Cheap, Thermally Efficient Non-Structural Wall Detail

[IanR]

Not sure where you are in the country, but I had very good Service from John Payne Insulation in Norfolk. They travelled to me in Essex, but I got the feeling they wouldn't like to travel much further.

 

Very approximately, expect to pay £25 / m2 of wall/roof area for installed blown cellulose fibre at 300mm/350mm thick..

 

Warmcell costs have gone up since manufacturing went abroad (I think Warmcel was bought out???). I know JPI are looking at and trialling alternative brands to reduce costs.

[Nick]

1 hour ago, IanR said:

Just taking a step back... The title of this thread is "Cheap, Thermally Efficient Non-Structural Wall Detail", however what you are developing here is a structural wall.

 

Have you talked yourself out of the steel portal frame à la "BarnHaus"? or is this still your current direction. If you are still considering the steel portal frame then you could consider a less structural options than the I-Joists to make of the external walls. At the moment there is a double-count in your strategy ie. both the steel columns (suitably braced) of your portal frame and your I-Joists of your wall detail are independently capable of holding up the roof.

 

My current build is a conversion of a steel portal framed cow-shed and in the early days I enlisted some help from Dr. Ed Green to develop early concepts on integrating the portal frame, due to his involvement with Barnhaus. I wasn't interested in the straw bale insulation, so we looked at several other options. and always the plan was to keep the steel outside the thermal envelope and ensure no warm moisture could get to it.

 

Integrating the steel structure into the timber frame structure adds an extra layer of difficulty. The portals of the the steel frame will require racking bracing at wall level (as well as the roof) and as you have drawn it the racking braces will run through the I-Joists. It would be much easier if the steel columns were external to the wall, but this looses you some floor space you are trying to retain. Mine are half in, half out - a compromise that adds complexity.

 

Your choice of steel portal frame is due to room size. What's the roof, ie. pitched of flat, and what way does it run. It's quite possible you could do this with Glulam and I-Joist to avoid the Steel Portal frame, and simplify your structure.

 

With regards to your I-Joist wall detail, and to play devil's advocate, if cost and wall thickness are critical then there's no need for a service void if you are only running suitably sized cables in the external walls. But you shouldn't run plumbing. Also, your VCL could be achieved with a foil-backed plasterboard if your wall build up is so designed. Also, your outer sheathing could be both racking support and the airtight layer with something like Egger DHF board, avoiding the need for any airtight or VCL membrane. Designed right, its a cost effective solution.

 

Ian,

 

Thanks a million, this is just the sort of info that will steer me the right way.

 

Still basing everything on the portal frame so far. Did wonder whether the span was doable with another solution. At this stage I'm trying to pin down prices for as many things as I can that I know with some confidence will work. Once I've got an overview, my plan is to try and chip away slowly at the design to get within budget.

 

The roof I have planned thus far is a mild pitch steel-barn style but I'm not wedded to the idea. Whatever can achieve the span will be fine by me although I'd imagine it would be easier to get PP for a pitched roof that could be disguised as a standard house construction. Roof runs parallel to the shorter dimension of the room (spanning the long dimension).

 

The Structural I-beams were there to take the wind-load right up to the eaves height (which I took of the JJI span chart) but I overlooked the diagonal steel members  Could technically run them through the OSJ webs I think but would be a nightmare. Was planning to try and run the beams to support the first floor on the inside of the steelwork so they're not clashing with the bracing. Any chance the I-beams could be replaced with 2x shallow stud walls allowing the steelwork to stay in the middle? Does your solution have the steelwork sat between the cladding and the insulated wall? 

 

Thanks for the airtightness solutions. Passivehaus guide seems to suggest the VCL can be achieved just by taping the joints of the inner OSB so that might also be an option. 

 

When you say the DHF board could be racking support, do you mean to the exclusion of the steel braces or just additional racking support to keep the timbers stiff?

 

Sorry to bombard you with questions but nice to meet someone who has already been there / done that with the design you're looking at.

