Author Topic: Unventilated cold roof condensation risk  (Read 24079 times)

Mark Pellant

  • AECBmember
  • Posts: 14
    • View Profile
    • Koru Architects
Unventilated cold roof condensation risk
« on: April 26, 2007, 12:17:48 PM »
On a number of occasions I have had problems persuading Building Control departments to accept German condensation risk calculations for a cold flat roof construction as follows:

Impermeable outer membrane (single ply or zinc etc)
Timber decking
Timber joists with full-fill natural insulation between (Warmcel)
ProClima 'Intello' variable vapour diffusion membrane
Plasterboard lining

UK software calculations by building Control consultants show it fails, German calcs show it doesn't. Can anyone help with a difinitive proof / calculation / certificate (LANTAC) etc, etc that can universally apply to this type of roof construction and can be accepted by Building Control without argument?

Nick Grant

  • AECBmember
  • Posts: 1312
    • View Profile
    • Elemental Solutions
Re: Unventilated cold roof condensation risk
« Reply #1 on: April 26, 2007, 07:44:53 PM »
Proving it works in theory aside are you confident that it will work? Presumably the timber gets wet in winter then hopefully the intelligent membrane sorts it all out in the summer?

I'd imagine this is sensitive to indoor humidity and obviously to detailing and airtightness of the membrane.

Nice to get rid of vent layer over yet use nice insulation and timber.

Interesting to hear what others think.

Nick

Beat Bergmann

  • AECBmember
  • Posts: 4
    • View Profile
    • NBT website
Re: Unventilated cold roof condensation risk
« Reply #2 on: June 05, 2007, 09:50:37 AM »
The Intello is a variable vapour membrane. There has been done some testing in Germany at the Fraunhofer Institute for Building Physics IBP regarding this topic:
http://www.hoki.ibp.fhg.de/ibp/publikationen/fachzeitschriften/smart_retarder_2.pdf
http://www.ibp.fhg.de/literatur/konfe/10dbmc_vapour_control.pdf

...and for those with language ability in German:
http://www.hoki.ibp.fhg.de/ibp/publikationen/fachzeitschriften/feuchtesicherheit.pdf

Hard to judge if all this theory really works - time will show. Interesting is that vapour barriers seems to be only a solution for builings with really high humidity impact (ie swimming pool). In all other cases the internal vapour check shouldn't be too much vapour closed. I tend to go for breathable construction without need for expensive variable membranes. I agree with Nick that air tightness is crucial.
Beat

Architype

  • AECBmember
  • Posts: 17
    • View Profile
Re: Unventilated cold roof condensation risk
« Reply #3 on: March 16, 2009, 10:27:15 AM »
Hi,

I was looking into using a similar build up in a small school that has minimal area of flat roof - has anyone found any precedents of this build up being used successfully (ideally with some post-occupancy testing!)

There are concerns in the office about the effect on the Warmcel if any water/condensation does end up within the structural layer....

Cheers!

Tom

Nick Grant

  • AECBmember
  • Posts: 1312
    • View Profile
    • Elemental Solutions
Re: Unventilated cold roof condensation risk
« Reply #4 on: March 16, 2009, 10:38:53 AM »
I wouldn't risk it Tom!

Nick

Architype

  • AECBmember
  • Posts: 17
    • View Profile
Re: Unventilated cold roof condensation risk
« Reply #5 on: March 16, 2009, 10:52:24 AM »
Really? what's your reasoning? seems to be alright for the Germans!

fostertom

  • AECBmember
  • Posts: 396
    • View Profile
Re: Unventilated cold roof condensation risk
« Reply #6 on: March 16, 2009, 12:54:24 PM »
In all other cases the internal vapour check shouldn't be too much vapour closed
My thinking too, but given that I don't currently want OSB on the inside face where it would serve as a moderate vapour resistor so everything outboard can be less resistant, I'm having trouble creating that moderate inboard resistance by other means. How about sloppily installed vapour check plasterboard? How critical is that 5:1 (some say 3:1) inboard to outboard resistance gradient? Why can't I find any papers on this (Warmcel inspired) breathability idea - not even from Warmcel HQ? Seems to be all hearsay. Can a simple interstitial condensation calc be relied upon?

Architype

  • AECBmember
  • Posts: 17
    • View Profile
Re: Unventilated cold roof condensation risk
« Reply #7 on: March 16, 2009, 06:08:19 PM »
So I'm getting the feeling that this hasn't actually been built anywhere - and that at the moment it's pretty theoretical....back to the vented firrings i suppose :(

Mark Siddall

  • Moderator
  • Posts: 1214
    • View Profile
Re: Unventilated cold roof condensation risk
« Reply #8 on: March 16, 2009, 08:18:29 PM »
A series of test roof were built by the Laboratory of Building Physics at K.U. Leuven, Belgium.

This study used tiled roofs. None of the roofs studies any particular moisture problems but the one with spunbonded polypropylene (which combined sufficient air and water tightness with a high vapour permeance) performed best. I've read some papers discussing the energy performance of the roofs studied. Here's the PHD (I've not read it but given that it is titled "Reliable Control of Interstitial Condensation in Lightweight Roof Systems" is may be of interest.)
www.kuleuven.ac.be/bwf/common/data/PhD_1998_Janssens.pdf

Another study examining Zinc roofs was also undertaken:
An evaluation of highly insulated cold zinc roofs in a moderate humid region—part I: hygrothermal performance, Construction and Building Materials. R. Zheng, A. Janssens, J. Carmeliet , W. Bogaerts , H. Hens, 2004

The "discovery" was that with out a perfect air-vapour barrier moisture problems abound! (NOTE: ventilation was provided between the zinc on timber backing and the spunbonded polypropylene membrane.) The perfect air–vapor system kept the moisture content in the wood deck and the rafters of the roofs below 20% by weight, which is usually regarded as a critical value for mould attack on wood products.

