Hi Everyone I am new to this forum and found it due to searching for answers for my clients.

I am a roofing contractor here in southern Ontario and today ran aacross 2 types of roofing i need help on.
Both are older buildings 50+  yrs. Both are flat roofs.
Building 1 had a 2x12 rafter to roof deck with r-12 bat insulation vapour barrier and air gap above bat. rafter is direct to underside roof deck(non ventilated)

Client asks about poly iso on top of roof deck before i add new roof, my concern is moving dew points into air gap.I wasnt going to put vapour barrier due to it already being on the warm side.

Roof 2 is different.
It is un-insulated, un vented  rafter under roof deck  then a plaster ceiling rafter seperate but below that so a double rafter attic cavity with a space in between. Client is asking for 3 inch poly iso min. there is no vapour barrier in this roof as it is original horse hair plasterworks. Again what is the best approach to insulate i am ripping roof to board deck. Kraft paper then 3" iso then my 2 ply roof? this is what we do on commercial steel deck roof in factories and such but i am concerned about the attic moisture I can add vents if need be but this will be a direct conduit to outside and the heat will just go out.

Both roofs are in row buildings and in middle units so rafters essentially run direct onto brick walls dividing the units. The second roof described has a neighbour wanting me to do his at same time but he is renoing the unit and gutted all the plasterworks. I suggested he rafter bat and vapour barrier from below and i will apply new roof with no insulation. Should it be best to give him a few vents as i was planning on several small dome vents for this one to extracate any trapped moisture. This is the way I always do them

Dew point is a temperature, not a location. The relevant dew point temperature is the dew point of the interior space air. For residential applications that's presumed to be 4C.

Unvented roofs don't need vapour barriers to be moisture save in most of southern Ontario as long as more than 50% of the total R-value is above the roof deck. In the temperate "Ontario Riviera" lake front regions along lake Erie & Ontario (eg Toronto) that can be safely reduced to 40%. At those ratios the average wintertime temperature of the moisture susceptible roof deck stays above 4C in a heated residential space, and it won't/can't draw enough moisture from the conditioned space to accumulate damaging levels. The roofing material itself is a vapour barrier- the roof deck can never dry toward the exterior, but venting under the roof decks in flat roofs doesn't provide any protection, since it won't purge the cavity air via convection.

Polyiso loses a bit of performance when the average temperature through the foam is under 10C, compared to it's performance at 25C average foam temp. In the US roofing polyiso is typically labeled R5.5/inch to account for that derating, which may be sufficient derating for southern Ontario, but for sure it's no lower than R5/inch in your stackups. So at 3" it would be no worse than R15 average performance during the cold months that count.

It's not clear what your stackups are. Parsing your sentences it sounds like:

---------great outdoors

roofing & underlayment

roof deck

air gap (of unspecified depth)

vapour barrier

R12 batt

Vapour barrier.

finish ceiling

-----------room air

Yes? No? How deep is the gap?

For roof #1 verify the details. If the ceiling is being gutted and there's access to the underside of the roof deck it changes the options.

For roof #2 I'm getting:

---------great outdoors

Roofing layup

roof deck

rafters

big empty space

secondary rafters

Horse hair plastered ceiling

-----------room air


For roof #2 it's perfectly fine to add 3" of polyiso on top without vapour barriers, since literally all of the R is exterior the susceptible wood, keeping the average temp well above 4C.

If code inspectors insist on vapour barriers despite favorable R-ratios of above/below roof deck insulation, use 2-mil nylon (eg Certainteed MemBrain) rather than 4 or 6 mil polyethylene. Sheet nylon at that thickness has variable vapour permeance, and meets NBC code when the air is dry (as it is in winter) but becomes more vapour open when the proximate air is humid enough to support fungus/rot etc, allowing it to dry toward the interior.