 

 

[IanR]

Ah, this is two storeys

 

Are you expecting a completely open ground floor, with no internal columns? You'll need a Structural Engineer to work out the most economic way of achieving that if you do. If I've understood your orientation correctly I see you needing an 8m steel at first floor (running parallel with the ridge) that will require structural gable end walls for it to sit on.

 

Maybe you need to work up a section. While you are thinking your timber walls will take the wind loading, where are they transmitting those loads to. Do you plan to structurally tie them into the steel frame at the eaves and first floor?

 

Personally I'd start by trying to do it in timber, with steels to reinforce if a timber member gets too big.

 

Having had to work around a steel portal frame I'd certainly try and avoid it in the future. Ridge height is very difficult to control across each of the portals. The wider the span and shallower the pitch the height variance is increasingly sensitive to the length tolerance of the beams and weight on the roof. I assume you'll need 3 portal frames to make up 2 x 5m bays, there's a risk of the ridge not being visibly straight. This goes unnoticed on agricultural buildings, but probably not on a house. Portal frames are also designed for a light-weight roof covering, what did you have in mind?

 

For mine the steels are outside of the thermal envelope. Due to a planning caveat they have to be externally visible in some areas, so also outside the cladding, but covered in others, so behind the cladding. [But: I do have 4 steel columns that are internal to my floor plan, helping hold up the roof, but due to how they are boxed and insulated they technically remains outside the thermal and airtight envelopes]

 

When I mentioned the DHF board that was for the structural timber frame scenario and no steel portal frame.

 

I found it was relatively easy to develop a hybrid (timber and steel) structure where timber walls, inline with and attached to the steel columns, where capable of supplying the racking support to the portal frame and the portal frame capable of dissipating the wind loads on the timber walls. But, if you've got that amount of structural connection between the two how do you then thermally isolate them and stop vapour getting to the steels, unless you can wrap the whole structure with external insulation, leaving just the floor connections to resolve.

[Nick]

5 hours ago, IanR said:

Ah, this is two storeys

 

Are you expecting a completely open ground floor, with no internal columns? You'll need a Structural Engineer to work out the most economic way of achieving that if you do. If I've understood your orientation correctly I see you needing an 8m steel at first floor (running parallel with the ridge) that will require structural gable end walls for it to sit on.

 

Maybe you need to work up a section. While you are thinking your timber walls will take the wind loading, where are they transmitting those loads to. Do you plan to structurally tie them into the steel frame at the eaves and first floor?

 

Personally I'd start by trying to do it in timber, with steels to reinforce if a timber member gets too big.

 

Having had to work around a steel portal frame I'd certainly try and avoid it in the future. Ridge height is very difficult to control across each of the portals. The wider the span and shallower the pitch the height variance is increasingly sensitive to the length tolerance of the beams and weight on the roof. I assume you'll need 3 portal frames to make up 2 x 5m bays, there's a risk of the ridge not being visibly straight. This goes unnoticed on agricultural buildings, but probably not on a house. Portal frames are also designed for a light-weight roof covering, what did you have in mind?

 

For mine the steels are outside of the thermal envelope. Due to a planning caveat they have to be externally visible in some areas, so also outside the cladding, but covered in others, so behind the cladding. [But: I do have 4 steel columns that are internal to my floor plan, helping hold up the roof, but due to how they are boxed and insulated they technically remains outside the thermal and airtight envelopes]

 

When I mentioned the DHF board that was for the structural timber frame scenario and no steel portal frame.

 

I found it was relatively easy to develop a hybrid (timber and steel) structure where timber walls, inline with and attached to the steel columns, where capable of supplying the racking support to the portal frame and the portal frame capable of dissipating the wind loads on the timber walls. But, if you've got that amount of structural connection between the two how do you then thermally isolate them and stop vapour getting to the steels, unless you can wrap the whole structure with external insulation, leaving just the floor connections to resolve.

 

Cheers Ian. Really appreciate you giving the benefit of your experience here.

