Electric+ASHP Underfloor Heating in passivhaus bathrooms - madness? =)

[markocosic]

19 hours ago, Gus Potter said:

The intrinsic beauty of UFH is it's simplicity.

 

With a slab temperature controlled by a weather-compensated / load-compensated heat pump?

 

(avoiding all those room stats and the associated zone valves)

[JohnMo]

20 minutes ago, markocosic said:

load-compensated heat pump

Don't bother to try using load compensation with UFH, been there and tried, just makes the temperature of the room yoyo. Ok for radiators only.

[markocosic]

Useful to know @JohnMo

 

Return temperature based compensation perhaps? (shove the water in hot enough that it comes back at X degC / poor man's slab temperature sensor)

[JohnMo]

14 minutes ago, markocosic said:

Useful to know @JohnMo

 

Return temperature based compensation perhaps? (shove the water in hot enough that it comes back at X degC / poor man's slab temperature sensor)

Just set up the weather compensation curve and balance the loops - job done. You don't need anything complex. If well insulated, floor temp is circa 1 to 3 (OAT dependant) hotter than room target temp, if the sun comes out heats the room the floor reduces or stops giving off heat.

 

There is no need for floor sensors etc, one thermostat per floor (operating as a limit stop instead of a controller, if you feel the need, or just one wireless thermostat, so you fine tune the position in the house.

 

The same method can be used for 24/7 WC or batch heating in the cheap hours of E7 (if you have it) or during the daytime when some solar may be available.  Just modify the WC curve to run 5 to 10 deg hotter, so it just the same amount of heat to the floor in 6 to 7 hours instead of 24.

[joth]

22 hours ago, Dan F said:

 

Technically it's a maximum heating load of 10W/m2 or max heating demand of 15Kw/m2.yr.  The heating load is typically always <12W/m2 though.

 

You have to allow some headroom though IMO because:

i) house might not perform as well as PHPP says

ii) outdoor temperature used for PHPP is quite modest.  In our case (SE England) it's only -1.6C and we've just had a winter where it was much colder than that for multiple days.

 

The next Ecodan down is likely 5kW which could well be too small for a large house if you were to allow for e.g. max 20W/m2. 

 

 

Yes. Also the ASHP maybe sized for DHW reheat time, with a eye to using it at a lower capacity for space heating. Also it may be designed to allow the daily heating load to be supplied in less than 24 hour block, e.g. allow for the DHW duty cycle, allow for boosting space heating during cheap rate energy, or simply not wanting the heat pump running 24/7 in an urban setting 

So many variables hard to categorically state one is too big. 

[puntloos]

23 hours ago, joth said:

Yes. Also the ASHP maybe sized for DHW reheat time, with a eye to using it at a lower capacity for space heating. Also it may be designed to allow the daily heating load to be supplied in less than 24 hour block, e.g. allow for the DHW duty cycle, allow for boosting space heating during cheap rate energy, or simply not wanting the heat pump running 24/7 in an urban setting 

So many variables hard to categorically state one is too big. 

Yep, Joth listed most arguments I heard myself from my installer, indeed I have a large DWH(400L), the house is about 290sqm (but with high ceilings, 2760mm, so more volume to heat). The total calculated heat loss for the house is 3000W. I think the 6kw device was considered but was seen as a little anemic. Not to mention that price differences between heat pumps isn't huge, as long as the device can throttle down enough without short cycling.

 

 

[puntloos]

On 07/07/2023 at 21:59, markocosic said:

 

Mmmhmmm

 

They're using an ecodan too.

 

So not the most performance oriented outfit.

 

https://heatpumpmonitor.org/

 

I'm not sure I get your point, the highest SCOP of the list from that site is ecodan based, and indeed, the ecodan had the highest (claimed...) SCOP of any device I looked at, and the features I wanted:

- High SCOP

- R32 or better coolant

- Low operating noise

 

On 07/07/2023 at 21:59, markocosic said:

Or ecological outfit (it's an fgas unit and not even a particularly low gwp fgas)

Well, 99% of modern domestic ASHPs are R32, I didn't accept R410 so that knocked a few contender out, but for example R744 aka CO2 is theoretically the 'nicer' (and safer) gas however it isn't as efficient - so the carbon cost to running a device based on it could well outweigh the carbon and climate upsides of the 'nicer' gas. It's too esotheric for me to work out in detail but it's certainly not clear to me as an amateur.

