0
\$\begingroup\$

I want to achieve a specific temperature using a semiconductor or an electrical resistance in general. The temperature I want to achieve is at about \$30^oC = 86^∘F\$. This experiment consider it to be local in space, just the semiconductor and me with a battery. The starting temperature is about \$24^oC−25^oC\$

Can you suggest any semiconductors that can do the job and are cheap?

How can I achieve this temperature? How much voltage, current should I use?

I don't want to use coolers to drop the temperature.

\$\endgroup\$
3
  • \$\begingroup\$ What device or component is intended to measure the achieved temperature? \$\endgroup\$
    – Anindo Ghosh
    Commented Nov 29, 2013 at 13:32
  • \$\begingroup\$ @AnindoGhosh, i don't know.Maybe if i use a thermostat? \$\endgroup\$
    – Dimitris
    Commented Nov 29, 2013 at 13:37
  • \$\begingroup\$ If you have a thermostat anyway, why not use it to control heating of a heater coil, i.e. a simple resistive wire? The heating current needs to turn on and off such as to approximately hold the desired temperature, many thermostats can support such an arrangement. The question is not clear: What engineering benefit is sought from the use of a semiconductor device? And, are you looking for a specific value of power to heat conversion (i.e. voltage x current) to automagically achieve a steady state at 30 degrees C? \$\endgroup\$
    – Anindo Ghosh
    Commented Nov 29, 2013 at 13:46

1 Answer 1

Reset to default
2
\$\begingroup\$

Your temperature specs were formatted to oblivion, but it seems you want to heat something as apposed to cool it.

There is no need for a "semiconductor". Resistors are excellent and efficient converters of electric power to heat power. You can also know just from looking at the voltage or current (and knowing the resistance) how much heat it is producing.

Again, we can't tell what temperatures you want to achieve, but a thermistor might be good. These are resistors with the resistance being a function of temperature. If you want one to be at a particular temperature, you measure the voltage and current accross it to determine the resistance, then vary the voltage (or current) it is being driven with to increase or decrease its temperature as appropriate. This will, of course, require some additional parts for temperature control. A microcontroller would be great for that since it can do the resistance calculation, which requires division.

If you want to actively regulate temperature, you are going to need more than a power supply and a single device. The closest I can think of is a bi-metallic element actuating a switch. This opens the switch when the element gets to a certain temperature, and closes it again when it cools. These will have some mechanical hysteresis, so the temperature will bounce between the two hysteresis limits. Technically, this is two parts, the bi-metallic element and a switch, but you can find these integrated into a single package. This is how automobile and christmas tree light flashers used to work. Old automobile voltage regulators also worked on this principle.

\$\endgroup\$
1
  • \$\begingroup\$ Old room heaters too worked on the bimetallic strip method. Old? Nahh, they still sell those things in my part of the world. \$\endgroup\$
    – Anindo Ghosh
    Commented Nov 29, 2013 at 13:47

Not the answer you're looking for? Browse other questions tagged or ask your own question.