Truly DIY SIP construction…

[G and J]

5 minutes ago, saveasteading said:

Booms are only one type of machine. the simpler version and much cheaper type has the concrete pump as a trailer, and the pipes go along the ground.

 

I can't see how that is a problem, except for the reduced insulation at studs, but that is only 5% of the area and so compensation / trade off should be findable elsewhere.

 

I suggest that you are set on sips panels and see anything else as inferior.  

If you have decided on a maximum wall thickness, and stick build is falling short on your target insulation, by say 5%, then it is easy to compensate for that in the roof or floor.

I’ve not investigated concrete pouring options for ICF as yet, but I will if when I look at ICF it looks like we can get a suitably warm and thin build up.

 

OK, guilty as charged, I am set on working out the best possible SIP design for us.  I think I’m close enough now to revisit the base case which is stick built.   Once I’ve repeated that process for stick built I’m thinking the choice will be obvious.  We shall see!

[Iceverge]

For thermal bridging mitigation nothing matches a continuous  layer of external insulation for ease of install and performance. 

 

 

Now some issues that are really really important in the practicalities of building. 

 

1. Local and completive availability of materials. 

2. Ease of making airtight and wind tightness.

3. Fire resistance. 

4. Noise protection

5. Ease of installing services. 

6. Lightly hood of finding appropriate tradespeople. 

7.  Need for specialist equipment. 

 

 

 

I reckon you should reconsider a stick build with a layer of PIR outboard and not get too tied up by the U value of the individual element but look at the wall as a whole. 

 

 

 

Or better still replace the mineral wool with blown cellulose. 

 

 

 

No messing with itchy mineral wool, tremendous airtightness with a taped OSB layer externally and cellulose. Plenty of room to deal with services, good noise performance, really good phase shift (decrement delay) for heat protection.  Lots of scope to deal with your chosen merchant for a competitive rate on materials. It can all be installed by 2 carpenters on site with no special materials. 

 

 

 

 

 

[saveasteading]

3 minutes ago, G and J said:

I’m thinking the choice will be obvious.  We shall see!

We may have different criteria. your obvious and mine may be different. 

 

[G and J]

9 minutes ago, Iceverge said:

You'd be surprised how the promised performance of a SIP wall worsens, once taking these all into consideration, as SIPS often require solid timber blocking in these areas.

 

I’ve wondered about that.  It does appear that a lot of SIP based designs use lots of timber splines rather than SIP ones.  However, I did also think that in comparing SIP and stick built each has the same timber framing round windows and stuff s that’s just the next problem to work, and therefore not highly informative in the making of the SIP/stick built decision.  
 

 

9 minutes ago, Iceverge said:

The essentials are you want a low U-Value Wall, with minimal thickness on a site that constrains the use of high overhead machinery. 

Spot on.  I also only want it to rain at night.  Oh, and be buildable within a sensible budget.   😉

[Iceverge]

Just saw the plans, 

 

That's a nice house, credit to your designer. 

 

It is a very tight site however! I'm going to double down on a stick build plan. You could buy every material at your local builders merchants yard, drop it in an 8x4 builders trailer and hand ball them into place completely eliminating any need for trucks of cranes etc.   

[G and J]

6 minutes ago, Iceverge said:

For thermal bridging mitigation nothing matches a continuous  layer of external insulation for ease of install and performance. 

 

 

Now some issues that are really really important in the practicalities of building. 

 

1. Local and completive availability of materials. 

2. Ease of making airtight and wind tightness.

3. Fire resistance. 

4. Noise protection

5. Ease of installing services. 

6. Lightly hood of finding appropriate tradespeople. 

7.  Need for specialist equipment. 

 

 

 

I reckon you should reconsider a stick build with a layer of PIR outboard and not get too tied up by the U value of the individual element but look at the wall as a whole. 

 

 

 

Or better still replace the mineral wool with blown cellulose. 

 

 

 

No messing with itchy mineral wool, tremendous airtightness with a taped OSB layer externally and cellulose. Plenty of room to deal with services, good noise performance, really good phase shift (decrement delay) for heat protection.  Lots of scope to deal with your chosen merchant for a competitive rate on materials. It can all be installed by 2 carpenters on site with no special materials. 

 

 

 

 

 

Wow, brilliant, thank you Iceverge.  I’ll use those as a starting point.  
 

I did read somewhere on here concerns about the permeable membrane being inside the PIR but other than that that build up might be perfect (with the addition of a 25mm service void for cables and a skin of ply or osb just under the PB).