 

On 07/07/2023 at 21:59, markocosic said:

I'd ask for a 2nd opinion. Most of those ecodans appear to get sold into use cases where the Mitsubishi rep has done all the homework) badly) for some eco grant funding application etc.

 

It wouldn't be the first choice unit of building a passive house where performance and low modulation etc are important.

It's worth a check but frankly I'm pretty far down this path with Mitsubishi and you're literally the first person I've ever heard bring mitsubishi up as a negative  

[Dan F]

23 minutes ago, puntloos said:

 

I'm not sure I get your point, the highest SCOP of the list from that site is ecodan based, and indeed, the ecodan had the highest (claimed...) SCOP of any device I looked at, and the features I wanted:

- High SCOP

- R32 or better coolant

- Low operating noise

 

Well, 99% of modern domestic ASHPs are R32, I didn't accept R410 so that knocked a few contender out, but for example R744 aka CO2 is theoretically the 'nicer' (and safer) gas however it isn't as efficient - so the carbon cost to running a device based on it could well outweigh the carbon and climate upsides of the 'nicer' gas. It's too esotheric for me to work out in detail but it's certainly not clear to me as an amateur.

 

It's worth a check but frankly I'm pretty far down this path with Mitsubishi and you're literally the first person I've ever heard bring mitsubishi up as a negative  

 

We used Vailant because they use R290 (lower GWP than R32), they can heat to 70C if needed, has good SCOP and low noise,  and because their 7kW unit was the right size for us while allowing some headroom (PHPP heat loss also just over 3000W). Vaillant modulates down to 30%.

 

That said, I haven't heard anything bad about Ecodan, this was our second choice and I know various others have Ecodan.  https://heatpumpmonitor.org/ is an interesting resource, but there is so much that contributes to efficiency beyond just the ASHP itself that I'm not sure it's a reliable source for deciding which manufacturers are efficient personally. @joth How is yours performing and what is its minimum modulation?

 

Edited July 9, 2023 by Dan F

[JohnMo]

27 minutes ago, puntloos said:

total calculated heat loss for the house is 3000W. I think the 6kw device was considered but was seen as a little anemic.

The problem isn't at or worse conditions than the design point, its if heat is needed in the shoulder seasons with you only need 15-25% of the design heat load. The oversized heat pump just can't modulate low enough. So you need a buffer to compensate and unless really big, they are designed (or not really designed) and you get a hit in CoP due to the mixing of the flow and return or a flow mismatch across the heat pump and heating circuit side.

 

If your spending a bucket of cash on a heat pump, get one that can cool also, otherwise you are missing a trick. ASHP plus solar equals a cool house in summer with next to zero running costs.

[Dan F]

1 hour ago, JohnMo said:

The problem isn't at or worse conditions than the design point, its if heat is needed in the shoulder seasons with you only need 15-25% of the design heat load. The oversized heat pump just can't modulate low enough. So you need a buffer to compensate and unless really big, they are designed (or not really designed) and you get a hit in CoP due to the mixing of the flow and return or a flow mismatch across the heat pump and heating circuit side.

 

You have to be careful though because a smaller kW, doesn't necessarily mean it will module any lower!  The Ecodan (PUZ-WM) is actually a good example of this as both Ecodan 6kW and 8.5kW seem to have the same minimum outputs.

 

[puntloos]

Also my house's thermal mass is pretty huge (it's block and beam ground and first floor) so while modulating can be a bit of an issue, I would imagine the cycles are not too frequent so a few "on and off" is not the end of the world? 

[Dan F]

10 hours ago, puntloos said:

Also my house's thermal mass is pretty huge (it's block and beam ground and first floor) so while modulating can be a bit of an issue, I would imagine the cycles are not too frequent so a few "on and off" is not the end of the world? 

It's also about the volume of water in the system.  If you aren't using per-room zoning you'll be OK, if you are using zoning need to ask i) why? ii) what size buffer do you need to minimize cycling.

 

UFH aside,  where you will almost certainly need a buffer is for using fancoils, especially if you have per-room control and they may be used without UFH.

[crooksey]

On 06/07/2023 at 15:35, Nickfromwales said:

In summer this will be nuisance heat anywhere other than the bathrooms, so for "chill removal" with tiled floors outside the heating season it's electric or nowt.