thanks so much for the detailed response, it answered my questions but rose 1 more. I have replaced flat roofing in the past and found that when i had deck rot i found the batt insulation touching the underside of the sheathing. So naturally i assume that if there is no air space it will condense. So the question becomes what allowable amount of air space is mininmally required to avoid condense. in an unvented roof rafter.
In the current situation number 1 there is 6 inches easy above so i look good for another 1.5 poly iso on the deck I was not going to add vapour barrier here since it is already under the drywall.but in situation 2 with about 16 inches total air space from deck to bottom of second set of rafters it appears it wont matter at all if 100% of the insulation is above the roof deck.
Thanks again for setting me straight i was under the impression that by creating 2 insulated areas with air gap in between moisture would trap and condense and cause problems for my clients down the road.

and it won't/can't draw enough moisture from the conditioned space to accumulate damaging levels

Unless air movement is blocked, then some common interior winter air %RHs can produces significant moisture levels at 4C. Air seal both sides of bat insulation (but leave at least one side (two is better) vapor permeable).

Posted By paulhutton on 22 Apr 2016 07:31 PM
thanks so much for the detailed response, it answered my questions but rose 1 more. I have replaced flat roofing in the past and found that when i had deck rot i found the batt insulation touching the underside of the sheathing. So naturally i assume that if there is no air space it will condense. So the question becomes what allowable amount of air space is mininmally required to avoid condense. in an unvented roof rafter.
In the current situation number 1 there is 6 inches easy above so i look good for another 1.5 poly iso on the deck I was not going to add vapour barrier here since it is already under the drywall.but in situation 2 with about 16 inches total air space from deck to bottom of second set of rafters it appears it wont matter at all if 100% of the insulation is above the roof deck.
Thanks again for setting me straight i was under the impression that by creating 2 insulated areas with air gap in between moisture would trap and condense and cause problems for my clients down the road.

With the fiber touching the roof deck there isn't sufficient air flow to purge moisture that gets in. With the air gap any condensation/adsorb from interior moisture drives begins on the cold ends of the insulation fibers, but unless the gap is vented to the exterior it will still get into the cold roof deck.  The amount of space required depends on the pitch of the roof, the amount of free air ventilation to/from the outdoors, and amount of air leakage at the ceiling plane.  If it's sealed up drum tight from the outdoors, it's at risk with or without the air gap.

But if you put sufficient insulation on the exterior of the roof deck to keep it's temperature at or above the dew point of the conditioned space air, moisture will not condense on (or more accurately, adsorb into) the roof deck, and no air gap with access to outdoor air is needed.

I agree, but with a well air sealed and not excessively vapor permeable interior side, it takes very little air flow on the exterior side to remove the exiting moisture. This is why things like stucco wrap (very very small air channels) can work. You don't need a steep pitch and a big gap to remove tiny amounts of moisture vapor. But if the materials are moisture susceptible, you do need some air flow.

Preventing ice dams is a separate issue and may be a source of legitimate recommendations for larger roof gaps/more air flow.

Flat vent gaps sound like a drainage nightmare (even though it could move enough air).

"Lots of foam on the cold side" is a moisture resilient design, but so is "breathable to both sides, more so to the cold side".

In the OP's cases, the former is the right choice.

Air gaps between insulating layers dramatically reduces the thermal performance of the building.

Torch down or EPDM flat roofs are not vapor permeable, nor are asphalt shingles on pitched roofs. Most standard roofing materials make drying to the exterior impossible (especially through a foot of snow ! :-) ), which is why roofs are vented from below. (Shingled or tile roofs can sometimes be set up for the roof sheathing to dry toward the exterior.) Roofs are not merely tilted walls. The venting efficiency on a roof is much less than a vented wall, which has convection working for it in a much stronger fashion. With FLAT roofs you get no convection driven air purge of the vent cavity at all.

On the Ontario Riviera assume the exterior is always vapor impermeable, so it's either vent the deck from below with ALL of the insulation below the vent space, or put enough R above the roof deck to keep it sufficiently warm in an air tight unvented roof assembly. In that location it takes 40% of the total R above the roof deck as the absolute minimum, 50% of the total R is better.

There are many problems with the ventilation in flat roofs. But thanks to the roofing contractor who is available anytime to solve this problem by just installing few heaters in the roof.