Home  5  Books  5  GBEzine  5  News  5  HelpDesk  5  Register  5  GreenBuilding.co.uk
Not signed in (Sign In)


Green Building Bible, Fourth Edition
Green Building Bible, fourth edition (both books)
These two books are the perfect starting place to help you get to grips with one of the most vitally important aspects of our society - our homes and living environment.

PLEASE NOTE: A download link for Volume 1 will be sent to you by email and Volume 2 will be sent to you by post as a book.

Buy individually or both books together. Delivery is free!

powered by Surfing Waves

Vanilla 1.0.3 is a product of Lussumo. More Information: Documentation, Community Support.

Welcome to new Forum Visitors
Join the forum now and benefit from discussions with thousands of other green building fans and discounts on Green Building Press publications: Apply now.

    Hello everyone, this is my first post, though I have read hundreds of threads and learned a great deal from the forum in the past.

    I am planning an extension to my listed cottage, and think I have just about overcome the planning & listed building consent hurdles, but now face the killer - building regs!

    In order to get listed building consent we agreed to demolish the existing cold felt roofed extension which eats into the side of the 200 year old granite cottage. A new extension is proposed, a few metres away, with a low, glazed link room connecting them. The challenge is this linking roof must sit fully below the eaves of the existing single story slate roof (around 2.4m high). Targeting a very modest 2m ceiling height leaves not much space for a roof, but its possible.

    However the good old "warm deck" roof issue has really thrown it. My plan had been 3 glu-lam beams spanning between old and new, with joists between them, filled with kingspan. This is a no-go, it seems the structure must sit fully within the warm zone, i.e. insulation all sitting completely on top. This is great where there is space, but there isnt here.

    I've followed all the usual routes, begging the building controller, googling magical thin foam insulations, etc.

    I can get away with this part of the roof having a U-value of around 0.3. I have some pictures I'll try and attach.

    My best idea at the moment is to still use glulam, but have the structural elements breaking throught the deck, and locally clad in super duper aerogel blankets. The other idea I have (being a mechanical engineer) is making a very slim custom steel fabricated roof structure with kingspan on top. It would be a challenge even with steel.

    Basically I want to bounce the idea around and hear your thoughts. I'm also pondering how they lay standing seam zinc on top of a foam roof - surely if they put a ply skin down then that is subject to the condensation that they are trying to avoid?

    Look forward to hearing your thoughts
    I'll try and add the file this time
    and the planning drawing to show what i'm on about...
    and another, for good measure (hope i'm not breaking any upload rules or etiquette)...
    • CommentAuthortony
    • CommentTimeNov 4th 2011
    could you/would you accept a lower floor level?

    could you use prefabricated laminated roof panels, still standing seam but with insulation in the sandwich?

    how far out can the roof height get higher? 200mm, 500mm, 1000mm ,2000mm?
    • CommentAuthorqeipl
    • CommentTimeNov 4th 2011
    My first instinct when presented with a problem like this is to think like a boatbuilder.

    Ideally, the roof would be a single cambered laminated panel, possibly around 200mm thick.
    Foam in the middle, maybe with plywood skins inside and out, certainly with glass/resin reinforcement layers.
    The outer finish could be marine quality gelcoat or, if that's too shiny, an epoxy/glass cloth with matt epoxy paint.
    Inner finish could be painted (might have to be intumescent)

    Doing it this way means that the available depth is fully employed as structure (no standing seam outside or plasterboard inside) and there are no cold bridges.

    The cambered panel will be much stiffer than the low-pitched structure in your drawing.

    The spans involved will determine what thickness the roof panel has to be, and for practicality it might have to be manufactured in two full-width panels and assembled on site.

    A connecting member would have to be built into the roof panel to attach it to the walls.

    It would be worth contacting a GRP fabrication specialist to get an opinion on feasibility.

    The other thing to consider is reducing the amount of glass in the walls and increasing the thermal resistance of the solid wall elements. BCOs hereabouts are flexible on how the heat loss requirement is met.