 

A well constructed house with proper floor insulation should never need "chill removal" even on tiled floors. Works fine for every job I have been on and had zero issues.

[Nickfromwales]

1 hour ago, crooksey said:

 

A well constructed house with proper floor insulation should never need "chill removal" even on tiled floors. Works fine for every job I have been on and had zero issues.

You can have a house acclimatise to the ambient, for a number of days, and the tiles will still feel cold(er) to human skin. The issue is different materials hold heat energy differently, which is why when you sit on a block of EPS during the tea break your arse warms up near instantly. Do the same on a stack of bricks and the heat from said arse will be absorbed and dissipated.

No internal floor should need insulation as it's already within the heated envelope, ergo an ambient will be achieved over (x) time, but this is why folk say to install LVT vs tiles, as it's the laws of physics at play here and not how well you've built.  

[puntloos]

16 hours ago, Dan F said:

 

We used Vailant because they use R290 (lower GWP than R32), they can heat to 70C if needed, has good SCOP and low noise,  and because their 7kW unit was the right size for us while allowing some headroom (PHPP heat loss also just over 3000W). Vaillant modulates down to 30%.

We were debating this and a similar Viessmann unit, R290 is pretty good but as I understand it slightly less good SCOP than R32, so you could argue over the entire lifecycle of the unit it contributes 'about the same' to the GWarming.. 

1 hour ago, Dan F said:

It's also about the volume of water in the system.  If you aren't using per-room zoning you'll be OK, if you are using zoning need to ask i) why? ii) what size buffer do you need to minimize cycling.

Interesting point, I'll verify, i think there's some zoning going on.

1 hour ago, Dan F said:

UFH aside,  where you will almost certainly need a buffer is for using fancoils, especially if you have per-room control and they may be used without UFH.

No fancoils - at least, no fancoils attached to the water unit (we have a A2A R32 device)

[markocosic]

On 09/07/2023 at 23:58, puntloos said:

you're literally the first person I've ever heard bring mitsubishi up as a negative  

 

The negative is more "the people who usually base designs on an Ecodan" - it's rarer to see one installed correctly / installed alongside a heating system / heating controls that allow it to operate correctly than it is other brands.

 

Loads with 4-pipe buffers and elevated heat pump flow temperatures for a given system flow temperature. Loads with zoned UFH systems with air thermostats that ensure short cycling etc. 

 

There are some good ones but more often than not the installs are nasty. 

 

e.g. on heatpumpmonitor:

 

- The top ecodan listing is space heat only (no hot water use that reduced the COP) into an underfloor heated old building. So it's sitting there pretty much all year (not just in the cold season) chucking in heat at the lowest possible temperature. A passive house will have a shorter space heating season and lots more hot water use. 

- The second one looks good but you'll not that it uses solar thermal for DHW so is again not particularly representative

 

 

Compare with say the Telford/Banbury/Oxford projects that are R290 Vaillant units actually providing DHW and running basic weather comp into rads or UFH etc.

 

Or look at "which are in the top 20" and "which are in the bottom 20" I think illustrates the "hmmm...the person doing the design likes Mitsi...they're more likely to be useless than useful based on the real world results observed..."

 

 

 

Your space heat loads will be "who cares" for a passive house. Small and a very short season. You do want to ensure that the heat pump can actually run long cycles though; which isn't a certainty if you're introducing zoning and effectively "decoupling" the heat pump from the thermal mass of the slab.

 

Your hot water loads will be more important. My understanding is that the published sCOPs don't include hot water generation. The performance of the units in (low temp) space heating mode will mask their performance at higher temperatures during hot water production that, in the real world, is more important to you with a passive house.

[Dan F]

38 minutes ago, markocosic said:

Loads with 4-pipe buffers

I think a lot of installs, including Vaillant, use 4-pipe buffers. I know ours does.  The Vaillant branded buffer is 4-port and their schematics show a 4-port configuration too.

[Dan F]

1 hour ago, markocosic said:

The negative is more "the people who usually base designs on an Ecodan" - it's rarer to see one installed correctly / installed alongside a heating system / heating controls that allow it to operate correctly than it is other brands.

 

If we assume that @puntloos chose his M&E consultant/installer based on their experience and credentials. If this is the case, and his system is designed correctly, then there is nothing wrong with using Ecodan (unless you are concerned about lower GWP).

