Roof diaphragm spaced from rafters

[Pichy]

I am interested in building my own house in the next year or so, and would like to minimize thermal bridging at framing members as much as possible. Is it structurally feasible to apply 2 1/2" rigid insulation over the rafters, and then screw 5/8" plywood over that, through the insulation, and into the rafters? This moves the diaphragm away from the rafters by the thickness of the insulation, and really puts the fasteners to the test. I would have 5/8" gwb directly on the underside of the rafters, and could even run galvanized strapping diagonally both ways to provide racking resistance. Where I live (central NY), wind generally governs, not seismic.

 

[MiketheEngineer]

I am sure this has been thought of before but I have never seen it done. I do indeed think that you may have some fastener problems. Also, you probably won't meet the local building codes and what happens when you go to sell??

You might look into SIP panels - foam glued and sandwiched between two sheets of plywood or OSB. The reputable companies have tested them and gotten various approvals.

 

[RHTPE]

Structurally, it would be better to attach a layer of sheathing directly to the roof framing, then bond a layer of rigid insulation on top of the sheathing, and then bond another layer of sheathing on top of the insulation - essentially creating a stress-skin panel. But there are other things to consider as well. How much heat loss will you really prevent? How will this kind of framing affect the life of your roofing material, as there's no ventilation directly under the sheathing to help keep the roofing material cooler in the Summer? Are you trying to create a cathedral ceiling below? Savings from one consideration creates costs in another.


Ralph
Structures Consulting
Northeast USA

 

[Pichy]

I considered SIPs panels, but using singular components would be easier for me to build myself than something that requires a crane. I also had hoped I would not have to put two layers of sheathing on the roof. Many roofs are designed to be without vents. Take SIPs for example. No vents needed (or possible, without another layer of sheathing). The key is to prevent moist air from moving through the assembly. And the rigid insulation effectively moves the first condensing surface away from the winter air. The roof, as I described initially, would be a "super insulated" roof. I have done retrofit roofs where I have added rigid insulation over the top of existing board sheathing, and then covered it with 2x4s flat, screwed through, and then plywood and shingles. This provides a "cold roof", which has advantages as well. But the whole diaphragm issue is a bit different in that case. I also wanted to explore this concept for use in low income housing, using my own house as a guinea pig. I checked codes briefly and did not find anything that precludes it. So if I wanted to pursue this, I suppose I will have to calculate shear in the fasteners...and of course get the local building inspector's blessing.

 

[RHTPE]

Pichy - You have to consider more than just shear in the fasteners. Rigid insulation is still compressible, so you will actually see shear & bending in the fasteners when the entire diaphragm resists lateral loads from wind or earthquake. I suspect the entire diaphragm will tend to "loosen up" over time as a result. I appreciate your goal, but I'm skeptical as to its durability.

Ralph
Structures Consulting
Northeast USA

 

[msquared48]

Put the rigid insulation over your roof sheathing material.

The difference in cost is not that much, but the structural performance MUCH BETTER.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[JAE]

It's not only better, I think its a must. There would be essentially NO diaphragm action with 2 1/2" gaps between the hard surfaces.

Not a good idea. I'd put sheathing down, then the insulation, then a hardboard over that.

 

[mfrad]

Besides no diaphragm action it is not a good idea to walk on top of rigid insulation. You may fall through.

 

[RHTPE]

Goog point mfrad - sometimes the obvious isn't. Constructability is essential for economy.

Ralph
Structures Consulting
Northeast USA

 

[Pichy]

Thanks for the replies, guys. Good points. Although walking on the roof would be ok, as the shingles would be on plywood directly. So no danger of falling through plywood. But certainly the compressibility of foam, having bearing only on the 1 1/2" wide rafters would be a concern over time. Time to cave, and add another layer of plywood. Thanks again.

 

[RockEngineer]

I think mfrad's point was that while building you will be walking on the rigid insulation when you are putting the plywood down.

 

[ornerynorsk]

Thermal conductivity of wood is very low to begin with. Your interior ceilings are a better place to begin with the heat issues rather than the critical structure.

Reference link for you.

http://www.engineeringtoolbox.com/thermal-conductivity-d_429.html

 

[Pichy]

Thank you for the link. Generally speaking, wood does have low conductivity, and certainly compared to most building materials. But when used as repetitive framing members, the way most houses are built, the cumulative effects of thermal transfer are significant in a very cold climate.

 

[boo1]

Consider sheathing directly to the trusses and two-component spray polyurethane foam insulation at the under side of the sheathing.