New Highland Home, ASHP size + calcs advice :)

[HighlandHopeful]

Hi all, this is a long one, but I’ve been storing this all up since June 😊

 

My profile name is 'Highland Hopeful' as I joined here when I was still gathering info for my planning application of a new house in the Highlands. I'm super pleased that we got planning approval and now crawling our way through the Building Warrant application. It is equal parts breakthrough and breakdown, as we are doing everything ourselves. We are 37 and 44 and very enthusiastic, both have engineering (sound, electrics, civil, transport) backgrounds and are hands on people.

 

I have found these forum threads so incredibly helpful, so thank you for existing, and haven't needed to post anything until today, but I would love to sense check a couple of things that have led to my ASHP sizing calculations as I’ve hit a wall.

 

I have been using info from the forum, as well as a very useful textbook from the library called 'Building Services Engineering' by David V. Chadderton.

The house is 19.2m x 4.8m, it has a 45degree pitched roof, it is one storey, so the ‘roof void’ is just open so that one can look up and behold it. It has one bedroom, and only my boyfriend and I will live in it, full time. The radiators and DHW will be heated by an air-to-water air source heat pump. We are not having underfloor heating. It will have triple glazed windows (I have already bought them – it was cheaper to design the house around what was available on seconds, than to spec them around a design).

 

The questions I would love your thoughts on 😊

 

1.      How to include DHW in ASHP Sizing Calcs

I have calculated the U-values of walls, roof, floors, windows and doors and used this to calculate the heat requirement. I’ve assumed 0.6 air changes per hour. Outside temperature of minus5C.

As a result, I got a heat requirement of 4kW. So I though woo great, I’ll get a 5kW heat pump.

But then I read in the textbook that this heat requirement is only part of the ‘boiler’ power calcs (I’m calling it a boiler as that is what the textbook from the library is based on) as I need to include for losses (10%, so 0.4kW, fine) and the power required to heat the water in the cylinder. This is where I got confused. To calculate the ‘boiler’ power you need to know the time to heat up the water in the cylinder, and for that you need to know the power of the ‘boiler’ (circular reference warning?!), so I used 5kW as that’s what I wanted to buy, but then obviously that tells you how long it takes to heat the cylinder if the power of the system is 5kW, which then would need adding to my heat requirement of 4kW and losses of 0.4kW to give a ASHP size of 9.4kW... I didn’t like how the equation was self referencing, i.e. if I’d said I was having a 12kW system, that’s what would go into he heating time calc, which would then go back into the ASHP size, and could just go up forever... Is it that I need to specify what kind of reheat time I’m happy with? Does 5kW based on the info above seem reasonable (U value of walls 0.131, roof 0.105, floor 0.21)? How do I incorporate the need for the DHW. Ha even this question is going around in circles!

A problem may be that I’m using ‘Boiler power’ calcs from the building services engineering textbook to try to work out ASHP size…

 

2.      U-value of Pitched Roof

I think there's a high chance I have done the U-value for the roof wrong. We are having a 45degree pitch, which is completely open i.e. doesn’t have a flat ceiling with a roof void above it. I have worked out the U-value of the roof as if it is just two flat pieces. In the textbook I am using it says for pitched roofs I should use the formula:

U = 1/(R_A cosβ + R_R + R_B)

where R_R is the resistance of the roof void (we don’t have one) and R_B is the combined resistance of the materials in the flat part of the ceiling including the inside (we don’t have a flat bit of the ceiling). I have tried to find online a way of calculating U value for pitched roofs, but not coming up with anything useful. Any ideas?

 

3.      Average Coldest Temperature

I’ve assumed an average coldest temperature for the heat requirement calcs of minus5C, but I’m really not sure about that. We are right on the north coast looking at the sea, but still seemed very low – any thoughts? I couldn’t find a good source for that, so just used what I found.

 

So sorry this is long and thanks for your time if you got this far! I’ve been working through this for months and have everything in a spreadsheet that will update if I change any of these numbers. I feel like I’m nearing the end so would love advice on these finer points.

