Warm roof with ventilation VCL needed?

[Stu84]

Morning all

 

Mono-pitch extension with knock through to make open-plan kitchen/diner.

 

Working on design for warm roof.

 

It's a proper warm roof with insulation (130mm) PIR entirely above the rafters.

 

I understand that if I incorporate a VCL below the insulation then I don't need ventilation at eaves/ridge. The thing is I'm not really sold on VCLs. If I install a VCL in a knock-through kitchen isn't there a chance that vapor/moisture gets into the existing dining room and potentially cause trouble there? Where does all the moisture go?

 

The roof will be vaulted. My plan was to install some plasterboard beneath the rafters that would act as a partial moisture control. Then have ventilation at the soffit and where the new roof abuts the wall. And hope that this air flow will draw moisture up through the insulation and on its merry way to the outside world.

 

Problem is most PIR blocks passage of vapor so perhaps that's where the plan fails?  I suppose I could not tape the PIR joins and this would provide a pathway through.

 

Thanks for your thoughts.

 

Stuart

[Mr Punter]

The VCL is to stop warm moist air getting into the joist space and condensing on the joists or deck.  The VLC will typically go on top of the flat roof deck, followed by firrings, insulation and waterproofing.  A warm roof should not be ventilated.

[Iceverge]

It's a confusing issue this. 

 

The vapour permeability of materials and house ventilation are two separate issues really. 

 

Add airtightness in there is you really want to make some soup from your brain. 

 

Vapour permeability has to do with the ability of airborne vapour (from breathing, cooking showering etc) to pass through a material, something like a cotton sheet is high permeability, something like a polythene sheet is low, something's like plasterboard is in the middle somewhere. 

 

Ventilation has to do with taking the stale air outside and bringing in fresh air outside, very important for the health of the occupants and the building. 

 

Airtightness has a dual mandate. It keeps the wind from blowing cracks and gaps and making the house very uncomfortable and hard to heat/cool.

 

The other important one is that it keeps that internal vapour from travelling into the structure of the house and condescending causing rot etc. 

 

 

The polythene VCL in your roof is an air barrier ensuring there are none of these gaps let wind blow through the structure. It should be taped diligently and returned to the airtighess layer on your wall.  It is a vapour blocker too as the microscopic holes in the polythene are too small for gaseous water molecules. Worry about airtighess before vapour as big holes are more of a problem than microscopica ones. 

 

You are right to want to get the stale vapour laden air out of the house. The best way is with continuous mechanical ventilation, something like a dMEV fan is cheap and very effective. Trickle vents on windows and extractor fans might work in theory with very diligent occupant's but in practice it doesn't.

 

Don't allow it pass through the roof as it will condensate and cause issues. 

 

[Iceverge]

1 hour ago, Mr Punter said:

 A warm roof should not be ventilated.

 

I think we're storing a mountain of problems with unventilated flat roofs in the long term. 

 

The grp/edpm is the only barrier to water ingress.  Its waterproofness is completely reliant on its total integrity. Being exposed to the temperature variations of the sun and weather it will inevitably fail at some stage. The first thing the occupants will know, is water will emerge from somewhere in their roof, with zero clue to where it started. Maybe months after it has failed, in the meantime causing significant damage. 

 

We all know how sloppy builders can be with membranes so I bet most VCLs are full of holes or nobody's taped the joints properly. The layer of OSB under the is GRP is the ultimate stopping point for the internal vapour to condensate. I think these will start to rot out and nobody will know until the roof leaks. 

 

I would far prefer something like this. 

 

GRP/EDPM

18mm OSB

25*75mm battens across the joists. 

25*75mm battens up the joist with the void fully ventilated at both ends. 

Breather membrane 

Rigid insulation.

Furring strips.

18mm OSB diligently taped as airtightness. 

Joists . 

Plasterboard 

 

I would also happily install insulation between the joists if I was doing it myself for a hybrid roof. However I  understand the issues and am picky about details. 

 

Edited Saturday at 11:55 by Iceverge

[Iceverge]

Also I would prefer something like metal cladding as instead of EDPM or GRP as it isn't weather dependent for install and as vulnerable to imperfect workmanship. 

[Mr Punter]

1 hour ago, Iceverge said:

The layer of OSB under the is GRP is the ultimate stopping point for the internal vapour to condensate. I think these will start to rot out and nobody will know until the roof leaks. 

 

 

That should not be there on a proper warm roof.  No timber on the cold side of the insulation.

[newbuild upnorth]

9 minutes ago, Mr Punter said:

 

That should not be there on a proper warm roof.  No timber on the cold side of the insulation.

 

what's a proper alternative though, in the instance of EPDM or GRP, im not convinced straight onto PIR is a good move.

 

I would rather have EPDM or GRP on OSB or ply personally.

