Posting from public library, so can’t put graphics into this post. Hope to add graphics later.
A residence has two roofs, one on top of the other. There’s a gap of what… 5 feet?… between the two roofs. The top roof has a hole cut into the vertex. A HAWT rotor that spins about a vertical axis is positioned in the hole.
The rotor is driven by two mechanisms – the ram air force that creates high pressure between the two roofs, and the suction created by wind that has to accelerate to go over the top of the roof. Imagine you are looking at the side of the house where you can see both halves of the roof sloping down at about 45 degree angles from the vertex. From this perspective, the vertex of the roof looks a little like the low pressure surface of the fattest part of an airfoil. The roof is smoothly curved to prevent turbulence and to maximize the low pressure that is created when wind flows from the right to the left, or from the left to the right. The top roof extends further from away from the outside walls of the house than does the bottom roof. This way, when the wind hits the side of the house, it will be diverted upward (on a nearly vertical path?), and the protruding lip of the upper roof will catch this “ram air” and direct it into the high pressure region between the two roofs. The lip is curved so that its edge is nearly vertical. This prevents the wind that goes over the top of the top roof from generating turbulence and vortices as it is forced to make a nearly 45 degree angle turn to conform to the top surface of the top roof.
As you probably already know, rooftop wind turbines usually don’t perform well because there’s too much turbulence over the roof for a typical wind turbine. I’m wondering if, properly designed, this idea could solve the rooftop turbulence problem. The worst turbulence is probably ulitmately generated by the wind that first “hits the side of the house”, but in contrast to a traditional turbine mounted on top of a house, the turbine described here actually makes use of this increased pressure, and controls the resulting flow so that it doesn’t result in a lot of turblence or vortices.
There are of course many details as well as possible variations not described here. Hope to fill in some more of this info later. In the meantime, note that a round house with a dome shaped roof might work best. Also, it seems like this idea might work for rooftops of retail buildings… like on the roof of a Walmart or something like that.
Okay folks, here’s a simple pic:
Maybe a hinged diffusor could be added to increase the amount of suction produced by the wind that flows over the top of the top roof: