Fabric and ventilation heat loss calculator

[TerryE]

On 17/01/2022 at 16:40, SteamyTea said:

If your house is that leaky, no need to put the MVHR in.

 

I would have cast that slightly differently.  If you are truly losing 4 ACHP with no heat recovery, then this will be by far the largest source of heat loss.  Think of your house as a system: it has to be in balance.  There no point in having fantastic windows, etc., to keep the heat in whilst at the same time letting it escape through every crack in the house if there is the slightest wind.

 

I have a passive-class house which is maybe 40% larger than yours, and our energy expenditure is just over 11 MWh p.a., (This is all electric, of which about 60% is resistive space heating – I can't make a case for installing an ASHP even though we've allowed for one in our initial  design).  Even with ~ 90% heat recovery, our MVHR circulation accounts for about a  ⅓ of our heat losses. 

 

I would suggest as an option that you consider retaining the MVHR, but look at what mitigation you can do to get your ACHP under 1.  You will find this will get you a more balanced system, and a more comfortable house as well as lower bills.

 

BTW, in my case my FP contract with OVO will more than double in October which tilts the cost benefit case and so we may be installing an ASHP over summer. ?

Edited January 20, 2022 by TerryE
Thanks Nick :-)

[SteamyTea]

12 minutes ago, TerryE said:

mWh

Milliwatthour

 

I was under 4 MWh last year (again).  When I moved here, 17 years ago, and even have a keen interest in energy usage, I was using 11 MWh.year^-1.

I still use resistance heating (no gas), have original timber windows and back door, though re-glazed to a 16mm gap, rather than a 4mm gap.  Added some insulation to the loft, fixed easy to find leaks (got one I just cannot seem to cure), the rest of the savings have been better management of my usage.

[TerryE]

Yup Nick, however we've got a largish detached 4 double bedroom + ward storage / drying room which we keep at ~ 22½ °C all day, all  year -- this may seem profligate, but we're getting old now and I just like being able to walk around in bare feet and shorts when I want without feeling cold; my wife and son ditto.

 

As we discussed on the forum previously, if you don't have fundamental design flaws in your build (unsealed air-gaps, missing insulation, cold bridges, etc.) then the macro heat-loss is pretty straightforward: just the sum-product of area × U-value × ΔT for external surfaces + the air heating of cold replacement air -- which is basically what JSHs spreadsheet represents.  The overall heat-loss is driven by the cost trade-offs on the coefficients in this calculation that you can control.  

 

As  you and others have shown, it is quite possible to significantly reduce heat losses by retrofit, but in my experience this is a lot easier to achieve in a new self build where the owner has far more control over the design itself and can monitor build quality.  In our case the extras incurred for going to passive standard were pretty much offset by savings elsewhere (e.g. no CH system other than the in-slab UFH loops -- my entire heating system cost about £2½K and is pretty much maintenance-free).

 

Even though 90+% of our electric heating is done at off-peak rate E7 tariff, with the expected tariff increases when our fixed price contract finishes, the ~4-5 MWh that I could save by installing an ASHP might just about cover its investment cost so long as I do the install myself. 

 

As Joe said on another post, IMO dot&dab boarding out counts as a major design flaw, as this is rarely done as the videos on YouTube demonstrate, and too many internal thermal images of such external walls show that these voids often act as a large plate heat-exchangers dumping heat into void behind the plasterboard where the external draft air circulating here carries it out as convective losses.  If you must use a conventional blockwork + insulation + external brick skin, then going for wet plaster on block if a far safer option, as well as making sure that all interfaces are properly taped during 1st fix.

[LiamJones]

How do i factor in additional losses from UFH using this spreadsheet? Should i adjust row 51 - Floor fixed heat loss to use the flow temperature, rather than the room temperature?

 

Floor area * (Room temperature UFH flow temp - ground temp) * U value

[LiamJones]

9 minutes ago, LiamJones said:

How do i factor in additional losses from UFH using this spreadsheet? Should i adjust row 51 - Floor fixed heat loss to use the flow temperature, rather than the room temperature?

 

Floor area * (Room temperature UFH flow temp - ground temp) * U value

Interestingly, for me, A flow temp of 45 would increase annual heating demand from 8875kWh to 10356kWh, or 16% 🤯 

A flow temp of 35 would increase annual heating demand to 9764kWh, or 10%

Edited October 12, 2023 by LiamJones
specify flow temps

[TerryE]

3 hours ago, LiamJones said:

Should i adjust row 51 - Floor fixed heat loss to use the flow temperature, rather than the room temperature?