 

assuming the VCL is a good one and fitted properly, assuming the PIR is foil faced and taped, and assuming the fixings holding it all together are thermally broken where is the moisture coming from to cause the interstitial condensation that is going to rot the top layer of treated wood?

[Stu84]

Massive thanks for the replies.

 

1 hour ago, Iceverge said:

Airtightness has a dual mandate. It keeps the wind from blowing cracks and gaps and making the house very uncomfortable and hard to heat/cool.

 

The other important one is that it keeps that internal vapour from travelling into the structure of the house and condescending causing rot etc. 

 

The polythene VCL in your roof is an air barrier ensuring there are none of these gaps let wind blow through the structure. It should be taped diligently and returned to the airtighess layer on your wall.  It is a vapour blocker too as the microscopic holes in the polythene are too small for gaseous water molecules.

 

 

> Is some vapor movement into a roof structure really an issue if there is some ventilation?  I would go further and say perhaps it is beneficial for a certain amount of vapor to escape into the roof provided there is some low>high air flow through the roof? 

 

 

1 hour ago, Iceverge said:

You are right to want to get the stale vapour laden air out of the house. The best way is with continuous mechanical ventilation, something like a dMEV fan is cheap and very effective. Trickle vents on windows and extractor fans might work in theory with very diligent occupant's but in practice it doesn't.

 

Don't allow it pass through the roof as it will condensate and cause issues.

> If you have a completely airtight structure how does fresh air enter the building when windows etc are closed?  dMEV is a great idea, but it seems like a fix for a problem that shouldn't exist in the first place if house designed well.

 

I'm engaging with a designer.  He initially said a VCL is a must and that I shouldn't have ventilation with a warm roof.  I think the problem is that warm roofs are a modern concept and have primarily been spec'd for flat roofs which don't normally have ventilation.  Much of the 'warm roof' documentation out there relates to flat roofs and people are applying that method to all designs.

 

Call me old school and I could be wrong, but in my scenario (an 18deg mono-pitch extension) I see only positives from having some low>high roof ventilation.  I can't see why that's a bad thing ever.  It might not technically need ventilation if you used a VCL with a high Sd value.  But why chance it all on a membrane?

 

And it seems as though (at least as far as the average homeowner is concerned) trapped vapor / stale air that could result from VCL is a bit of an after thought and maybe a problem that could arise in 15 years time when these things have been truly tested in various scenarios.

 

NHBC chapter 7.2 deals with ventilation for pitched roofs.  Table 14 (see attached) suggests that where a VCL cannot be used then ventilation is acceptable.

 

Admittedly it seems worded in favor of a VCL but doesn't seem absolutely necessary if have ventilation.

 

I'm wondering if the ultimate solution is to have a VCL with a low Sd value (i.e. it's not a barrier but allows some vapor to escape into the roof) and combine this with some low > high roof ventilation.  In my scenario space is very restricted, but I can have 10,000mm^2/m at the eaves and 5,000mm^2/m where the roof abuts.  I'm hoping that this combination will allow the escape of some condensation/vapor through the VCL, but not enough to cause damage as the air flow will help dry.

 

I see little wrong with the above until..... enter PIR!  I have to use PIR due to required 0.15 U-value and restricted vertical height.  But (please correct me if wrong!) I don't think PIR will allow for vapor transfer very easily, so there may be a risk that using a low-valued VCL causes condensation to rise and hang around/within the PIR layer damaging it....

 

Ah tis a challenge!

 

Thanks for knowledge please keep sharing.

 

Stuart

 

 

 

[Stu84]

Again everyone thank you for replies!  Just a friendly reminder that this concerns warm pitched roofs (sorry should have stated in subject title).  Thanks.  Stuart

[Mr Punter]

2 minutes ago, Stu84 said:

Again everyone thank you for replies!  Just a friendly reminder that this concerns warm pitched roofs (sorry should have stated in subject title).  Thanks.  Stuart

 

Ahh OK.  I assumed it was a shallow pitch lean-to, thus treated as a flat roof.  What is the roof covering?  Why are you against eaves and ridge ventilation?

[Stu84]

Pitch is 18 degrees. So don't think that counts as a flat roof?

 

Not against eave and ridge ventilation I just don't have the horizontal space (yes really) to have 25,000mm/m^2 grills and therefore can't meet building regs but can fit 10,000mm/m^2 which may be better than nothing.

 

Think main question remaining is will PIR allow any vapor/moisture to move through? If not then little point in ventilation as moisture won't even get past in the first place. 

 

If having no ventilation and a high-rated VCL I'd go with an LR underlay to allow small amounts of vapor that may get through to escape.

[Mr Punter]

Yes your 18 degrees would count as pitched.  Any chance of a section drawing of the eaves and ridge?  Also what is the roof material?

[Iceverge]

23 hours ago, Mr Punter said:

 

That should not be there on a proper warm roof.  No timber on the cold side of the insulation.

 

It's there in pretty much every detail. That comes up on Google so I assume that's how the majority are built too. Ply or OSB above the insulation. 