    Another option is to make the glass room an 'unheated space', which removes it from heat loss calcs.
    Hi Tony

    A lower floor level could be accepted, although the kitchen is in the old bit, so there are other regulations about the accessibility between kitchen/bathroom/bedroom zones. I think we could get away with ramping gently down as we went through from old into the link, but it would have to be minor and I can't seeing it giving us the 150mm needed to simplify the issue.

    I have no idea if I could use the pre-fab panels, could be an interesting option. I'm guessing they'd still need some structural supports that spanned from old wall to new wall, and they would sit on top of that? I also considered SIPS, but in my brain at least 50% of SIPS is cold, so might not be allowed in Scotland (as a flat roof).

    I think as long as the roof doesn't visually cut into the old roof it would be fine - you thinking rotating the pitch so that the ridge runs parallel to the wall of the old building?

    • CommentAuthorSaint
    • CommentTimeNov 4th 2011 edited
    Thomas, if by Kingspan you're referring to their Koolphen phenolic product then on account ever let in come in contact with the zinc.
    A structural sandwich panel would be a good option as an example a coldstore ceilng panel with 0.5mm steel skins on a 150-200mm rigid foam core will span 6m and act as a walk on platform
    Thanks for the thoughts, qeipl

    I really like the idea of a single stiff panel, with a constant arc across it. for listed-building we'd definitely need to clad with zinc/lead - but that needn't be too much of an issue.

    Does anyone know if sandwich constructions (such as described) are acceptable in terms of warm deck roofing, given that a fair portion of the structure (the upper skin) is cold. Is it ok for the upper skin to be cold if it is a material unaffected by condensation?

    I am tempted to start playing around with compensatory insulation, but I also plan to live there forever so want it to feel warm. It gets pretty cold in winter round these parts.

    Finally, the unheated space approach would solve all my problems, except for having to go through (at least) 2 external spec doors every time I needed a pee. :confused:

    I'll definitely ponder the boat-hull approach some more...

    Hi Saint

    Ah that's one to watch. I've also learned recently that larch cladding also shouldn't come near zinc, which might be another issue.

    I'm starting to think sandwiches are the future - the spans/thicknesses all seem to add up. Its just a matter of confirming that they are not considered cold-deck and finding someone to make a strange bespoke one.

    Out of interest, when does a sandwich panel become a SIPS?
    • CommentTimeNov 4th 2011
    qeipl has the right idea here, I think. A laminated curved structure could be made quite simply, with the necessary insulation forming an intrinsic part of the sandwich structure.

    I'd be very inclined to look at using laminated ply for the inner skin (laminated up from thin sheets of ply in a curve, over a male form). Bond PU foam sheet on top of this, using PU adhesive (the foaming stuff) filling any gaps with can foam. It'll need weighting down evenly to get it to follow the curve and resist the foaming action of the PU adhesive curing. Finally, lay glass fibre up as the outer skin. This will bond exceptionally well to the PU foam (it's the way sandwich boat hulls are made) and give you a combined structural and watertight outer skin. If you don't feel confident about doing this yourself, then a boat builder or glass fibre roofing company could probably do it fairly easily.

    As qeiple says, you may need to bond in some stiffening timbers at the wall plate junction, although I'm inclined to think that you would be OK with timbers bonded to the underside of the ply layer.

    The net result would be a pretty tough roof that was strong yet thin.
    Thanks JSHarris

    I'm fairly handy - well I'm ambitious and keen to try things, with mixed results!

    You think this could be built in situ? I can imagine it working well.

    2 questions, though:

    1. can I lay a zinc/lead roof onto GRP - as a purely cosmetic trim?

    2. would you see any issues with hanging a horizontal plaster board ceiling beneath it (in terms of ventilation etc). there are a few reasons for this, mainly routing of services, lighting etc?

    • CommentTimeNov 4th 2011 edited
    Building in situ should be OK, but would mean making some temporary curved forms to fit between the walls to laminate the ply sections up. I'd suggest using an underside ply skin made up from three or maybe four layers of 4 or maybe even 6mm ply, with all the joints overlapping (cut sheets in half lengthways/widthways to stagger all the joints by half a sheet thickness if you can).

