Andrew Morris was frustrated by the inadequate thermostats on most space heaters. The thermostat is typically built inside the heater and has little contact with the outside world. It switches on and off according to the temperature inside the heater, not the ambient air temperature. This makes it little more than a duty-cycle controller, similar to the nonthermostatic control on an electric stove or the temperature control on an electric blanket.
Morris decided to design an electronic thermostat that senses the actual room temperature. He tested and tweaked it to get the proper control range and hysteresis. The new thermostat has an internal adjustment to set the center temperature. It has a range of 20F, centered at 70F. In creating the new thermostat, he did not alter any of the safety functions in the heater.
Wall mounting is more convenient to use and gets a better sense of the room temperature.
Andrew Morris' gadget saves electricity by using the actual room temperature to control the heater.
I have built three of these and they serve my needs quite well, but I will accept no responsibility for any injury or damages arising out of the construction of or the use of this gadget. BUILD AND USE IT AT YOUR OWN RISK! I was not aware of it when I designed this gadget, but there are commercial, UL approved products available that do what this device does. This gadget posting is for educational purposes only. Use any good ideas found in it, but DO NOT BUILD IT! Use of this device may void your fire insurance.
Thanks! I mounted the sensor inside the box because the circuit is not isolated from the power line. The circuit does not require a regulated 24 volt power supply. The power supply voltage only affects the heat generated by the 6 volt zener regulator, being used as a thermal air pump. Using a microcontroller and a thermistor would not affect this situation. If the sensor is mounted outside the box, or fast sensing response is not required, the 6 volt regulator can be moved to the power unit and the problem goes away entirely.
I frequently work with PIC microcontrollers (see GF#192 and GF#198) and could have gone that route. If I can do the project simply without a microcontroller, I prefer to do it, because not all Gadget Freak readers have the means to program a microcontroller.
This analog circuit is closed-loop and is capable of very high precision.
A microcontroller with a thermistor circuit definitely would resolve the voltage regulator temperature sensitive because temperature monitoring closed loop detection can be embedded within the software code. The closed loop circuit will definitely impact the project BOM cost. Very nice project though!
Note that when the unit is first plugged in, you must wait at least a half hour for the temperature setting to stabilize. The voltage regulator circuit beneath the hole in the top of the box is located there in order to pull ambient air through the box, but it also heats the box up a bit. This "thermal air pump" was created to speed up the thermostat's response. This wasn't an issue when the circuit was an open breadboard, but enclosing the sensor in the box greatly increased the delay and therefore the temperature overshoot (using a forced-air heater). With the thermal air pump, the response of the thermostat is quite good. With a slow heater, like an oil-filled radiator, the thermal air pump won't be needed and the voltage regulator circuit can be moved to the power unit, if desired.
I use my thermostats with the hysteresis mostly bypassed. That means that the jumper is shorting pins 2 and 3 of J2. This means that the hysteresis is determined by the Thevenin equivalent of R2 through R6 and by R9. This is sufficient for my needs, as I want precise control. I left the option available for increasing the hysteresis in case a user finds that the heater is switching on and off too often.
The SCR voltage regulator is a bit temperature sensitive, which could slightly affect the thermostat's set-point. This will affect the internal heating of the control unit, due to the thermal air pump. The temperature coefficient of the SCR regulator is -80mV/degC. Once the thermostat is stabilizing the room temperature, this won't be an issue at all. If you turn the heater on in a cold room, the thermostat will think the room is a few degrees warmer than it really is and shut off prematurely. Within a half hour, the set-point will stabilize at its proper value. I didn't find this to be a problem because I keep my rooms regulated, but if you do have a problem with it, replace the SCR, R19 and R20 with two 1N4742 (12V, 1 watt) zeners in series. Of course, the standby power consumption will be increased a little bit and the zeners won't be as robust against power surges as the SCR. Definitely use the zeners if you intend to put numbers on the temperature dial.
What's maddening is you can't anticipate where trouble will come from. In Chuck's Toyota floor mat story, the car company is running a recall because an floor mat not designed for the car model could bump into the accelerator.
Never say never when it comes to lawsuits and ridiculous awards. What do hot coffee, circular saws, lawn mowers and highchairs have in common? All were at the base of some outrageous jury awards. Who would have ever guessed that you can get burned by hot coffee?
I would not lose any sleep over it though, because you do not have enough cash to attract one of the real blood sucking attorneys.
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.