 

[ProDave]

1)  What U values are you expecting to achieve for the walls, roof etc?  your house is about 92 square metres.  Ours is 150 square metres and has a heat demand of about 2.3Kw at -10 outside.  What dis you use to calculate your heat loss?

 

Why not UFH?  Easy to design and build at construction stage, leaves all your walls free of clutter and runs at a lower temperature than radiators so more efficient for the ASHP.

 

Typically out HP will spend an hour per day heating DHW, so your HP would need to provide the required heat in 23 hours to allow that. Heat pumps generally do heating or hot water, never both together, and most allow different flow temperatures from the HP for each.

 

2) heat loss through the roof is exactly the same calculation as the walls, you just have to work out the surface area of the sloping roof, which will be more than if you had put a flat roof on the building.

 

3) depends where you are.  I calculated worst case heat loss for -10 outside and +20 inside.  The cold spell a couple of weeks ago we got to -12 here and some parts to -16.  You don't want average coldest unless you are prepared to be cold, or use another heat source (e.g. wood stove) when it gets really cold, which is why I designed for -10.  the average is much higher than but a winter anticyclone and the cold can hang around a long time.

[JohnMo]

11 minutes ago, HighlandHopeful said:

How to include DHW in ASHP Sizing Calcs

With a 5 to 6kW heat pump, on a cold day AND if you need to heat cylinder twice, will take about 1.5 to 2 hours in total. Sometimes more sometimes way less.

 

So the easy way to size for is assume you can only heat the house for 22 out of 24 hrs.

4kW x (24/22), 4.4kW heat source is required. So 5kW is fine, as long as the heat output is that at your design temperature. The above assumes 210 to 250L cylinder with 3m² coil for heating.

 

We are on the Aberdeen side of Elgin, we have had full days at -9 to -10. Local weather stations are few and far between, our local one (Kinloss) and our temps can be 5 degs difference sometimes.

 

Roof U value, just do the same as the wall (1/r1...) and then apply the heat loss to the total area of the vaulted ceiling.

 

23 minutes ago, HighlandHopeful said:

assumed 0.6 air changes per hour

How are you doing ventilation, this changes this figure hugely?

[HighlandHopeful]

19 minutes ago, ProDave said:

1)  What U values are you expecting to achieve for the walls, roof etc?  your house is about 92 square metres.  Ours is 150 square metres and has a heat demand of about 2.3Kw at -10 outside.  What dis you use to calculate your heat loss?

Thank you for helping!

I've calculated our U value of walls 0.131, roof 0.105, floor 0.21. We have a lot of windows, which are 1.1 ish.

For heat loss I used:

Qp = ((F1 x Sigma AU)+(0.33F2 x NV)) x (tc - tao).

F1 and F2 were taken from the textbook for panel radiator systems and were 1.00 and 1.10 respectively.

Sigma AU was the sum of all the areas of elements multiplied by their U values

N was the air change rate - I assumed 0.6

V volume of the house

tc 21 degrees and tao WAS -5 but I've edited to -15 as I get now that I need to design for worst case as we won't have supplemental heating, not for an average case.

 

23 minutes ago, ProDave said:

Why not UFH? 

I thought about it for a long time but I really want wooden floors and lots of rugs and carpet in a couple of rooms...

 

Thanks for the other info, incredibly helpful!!!

[saveasteading]

Welcome @HighlandHopeful

 

I agree with @ProDavethat UFH is preferable to radiators.

The radiators would have to be very large areas when compared to gas or oil fired, which run at much higher temperatures.

Also I think they are more expensive than the basic steel ones. That goes a long way to equalising the cost.

 

22 minutes ago, JohnMo said:

Elgin, we have had full days at -9 to -10.

 

 

That is something that needs more local knowledge perhaps.   How many days of not quite enough warmth do you tolerate?

 

Plus there is the comfort issue. Warm feet makes you feel good, and so less energy is needed. And of course you are walking and sitting down there, not on high.

You are going to have  a lot of heat rising to your exposed roof. It might be worth sucking it back down.

 

Also that is all the more reason for having good roof insulation.