    If you can find a means of attaching the metal roofing then it should be fine. Lead would probably be easier (but undoubtedly more costly) than zinc, but a chat with one of the zinc roofing companies might give you an idea of the art of the possible. I believe that the on-site rolling machines for standing seam can roll slight curves OK, so that might be an option.

    Fitting a flat ceiling underneath would be easy, as you could utilise the forms as rafters to hold it. It would add further strength as well, although to be honest I don't think you'll need it. Years ago I made a curved deck for my old yacht like this. The deck was six feet wide and had a curve that was around 3 inches up in the centre. The lower skin was two sheets of 3/16 in ply laminated up over forms on the floor of the workshop. The foam was 1 in thick PU sheet, bonded to the ply and the deck was a modest layup of glass fibre. I could walk about on this with barely any deflection at all, it proved to be very stiff indeed. With a foam core that's maybe two or three times thicker I think you'd have a very stiff structure indeed.
    • CommentAuthorRobinB
    • CommentTimeNov 4th 2011
    Posted By: thomasjamesrhodesThis is a no-go, it seems the structure must sit fully within the warm zone, i.e. insulation all sitting completely on top. This is great where there is space, but there isnt here.

    I thought you'd be OK as long as the major part of the warm roof was above the rafters. Ours has 75mm Kingspan above the rafters and cheap recycled glass fibre inbetween the rafters. I'd welcome members comments.

    Given the location dont forget to factor in snow loading. With little roof line fall it will build up quickly and be a consderable load as seen in past 2 winters.

    Mike up North
    @RobinB I'm not sure about the spectrum of possibilities between warm deck roof and cold deck roof - I pushed my local building control chap on the subject and he said if it was unclear they'd request a dew-point-analysis, which I guess is their get-out. Sounds expensive, to me!

    I guess the main thing is so long as you have U-value of 0.35 above the structure, you'll probably be ok to infill behind it.

    The engineer in me is still bothered by the sandwich panel roof construction in that all of the compressive (upper) elements are effectively cold. I'd really like some clarity on this if anyone can help.

    this kind of thing what you meant, JSHarris?
      cold deck thinking 03 copy.png
    I like that. It shows a panel of 200mm thickness, with an arc to suit the existing bits. That 200mm would need to include the top and bottom decks of the sandwich, so would probably only be 150mm of foam, but thats still more than enough to meet regs.

    The problem is 1m from the apex (on either side) the roof equivalent pitch is around 2degrees, less than the 3degrees minimum specified by VMzinc. Perhaps that is fine if we have a fibre-glass layer under it anyway - its not exactly going to leak!?

    Perhaps lead might be better for this anyway... can anyone advise?

    • CommentTimeNov 4th 2011 edited
    <blockquote><cite>Posted By: thomasjamesrhodes</cite>this kind of thing what you meant, JSHarris?</blockquote>

    Yes, that's exactly it. Main structural issue to watch is the point load at the wall attachments, where the bearing load is taken on the lower skin. Some local reinforcement of this area (maybe just a bonded on load spreader lamination on the underside at the edge) might possibly be needed, depending on the total weight plus worst case roof loading.

    The roof loading stress isn't going to be very high, though, even with a couple of feet of snow on it, as the span doesn't look as if it's very wide (maybe 4m?).

    I can't see the temperature differential causing a significant problem, although that depends very much of the relative thermal expansion of the inner and outer faces. As the outer face is free to move laterally and longitudinally, restrained only by the slight flex in the foam core, I doubt that it would create any unwanted buckling. In terms of degradation, then UV protected GRP as the outer skin should last a long time, provided it doesn't get excessively hot in summer. If you use a partially reflective metal roof on top I doubt that it'd get warm enough to cause any degradation.
    • CommentAuthorCWatters
    • CommentTimeNov 4th 2011
    another route is perhaps this one - its a flat panel sandwich, with pitch being introduced in the metal roof only. I have no idea how feasible this might be, but it looks a lot more traditional, and looks identical to previous plans, so might not need a non-material-variation into planning/listed-building etc.