 

My fault, I should have made the design clearer. Very roughly, Imagine an 18m long x 12m wide steel-barn. Split it into two so you end up with 2 areas at 9m x 12m. I'm looking to get one of those areas full-height without columns, and one 2-storey. If 12m was just too far to span, I could reduce the design to 9m width and maybe add an extra bay or two to get that floor area back. Those sizes are why I assumed (probably wrongly) that the portal frame was the only way to go. Not sure on the number of bays - was under the impression I would have to bring them closer than the agricultural-standard 5m for stability - the BarnHaus uses 4 bays at around half my build length so I was thinking I would end up needing 8 bays. This smaller span was why I didn't think about the braces.

 

Yes, the rough plan so far was to line the underside of the eaves with a timber header and run suitably strong beams (maybe gluelam) between each steel at ground floor level. That would have allowed the I-beams to tie-back at these points with the standard brackets and span ground to eaves all the way around the building in continuous lengths thereby making getting the envelope up and watertight fairly quick & straightforward. Needed the I-Beams to be 300mm min anyway for insulation reasons so being able to run them full height seemed like a good bonus.

 

The first floor (also I-Beam) was going to have a perimeter beam fixed to the inside of the steelwork at first floor level to hang from, and then rest on two rows of additional steel stanchions coming from foundation level so that instead of having to span 12m, it only had to span 4.75m, 2.5m & 4.75m. These additional steels I was going to box in / insulate and hide within internal walls (seems similar to what you described).

 

Still would be great to go with enclosing the steel within the wall from a simplicity point of view if there was a way to get the racking support without changing the wall position. Do you think there's any way that sheet material (DHF? Ply?) could give enough support if it was beefed up and fixed at regular centres? The more I think about it, it might not be that bad running a diagonal brace through the OSJ web - just a case of a hole at a different height for each of the 4 I-Beams that make up each bay - should be consistent across every bay so could be measured & drilled on the ground alltogether and just lifted into place.

 

Working on detailing this in CAD now so will update the thread with it soon but hopefully you get the idea?

 

If the portal frame is going to be too complicated I may just try to start again in timber and use steels only where needed for spans as you've suggested. If/when you get the chance would be great to see an example of the timber build on your place.

[PeterW]

10 hours ago, Nick said:

 

Thanks for the suggestion. Looks like solid stuff but TBH, as I'm trying to save cost and do as much as possible solo, it makes more sense to stick with two lighter, cheaper boards I'll actually be able to lift

 

Ok so I did a whole house with Fermacell on my own and it's easy to do and very forgiving. Much better than PB and it doesn't need skimming or finishing other than filling the gaps and some FST over the top. 

 

Its very DIY and they even do a 1 man board size. 

[Nick]

9 minutes ago, PeterW said:

 

Ok so I did a whole house with Fermacell on my own and it's easy to do and very forgiving. Much better than PB and it doesn't need skimming or finishing other than filling the gaps and some FST over the top. 

 

Its very DIY and they even do a 1 man board size. 

 

I'm slowly weakening to the idea.. though still can't shake the suspicion you all just want me to break my neck

 

It really does look like good stuff and the 1200x1200 board should be doable but it's nearly double the cost of one 12.5mm plasterboard and one 11mm OSB by my reckoning. If by some miracle there's any money left in the pot by that stage I might go for it.

 

The main lesson I've learnt so far with costing is that every time I see a better way of building something, the best thing to do is close my eyes and run from the screen before it sinks in as sooner or later I'll convince myself that I need to use it - the m2 cost just creeps up and up even though deep down I'd be happy enough just to be able to have the place built with no insulation and no plasterboard.

[PeterW]

So how are you going to finish your plasterboard..??

 

Skim coat adds £££ and is not very DIY - add in double the fixings and double the labour time (even at £1/hr you cost something ..) and Fermacell starts to make sense. 

 

Its also much cheaper by the pallet through a distributor.  

[Nick]

14 minutes ago, PeterW said:

So how are you going to finish your plasterboard..??

 

Skim coat adds £££ and is not very DIY - add in double the fixings and double the labour time (even at £1/hr you cost something ..) and Fermacell starts to make sense. 

 

Its also much cheaper by the pallet through a distributor.  

