In reply to "what is a heatilator", see the company's link at http://www.heatilator.com/. My parents had one in the house I remember as a child, from the 1950s. The simplest description of that one, which appears to have been comparable to the one in this article, would be a gravity furnace in the form of a fireplace. Ours had vents on both sides near the top and bottom of the fireplace opening, and the heat of the fire would encourage convective air flow around the firebox--it greatly increased heating of the room when there was a fire burning.
I can see that this scheme would work--but realistically, if we currently had a heatilator and I tried to install something that looked like the pictures for this, my wife would remove it!
I burned wood for years. Rebuilt an old coal parlor stove and modified it to be similar internally and air control wise to what a better Franklin would have.
Splitting the 3 to 5 cords a year to feed it was some of the best exercise I've ever had. Harvested some of that wood myself too from local woodlands, great fun on a good day. Not too bad ever, but then that's just me.
I bought 5 acres 5 years ago and one of the first things I did was plant some fruit and a weeping willow.
Mixed results on the fruit, but the willow went from being a 4 ft high stick to 15 feet of lush in those 5 years. The water table is only a couple feet down all year round.
Talk about maximizing your heat efficiency! @ 78RPM, some emergency room hospitals have something similar for their drive paths, which use heated water that circulates through piping underneath the asphalt to keep the path ice free in winter months.
Thanks for the link. From the looks of your circuit and your construction, it looks like you're an old-timer, like me. I recognized some really old capacitors, there. I take it you constructed this circuit a long time ago. I am also a geezer, who has been designing electronic circuits since adolescence.
I think PUTs are still found in motor speed controls for small appliances and possibly in light dimmers. SBSs are still available too and used in these applications. I used the same PUT 7 years ago in a more conventional-looking topology: <http://www.designnews.com/author.asp?section_id=1386&doc_id=219772>.
Well, the photo DN chose for the title could be made worse only by showing all the insulation packed around the fan in the final installation. When they finally get the video up, you'll see what it looks like installed. Could have made it fancier if I'd considered consumer appeal, but the really ugly parts are hidden in the air duct. You'll see those too in the video, "warts and all"!
Using a MOSFET could improve efficiency if I were using a switch-mode output, but this is a purely analog Class A output stage. Any series device (BJT, MOSFET, rheostat, etc.) would waste the same amount of power at any given operating point. I considered whether I wanted to add the components to make a PWM output, and noted that (1) the max dissipation in the output stage is around 6W, and it's only used during the heating season, and (2) the cheap transformer I used (can't speak for the one I specified since I didn't test one) wastes more power than that. The fans have brushless DC motors, so can't run directly on PWM output - I'd have to add an LC filter.
The flaps do add noticable aerodynamic load at low speed when they're barely open, but at high speed I doubt the load is significant. Bear in mind they're held closed by magnets, not springs - once they're free of the magnets it doesn't take much to hold them open. The added load at low speed does have one useful consequence: it extends the flow rate range at the low end, since the brushless DC fans can't be electrically slowed down any further. You'll see the flaps in operation if DN ever gets around to publishing my video.
As for using solenoid-controlled louvers, that's worth considering for a couple of reasons. When the fire first gets going, there's about a 5-15 minute wait before the fans start up. The reason is with the flaps closed, the normal convection path is completely blocked and the warm-air sensor gets heat mainly via a secondary circulation path. Air goes in the bottom of the top register, warms up a little and comes out the top, where the sensor is. If I had a way to manually trigger the flaps/louvers, that delay would be much shorter. I can also shorten it by blowing a puff of air into the bottom of the top register. Another advantage, assuming they can be triggered manually, would be seen during a power failure. With my design as published, I'd have to either provide battery power, prop the flaps open with sticks inserted through the registers, or take the bottom registers out.
The hysteresis is only to prevent chattering on and off when near the threshold. There's actually an inherent feedback loop involving the temperature sensors, which is negative feedback during steady-state or pseudo-steady-state operation, but positive feedback when the flaps open or close. The latter normally prevents unwanted on/off cycling. As I mentioned above, when the flaps are closed the warm-air sensor doesn't get much heat. The moment the fans start and the flaps open, that sensor starts to heat rapidly. Likewise when shutting down, the moment the fans stop and the flaps close that sensor cools down further for lack of warm air flow. The heat exchanger warms up a little but the warm air inside can't go anywhere. In fact, if I manually insert a stick to open the flap on that side, the fans will start up again in a couple of minutes (or try to, the sitck is blocking a fan). These events involve positive feedback from the temperature sensors.
Negative feedback occurs when the system is near steady-state operation. Let's say the fire has died down and only a small flamelet or a few embers remain. Heat is still being produced, slowly. The temperature difference decreases with increasing fan speed. This may need elaboration for some readers. Increasing fan speed increases the heat transfer rate (HTR). HTR is the product of temperature rise, mass flow rate, and specific heat of the medium. Using mass rather than volume flow rate avoids the need to think about the air density change. The sensors are monitoring temperature difference (rise), not HTR. The temperature difference decreases because (1) the air spends less time in the heat exchanger, and (2) with heat being removed more rapidly, the heat exchanger itself runs cooler. The fan speed seeks an equilibrium that balances this loop.
The above points out another advantage of using solenoid-switched louvers. When there are only a few hot coals left, it may be preferrable to have the fans start back up after the heat exchanger warms up again. It would if the flaps didn't close immediately when the fans stop. Louvers could be set up to close after a delay.
It will be interesting to see how other builders incorporate all this input!
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.