 

The invisible benefit is the absence of rads and pipes on the walls.

 

The heat loss calculation is simply a function of area and U value, so you are on the wrong track.

Roof, walls, floors with adjustments for windows etc and it is done.

What constructions are you having,, if decided?

I am sure one or more of us can send a spreadsheet for this... I'd have to find one of mine somewhere.

 

Controversially?  Do you need forced air changes? You have a very big volume and 2 of you.

You will be opening doors sometimes, and the cooker hood and wc fans vent whether you want them to  or not. I am not convinced they are an essential , have pushed my luck (with client agreement) on much more sensitive buildings than this, and had no issues.

 

45 minutes ago, HighlandHopeful said:

both have engineering (sound, electrics, civil, transport)

Excellent. we may have questions back to you.  Car radios a speciality? Have a first heart in anticipation.

 

[HighlandHopeful]

24 minutes ago, JohnMo said:

How are you doing ventilation, this changes this figure hugely?

 

Thank you for all of that, really helpful, I've updated my calcs to assume a much lower temperature, I'm not sure why but had it stuck in my head that you design for average lowest, not actual likely lowest.

 

Re: how we are doing ventilation. I assumed 0.6 air changes per hour for the heat loss calcs as I read that passivhaus should aim for that. We are using heat exchange MVHR. Haven't got as far on that yet, but let me know if you think 0.6 is a stupid number to have put into my calcs!

 

Thanks again, I'm chuffed you replied. 

[ProDave]

Download the heat loss spreadsheet from this post, it proved very accurate for my build.

 

 

I would be looking to improve floor insulation in your build.  We have suspended wooden floors with UFH but wood or slate flooring throughout and that achieved a U value of I think it was 0.13 and works well with the UFH.  Carpets will slow down heat delivery 

 

 

 

[saveasteading]

3 minutes ago, HighlandHopeful said:

I really want wooden floors and lots of rugs and carpet in a couple of rooms...

wooden floors is not a problem, unless very thick.  Rugs likewise, as the heat will move on and emerge elsewhere. 

Carpets no.   On our latest family project for the rooms where ufh isn't appropriate I am pushing for aluminium skirtings with the ufh pipes in them. No clutter. But you could still put rads in these rooms.

 

Just now, HighlandHopeful said:

it stuck in my head that you design for average lowest, not actual likely lowest.

'They' want you to buy bigger kit so push us in that direction.

[HighlandHopeful]

1 minute ago, saveasteading said:

wooden floors is not a problem, unless very thick.  Rugs likewise, as the heat will move on and emerge elsewhere. 

Carpets no.   On our latest family project for the rooms where ufh isn't appropriate I am pushing for aluminium skirtings with the ufh pipes in them. No clutter. But you could still put rads in these rooms

Interesting, love the sound of the aluminium skirtings with UFH (although BFs dad very allergic to aluminium and will be helping with the build - does any other metal work?).

Another reason against UFH for us, which I should have said, was that I thought you need a concrete slab for UFH, but we can't get a concrete tanker to our plot. We are having strip foundations so that we don't need a tanker. 

[saveasteading]

1 minute ago, HighlandHopeful said:

we can't get a concrete tanker to our plot. We are having strip foundations so that we don't need a tanker

Tell us more please. The foundations and floor will be made of????

other lorries can't get there either? 

 

A bracken on earth floor, driftwood walls infilled with heather and mud. Thatched roof harvested adjacent?  

 

 

OR  small loads of concrete. pumps. dumpers.

 

[saveasteading]

7 minutes ago, HighlandHopeful said:

does any other metal work?).

It will but I don't think it is available. The aluminium product is moulded so that the ufh pipe clips into it. can be painted.

   

10 minutes ago, HighlandHopeful said:

BFs dad very allergic to aluminium and will be helping with the build

We don't do family relationships and politics on here. But the appropriate radiators are made of aluminium too.