 

If as you suggest, most Ecodan systems are badly designed or installed, then this would explain why they do poorly on heatpumpmonitor.org.  But I'm not sure why you say that Ecodans are not performance oriented.  That's my take anyway.

[markocosic]

My take on the Ecodan:

 

 

Ecodans do fine on the "euro tests" for sCOP. The euro tests for sCOP test the unit at a few fixed points that are representative of perfectly weather compensated heating of an old-build in one of three climate zones.

 

This means:

 

- They have a "long and low" space heating season (heating well into the shoulder seasons and relatively more of this vs heating during the coldest snaps)

- They don't capture domestic hot water production

- They don't capture how well the built in controls on the unit do, or don't, work (the tests are at operating points fixed by the test rig)

 

It isn't the highest performance unit on the market. And it's also an FGas based unit. High temperature capabilities are limited.

 

 

In the real world you see a very mixed bag on the results.

 

The majority are fairly poor. (if we believe heatpumpmonitor results) This could be because the people who typically choose Ecodans don't know how to install and operate them. This could also be because the controls on the units are poor and they don't operate anywhere near the lab test conditions in real life.

 

 

If you dig into the results, the top ecodan is...delivering space heating ONLY (so no high temperature operation to produce domestic hot water) in an old build (so extended heating season) with UFH (so in effect it never operates hot). It comes close to the test results.

 

The second ecodan...essentially uses uses solar thermal for the domestic hot water and tops out at 38C at design condition in an old build property. So again t's doing nothing hot.

 

The third ecodan...is more normal (providing heating and lukewarm water - 42C) but read further and it's operated by somebody who gave up with the OEM controls being being excrement and wound up rolling their own to try get the thing to perform.

 

Only by the time you get to the fourth ecodan do you see something "normal" but again I'd caution that it's an old build with extended space heating.

 

The rest get progressively worse; with Ecodans occupying too many of the bottom rungs on the list for there to be no correlation between the unit (or the people who choose this unit) and performance.

 

 

You have a passive house. The heating season will be MUCH shorter and harder (there won't be the long shoulders of low grade heat and relatively high ambient temperatures) and hot water will be a MUCH higher proportion of overall demand.

 

You will want to pay a disproportionate amount of attention to the high flow temperature DHW production performance of the units. That will likely favour the newer R290 based units over R32 units. 

 

You'll want to be MUCH more careful about designing an overall system (both heat pump itself and the heating system controls that couple this to the thermal storage of your pipes/floors - no stupid hysteresis stats on 101 zones etc) so that it can operate at part-load.

 

I don't think an Ecodan ticks those boxes particularly well. It's long in the tooth. Naff OEM controls (read through the trials and tribulations of the openenergymonitor folks trying to control these) and what looks  like narrower operating window (look at the minimum flowrates etc required) compared with other units. 

 

 

The Vaillant units aren't anything to write home about on paper. They look like they don't work as well as an Ecodan. Yet they're disproportionately represented on the upper part of the heatpumpmonitor charts. Probably not a coincidence. Perhaps due to an "ok" unit with superior control when it comes to heating. Perhaps due to those designing these installations being better at it.

 

 

Performacne oriented units?

 

 

Perhaps Nibe S2125? Rated with a sCOP of 5.0 on the average climate and 6.4 on the warm climate for underfloor systems (35C peak flow temp) on the euro test cycle.

 

https://www.nibe.eu/en-gb/products/heat-pumps/air-source-heat-pumps/s2125

https://assetstore.nibe.se/hcms/v2.3/entity/document/319563/storage/MzE5NTYzLzAvbWFzdGVy

 

(The south of the UK is warm climate. The north of the UK is average climate.)

 

 

Don't get confused by the nameplates. 8 kW nameplate on Nibe is about the same at 5 kW on Vaillant. (Nibe quote the "max" heat output in warm conditions; Vaillant quote the "design" heat output at the cold/design condition; both are similar)

 

 

Perhaps a Lambda unit if you're feeling brave enough to try a low volume unit?

 

https://lambda-wp.at/luft/

https://zewotherm.com/de/produkte/waermepumpen/zewo-waermepumpe-lambda.html

 

Rated at sCOP 5.7 on the 35 degree *average* climate cycle. Similarly impressive sCOP on the 55 cycle infers decent DHW performance too.