[G and J]

1 minute ago, Iceverge said:

That's a nice house, credit to your designer. 

I’ll tell Julia you said so.  In fairness our architect is brilliant in that they accepted we wanted to design it, they’ve tested the design firmly and respectfully, and they are truly working with us.

 

3 minutes ago, Iceverge said:

I'm going to double down on a stick build plan

Which is what the architect said when we interviewed him on site before engaging him.  But me being me I want to be an informed decision maker partly to ensure that we aren’t unconsciously falling foul of other people’s prejudices. 

[Iceverge]

15 minutes ago, G and J said:

I did read somewhere on here concerns about the permeable membrane being inside the PIR but other than that that build up might be perfect (with the addition of a 25mm service void for cables and a skin of ply or osb just under the PB).

 

Ah, there's a convention to have a vapour control layer (VCL) inboard of any timber structure in this neck of the woods. It's yet another of those things that based on worse case scenario with terrible workmanship assumed. Applied without thought and understanding is risky. In this case you would risk trapping moisture in the structure between two materials of very low vapour permeability (VCL and PIR). Almost all moisture is carried into a wall via air paths (read poor airtightness). If you were to build with cellulose and an external airtightness layer you could expect to see very air test results is done diligently and it's simple to implement. Omitting the VCL would allow any moisture to dry to the inside again. 

 

11 minutes ago, G and J said:

Which is what the architect said when we interviewed him on site before engaging him.  But me being me I want to be an informed decision maker partly to ensure that we aren’t unconsciously falling foul of other people’s prejudices. 

 Sounds like a good architect. No harm to be diligent however.  

 

20 minutes ago, G and J said:

lots of timber splines rather than SIP ones.  However, I did also think that in comparing SIP and stick built each has the same timber framing round windows and stuff s that’s just the next problem to work, and therefore not highly informative in the making of the SIP/stick built decision.  

 

Those pics are from the kingspan design details which use SIP splines but solid timber around the openings etc. 50mm of PIR outboard of any framing would all but eliminate any bridging concerns. 

Edited June 8 by Iceverge

[SteamyTea]

39 minutes ago, G and J said:

2 hours ago, SteamyTea said:

 

I’ve noted that many peeps seem reticent about posting their plans so I haven’t posted so far, but I don’t see why not. 

Because we like to rip other peoples designs to shreds, then redesign it as the place we want.

I usually never comment on the layout, unless it is hideous (only commented once).

 

39 minutes ago, G and J said:

learning tons of stuff too.

That will be 'to' not 'too', or so I am told.

You seem to be getting a good handle in it, so what ever you make will probably be a better standard than an off the peg design.

 

Edited June 8 by SteamyTea

[G and J]

6 minutes ago, Iceverge said:

Omitting the VCL would allow any moisture to dry to the inside again. 

I think this means not having a plastic sheet directly behind my service void.  I thought I had to have one.  
 

Do I?

 

[G and J]

1 minute ago, SteamyTea said:

better standard than an off the peg design

Fingers and toes crossed.  I see it as a simple equation,

 

listen+think+listen+check*think more=hope.  

 

I’m happy for our designs to be criticised, once I’ve stopped sulking I’d have learnt something or embraced new ideas.   I though peeps were worried about privacy (so far our plans have only been seen by a very select few of just about everyone who cares to look at our planning application lol)

[SteamyTea]

3 minutes ago, G and J said:

I think this means not having a plastic sheet directly behind my service void.  I thought I had to have one.  
 

Do I?

 

That is a very difficult one to answer in isolation.

Condensation risk analysis is a huge topic, but one that will need explaining sometimes.

 

My view is that for any material that can absorb moisture, AND then give problems, you need a VLC.

[SteamyTea]

4 minutes ago, G and J said:

worried about privacy

Partly, why it is best to scribble out anything that makes it identifiable.

The other problem is that search bots trawl websites and pick up on names address, phone numbers and payment card details, which is a problem.

[Iceverge]

1 hour ago, SteamyTea said:

That is a very difficult one to answer in isolation.

Condensation risk analysis is a huge topic, but one that will need explaining sometimes.

 

My view is that for any material that can absorb moisture, AND then give problems, you need a VLC.

 

I would limit any material of low vapor permeability to one layer in a buildup. This will aid "drying", either inwards or outwards. 

 

The important thing is to have a good airtightness layer somewhere. This will limit "wetting". 

 

It's a tricky concept to get your head around but so long as "drying" exceeds " wetting" you're sorted. 