[HighlandHopeful]

8 minutes ago, ProDave said:

I would be looking to improve floor insulation in your build.  We have suspended wooden floors with UFH but wood or slate flooring throughout and that achieved a U value of I think it was 0.13 and works well with the UFH.  Carpets will slow down heat delivery 

 

Yes I am perplexed about why floor U value is so high, I wonder if I've calculated it wrong. I've used this for advice: 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

 

The build up the floor is:

On top of the soil, 50mm of sand, 50mm of concrete.

Then 150mm air gap - as part of the calcs I had to imagine this sucking heat out through the floor and transmitting to the uninsulated parts of the foundations around the perimeter of the building. 

Then 150mm joists with 100mm rockwool in between (and a 50mm airgap)

Then vapour barrier

Then 2 layers OSB at 11mm each

Then the wooden flooring but I haven't included the thickness of resistance of this yet, so U value might go down slightly, but not significantly.

 

I used

Resistance of the floor, Rf = 1/Uf - 0.17 - 0.17

And U-value of the floor, Uf = 1 / [(1/U0 ) – 0.2 + Rf]

U0 was derived from the table on page 6.c.3 at the above link - our area to perimeter ration is 0.5 and the "Ventilation opening area per unit perimeter of underfloor space (m²/m)" is 0.003ish assuming that I calculate this by doing depth of air gap between soil and joists divided by perimeter.

[saveasteading]

Sorry, but that floor construction doesn't seem right to me. Are you trying to invent a method from scratch?

 

My garden shed has a  100mm concrete on earth.

For a house as an absolute minimum, 100mm stone then 100mm concrete with mesh, then dpm. after that a timber batten base still feels like a shed to me.

THis sort of detail did exist in the distant past, but really isn't right.

 

That's without knowing your site conditions. What is the ground?

 

The U values is way down the list and can follow on from agreeing a structure that will last more than 30 years.

 

 

So, can you go back a few steps?  What is the access for materials like?  What is the ground?

What is the intended construction, in principle not detail.

[JohnMo]

1 hour ago, HighlandHopeful said:

how we are doing ventilation. I assumed 0.6 air changes per hour for the heat loss calcs as I read that passivhaus should aim for that.

Airtightness and ventilation heat loss are very different. An airtightness test of 0.6 will give an infiltration ventilation rate of nearly zero. So all the ventilation comes from the MVHR, which in turn recovers around 90% of ventilation heat loss.

 

Scottish regs require 0.5ACH or thereabouts. So your ventilation heat loss is 0.5 x 10%, 0.05 ACH. NOT 0.6. So ventilation heat loss drops by over 90% from your calculations.

 

Based on what you have said your -10 heat loss is going to be less than 2kW.

 

Low heat losses drive more issues as you struggle to get heat into the house without lots of cycling of heat source. UFH is way more forgiving for oversized heat sources, it's cheap to install and acts as a big buffer for heat just like an oversized storage heater.

 

Would rethink your floor design - go concrete, 200mm PIR insulation or 300mm EPS, the UFH on 200mm centres, then concrete over the lot. No screed needed

[HighlandHopeful]

Thanks everyone for all your help today. I will go away and think about the floor.

[sharpener]

On 19/01/2025 at 16:53, saveasteading said:

It will but I don't think it is available. The aluminium product is moulded so that the ufh pipe clips into it. can be painted.

   

We don't do family relationships and politics on here. But the appropriate radiators are made of aluminium too.

 

Aluminium is used for skirting rads bc it can be extruded to the exact shape along with all the internal detail so there is no waste or post-finishing needed. AFAIK it is the only metal you can do this with economically.

Unless you want "designer" radiators then ordinary Stelrad or equivalent steel radiators are significantly cheaper and just as good. With proper inhibitor in the system they will last 25 yrs +. Ignore claims that aluminium rads heat up quicker or are more efficient, this is marketing BS.

Heatpunk is a convenient and free design tool.

https://heatpunk.co.uk/

 

Edited Monday at 22:16 by sharpener

[ProDave]

On 19/01/2025 at 16:41, HighlandHopeful said:

Another reason against UFH for us, which I should have said, was that I thought you need a concrete slab for UFH, but we can't get a concrete tanker to our plot. We are having strip foundations so that we don't need a tanker. 

