Q. Is it necessary to provide ventilation under shingles installed over SIP (structural insulated panel) roofs? I’ve heard that shingles installed directly over SIPs, without a vent cavity, fail prematurely.

A.Bill Rose, a researcher with the Building Research Council of the University of Illinois in Champaign, responds: Unfortunately, the best answer I can provide at this time is "I don’t know." I am ten years into a side-by-side comparison research test, and so far the performances of the fully vented roof and the unvented SIPs-equivalent roof appear to be about the same.

Our research has turned up other relevant information, though. In a vented cathedral ceiling (SIPs or not), with a continuous slot, say 11/2 inches deep, with openings top and bottom and the sun shining on the surface, air will be buoyed upward in the cavity. The air flow may be considerable. It has a distinct cooling effect on the lower part of the roof. However, the engine that drives this cooling is the temperature difference between the top and bottom of the slot, and the upper part of the slot may be quite hot. In fact, our measurements indicate that the top of the vented slot is at about the same temperature as an unvented cavity. Conclusion: In a long shallow vent slot in cavity roof construction, venting can cool the lower part of the roof but not the upper part.

We’ve also found that a vented cathedral ceiling cavity is much hotter than an unvented full attic, other things being equal. In addition, our research shows that some ridge vents are so well designed to exclude snow that air barely moves through them.

So far I find little technical support for the claim that ventilation enhances the service life of shingles. One argument — that higher temperatures cause faster diffusion of the compounds that keep shingles pliable — makes sense, but the temperature difference between vented and unvented is awfully tiny. Shingle color and latitude make for a much greater temperature difference. In the absence of technical support, I believe that either course of action — venting or not — is viable. Those who vent encounter fewer problems with code officials and shingle warranties. Those who do not are working in the fine tradition of craftspersons who use common sense to challenge accepted wisdom.

Venting has been presented as the key to blissful service life of shingles. My opinion is that service life is first and foremost a matter of shingle quality. If ventilation has one ill effect, it is that it gives shelter to manufacturers who are not investing in product improvement.