[G and J]

So a build up like:

 

1. Skim

2. PB

3. OSB/ply

4. 140mm frame filled with rockwool

5. OSB3

6. water barrier membrane

7. PIR

8. Batten

9. Counter batten

10. Larch cladding

 

Would potentially create a condensation pocket between layers 3 and 7.  Fast forward to a rotten frame.  Damn.

Edited June 8 by G and J
Forget the OSB3 sheathing layer. D’Oh!

[SteamyTea]

31 minutes ago, G and J said:

Would potentially create a condensation pocket between layers 3 and 7

In the UK where we generally have a greater relative humidity inside than outside, and generally have lower temperatures outside than I side, moisture needs to travel outwards.

This is generally achieved by modelling it via the partial vapour pressures.

All that means is that as long as the associated temperature and RH barrier allies water vapour to move outward, then there is generally not a problem.

Any condensation within the structure is expected to evaporate and travel outward in a short space of time.

The VCL is placed inside because that is where the warmest and highest RH air is, ventilation (MVHR) is used to replace this air with lower RH air.

 

So when looking at the different layers, you need to know the vapour transmission rate and make sure they can allow more vapour though then as they get colder (outside).

 

The theory is based on the ideal gas formula PV/T where P is vapour pressure rather than absolute pressure. V is obviously a fixed volume and T is absolute temperature.

 

(A lot of thermodynamics is based on modifying the idea gas formula)

[G and J]

3 minutes ago, SteamyTea said:

In the UK where we generally have a greater relative humidity inside than outside, and generally have lower temperatures outside than I side, moisture needs to travel outwards.

This is generally achieved by modelling it via the partial vapour pressures.

All that means is that as long as the associated temperature and RH barrier allies water vapour to move outward, then there is generally not a problem.

Any condensation within the structure is expected to evaporate and travel outward in a short space of time.

The VCL is placed inside because that is where the warmest and highest RH air is, ventilation (MVHR) is used to replace this air with lower RH air.

 

So when looking at the different layers, you need to know the vapour transmission rate and make sure they can allow more vapour though then as they get colder (outside).

 

The theory is based on the ideal gas formula PV/T where P is vapour pressure rather than absolute pressure. V is obviously a fixed volume and T is absolute temperature.

 

(A lot of thermodynamics is based on modifying the idea gas formula)

So that’s a ‘maybe’ then, I think.  🤷‍♂️

[SteamyTea]

1 minute ago, G and J said:

So that’s a ‘maybe’ then, I think.  🤷‍♂️

Um, yes, or no.

 

Raoult's Law

 

Years since I studied this properly, but seem to remember that the general idea is basic.

When looking at this sort of stuff I use statistical models, then the real risk can be quantitative. 

 

[G and J]

The principle that an impermeable barrier is placed on the warm side so it doesn’t get condensation on it makes sense.  
 

The principle that there is only one impermeable barrier also makes sense, even if that barrier is double layered (e.g. plastic sheet and PIR).

 

Which says to me the only safe option is to not have PIR outside my sheething layer. 
 

Are there practical alternatives?

[SteamyTea]

You can get PIR boards without the foil vapour barrier.

https://www.secondsandco.co.uk/product-page/pir-boards

 

They would probably be OK as by the time the VCL/MVHR has done there business, and the fact the PIR is on the low temperature/Lower RH side, there would be virtually no mass of water vapour to condense/freeze.

 

This does raise the question if what sort of roof you want to fit, Warm (would be my choice), Cold (just why) or the Hybrid (horrible).

The reason this makes a difference is because it can change where the effective moisture release areas are.

 

My feeling is if you control ventilation and heating, which is really just another way of saying keep the internal RH as low as comfortable, once the building has dried out fully, you will not have a problem.

[Iceverge]

4 hours ago, G and J said:

So a build up like:

 

1. Skim

2. PB

3. OSB/ply

4. 140mm frame filled with rockwool

5. OSB3

6. water barrier membrane

7. PIR

8. Batten

9. Counter batten

10. Larch cladding

 

Would potentially create a condensation pocket between layers 3 and 7.  Fast forward to a rotten frame.  Damn.

 

Not necessarily, I would omit the inner OSB layer in any case as it's not needed and will just add to the thickness which you want to avoid. 

 

You may want some Plywood or noggins/blocking if you plan on hanging furniture from the wall etc.im certainly areas. This localised ply inner layer won't be airtight as there'll be many penetrations for wires etc so I wouldn't consider it much of a barrier of any kind. 

 

 

 

 

 

[Iceverge]

4 hours ago, SteamyTea said:

the UK where we generally have a greater relative humidity inside than outside, and generally have lower temperatures outside than I side, moisture needs to travel outwards.