No.  You are partially correct in that heat loss through the slab is really dependent on the average slab temp and not the room temperature, but the slab will pass about 7ΔtA W into the living space, so by example if you have a 100m² slab and need 24 kWh of daily heat input, say, to keep the house in thermal balance, then this is an average 1000W so the slab needs to be roughly 1.4 °C warmer than room temp across the day to do this, and you should add this 1.4 into this term, but given that this is a small delta and the estimate is ballpark, then it's just easier to use room temperature.

 

The average slab temperature can be very different to the boiler / heater flow output.  (See my blog posts on this.)  Also because of this sort of simplification, this spreadsheet really only works if you have a (near) passive class build (i.e EPC A class).  In this sort of build, there is also no way  a flow temp of even 35°C is suitable for this type of house as you'd end up with a 1:10 on/off cycling of your boiler / ASHP.

Edited October 12, 2023 by TerryE

[YYT]

Hi All,

 

just filling the heat loss sheet out for my house. Where it states floor area is that ground and first floor? I only have 2 floors. Also I have no MVHR so do just put 0 in there. Air changes per hour from my sap is 8. That seams very high?

[JohnMo]

5 minutes ago, YYT said:

Hi All,

 

just filling the heat loss sheet out for my house. Where it states floor area is that ground and first floor? I only have 2 floors. Also I have no MVHR so do just put 0 in there. Air changes per hour from my sap is 8. That seams very high?

Area in contact with the ground is the area.

 

Air changes an hour, in you sap report in the ventilation section there there is a part named "Infiltration rate" use that or 0.5 which ever the higher. And reduce MVHR efficiency to 0.

 

 

[nezdub]

Another noob question from me, apologies. For wall area - is this only walls exposed to the outside? and does it include internal walls of rooms?

 

I'm trying to do calcs for my terraced house and appreciate any input.

[Mr Punter]

4 minutes ago, nezdub said:

Another noob question from me, apologies. For wall area - is this only walls exposed to the outside? and does it include internal walls of rooms?

 

I'm trying to do calcs for my terraced house and appreciate any input.

 

Yes, external heat loss walls only.

[Chris D]

Hi All, after many hours reading and investigating I thought perhaps easier to actually post my query on here so please be gentle with me if I'm being stupid. 

 

We're planning a large refurb of a ~250m2 bungalow which is currently heated with 70's hot air blown heating which is noisy and unpleasant. 

Floors are currently solid concrete. 

Proposal includes lots of glazing and external cladding and insulation but nothing on the floor. 

 

As no radiators or pipes at present we'd much prefer UFH but don't have much height so are looking at the low profile matting systems like nu-heat, ambisolo, etc.

 

I've read the hundreds of posts saying you need insulation under UFH to minimise heat losses and am trying to quantify this to make an informed choice and there are lots of competing elements, heat rises, flow temps, floor covering, etc. 

 

Using Jeremy's amazing heat loss spreadsheet I think I've worked out that approx 20% (2kW) could be lost into the floor with no insulation (assuming concrete u value of 0.8). I think my absolute numbers are wrong as I think the spreadsheet concludes 50,000 kwh/yr and currently with v poor insulation it's still only about 25,000 kmh/yr.

 

So my dilemma which I'd appreciate any advice on is that does 20% heat loss into an un-insulated floor sound about right? if so even with current bills that would be ~£700 per year into floor heat losses which whilst bad, compared to the cost of digging up and re-laying 250m2 of concrete, new skirting boards, etc. would still take a very long time to pay back even if you saved the majority of this. .

 

Any suggestions or ball park estimates to quantify proportion of losses into the concrete floor pre and post insulation would be much appreciated. 

 

Thanks in advance,

 

[ProDave]

28 minutes ago, Chris D said:

So my dilemma which I'd appreciate any advice on is that does 20% heat loss into an un-insulated floor sound about right?

No.  Jeremy calculated with his really well insulated floor and a low UFH temperature, he was losing 8% through the floor.  So a completely un insulated floor would lose a lot more than 20%

[JohnMo]

56 minutes ago, Chris D said:

heat rises

Hello there - to clarify conducted and radiated heat moves in every direction - from hot towards cold 

 

56 minutes ago, Chris D said:

assuming concrete u value of 0.8

For concrete the k-value depends on its bulk density and the moisture content in service. (CIBSE guide Table A3.1., 1980)

 

Bulk density kg/m3	k- value W/m.K Internal (3% moisture by vol.)	k- value W/m.K External (5% moisture by vol.)
2000	1.13	1.24
2200	1.45	1.60
2400	1.83	2.00

 So assuming the 3% moisture and density of 2200 and 100mm thick concrete, your U value would be

 

0.1 / 1.45 = 0.069

1 / 0.069 = 14 = U value. So leaks heat like a sieve. So 18 times the heat loss you used. Re do your calculations and think again!