 

Quiet courtesy of both the sheer size and blowing through the heat exchanger to muffle fan noise rather than sucking through it and giving you direct fan noise. (at the expense of being somewhat more difficult to clean)

 

Three options for PV diversion

1) run at a set output when digital contact bridged 

2) measure the actual excess export and run at as close to this as possible using a meter

3) get told what to run by an external device using a modbus input

 

Also tall to avoid getting buried in snow; buried in cold lakes of air; and designed to thermosyphon from the house (needs heat pump higher than the house) in the case of a power cut etc.

 

 

Relative to this kind of thing the Ecodans aren't a performance oriented unit. They're so so. Robust but not lighting anybody's pants on fire. And in practice most appear to be towards the bottom of the league tables in real world performance vs other similarly "so so priced" units. I'd at the very least want to know why my designer were picking this over other units and what they do differently to ensure that their designs aren't at the bottom of the league tables in terms of performance. It doesn't appear to be a given that these perform well.

[puntloos]

9 hours ago, markocosic said:

My take on the Ecodan:

 

 

Ecodans do fine on the "euro tests" for sCOP. The euro tests for sCOP test the unit at a few fixed points that are representative of perfectly weather compensated heating of an old-build in one of three climate zones.

 

This means:

 

- They have a "long and low" space heating season (heating well into the shoulder seasons and relatively more of this vs heating during the coldest snaps)

- They don't capture domestic hot water production

- They don't capture how well the built in controls on the unit do, or don't, work (the tests are at operating points fixed by the test rig)

 

It isn't the highest performance unit on the market. And it's also an FGas based unit. High temperature capabilities are limited.

 

 

In the real world you see a very mixed bag on the results.

 

The majority are fairly poor. (if we believe heatpumpmonitor results) This could be because the people who typically choose Ecodans don't know how to install and operate them. This could also be because the controls on the units are poor and they don't operate anywhere near the lab test conditions in real life.

 

 

If you dig into the results, the top ecodan is...delivering space heating ONLY (so no high temperature operation to produce domestic hot water) in an old build (so extended heating season) with UFH (so in effect it never operates hot). It comes close to the test results.

 

The second ecodan...essentially uses uses solar thermal for the domestic hot water and tops out at 38C at design condition in an old build property. So again t's doing nothing hot.

 

The third ecodan...is more normal (providing heating and lukewarm water - 42C) but read further and it's operated by somebody who gave up with the OEM controls being being excrement and wound up rolling their own to try get the thing to perform.

 

Only by the time you get to the fourth ecodan do you see something "normal" but again I'd caution that it's an old build with extended space heating.

 

The rest get progressively worse; with Ecodans occupying too many of the bottom rungs on the list for there to be no correlation between the unit (or the people who choose this unit) and performance.

 

 

You have a passive house. The heating season will be MUCH shorter and harder (there won't be the long shoulders of low grade heat and relatively high ambient temperatures) and hot water will be a MUCH higher proportion of overall demand.

 

You will want to pay a disproportionate amount of attention to the high flow temperature DHW production performance of the units. That will likely favour the newer R290 based units over R32 units. 

 

You'll want to be MUCH more careful about designing an overall system (both heat pump itself and the heating system controls that couple this to the thermal storage of your pipes/floors - no stupid hysteresis stats on 101 zones etc) so that it can operate at part-load.

 

I don't think an Ecodan ticks those boxes particularly well. It's long in the tooth. Naff OEM controls (read through the trials and tribulations of the openenergymonitor folks trying to control these) and what looks  like narrower operating window (look at the minimum flowrates etc required) compared with other units. 

 

 

The Vaillant units aren't anything to write home about on paper. They look like they don't work as well as an Ecodan. Yet they're disproportionately represented on the upper part of the heatpumpmonitor charts. Probably not a coincidence. Perhaps due to an "ok" unit with superior control when it comes to heating. Perhaps due to those designing these installations being better at it.

 

 

Performacne oriented units?

 

 

Perhaps Nibe S2125? Rated with a sCOP of 5.0 on the average climate and 6.4 on the warm climate for underfloor systems (35C peak flow temp) on the euro test cycle.

 

https://www.nibe.eu/en-gb/products/heat-pumps/air-source-heat-pumps/s2125

https://assetstore.nibe.se/hcms/v2.3/entity/document/319563/storage/MzE5NTYzLzAvbWFzdGVy

 

(The south of the UK is warm climate. The north of the UK is average climate.)