 

I would argue that absolute humidity tends to equalise rather than travel in any particular direction. Indoor absolute humidity tends to be higher and therefore the trend to migrate outwards. 

 

By keeping a layer of continuous insulation outboard of any organic materials like timber then it will remain above the dew point for almost all of the year and condensation cannot happen.

 

As there's only space for 50mm of PIR in this case, below about 5 Deg there is a risk of moisture accumulating on sheathing/PIR interface. This assumes that room air comes into direct and regular contact with the sheathing AKA terrible airtighess. 

 

In reality in our neck of the woods we are very rarely exposed to prolonged cold temperatures and if you plan on doing a good airtighess job then the risk is truly minimal as ,like I've said before, almost all moisture gets into a structure via air paths and cracks and not through diffusion through materials. 

 

3 hours ago, SteamyTea said:

does raise the question if what sort of roof you want to fit, Warm (would be my choice), Cold (just why) or the Hybrid (horrible).

 

I'm afraid , probably quite realistically in many cases, you've assumed terrible workmanship. 

 

Flat roofs go wrong when people take a token (if even that) approach to airtighess internally, have multiple down lighters etc, and token offcuts of PIR jammed between joists as a insulation. 

 

We have a cold roof. 450mm blown cellulose at ceiling level. However the external breather membrane was fully sealed so there's no ventilation in the roof. This was only possible because of the really diligent job we did on the airtighess layer below the trusses ensuring there was no way for air currents to take moisture up there. The carpenters were all spitting with indignation when I asked them to do it, telling me our roof would rot . But 4 years on and the timbers are as dry as a bone. 

 

Hybrid roofs are also an option if you aren't an idiot installing them and will save a lot on extra long fixings the thicker insulation in a pure warm roof of the same U value. 

 

 

 

 

 

 

 

 

 

 

[eandg]

10 hours ago, G and J said:

I’ve not spoken with an ICF company or a concrete delivery company but the fact that the boom would have to arch over electric cables would be a worry.  My target is a total wall thickness of ~300mm (inc. 70mm of battening and cladding on the outside skin) and I’m under the impression that ICF would struggle with that.  
 

Stick built is the default option, I’m exploring alternatives in part as a validation exercise but I’m learning tons along the way.

 

My investigations so far indicate that a SIPs on site blank panel build would use 175mm PU SIPs or 200mm EPS SIPs to yield a u value of better than 0.17.

 

Cost wise, and this is very broad brush, (and includes walls, joists, chipboard flooring at first floor only), it appears that:

 

1. Get a company in to do the lot and it’s just over 100k.  Game over.

2. Have a different company make it in factory and deliver it for me to erect and it’s about 52k.

3. Buy blank PU SIP panels and build it myself, buying in joists etc. from other suppliers will cost about 35k.

 

Each of these options require a fire board skin, cladding, internal battening and osb/plasterboard (I want good solid walls).

 

So, next step is to repeat the same process for stick built.  At present I’m struggling to work out a stick built wall build up that yields a decent u value and the right thickness.  

 

 

I thought our joiners did very well and they were on a bit less than a quarter of the £48k you're allowing for erection... Plus if you go supply and fit then you aren't out the VAT which helps your cashflow, and you pass much of the risk onto the supplier. 

 

I've no idea how you think you can handball sips panels into place, particularly at height and on the roof. It's not necessarily complicated but if you've anything a bit unusual, steels/glulam then you really want someone that's got a bit of experience erecting for you. 

[G and J]

2 hours ago, Iceverge said:

I would omit the inner OSB layer in any case as it's not needed and will just add to the thickness which you want to avoid. 

My thinking is that a layer of ply or osb would give a better feeling of solidity, better acoustic performance, can be done by me so it’s only the cost of materials and it has the spin off benefit that one can hang stuff wherever (as I’m used to doing in my blockwork house).  I’ve been told by a couple of peeps now that they are glad they did that, but I grant you, it’s not technically necessary.  It’s so hard as we all have a natural tendency to view our own past decisions with rose tinted glasses.

[G and J]

2 hours ago, Iceverge said:

As there's only space for 50mm of PIR in this case, below about 5 Deg there is a risk of moisture accumulating on sheathing/PIR interface

300mm is ideal but we won’t be silly about it. If there’s a really good reason to sacrifice some room space for better walls we will -better in this case meaning greater reliability and longevity, insulation is, as has been mentioned, capable of being ‘traded off’. 
 

So if losing 50mm made a big difference we’d do it.  But I’d look for alternatives first.