    It doesn't look as warm, won't be as stiff, isn't as cool. But sometimes that's how it goes. I'll need to speak to my building standards controller and hear his thoughts.

      cold deck thinking 04 copy.png
    Posted By: CWattersGoogle found..
    http://www.sips.uk.com/system_curved_panel.php" rel="nofollow" >http://www.sips.uk.com/system_curved_panel.php

    Ask them about what cladding is required.

    http://sips.uk.com/images/img_curve2.jpg" rel="nofollow" >http://sips.uk.com/images/img_curve2.jpg

    Wow that certainly looks the business. I'll drop them an email.
    Cheers CWaters
    • CommentAuthorCWatters
    • CommentTimeNov 4th 2011
    Google "curved SIP" or something like that. I expect most SIP manufacturers can make them for a price. Perhaps shop around.
    hopefully they can answer my SIPS-coldish-roof concerns too...
    • CommentAuthorCWatters
    • CommentTimeNov 4th 2011
    Yes that's what I meant about asking about cladding (and any need for ventilation under the cladding).
    • CommentAuthorSteveZ
    • CommentTimeNov 5th 2011
    What an interesting thread this is turning into! I suggest using engineered timber joists. The thing they do well is span wide spaces with very little flex. How about a normal flat roof, using almost the full depth available. Imagine a beam and block construction with the beams made of engineered timber I-beams with the blocks made of your choice of solid foam fitted tightly in the gap and butted/sealed to each beam, and level with the top and bottom of the joists. The underside could be plasterboard with a vapour barrier. If necessary, a layer of foam could laid on top, although I don't think it would be needed as I doubt the timber/OSB/timber joists would conduct heat well enough to be a problem. I'll wait for ST or Viking House to point out where that's wrong :-)

    It would make fitting the new roof to the complex existing walls a lot easier, and it would be easier to assemble on site. You could lay a glassfibre roof on top but modern flat roof membranes (butyl?) should be good for 20+years, and if you have the outside joists as high as allowed (if height is not part of the planning problem, then they could be stepped upwards once they leave the old roofline and could be clad with zinc or lead if thought necessary) with the actual roof dropped slightly using slimmer joists. If the actual roof is concealed behind the side joists you might not need to show a zinc or lead roof. I can't see the details of the depth on my screen for some reason, but surely using the maximum depth available (10 inches??) is better for insulation and strength than the very strong, but thinner, curved roof.

    PS I think the curved roof would look great - making and fitting it would be an interesting experience..............
    • CommentAuthorCWatters
    • CommentTimeNov 6th 2011
    I'd consider buying the curved roof first and building the rest to fit it :-)
    • CommentTimeNov 7th 2011
    Used to make insulated curved GRP roofs for steam rooms. Largest I made was a 5.5m span. Never had to worry about snow loading, just a fitter/maintenance person crawling over them. Not something I would want to do on site though.
    Hi Steve

    I started off thinking along the same lines as you, with insulation between structural members, both of which utilised the full thickness making for a nice warm and strong roof. However my building controller told me that this was not allowed at all, and that there basically had to be a dotted line between where insulation stopped and structure began, with all structural elements being within the warm zone. I think he simplified it somewhat for my uneducated sake. I also think this is only applicable in Scotland.

    I've spoken to a few SIPs people, not had any clear feedback yet, but here's hoping.

    I've also done a few calculations, and a curved roof as discussed, only counting the strength of a 75mm thick ply skin is plenty strong enough to cope with snow loading (I worked out less than 3mm of deflection under 1.5kN/m^2). This would leave a good 100mm for foam insulation, then fibreglass and cladding on top of that.

    I am aware that including the foam and fibre glass will probably triple or quadruple the strength of the roof, but I was interested to work out if it could be made strong enough when considering only the 'warm' elements.

    I'll keep updating this thread with any progress. I've pondered the arced panel all weekend, and now think it would indeed be a thing of beauty, and I'd be daft to hide it above a horizontal plasterboard ceiling.

    Thanks again everyone for your comments and suggestions. I've had my £6 worth already :)

The Ecobuilding Buzz
Site Map    |   Home    |   View Cart    |   Pressroom   |   Business   |   Links   

© Green Building Press