Flat roof detail - warm or cold?

[SBMS]

22 minutes ago, G and J said:

EPDM appears to be the standard approach at the mo.  

Fair enough. I think it depends on application, we are putting a green roof on so EPDM not ideal and more susceptible to tears etc. 

[Mr Punter]

5 hours ago, G and J said:

PIR (tapered to save the thickness of furring as it is only a small area.

 

I don't think you will be able to get hold of this for a small area.  Probably min order of a pallet and you may need more than 1 size.  Just use firrings.

[saveasteading]

32 minutes ago, G and J said:

PIR and Superfoil is a good option

I am genuinely open to conversion. Can anyone show me superfoil or similar tested in a real life construction?

Last time I researched it, the figures were from some obscure method of insulating a box. Not in a section of wall in a lab.

 

 

Look at what it is: several layers of foil and plastic and foam. Yes each layer works a bit. but foil works by reflecting energy back into an air space. In multiple layers it then hits another face and bounces back again. It depends on being loose so that the air does the work.  

I went to a talk by the manufacturers, sponsored by the local BCO, but they didn't mention superfoil, only their other products. That confirmed my suspicions.

Thus I think their marketplace is for non-industry persons.

 

Btw, the plasterboard manufacturers used to claim great things for foil-faced boards as insulation, but now don't seem to.

 

I've just had a skim through the website.  A masterclass in avoiding any solid information.  I have gleaned though 'It used not to be as good as was claimed but it is a bit better now because it is thicker.

 

It does work. But I think there are usually much better solutions.

[Iceverge]

Sorry no sketch. My family dragged me off before I got my drawing finished. 

 

As ever durability, cost and buildability are top of my agenda.

 

Couple of points to discuss with your designer.

 

1. The steel looks like a 305*102 UB of some description. I would try to get it swapped out for 2 x C24 300*75mm timbers or maybe a Gluelam. For a TF building they would be far easier to fix to and present less thermal bridging issues and be about half the steel price. 

 

2. Those pozi joists look to be only about 180-200mm which really isn't the best application of them in my opinion. You are gaining little of the potential strength of them by having such a thin depth . If you were to replace them with 225*75 solid joists you'd save a ton of cash and be able to collect them from any builders merchant. They'd be simpler to trim and adjust on site for chippies too. 

 

3. To get the desired slope I would just rip a 100x75 timber onto a wedge and hammer it on top of the gable walls and use them to slope all the joists. Accept a 2 degree fall in the ceiling inside too. It'll be imperceptible. No need to source tapered insulation or multiple fitting strips. 

 

Here's my proposal. 

 

 

225*75 joists in lieu of posijoists. Cheaper and easier to source. 

 

100mm mineral wool batts between joists. Easily sourced and cheap. 

 

11mm OSB layer returned to internal airtighess layer at perimeter with strips of membrane. Joints all taped externally for ease of construction. 

 

2 X 50 mm layers of PIR staggered taped.and foamed joints, to mitigate gas migration and board shrinkage. 

 

Breather membrane. 

 

38 x 50 mm battens at 90 Deg to joists for ventilation/drainage plane to be continuous with wall ventilated cavity 

 

18mm ply deck to support EDPM or GRP. 

 

Build the parapet on top of this layer of ply allowing the ventilation to go under it and down the wall cavity. 

 

 

Edited October 5 by Iceverge

[Iceverge]

One key point about warm deck roof construction that I dislike is that plywood gets trapped between multiple layers of impermeable material with no opportunity to dry through ventilation. 

 

If there's any excess moisture or construction moisture it can't go anywhere, potentially storing issues for freeze thaw action or organic matter decay. 

 

In the above diagram every piece of timber has the chance to dry to air. 

 

 

[Gus Potter]

I've had a read through this thread and what is apparent to me is that a lot of folk are focusing on the detail and not appreciating just what they are doing.