Timber frame build, strip foundations, I beam joist floors, insulated and covered in OSB.

 

Lay battens (following joists)  Lay UFH pipes.  Fill gap with dry pug mix sand / cement as a heat spreader.  Lay Engineered Oak floor as structural floor spanning battens.

 

 

 

[saveasteading]

1 hour ago, sharpener said:

steel radiators are significantly cheaper and just as good

Yes, I was assuming that those aluminium rads with prettier shapes and fins will have more surface area for a less huge rad.

The main issue is the sizes of rads required in any material. They have to be very big as compared to rads with much hotter water from oil or gas, and a lot of wall is needed.

 

1 hour ago, ProDave said:

Timber frame build, strip foundations, I beam joist floors

I suppose that can all be carried or barrowed up  a track where a lorry can't go.  And the footings mixed on site x 100 or 200 mixes, and barrowed to the hole.

 

Dry mix likewise. I've seen an 800m2 floor done in dry screed. At least the guy knew what he was doing tomorrow...and tomorrow...

 

UFH would be perfectly feasible.

[Beelbeebub]

How far from the road is your plot?  A mini concrete pump can sit on the road and push you concrete quite some distance. 

 

One visit for the foundations, then another for the floor slab. 

[saveasteading]

9 hours ago, ProDave said:

I beam joist floors, insulated and covered in OSB

 An aside perhaps. In my experience of these floors,  the concrete beams are ultra hard, and drilling requires that you don't hit a steel tendon. Did you use short screws?

And they have a camber to the centre.

Yours don't appear to have this camber, but perhaps you packed the battens?

[saveasteading]

@HighlandHopeful you see already how much sound advice is available. 

Tell us two critical things please. 

What access do you have?

What are the ground conditions?

[JohnMo]

2 hours ago, Beelbeebub said:

A mini concrete pump can sit on the road

Or a big one

[saveasteading]

21 minutes ago, JohnMo said:

mini concrete pump can sit on the road

You could even have one on the road feeding another half way. Or a couple of dump trucks.

[ProDave]

1 hour ago, saveasteading said:

An aside perhaps. In my experience of these floors,  the concrete beams are ultra hard, and drilling requires that you don't hit a steel tendon. Did you use short screws?

Sorry I did not describe it well.

 

JJI 300mm deep engineered timber I beams for the floor, as light as a feather to carry.

 

Here they are being insulated before being covered.

 

 

 

[SteamyTea]

On 19/01/2025 at 15:36, HighlandHopeful said:

3.      Average Coldest Temperature

I’ve assumed an average coldest temperature for the heat requirement calcs of minus5C, but I’m really not sure about that. We are right on the north coast looking at the sea, but still seemed very low – any thoughts?

Mean temperatures, and these can also be mean lowest temperatures, tend to follow a normal distribution.  You can see if you have a local Met Office weather station nearby here.

What you need to know is the base temperature you need to start heating at, for me that is around 9°C OAT (Outside Air Temperature), then look at the fraction of time (there are 8760 hours in a standard year) that your heating needs to be on.

Then, this is getting more complicated, what power you need at the specific ΔT (temperature difference between inside and outside) to hold the inside temperature i.e. 100W.K^-1.

And now to make it much more complicated, you need to know what power the ASHP can deliver at those OATS.

And this is the real hard bit, what the CoP (coefficient of performance of the heat pump will be at those temperatures).

Then it is easy to sum the lot up and you know the energy used, the CoP and therefore the bought in energy price.

DHW (domestic hot water) is an add on and needs to be treated as such, space heating and DHW are different things, and different times and at different temperatures.

So some allowance needs to be made for the time that your DHW is being recharged, with that time subtracted from the space heating times i.e. you cannot use the 24 hour average power delivery for space heating as your DHW may need 2 or 3 hours of that 24 hours, so that ups the power that needs to be delivered to the space heating by around 10%.

 

Initially you need to make some assumptions.  These can usually be calculated from your existing usage for DHW.