 

I think you can get some types of insulation with specific fleeces bonded to one side to take grp/EDPM. 

 

 

 

 

Edited Sunday at 13:17 by Iceverge

[Iceverge]

23 hours ago, newbuild upnorth said:

assuming the VCL is a good one and fitted properly, assuming the PIR is foil faced and taped, and assuming the fixings holding it all together are thermally broken

 

Reality unfortunately gets in the way of these assumptions. Mistakes and sloppyness are a part of building and I think any design should be somewhat tolerant of non perfect workmanship. 

[Iceverge]

23 hours ago, Stu84 said:

Is some vapor movement into a roof structure really an issue if there is some ventilation?  I would go further and say perhaps it is beneficial for a certain amount of vapor to escape into the roof provided there is some low>high air flow through the roof? 

 

 

I would argue this is exactly what you need to avoid, primarily through good airtighess. 

 

Buildings die when water (in whatever medium ) gets into the structure and cannot get out. 

 

Stop all drafts taking vapour through a structure. Also at all costs avoid sandwiching materials of very low permeability.  That's why I don't like the polythene VCL, Foil faced PIR and EDPM all together. If water gets in then it's never getting out. 

 

Always better to stick to one impermeable layer maximum.

 

 

23 hours ago, Stu84 said:

> If you have a completely airtight structure how does fresh air enter the building when windows etc are closed?  dMEV is a great idea, but it seems like a fix for a problem that shouldn't exist in the first place if house designed well.

 

Are you suggesting that just having poor airtightness is a solution to ventilation? 

 

That's much the same as having a hole in the roof as a solution to getting drinking water into your house. 

 

Beware your designer and NHBC aren't gospel either. It's a woefully misunderstood topic this. 

 

Stick up a section of your roof and we'll have a look at it. What roof covering is on your current house? 

[Stu84]

Useful info thank you all.

 

Sorry it's taken a while to post back.

 

I've attached a section showing the roof build up:

 

Renown concrete tiles

battens

underlay (HR or LR?) draped over counter battens but with a clear gap above the PIR for airflow.  (Underlay not really visible on drawing attached)

40mm counter battens

130mm PIR

125mm C24 battens

Regular plasterboard on underside of rafters (not shown)

 

Guess any VCL would have to sit either immediately below the PIR or perhaps above the plasterboard...  That said many reckon foil-backed and taped PIR basically is a VCL.

 

Full plans are being drawn as we speak so excuse the total lack of detail shown at the eaves.  Indeed, I'm not even sure if I can possibly set up a good air flow with the way its currently drawn, what with the air needing to get into the low>high channels formed by the counter battens, this would be difficult given the vertical distance from the air entry point (tiny 10mm gap between rear of fascia and wall) up to the counter battens.

 

I'm starting to think a VCL with no low>high ventilation may be the easiest way to go.

 

7 hours ago, Iceverge said:

 

I would argue this is exactly what you need to avoid, primarily through good airtighess. 

 

Buildings die when water (in whatever medium ) gets into the structure and cannot get out. 

 

Stop all drafts taking vapour through a structure. Also at all costs avoid sandwiching materials of very low permeability.  That's why I don't like the polythene VCL, Foil faced PIR and EDPM all together. If water gets in then it's never getting out.

 

I'm just saying that the roof structures of older houses (80s for example) I've been in where there is air flow seem OK. And they haven't had VCLs installed nor LR underlays so air ventilation seems effective enough in many instances to solve rising vapor.  But I appreciate that we are now trying to retain more heat etc.

 

Worrying but no doubt correct thoughts regarding foil-faced PIR.  This is why I'll probably end up going with a high-valued VCL as worried the PIR won't allow movement of any vapor through it should some get in.  Mind you... I did see a PIR product by Unilin that claimed to be vapor-open.  Perhaps that would be lower risk?

 

Thanks all

 

Stuart

 

 

RN22TOP1 - Top Edge Abutment (Ventilated).pdf

[JohnMo]

You have a very thick roof profile with minimum insulation. Have you looked an outer membrane that requires no ventilation air gaps, then use a deeper profile rafter, and full fill with 250-300mm of mineral wool? Then underdrawing with 25mm PIR or insulated plasterboard? Way more simple to construct and less reliance on the workers doing a good job of installing the PIR into all the gaps.

[Iceverge]

Top down, 

 

Tiles

38mm Tile Battens across the rafters

25*75 Battens up the rafters

Glidevale VP400 membrane. 

11mm OSB taped at all joints. 

200*44mm rafters at 600cc full fill with blown cellulose ( or mineral wool batts)

Airtight membrane,

45*50mm battens at 400cc across the rafters with 50mm mineral wool in the service cavity . 

 

12.5mm plasterboard a skim.

 

U value about 0.17. 

 

 

 

 

Edited 10 hours ago by Iceverge