Quote Originally Posted by mark beiser View Post
I've played around with a few water heaters in attics that have one or more PAVs.
On the ones I checked out, the draft hood provided enough of an exit point for the air being pulled down the over fire draft in the water heater wasn't reduced all that much.
Under the right conditions, like in a case of a severe backdraft, I can imagine the velocity of the air coming down the vent pipe being high enough to stop the over fire draft and cause flame rollout though.

One house I found had 3 PAVs, 3 water heaters and 3 furnaces in the attic, with absolutely no soffit ventilation. The water heater vent pipes were backdrafting bigtime, the airflow was actually clearly audible, but the water heaters still had enough over fire draft to work. The 80% furnaces were backdrafting so bad that the draft inducers couldn't close the pressure switches though, lol.
The house itself was being pulled into a 24 Pa negative pressure!
I suppose, then, that the PAV's were pulling the stack gasses out through the draft hoods, and since the house was 24 Pa negative to atmosphere, there was little risk of stack gasses entering the house...correct?

Even so, what in your mind would create a backdraft severe enough for rollout to occur with PAV's in operation? A sudden gust of wind with the stack on the windward side of the roof?

I had a separate thought today regarding soffit venting/ridge venting or whirlybird venting near the peak of the roof. I have stood under soffit vents on hot days and felt hot air spilling out of them. I can't recall at the moment how windy the day was when I observed this, but probably not all that high. The house in my memory where I last observed this occurring has a hip roof with soffit vents all around, and whirlybirds near the peak.

I see two causes, and your experience can either confirm or nullify my thinking...
  1. With little to no wind, the heated air in the attic expands to the point where it drops out the soffit vents. To do this it would need to draw from the whirlybirds or ridge venting, or from the house interior, unless it merely dumps at any available opening due to sheer expansion. Regarding the house interior, being that it is cooler and the air is more dense (hence a slightly lower pressure), the "reverse stack effect" through ceiling penetrations would conceivably allow attic air to enter the house. The point of make-up air for the attic air as it heats, expands, and dumps out the soffit vents would need to come from somewhere.
  2. With adequate wind pushing passing over the roof, air could conceivably flow down ridge venting or whirlybird openings and then pressurize the attic, hence causing heated air to dump out the soffit vents. But this same air is also pushing against windward facing walls, whereby air would be encouraged to flow into soffit vents due to pressure difference between windward wall and attic conditions. Hence, it may cause hot air to dump out of soffit vents on the leeward side of the house, and any roof venting on the leeward side of the roof.
Food for thought, yes?