 

Was just going to dry line TBH. We aren't fussy people

 

Will look at the pallet prices nearer the time for sure. The other aspect is that I have a few friends in the building trade so stuff that is fairly standardised and can be bought at every builder's merchants/wholesaler (like plasterboard & OSB) I should be able to get an even better deal on if i really try by pinching their trade discounts or finding someone who has some unused pallettes. Proprietary/exotic materials tend to be harder to get bargains on so I'm trying my very best not to design them in unless I really have to.

[IanR]

A 12m span is easily achievable in the typical steel sizing used for Agricultural barns, without stepping up to the monsters used these days in warehouses and distribution centres.

 

The sketch below was for the structural changes I made to the primary frame of my barn. The high lighted yellow portion is the main barn span and is what your are describing and the areas crossed out on the first portal are lean-to bays that are add-ons. The distance between the two high-lighted vertical columns is 45 feet (13.7m). The width of each bay 19'6" (6m). Ridge height around 6.5m. Columns are 10" x 5.75" and the roof portals are 8" x 5.25". Not shown are the 173mm Z Purlins and 9" x 3" Eaves Beam timbers similar to what you describe.

 

 

Typically the first bay is braced as shown and the follow on bays lean against the braced one. I removed the low level cross-bracing (simple 2" tube) as it got in the way of an open area I wanted and had to replace it with the giant goal-post arrangement you can see between the internal columns of the centre bay.

 

As I said previously, the portal frame structure is designed for a light-weight roof, ie. profiled sheet steel, corrugated cement board or insulated profiled panels. I'm not sure it will hold up the weight of structure required for a domestic warm roof that you'll be after in your double volume area.

 

I'm not sure how much this image shows:

 

The steels are half in - half out the external wall. The sheathing on the timber structure actually notches around the steels so the wall narrows across the width of the steel.

 

And then when the cladding does its bit, the lower portion of the steels are visible and the upper covered, as per planning requirement.

 

I know I sound like a broken record, but in your shoes I'd do it with a large glulam ridge beam and I-Joist roof structure. I'd hate to add up what it's cost me to integrate the steel frame. But I had no choice, my planning is for a conversion, not a new build.

Edited January 19, 2017 by IanR

[Nick]

11 hours ago, IanR said:

A 12m span is easily achievable in the typical steel sizing used for Agricultural barns, without stepping up to the monsters used these days in warehouses and distribution centres.

 

The sketch below was for the structural changes I made to the primary frame of my barn. The high lighted yellow portion is the main barn span and is what your are describing and the areas crossed out on the first portal are lean-to bays that are add-ons. The distance between the two high-lighted vertical columns is 45 feet (13.7m). The width of each bay 19'6" (6m). Ridge height around 6.5m. Columns are 10" x 5.75" and the roof portals are 8" x 5.25". Not shown are the 173mm Z Purlins and 9" x 3" Eaves Beam timbers similar to what you describe.

 

 

Typically the first bay is braced as shown and the follow on bays lean against the braced one. I removed the low level cross-bracing (simple 2" tube) as it got in the way of an open area I wanted and had to replace it with the giant goal-post arrangement you can see between the internal columns of the centre bay.

 

As I said previously, the portal frame structure is designed for a light-weight roof, ie. profiled sheet steel, corrugated cement board or insulated profiled panels. I'm not sure it will hold up the weight of structure required for a domestic warm roof that you'll be after in your double volume area.

 

I'm not sure how much this image shows:

 

The steels are half in - half out the external wall. The sheathing on the timber structure actually notches around the steels so the wall narrows across the width of the steel.

 

And then when the cladding does its bit, the lower portion of the steels are visible and the upper covered, as per planning requirement.

 

I know I sound like a broken record, but in your shoes I'd do it with a large glulam ridge beam and I-Joist roof structure. I'd hate to add up what it's cost me to integrate the steel frame. But I had no choice, my planning is for a conversion, not a new build.

 

That's a great looking build and an amazing amount of space! The photo definitely helps btw.

 

I completely hear what you're saying about doing a timber build. I went for the steel originally just because of the span but if it can be done with engineered timber than that's my preference. Guess I won't know 100% if the Glulam will do the job exactly the way I want it until I'm at the point of paying a structural engineer to design everything properly so I'll keep the steel portal version in my back-pocket just in case there's an issue, and make sure my budget covers either option.