 

The main reason I think this is that folk just don't have a grasp of the basics and some think they can get a free lunch. Do your best to understand how the different types of roof work. Flk are using software that provides a nice number..and a pretty picture.. big mistake! Have you set your perimeter area for example? Have you allowed for noggings / SE requirements? you won't have as unless you have a complete model you can't. Even then a model is just a model... common sense and engineering judgement must always prevail.

 

If you have an extension the detailing of the interfaces are critical.. that is where water gas will get in. You can't detail this unless you understand the basics of the diffferent types of roof construction. Some are trying to save money by not paying a professional.. but to do this well you need to go through a learning curve.. and put a lot of effort into it.

 

Please folks..spend a bit of time learning about how the different types of roof work fundamentally before you try and be clever. Go for the simple stupid and see if you can get all the interfaces to work.. then refine your detail and value engineer. This is where I think folks are getting it wrong... a lot of folk are jumping the gun and looking for shortcuts. Self building / extending is hard work and there is no free lunch these days if you want to actually build what you have designed and are paying for.

 

Always think... can I get the builder to actually do this on site and how much will they charge me.. always design for the simple stupid that a local builder can actually "make it happen" on site.

 

With my SE hat on now.  Often flat warm roofs are a godsend as if they are simple we can transfer loads of wind load to them which can save a pile of cash.. avoids goal posts for wind loading say if you have big areas of glass. Firring pieces for example and their fixings can be problematic if the main roof structural diaphragm is above.

 

The way I do it is to say.. what kind of roof may work in principle.., pitched or flat say. Then look at the interfaces.. how do I still ventilate a roof I may be tying into? what Architectural envelope do I have to work to (usually the depth of the roof)... and what are my SE requirements. This takes a bit of experience.. I know what the different roof depths are conceptually.. warm, cold.. don't forget inverted roofs (can be an elegant solution).. its an iterative process.

 

I'm always thinking like a good Architect also should.. can I make my design SE stable so I don't get stuffed later on when the builders costs come in. Have I dealt with the cold bridges in the most pragmatic way. How do I design so it is buildable by a local builder with off the shelf materials and components.

 

 

[G and J]

Crumbs guys, thank you.  Loads to think through. 
 

Standing back from this a moment it’s interesting to contemplate how we’ve got here.  
 

We drew a picture, including a flat roofed bit for preplanning which got a thumbs up.  We thought we’d drawn something pretty standard  from a buildability point of view.  
 

Architect then turned our picture into a planning submission and we assumed any buildability issues would be flagged then. Some where but not one with the flat roof.  
 

Planning obtained architect went into building regs design.  This flat roof is the last bit, maybe because it’s the worst bit or maybe because it just happens to be last.  
 

We’re on a fixed price contract for getting us through building regs, and perhaps the architectural technician who’s doing the work is aware that the amount of time spent on our design is now coming to an end. 
 

We could simply accept the initial design, though I have concerns about how insulative it is and how easy the little upstand vents will be to get right and keep open.  So my instinct tells me that research here may bear fruit.  
 

Certainly I feel my many hours spent considering wall options/build ups/etc. helped me feel informed in the final debate that decided what to go for.  I even kid myself that the design is a little better for it (but I’ve been called deluded many, many times so who knows?). 
 

And that’s where I am right now, back gathering understanding of factors and options.  Perhaps the best I can hope for is to balance the slightly over conservative approach of the architect, but I’m driven to do my best.  
 

But how do I do that given where I am right now?  Do I simply accept the first design, with the intention of finessing later? Do I pause the process now and engage another party?

 

I’m not 100% comfortable with the process.  It seems to be that key design elements are ‘done by others’ (steels for racking strength, to hold up rear gable and to allow thinner posijoists; foundation design; beam and block design; etc.) only after building regs submission which seems daft to me.  
 

And there seems to be so much focus on ‘getting past’ building regs/control.   They should be our friends, a vital asset to help us improve our build quality yet more.  
 

Sorry for the quasi-philosophical dump, but I woke up early and sometimes it’s hard to work out how to think about the problem at hand.  

[Roger440]

All very interesting stuff.

 

Especially as i have a small flat roofed extension which i want to enlarge. 