 

 

Don't get confused by the nameplates. 8 kW nameplate on Nibe is about the same at 5 kW on Vaillant. (Nibe quote the "max" heat output in warm conditions; Vaillant quote the "design" heat output at the cold/design condition; both are similar)

 

 

Perhaps a Lambda unit if you're feeling brave enough to try a low volume unit?

 

https://lambda-wp.at/luft/

https://zewotherm.com/de/produkte/waermepumpen/zewo-waermepumpe-lambda.html

 

Rated at sCOP 5.7 on the 35 degree *average* climate cycle. Similarly impressive sCOP on the 55 cycle infers decent DHW performance too.

 

Quiet courtesy of both the sheer size and blowing through the heat exchanger to muffle fan noise rather than sucking through it and giving you direct fan noise. (at the expense of being somewhat more difficult to clean)

 

Three options for PV diversion

1) run at a set output when digital contact bridged 

2) measure the actual excess export and run at as close to this as possible using a meter

3) get told what to run by an external device using a modbus input

 

Also tall to avoid getting buried in snow; buried in cold lakes of air; and designed to thermosyphon from the house (needs heat pump higher than the house) in the case of a power cut etc.

 

 

Relative to this kind of thing the Ecodans aren't a performance oriented unit. They're so so. Robust but not lighting anybody's pants on fire. And in practice most appear to be towards the bottom of the league tables in real world performance vs other similarly "so so priced" units. I'd at the very least want to know why my designer were picking this over other units and what they do differently to ensure that their designs aren't at the bottom of the league tables in terms of performance. It doesn't appear to be a given that these perform well.

That's.. quite an unexpected amount of information  - but thank you, it's worth thinking about and I might be able to still go for Nibe if I wanted to (my installer carries the brand, but it does make me wonder why they didn't suggest it.. )

 

Your point about 'mostly hot water' is interesting, I think it is probably true that a well-insulated house needs fairly little heating but a "normal" amount of hot water so optimizing for DHW is not a crazy idea. And yes while I don't need 75C (does anyone heat their DHW to 75?) clearly a coolant that can reach those levels is likely to be more efficient near 60-ish?

 

R290 is interesting. I don't think R32 is too bad as things go, R290 is better in theory but don't know about externalities, maybe it is harder to process (and therefore costs energy?) But I seem to recall there were a bunch of downsides to R290 once you have it too. How does it perform in bad cold?

 

And yes I'm in south-ish england. Hertfordshire. It would probably qualify as a 

[puntloos]

... warm climate, if not now, then soon (yayy climate change)

 

R290 stuff is a head scratcher. Quoted COP and SCOP vary quite a bit. I don't disagree in theory having a gas that handles high temps better might be a good idea given the focus on DHW but the numbers don't seem to bear it out as far as I can tell. For one most R290 device manufacturers list 55C performance, not 65ish that you would set it to.. But just FYI some last minute research:

 

Mitsubishi R32 Ecodan [source]

- SCOP 55: 3.48

- SCOP 35: 4.84

- COP A-7/W35: 3.20

- COP A7/W35: 4.80

 

Valliant R290 Arotherm Plus [source]

- COP A-7/W35: 2.80

- COP A7/W35: 4.80

 

Nibe R290 S2125 [source]

- SCOP 55: 3.80

- SCOP 35: 5.0

- COP A7/W35: 5

 

End of the day I don't think I disagree with @markocosic that Ecodan is 'OK, not great' but Nibe is great in many things including size. Plus, for a passivhaus, according to https://heatpumpmonitor.org/ - the Derby 315m2 passivhaus halfway down the list is similar to mine in size, and they probably pay ~300 GBP per year (900 kwh). They have a ground source pump so a little apples oranges, but the SCOP they reach is 3.7, while if I compare with the SCOP of the lower ecodans (3.1) I'd save 50 quid a year (3.1->3.7) .. not worth too much headache.

 

 

Edited July 17, 2023 by puntloos

[markocosic]

The issue with R290 is flammability/explosion risk.

 

Can't safely use vast quantities of it indoors. Which is most VRF/VRV air conditioning systems. So the vendors aren't geared up for using it.

 

All are headed in that direction for small packaged units now though.

 

It'll be the units rather than the gases that are optimised for heating operation.