 

Just realised I never answered your question about the roof - given that budget is priority number 1, I'd happily use cement particle board or another lightweight board if planning don't raise any objections. I didn't think the weight would be too bad. Neither me or my other half are worried about kerb appeal or having a house to show off on Pinterest so a neat & tidy design that's warm with plenty of space is all we're aiming for.

 

Am re-doing all the basic details now to suit timber only and already it feels ten times easier - no worrying about how do I fix A to B, how do I stop X cold-bridging Y, e.t.c.

 

They only thing I'm not sure about is how to design the load bearing external and internal walls that will carry the roof, floors, e.t.c in timber. Will have to start doing some reading on that later on. Broadly, given the choice would you design the whole envelope using I-Beams and adding Glulam joists for large spans or is there a more cost effective way of doing it by mixing/matching standard timbers or other materials?

 

 

 

 

[PeterW]

11 minutes ago, Nick said:

 

That's a great looking build and an amazing amount of space! The photo definitely helps btw.

 

I completely hear what you're saying about doing a timber build. I went for the steel originally just because of the span but if it can be done with engineered timber than that's my preference. Guess I won't know 100% if the Glulam will do the job exactly the way I want it until I'm at the point of paying a structural engineer to design everything properly so I'll keep the steel portal version in my back-pocket just in case there's an issue, and make sure my budget covers either option.

 

Just realised I never answered your question about the roof - given that budget is priority number 1, I'd happily use cement particle board or another lightweight board if planning don't raise any objections. I didn't think the weight would be too bad. Neither me or my other half are worried about kerb appeal or having a house to show off on Pinterest so a neat & tidy design that's warm with plenty of space is all we're aiming for.

 

Am re-doing all the basic details now to suit timber only and already it feels ten times easier - no worrying about how do I fix A to B, how do I stop X cold-bridging Y, e.t.c.

 

They only thing I'm not sure about is how to design the load bearing external and internal walls that will carry the roof, floors, e.t.c in timber. Will have to start doing some reading on that later on. Broadly, given the choice would you design the whole envelope using I-Beams and adding Glulam joists for large spans or is there a more cost effective way of doing it by mixing/matching standard timbers or other materials?

 

 

 

 

 

@Nick if you PM me your email address I have the construction manual for one of the major engineered joist manufacturers on my laptop. It gives some decent detail on how to do rim boards and other floor and roof details. 

 

Glulam can be as big and as strong as you like - 40m span is about the limit ..!

[Nick]

5 minutes ago, PeterW said:

 

@Nick if you PM me your email address I have the construction manual for one of the major engineered joist manufacturers on my laptop. It gives some decent detail on how to do rim boards and other floor and roof details. 

 

Glulam can be as big and as strong as you like - 40m span is about the limit ..!

 

Cheers Peter, will email you now.

 

40 metres is pretty amazing. Are we talking 40m span on its own or as part of a roof construction including joists & other supports.?

[PeterW]

If you search the UK Timber Engineering Association (I think..) they have a load of stuff on glulam beams and their use.

 

From memory, @JSHarris has a glulam beam in his MPC build - I've gone for traditional steel as its easier to get the calcs on a single element.

[Ferdinand]

1 hour ago, PeterW said:

If you search the UK Timber Engineering Association (I think..) they have a load of stuff on glulam beams and their use.

 

From memory, @JSHarris has a glulam beam in his MPC build - I've gone for traditional steel as its easier to get the calcs on a single element.

 

This studio bungalow of mine has an approx 14m Glulam Beam as the ridge beam.

 

It has intermediate support, though no idea whether this is necessary ... the longest span is about 9m.

 

Presumably at some length road transport considerations come into play !

 

 

Ferdinand

 

Edited January 19, 2017 by Ferdinand

[Nick]

Just now, Ferdinand said:

 

This studio bungalow of mine has an approx 14m Glulam Beam as the ridge beam.

 

It has intermediate support, though ... the longest span is about 9m.

 

Ferdinand

 

Thanks Ferdinand - I'm looking for about 9m unsupported too so that's really good to know.

 

I take it it was built recently? All engineered timber build or other materials too?

