump?
  • should I ask for a pump from a specific manufacturer (recommendations welcome) or are all the R290 pumps similarly efficient?

Thanks in advance for any help and advice

2

Hi Louie, and a warm welcome to the community :slightly_smiling_face:.

The amount of risk depends partly on the meteorology at your house location. Your prospective installer will suggest a design minimum ambient temperature and specify the kit to achieve an adequately warm house for that temperature. There will probably be a few days per year when this design temperature is exceeded, so you have the choice of selecting a larger nameplate HP size (and risk it being oversized for more of the time, with the consequent cycling problems), or choosing a more sensible design temperature (and maybe putting on a sweater occasionally).

If you aren’t sure about your local meteorology, there’s an excellent tool to help you see how many hours per year are below any selected temperature - https://www.degreedays.net/. You simply choose a local weather station (there are hundreds available in the UK), enter a design temperature T, select a suitable time period (e.g. the last 12 months) and hit “generate”. It will create an Excel spreadsheet showing “hours below T” each day. By trying different values of T you can get a feel for how many days you might need that sweater.

This is assuming, of course, that your installer doesn’t include any margin in his calculations. If he says your heat loss will be 6-7kW at TdegC ambient, it may in reality be considerably less - in the UK the MCS database is known to be conservative in some areas, partly down to natural industry conservatism (they don’t want to risk litigation if the house isn’t warm enough) and partly down to recent technology improvements (for example some double glazing OHTCs achieved nowadays are at least 30% better than the database).

The consensus within this community seems to be to go for a somewhat lower nameplate, and minimise the cycling problems (it’s hard to avoid them entirely during lower heat demand months).

As for HP vendors, I suggest you just spend a couple of hours trawling the posts on this community - you’ll soon get a feel for which draw most criticism (or recommendations). Personally, after 18 months use I’ve been very happy with my Samsung (though it’s R32 not R290).

Sarah

1 Like
3

Use your EnerPhit calculations to detemine the real world heat losses.

Ignore the MCS heat loss calculations. They make absurd assumptions with regards ventilation losses etc.

Look at what the MINIMUM output of a heat pump is when choosing between models / sizes.

For example in the Vaillant range the 5 kW will turn down to about 2.3 kW and the 7 kW down to about 2.8 kW (air 7 deg / water 35 degC). That’s much of a muchness (the cycle times when the unit can no longer turn down below what the house / radiators demand will be long enough not to knacker the heat pump / not to cause uncomfortable temperature swings.) They also run efficiently at low flow temperatures so you an have radiators ticking over lukewarm in the shoulder season.

If an installer suggests a 9 kW unit from the Daikin range then be aware this is actually a 16 kW unit wearing a different hat and will struggle to run at below 4 kW. That might well either cycle too frequently or cause uncomfortable temperature swings when the house doesn’t need much heat.

Look at the controls on offer. Are you running radiators or underfloor? Does the unit weather compensate / load compensate well? Do you like the local user interface which you will be relying on long after the vendor pulls the plug on the app due to the equipment begin 20 years old and too insecure to leave connected to the internet.

Are there units in stock? When it goes bang can you get parts from various sources or are you stuck getting parts direct from the OEM only? Are heat pumps THE business that the OEM is in and will struggle to exit or are the heat pumps an experiment that might be abandoned? (I perceive this as a risk with something like the Octopus unit)

Is the outdoor unit quiet? Does it look like vomit? Can you find somebody to install it where you want it or will they whinge about R290 and windows/doors? Is it small enough to be permitted development if installed by an MCS union member?

And given the relatively low heat demand of the property; can you justify the capex uplift some something like a Viessmann vs a more mainstream unit with marginally lower performance but at half the cost?

Top of heatpumpmonitor might be Nibe/Viessmann; and most of the upper parts of heatpumpmonitor is full or Arotherms; but there are also a couple of well installed Grant and Samsung units in there that are at a lower price point than even the Arotherms.

5 Likes
4

This seems a little on the high side for a “low energy home”, though you’ve not said how big it is. MCS installers tend to assume the air change rate to be higher that it really is, so watch for that. Recommend finding a HeatGeek if you can.

There is an argument for deliberately undersizing the heat pump, as it’s usually only very cold for a few days each winter → Deliberately under-sized heat pumps. | Protons for Breakfast

5

@Timbones Hi Tim, the finished home will be roughly 200m2.

Thankyou everyone - really useful advice and definitely food for thought. Makes me tempted to go for something more ‘mainstream’ like a Vaillant.

1 Like
6

I chose to go for a Vaillant Arotherm Plus, for the following reasons:

  • Excellent UK service, support and parts network. This is miles ahead of all other manufacturers in my experience, and counts for a lot. Vaillant pick up the phone, parts and engineers are readily available.
  • Controls: the Sensocomfort unit looks inoffensive enough and is straightforward to use. The myVaillant app is excellent, better than most, although does rely on the cloud.
  • R290 with high flow temp capability meaning zero reliance on immersion for DHW/legionella cycle
  • Big UK & European consumer install base, meaning loads of dorks online figuring out monitoring & control integration (ebusd, myPyllant etc), optimal running configurations, troubleshooting etc.
  • Robust build quality with stainless fasteners, inoffensive fan & compressor noise profile etc

Do not ‘size up’ your unit, whichever brand you end up choosing. Avoid a buffer vessel and any zoning or third-party controls/thermostats.