 

 

Suspect the difference between the Nibe unit and the Vaillant unit is stuff. 115 kg vs 165 kg of stuff. Some will be the case. The rest will be larger heat exchangers with smaller temperature drops that bump COP. 

 

 

When it comes to pure £s though it probably isn't worth spending on a premium unit. £1000 over 20 years saving? Meh. Hence the proliferation in so so units that if you're purely £ driven will work out annoyingly more cost effective. Assuming no change in energy costs that is.

 

 

I haven't seen good explanations for why the ecodans that are monitored do so badly (typically) than notionally lesser performing units by vaillant though. And that's what bugs me about them.

[markocosic]

10 hours ago, puntloos said:

does make me wonder why they didn't suggest it..

 

Price. Not cheap to buy. Like, really not cheap by a factor of almost two. Be you deciding everything on £?

 

Margin and credit terms. Mitsubishi offer good discounts to the trade and extended credit. Be these more important to the installer than performance?

 

Newness? Not willing to fit the S2125 until it's been in use a few years? (same reason to never buy first year production of any new car e.g. early S2125 had some exploding inverters)

 

Ask the question - I'd be interested in the answer!

[Dan F]

6 hours ago, puntloos said:

... warm climate, if not now, then soon (yayy climate change)

 

R290 stuff is a head scratcher. Quoted COP and SCOP vary quite a bit. I don't disagree in theory having a gas that handles high temps better might be a good idea given the focus on DHW but the numbers don't seem to bear it out as far as I can tell. For one most R290 device manufacturers list 55C performance, not 65ish that you would set it to.. But just FYI some last minute research:

 

Mitsubishi R32 Ecodan [source]

- SCOP 55: 3.48

- SCOP 35: 4.84

- COP A-7/W35: 3.20

- COP A7/W35: 4.80

 

Valliant R290 Arotherm Plus [source]

- COP A-7/W35: 2.80

- COP A7/W35: 4.80

 

Nibe R290 S2125 [source]

- SCOP 55: 3.80

- SCOP 35: 5.0

- COP A7/W35: 5

 

End of the day I don't think I disagree with @markocosic that Ecodan is 'OK, not great' but Nibe is great in many things including size. Plus, for a passivhaus, according to https://heatpumpmonitor.org/ - the Derby 315m2 passivhaus halfway down the list is similar to mine in size, and they probably pay ~300 GBP per year (900 kwh). They have a ground source pump so a little apples oranges, but the SCOP they reach is 3.7, while if I compare with the SCOP of the lower ecodans (3.1) I'd save 50 quid a year (3.1->3.7) .. not worth too much headache.

 

 

 

Vaillant quote following SCOP for 55C:  7kW - 3.39,  10kW - 3.58. 

 

Few points on DHW:

- When you look at temperatures, you need to consider ASHP flow temp will need to be 60-62C to heat the tank to 55C. So check what the max temp on non-R290 models is.

- If you run a Legionella cycle weekly, the R290 AHSPs won't use immersion for this.  They can heat the tank to 65C with COP of, at worse, around 1.8.

 

SCOP and COP vary because they are two different things.    The COP of all three is roughly the same (with Nibe a touch better).  Given how close they are though I'd be looking for the best controls including weather compensation (which can have a significant impact on performance), the best integration with Loxone and support for cooling and price. 

[Dan F]

11 hours ago, puntloos said:

Your point about 'mostly hot water' is interesting, I think it is probably true that a well-insulated house needs fairly little heating but a "normal" amount of hot water so optimizing for DHW is not a crazy idea. And yes while I don't need 75C (does anyone heat their DHW to 75?) clearly a coolant that can reach those levels is likely to be more efficient near 60-ish?

I don't know that it does make them more efficient at 60C to be fair.  Also, as I said elsewhere 75C doesn't mean DHW at 75C.  My aroTherm gets up to around 73C, so the max DHW temperature is really not much more than 65C.  We sometimes heat the tank to 65C if we know DHW is going to get a lot of use, and it makes the 300L UVC go further.

 

11 hours ago, puntloos said:

 

R290 is interesting. I don't think R32 is too bad as things go, R290 is better in theory but don't know about externalities, maybe it is harder to process (and therefore costs energy?) But I seem to recall there were a bunch of downsides to R290 once you have it too. How does it perform in bad cold?

R290 doesn't have any specific issue with the cold, it does have some additional installation guidance regarding clearances.

Edited July 17, 2023 by Dan F