 

 

 

 

[TerryE]

Picking up @JohnMo's point, your floor is acting as a near to perfect heat sink.  However the ground temp under the house will be in the 8-10°C range.  Hence a bare or tiled floor will soak up about 6-7 W/m²K so if the Δt air:floor is 10°C that's 6×10×250 W or 15 kW and that's 24×7. 

 

That's nearly all radiant, so the "heat rises" adage, as John says, just doesn't apply.  However, adding a decent fitted carpet and underlay throughout will make a big difference as this in practice is the main insulation layer.  How much is moot.  I suspect maybe 2 or 3 × factor reduction. 

Edited January 4 by TerryE

[ADLIan]

The heat loss from a ground floor depends upon its size, shape, edge condition and ground under it. Assuming your floor is a simple rectangle it will have a u-value of approx 0.65 W/m2K without any insulation. Try inputting this value into the heat loss spreadsheet

[TerryE]

5 minutes ago, ADLIan said:

it will have a u-value of approx 0.65 W/m2K

 

Eh? This is about 10 × smaller than the radiant equivalent. It doesn't pass the sniff test.  What is the source for this statement?  

[Redbeard]

 This:(https://warmafloor.co.uk/support-centre/u-values/#:~:text=U-Values of uninsulated ground floor&text=In order to meet the,by the addition of insulation.) has a useful 'ready reckoner' based on the IP3/90 Information Paper.

 

This (https://www.gov.scot/binaries/content/documents/govscot/publications/advice-and-guidance/2020/02/tables-of-u-values-and-thermal-conductivity/documents/6-c---u-values-of-ground-floors-and-basements/6-c---u-values-of-ground-floors-and-basements/govscot%3Adocument/6.C%2B-%2BU-values%2Bof%2Bground%2Bfloors%2Band%2Bbasements%2B%2B.pdf) from Scottish Gov't, usefully separates out solid and suspended floors which the other one does not.

 

 

Edited January 4 by Redbeard

[TerryE]

OK, my bad.  Having scanned the BRE paper and the CIBSE paper it references, this treats ground leakage as a semi-infinite heat flow to ground and that, given a large enough area, this is negligible (in the limit). The IP3/90  (P/A) and (P/A)² terms are curve fit approximations which reflect edge and corner leakages and these heat flows dominate the losses. Note this is all rough order empirical estimation, and as @Redbeard's reference state "On a cautionary note it must be stressed that the following table is based on approximate calculations as detailed above and is for guidance only. It is intended to help give some ‘feel’ for insulation requirements".

 

So @ADLIan's advice is better than mine. 🤒

Edited January 4 by TerryE

[JohnMo]

My advice was also flawed. However when you use UFH the floor temperature is increased. The water through the pipes could be up to 45 degs and the floor temperature will settle out at a higher temperature than the room.  Well insulated it could be 2 to 4 degs warmer than the room. Other cases could be higher floor temperature. 

 

When using the calculation you need to take floor temp into account for downward heat loss, not the room temperature.

 

So a room temp of 20 - ground 6 degs. 14dT.

No under floor heating downward heat loss is room temp, ground temp difference against overall floor U values including carpets and underlay.

 

With UFH

Floor temp to ground difference. So very good insulation your temp may be 24 for a 20 deg room, so already 18% greater downward heat loss. The more insulation the floor covering provides the warmer the floor becomes as the heat has to overcome the upwards resistance before getting to the room. Flow temperature is a balance of upward and downward heat profile. So heat losses could be in 30 to 40 range compared to radiators if you are not careful.

[Chris D]

Hi All, thanks for the quick responses.

 

All understood on the heat flow, was being stupid as all radiant heat. 

 

Looking at the Scottish paper and including my P/A calc it places our U value at around 0.5 so actually a reduction compared to the 0.8 I used previously however take the point about heat losses. 

 

on previous forum posts above from @TerryE it was suggested not to use flow temperature for the floor temperature in @Jeremy Harris spreadsheet and that this would only be a few degrees higher when calculating floor losses.

 

If I increased dT by 60% which would increase room / floor temp to 27C in in the calculation this would balance out the reduction in U value from the Scottish calculation and leave it as the ~20% of heat going into slab as calculated previously. 