1 Like
7

Is somewhere between EnerPhit (25 kWh/m2.year) and AECB (50 kWh/m2.year) around 40 kWh/m2.year? So maybe 8000 kWh/year space heating demand?

My space heating demand is ~7000 kWh/year and my accurate heat loss / demand is 3.3 kW. There are a few systems on HeatpumpMonitor.org with a space heating demand of 8000 kWh/year and these have a measured heat demand on the coldest day including hot water of just over 4 kW…

Do you have a heat loss calculation of the house that you could share with us?

8

+1 for Vaillant, especially if your not totally confident in what the heat loss will be, there does not seem to be much penalty for performance if you have a vaillant arotherm that’s a little bit too big. The 5kW might be fine as it can actually output over 6kW in practice. The 7kW (~8.5 kW in practice) is probably a bit big but does it matter? (see Is an oversized Vaillant a problem?).

1 Like
9

@TrystanLea thanks Trystan, kwh/m2 is forecast to be somewhere around 35 but that number is not fixed and could well move upwards due to budget pressure in terms of things like window choice. So the 7kw sounds like a perfect fit.

Really appreciate the advice, will start contacting some Vaillant installers today.

1 Like
10

@maxysm dont rule out the 5 kW, it can go up to 6.4 kW at -3C, and if your actual head demand ends up closer to 3-5 kW that’s should be plenty. The 5 kW is also a little cheaper and smaller…

1 Like
11

Complete your own heat loss calculations. Heat Punk is very good. Setting air changes/ hour is critical. The deafult value will probably be too high if you have modern windows and doors.

1 Like
12

I swat the budget constraint notes. Some unsolicited off topic advice from lessons learned by making mistakes:

Splurge on the windows. They’re the weakest link. I regret some life choices in this regard even if at the time of our purchase (peak Biden inflation stimulus in 2021) it was “rubbish windows in 3 months or nothing” I think I would have chosen to wait…

A window with the glass at say 0.7 W/m2K but an overall window U-value of say 1 W/m2K will be sad - in order for the small amount of frame to tank the overall U-value that much the frame will be terrible.

Their U-values are also measured/calculated without any pressure difference. As soon as you get some wind or some stack effect you have a pressure difference and will draw air through the seals. Once air is passing the seals the effective U-value of the window might as well not be there.

The outer seal is for “worst of the rain” not “air”. The first chambers of the profile are for “draining water” and are open to outside air / will be at outside air temperature due to drain holes.

The second seal (and there may only be two in a profile such as Rehau Euro 70) is for the air. If you have a tilt and turn window then the mechanism goes straight through this seal and there is nothing there.

I’m buttoning up our windows/checking for leaks before an airtightness test (was about -2C when photos taken) and you can see what a bad frame looks like:

FLIR_20241123_112823_559
FLIR_20241123_112823_5591080×1440 69.4 KB

Glass is at about +17C with air -2C (so glass is not generating any horrible convection currents - not the air leak visible at the top and right tho)

FLIR_20241123_112835_957
FLIR_20241123_112835_9571080×1440 81 KB

Frame is 13C (so guess where you will get condensation first if the humidity is high enough - I will be using ply for the window reveals not plasterboard and leaving a small gap to the frame

FLIR_20241123_113025_502
FLIR_20241123_113025_5021080×1440 78.7 KB

FLIR_20241124_090942_471
FLIR_20241124_090942_4711080×1440 91 KB

Still finding the odd leak - again something like a screwhole not filled in or a tongue and groove not completely bubble glued is enough to draw cold air in through all the insulation and drop the temperature very locally to the point you’d be neat the dew point.

e.g. If Rehau cheapo UPVC; ignore the Euro70 profile. The Synego 80 is acceptable in terms of U-value / notional airtnightness but has the structural integrity of wet spaghetti. The Geneo 86 is the one to have - same thermal performance but is fibre reinforced so the nothing droops/binds in the larger sizes.

https://www.rehau.com/us-en/geneo-passive-house-certified

RRP for say 900 x 2100 mm tilt/turn with warm edge triple glaze 373 vs 462€ - very little - with actual price say €220. Not. Worth. Skimping. When you can have passive house certified windows for very little.

Screenshot 2024-11-26 at 12.10.05
Screenshot 2024-11-26 at 12.10.051646×2456 191 KB

I would also spluge €30 nominal (€16? after discount) and go with a hidden tilt/turn hinge for (a) keeping it INSIDE the air seal and (b) giving more clearance between the window and the window reveal when it opens.

You may also want to learn how to measure / fit windows yourself. The markups applied by people who are doing door to door sales; to cover their backsides when they order the wrong sizes etc; are pretty damn huge. Possibly the largest you can make in a retrofit and even if you mess up…half the windows…you’ll still save overall and will have an amazing set of windows for the garden shed later lol

https://www.youtube.com/playlist?list=PLlEDeDWZazY2G-T8SZ974gdzeoeQeyMs0

Don’t do low threshold doors. They just don’t seal properly. We don’t need to keep the thresh indoors any more but you still want the water to be kept out…

2 Likes