Mark
« Last Edit: March 16, 2009, 08:23:57 PM by Mark Siddall »

Nick Grant

  • AECBmember
  • Posts: 1312
    • View Profile
    • Elemental Solutions
Re: Unventilated cold roof condensation risk
« Reply #9 on: March 17, 2009, 04:24:53 PM »
Does that answer your question Tom M?

All feels far to clever both limiting moisture entering the insulation (ideally completely vapour and air tight) but then wanting to let it back out the same way it got in - clever membranes. If roof faces North and is shaded and the RH is a tad high and the builders messed up the membrane . . .

Nick
« Last Edit: March 19, 2009, 05:52:15 PM by Nick Grant »

Architype

  • AECBmember
  • Posts: 17
    • View Profile
Re: Unventilated cold roof condensation risk
« Reply #10 on: March 18, 2009, 09:27:15 AM »
Thanks for your comments everyone! just a couple of things....

In all other cases the internal vapour check shouldn't be too much vapour closed
My thinking too, but given that I don't currently want OSB on the inside face where it would serve as a moderate vapour resistor so everything outboard can be less resistant, I'm having trouble creating that moderate inboard resistance by other means.
Tom, is the reason for not wanting the OSB so that you can achieve a better level of vapour movement from inside to out? Is OSB  too vapour closed?
We're currently suggesting the use of OSB as VC layer on the same school project (in order to reduce cost and attempt a greater level of air tightness-a move towards PH) The exact build up is from outside to in:
Untreated douglas-fir cladding/50mm vent zone/22-60mm bitroc/302mm Warmcel in I beam cassette/15mm taped OSB/25-50mm service zone potentially filled with a cheap generic insulation/fermacell..any comments anyone?

Nick, Yeah, had a scan through the paper and even though it's not dealing with the exact roof build up; the conclusion seems to be that if there is any failure/leakage in the VC layer that problems will occur...

Oh well back to the drawing board : )

cheers all!
Tom
« Last Edit: March 18, 2009, 09:45:14 AM by Architype »

fostertom

  • AECBmember
  • Posts: 396
    • View Profile
Re: Unventilated cold roof condensation risk
« Reply #11 on: March 18, 2009, 09:58:29 AM »
Hi Tom from Tom

Not that I don't want OSB, just not on the inside. My proposed build-up is:
extg 6x2 rafters, blown-in Warmcel or sprayed expand-in-place Icynene between. Pbd and skim beneath, 9OSB over. On top of the OSB, 100 EPS, breather felt laid down-slope. laps clamped by down-slope battens (Mark Siddall's good idea - but not sure now that stands with BBA on watertightness), battens and slate.

No principal airtight membrane, because I'm suspicious of relying on a single membrane and the perfection of its taped joints forever. Instead, a series of failry-airtight things - the blown-in Warmcel with its gapfilling ability, the OSB allowing for imperfection in many of its joints, and the trapped-edge breather felt. Seems long-term robust.

The question is, the water vapour resistance gradient from inside to out. The main resistor, the OSB isn't on the inside but half way through the sandwich. At least it's not on the outside. Matbe I do need a stronger resistor on the inside - perhaps leaky vapour check plasterboard? I'm wondering whether a simple interstitial condensation check would be reliable to OK this build-up. So far I've found no solid technical guidance on the breathing wall principle, even from Excel, just rule-of-thumb/hearsay.

Architype

  • AECBmember
  • Posts: 17
    • View Profile
Re: Unventilated cold roof condensation risk
« Reply #12 on: March 19, 2009, 12:29:11 PM »
Hey Tom from Tom,

Ah ha - ok you're concerned with the position of the pretty vapour impermeable osb half way through your build up. Have you thought about using Paneline/Panelvent from Excel industries?
http://www.excelfibre.com/building/products4.html these are a lot more vapour permeable than osb...still not sure about it's position however, as we tend to use Paneline on the inside layer (with a proclima membrane - INTELLO, I think it's possible to dump the membrane though) and Panelvent on the outside; with a Warmcel filled cassette sandwiched between (we then also use a breather membrane on the outside for belt and braces)


Tom
« Last Edit: March 19, 2009, 12:34:05 PM by Architype »

fostertom

  • AECBmember
  • Posts: 396
    • View Profile
Re: Unventilated cold roof condensation risk
« Reply #13 on: March 19, 2009, 12:44:35 PM »
Sounds like you're not concerned with that 5:1 inboard to outboard ratio of vapour resistance, if putting low-resistance Panel* both inboard and out, or even just inboard. I'm favouring OSB because it does have slightly higher resistance, tho still low. My data http://www.greensteps.co.uk/tmp/assets/1163178050906.pdf says 20thk wood fibre 0.5MNs/g, 9thk OSB 1.95 - just right. Do you think that that 5:1 is unimportant? I know it as the 'breatheable construction' rule-of-thumb, but where can I find the authoritative basis for that?

Would the keepers of AECB wisdom say that the 'breatheable construction' idea, in this form, is unproven/unsafe/just a marketing puff?
« Last Edit: March 19, 2009, 12:52:41 PM by Tom Foster »

Nick Grant

  • AECBmember
  • Posts: 1312
    • View Profile
    • Elemental Solutions
Re: Unventilated cold roof condensation risk
« Reply #14 on: March 19, 2009, 05:57:22 PM »
Tom and Tom

Wires seem crossed.

One discussion is a breathable roof vapour permeable on outside (Tom F) and other is a flat roof vapour impermeable on the outside, this is the scary one and subject of original post and I think Tom M's thread.

nick
« Last Edit: March 22, 2009, 10:26:09 AM by Nick Grant »