 

Any guidance on what floor temps to use in the calculation or corrections much appreciated. 

[TerryE]

@Chris D, I have a warm slab (essentially 17 tonne of slab + another 10 tonne ring-beams wrapped in 200-300 mm EPS) with the UFH loops running in the slab itself. The house is a near-passive design so the slab only needs to run about 3°C warmer at its surface to provide enough space heating to keep the house at ~ 22½ °C which is what we like.

 

Here is a plot of the last 7 days manifold out & return temps:

I do my heating using Octopus Agile.  The CH algo calculates the total heat required daily and heat the slab at the cheapest ½hr price slots, which are mostly (but not always) in the overnight period -- hence the jaggies.  The house is treated as a single zone.  Currently the return peaks at 26-27 °C and decays to room temp through the day.  The outs are ~4°C hotter during heating.

 

With your house, you will need the slab to radiate more heat, and a conventional heating control will tend to keep the temp in tighter tramlines so the core of the UFH will run hotter, say 30°C, with heat conducting up the surface and then radiating into the room from the floor, but also a lot conducting down through the subsoil because you don't have that 300mm EPS layer.  

 

Given that the P/A U-value estimate is relative to external temperatures, then you could be looking at a Δt of maybe 25 °C  when you are heating the slab, or 25 × 0.5 × 250 W loss or 3 kW or up to 72 kWh / day of ground heat loss on top of what goes into the house itself.   So you need to do your figures carefully.  The bottom line is that putting UFH in an uninsulated slab is nearly always a very bad idea. ☹️

 

Edited January 5 by TerryE
Update graph

[Adrian Walker]

7 hours ago, TerryE said:

@Chris D, I have a warm slab (essentially 17 tonne of slab + another 10 tonne ring-beams wrapped in 200-300 mm EPS) with the UFH loops running in the slab itself. The house is a near-passive design so the slab only needs to run about 3°C warmer at its surface to provide enough space heating to keep the house at ~ 22½ °C which is what we like.

 

Here is a plot of the last 7 days manifold return temps:

 

I do my heating using Octopus Agile.  The CH algo calculates the total heat required daily and heat the slab at the cheapest ½hr price slots, which are mostly (but not always) in the overnight period -- hence the jaggies.  The house is treated as a single zone.  Currently the return peaks at 26-27 °C and decays to room temp through the day.  The outs are ~3°C hotter during heating.

 

 

☹️

 

I have been following your approach and solution to UFH with interest over the years. What is the slab temperature over the same period? Do you get jaggies or does the mass smooth it out?

[jack]

31 minutes ago, Adrian Walker said:

I have been following your approach and solution to UFH with interest over the years. What is the slab temperature over the same period? Do you get jaggies or does the mass smooth it out?

 

I also have an MBC slab and try to do as much heating overnight as possible. I have a typical daily ripple of about 0.2 degrees as measured at the top of our concrete floor.

[TerryE]

@Adrian Walker, a bit off topic I discuss this in my blog posts, but to summarise for your Q, I have added the manifold out temperatures.  This is a Home Assistant plot.  My CH system is a standalone RPi running Node-RED.  It publishes temperature and other stats ½ hourly to my local MQTT instance and these are tracked by HA, so the jaggies in the plot are partly because of the 30 min subsampling, and partly because the heating is only on when the electricity is cheapest. (Octopus actually paid me to heat the house on a couple of days over the Xmas break.)

 

As you can see the out whilst heating is ~4°C hotter than the return and the temp gradient is pretty even along the loops so the UFH pipework peaks at ~6°C hotter than the slab surface during heating.  The heat flows out radially from the UFH loop throughout the slab and within 15 mins of so of the heating stopping the bulk of the heat has spread pretty uniformly throughout the slab at around just over 25°C or 3 °C hotter than room temp. 

 

It is noticeable and Jan really like padding around in the kitchen in her bare feet in the morning when the slab is at its warmest.

 

My ripple is was little more that J's: when I measured it with a spot IR temp meter -- about 1°C.   

 

Edited January 5 by TerryE

[Bethany]

On 15/10/2023 at 14:08, YYT said:

Hi All,

 

just filling the heat loss sheet out for my house. Where it states floor area is that ground and first floor? I only have 2 floors. Also I have no MVHR so do just put 0 in there. Air changes per hour from my sap is 8. That seams very high?

Hi, do you have a copy of the spreadsheet you used please? I tried to follow the link for the JS Harris one but it says unavailable. I'm trying to calculate my whole house heat loss so I can work out what size boiler I need... thanks!