 

However, i only have circa 175mm to play with otherwise it will foul the door on the flat roof.

 

It strikes me there is no "compliant" roof build up that i can use.

 

So currently looking at composite steel cladding!!! Which makes all the problems of having timber in the build up go away. It does however creaye some other issues....................

[saveasteading]

1 hour ago, Roger440 said:

composite steel cladding

No. Not for a small area. 

1 hour ago, Roger440 said:

there is no "compliant" roof build up that i can use.

Of course there are. Others on here will know more.

First keep the rain out and get it away. Get all your flashings thought out.

Next insulate what results.

[Roger440]

3 hours ago, saveasteading said:

No. Not for a small area. 

Of course there are. Others on here will know more.

First keep the rain out and get it away. Get all your flashings thought out.

Next insulate what results.

 

Why not for a small area? Its only 3.6m deep, so can span that without any supporting structure at all, other than the wall it attached to (existing) and the outer wall of the extension. 150mm panel would meet the regs i believe from an insulation perspective.

 

Your latter suggestion is fine, except that it wont meet regs  with the depth i have. Currently its not even insulated but has no ventilation as it has existing walls in 2 sides, so no easy route to do so. Ive yet to look in there and see how much mould there is.

 

[saveasteading]

8 minutes ago, Roger440 said:

Why not for a small area?

Because cutting is tricky as is sealing. What slope have you got?

A single span of 3.6m will deflect a lot. Have you checked?

[Roger440]

4 hours ago, saveasteading said:

Because cutting is tricky as is sealing. What slope have you got?

A single span of 3.6m will deflect a lot. Have you checked?

 

Im good with the installation side. Used it before, so im not concerned about that aspect. Pretty much got everything i need both tools and consumables etc, just not the actual panels.

 

Yes, checked the Kingspan load tables, and, handily, 3.6m was the max span for the sheets i considered using,  though i will confess the load data im less clear on. Not my strong point.

 

I appreciate its not a common solution for a house, but it works ok on commercial buildings and my own garage i built way back.

 

 

[Iceverge]

I don't know why we don't use more insulated metal panels on houses. 

 

They're extremely fast and perform very well.

 

 

[G and J]

3 minutes ago, Iceverge said:

I don't know why we don't use more insulated metal panels on houses. 

 

They're extremely fast and perform very well.

 

 

I almost fell for that.

 

It would be like living in a biscuit tin when it rains.  I’ve spent time in a ‘72 bay tin top.  You hear every drop, big time.

[Iceverge]

Lol, wasn't actuallly ment to be a joke!

 

Yes noise transmission would be a potential downside.

[Roger440]

48 minutes ago, G and J said:

I almost fell for that.

 

It would be like living in a biscuit tin when it rains.  I’ve spent time in a ‘72 bay tin top.  You hear every drop, big time.

 

Not really, not on a decent thickness composite panel. We are not talking about single skin steel roofs.

 

 

[Roger440]

55 minutes ago, Iceverge said:

I don't know why we don't use more insulated metal panels on houses. 

 

They're extremely fast and perform very well.

 

 

 

Dead easy to avoid thermal bridging at foundation level if using an insulated raft too.

[saveasteading]

2 hours ago, Roger440 said:

i will confess the load data im less clear on. Not my strong point.

That's what the span tables are. I'm not getting into it for you, but 3.6m seems a lot. deflection will be designed to be  1/180mm , so 20mm up and 20mm down.

 

27 minutes ago, Iceverge said:

noise transmission would be a potential downside.

It isn't bad at all. less than rain on a skylight. I've asked clients what they think of rain noise on a metal roof (in offices and schools) and they either don't notice, or like it.

[G and J]

1 minute ago, saveasteading said:

31 minutes ago, Iceverge said:

 

It isn't bad at all. less than rain on a skylight. I've asked clients what they think of rain noise on a metal roof (in offices and schools) and they either don't notice, or like it.

That’s good news for my man cave then.

 

Mind you, the sound bar should drown just about everything else out: low flying fast jets; earthquakes; even the music from an Amazon delivery driver…., etc.