Hot Air Doesn’t Always Rise

When we were designing the house I wanted a recirculating duct from the living room on the parlor floor to the bulkhead 4 flights up. I figured in the winter we could pull the hot air from the top of the stairs down to the parlor, and in the summer we could push cool air up where it was hot.

Problem was, the architect wasn’t too enthusiastic about the idea, and his mechanical engineer told me air in buildings didn’t work that way – that the hot air wouldn’t rise. That contradicted everything I had been taught in grade school. He said hot air only rises when there’s moving air. The air in a well-sealed house with the windows shut doesn’t move.

I didn’t believe him and when we got to the end of the project the recirculating duct was one of a short list of things I felt like I would have done differently if I could do it all over again.

Well, who’d have guessed – but the mechanical engineer was right. It’s winter and the top floor is at least 5 degrees cooler than the lower floors. It’s not because of poor insulation – we’ve got an R62 roof and R22 walls and great windows. At first I thought it might be just be that the room where I was when it struck me had had the door closed. But then I moved into the stairwell and it was the same there.

In the summer the area up in the bulkhead would get incredibly hot, but apparently that was because of solar heat gain from the large bulkhead windows (and glass door). But in the winter there’s less solar heat gain even – though the bulkhead is designed to capture winter light and discourage solar heat gain in the summer.

Now I’m perfectly happy the recirculating duct wasn’t put it. The air is cold up in the bulkhead – there’s no point of pulling cold air down into the living space in the winter. And likewise with the summer – we don’t need the bulkhead air conditioned.

The moral of the story is engineers know what they’re talking about. Who’ve guessed? lol

2 thoughts on “Hot Air Doesn’t Always Rise

  1. Hi, Jay!
    I am glad to hear that you obviously survived Sandy without major trouble, since otherwise air circulation would have been the least of your worries.
    While I am not familiar with the R ratings, as it is counted differently here, I suspect is that what you are describing may be actually heat loss at the top floor- even with the best insulation some heat loss will occur, it is just a question of how much.

    • The insulation we have in our roof (R62) is typically what “zero energy” homes have and the ceiling is the only thing that’s different from the floors below where it’s warmer.

      The walls have considerably less insulation – I think zero energy homes typically have R40 walls, compared to our R22. However, we have a warm building next to us on an uninsulated brick party wall – so typically what we lose through exterior walls and windows is less than what we gain through the party wall.

      If you’re used to insulation in U values – U is roughly the inverse of R. So R62 = U 0.016, and R22 = U 0.045.

Leave a Reply

Your email address will not be published. Required fields are marked *