Non-vented metal roof on insulated deck – ice dams?
Eric_Powell
| Posted in General Questions on
Does anyone have any thoughts about metal roofing above an insulated unvented deck – for ice dams in snow country?
What I have: rustic cabin with 2×6 tongue and groove boards on top of rafters that serves as both ceiling and roof deck. Presently have shingles on top of this and am experiencing ice dams and leaks when the space is occasionally heated and used in the winter. Roof has hips and valleys and some big overhangs.
Location is in the mountains in NY state, there can be a foot of snow or more on the roof and a few feet of snow on the ground.
I am proposing to remove the shingles and install ice and water shield over the entire deck, then 2″ of rigid foam, then 1×4 furring 24″ OC screwed through foam into deck, then 26ga 1-1/4″ rib exposed fastener metal roofing panels, and using closures at the ridge, eaves, hips and valleys to make a an airspace under the metal that is not vented to the outside.
I am seeking to minimize the weight of the snow on the roof due to the building’s lack of rafter ties, and am thinking that by not venting the airspace between the metal and the foam, it will help melt some snow without creating ice dams because the temperature of the airspace will somewhat equalize due to convection – the overhanging eaves would be warmer than otherwise due to the airspace and insulation.
Does anyone have any thoughts or experiences about this?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
> the overhanging eaves would be warmer than otherwise
Agreed, but I'm pretty sure that it won't be enough to prevent ice dams.
You may be able to box in, insulate and heat the eave area. Then the eave upper surface will receive about the same amount of heat as the rest of the roof.
Jon thank you for your reply. I will build it as a vented roof.
How well that does with ice dams depends on the R value and how much air the vent moves. The idea is that there will be so little melting of the roof snow (from building heat) that it won't cause a problem when it refreezes on the cold eaves.
Eric,
Jon R is right about vent rates and R values. I have metal roofing on an unvented insulated roof deck. It is however an R54 profile.
If you are only putting 2" of foam over an existing roof deck in the NY mountains, I hope it is the Poconos not the Saranac Lake region. Even with limited periods of heating, the losses through the roof will likely warm up the underside of the metal the longer you are there. The eaves will not be receiving any heat when you are being cozy inside, so the venting you now propose might keep things tenable. Or it might not. If the ridge point is capped in snow you will not have much flow. The valleys will be tricky to vent as well. The large over hangs will be cold either way.
The water shield over the whole deck is paramount to surviving any ice dams that occur. The over flow height for face screwed materials is often pretty low. You might consider placing the sleepers on a diagonal pattern or inserting thin pads under the anchor points to allow any water to slide down the foam face. I would really recommend more foam - 2" at least for less heat loss. Do also consider your screw length choices very carefully. Over driven screws peeking thru the ceiling won't enhance the rustic look.
One thing you might consider as well is contacting an engineer to advise on alternate forms of rafter ties you could use to make the roof less risky. I have seen cabin style structures with cables acting as rafter ties which sorta melt into the background more than a 1x8. Some timber frame situations do the same thing. Turn a bug into a feature. It may even work out that only every third rafter needs a cable.
The snow on my roof is generally 6-10" drops though we have has 28" a few times. There is no venting under the snap seam type panels and very little heat loss from the house. My two foot plus overhangs are virtually the same temperature. The snow generally evaporates here due to the very low humidity and bright sun. Sunlight warming up parts near the peaks will more often determine melting activity.
When we are hitting spring, the melting can get advanced enough for the snow to just slide off on a thin film of water. The bottom of the snow overlay may be a thin crust of ice, but the smooth metal seems to release it pretty well. I did not put snow brakes on, partly to avoid possible icing issues, but mostly I didn't need to be careful of unexpected avalanches killing someone coming to the front door. Just think about where stuff drops before being stampeded into snow brakes.