[Ferdinand]

10 minutes ago, Nick said:

 

Thanks Ferdinand - I'm looking for about 9m unsupported too so that's really good to know.

 

I take it it was built recently? All engineered timber build or other materials too?

 

It was a new roof put on about 8 years ago.

 

The walls are brick and mortar, with joists across about 4.5m at ceiling height, and the glulam along the spine above.

 

I think you just have to run the numbers and see what comes out.

 

These chaps keep 12-15m lengths in stock, for an example:

http://glulambeams.co.uk/about-glulam/specifications

Ferdinand

Edited January 19, 2017 by Ferdinand

[IanR]

21 hours ago, Nick said:

They only thing I'm not sure about is how to design the load bearing external and internal walls that will carry the roof, floors, e.t.c in timber. Will have to start doing some reading on that later on. Broadly, given the choice would you design the whole envelope using I-Beams and adding Glulam joists for large spans or is there a more cost effective way of doing it by mixing/matching standard timbers or other materials?

 

I would have thought "standard" timber studs of a suitable depth to enclose sufficient insulation to meet Buildings Regs would be the most cost effective. Any chippie would be able to build it without it being Designed and Engineered, and Building Control would most likely pass it without needing a full set of calcs to prove it's strong enough.

 

I-Joists (and twin stud) construction however comes into its own if you want to insulate to a higher level than building regs. They both offer a cost effective deeper wall (to house more insulation) with a lower timber fraction than standard stud.

 

In my view an I-Joist wall could potentially give a "cheaper" wall than twin stud (but very difficult to get like for like costings to know for sure), but would possibly require a slightly higher cost slab/foundation to eliminate the cold bridge at the wall floor junction, IF that's what you were looking to achieve.

[Nick]

9 hours ago, IanR said:

 

I would have thought "standard" timber studs of a suitable depth to enclose sufficient insulation to meet Buildings Regs would be the most cost effective. Any chippie would be able to build it without it being Designed and Engineered, and Building Control would most likely pass it without needing a full set of calcs to prove it's strong enough.

 

I-Joists (and twin stud) construction however comes into its own if you want to insulate to a higher level than building regs. They both offer a cost effective deeper wall (to house more insulation) with a lower timber fraction than standard stud.

 

In my view an I-Joist wall could potentially give a "cheaper" wall than twin stud (but very difficult to get like for like costings to know for sure), but would possibly require a slightly higher cost slab/foundation to eliminate the cold bridge at the wall floor junction, IF that's what you were looking to achieve.

 

OK I follow. So the options are basically a standard post & beam construction or an all engineered timber construction but there's no sense trying to mix & match (plus potential issues with different rates of expansion between the two types).

 

Have done a fair bit of research over the last day and I think I mostly understand the basics now. The detail the TJI manufacturers seem to show for the first floor has the floor joists sitting directly onto a laminated header / beam, an engineered rim board all around and the second floor cill plate fixed on top of that with squash plates to support the joists where there are point loads.

 

Playing around with drawing that detail for my build it seemed like there could be some cold bridging through the joists as they were now in contact with the ventilated cavity of the rainscreen, bypassing all the insulation. Also looked like it would be very challenging to keep the VCL continuous having to tape around all the beam ends where they penetrated the inside sheathing of the wall.  Putting the joists on hangers seems a good enough solution to both. Then decided to have a go at drawing a solution that used the I joists everywhere instead of glulam beams for (potentially) less bridging and a structure that could be unitised so I could make frames off site that would just stack together for quick assembly. Basically instead of using a solid beam everywhere, the idea was to use I-joists so they can be fully stiffened with OSB at the fixing centres (where the studs attach and the floor joists hang), and fill them with 2x40mm PIR everywhere else. That way 2/3-ish of the solid timber is replaced by high quality insulation to compensate for the bridging.

 

About halfway through drawing it it started to look like overkill (assuming the joists are even strong enough when loaded that way), but I thought I may as well finish the detail to see what you all think - it's only a quick swap to put the glulam beams in.

 

D-Foundation & Ground Floor Wall Junction.pdf

D-External Wall & First Floor Junction.pdf

D-External Wall Detail.pdf