temperature control for fan

Discussion in 'The Projects Forum' started by iepaul, Jan 26, 2011.

  1. iepaul

    Thread Starter New Member

    Jan 26, 2011
    8
    0
    I need to put a circuit in to my satellite receiver to control the fan. other people have successfully used the velleman mk138 but I wonder if anybody had a circuit I could build on some strip-board to develop my own skills.
     
  2. Gundalf

    New Member

    Jun 22, 2010
    13
    0
    My usual circuit for this sort of thing consists of an op-amp, a pair of transistors and assorted bias resistors. If you are interested then I'll dig the circuit out but essentially you use one transistor such as a 2n2222 or a 2n3904 as a temperature sensor and the op-amp as a comparator against a fixed voltage.

    The opamp then just switches a moderate current transistor like a BFY51 and that lets the fan run. Easy and simple and doesnt require mucking about looking for a thermistor which never lices in my junk box (I have lots of odd transistors though!)
     
  3. iepaul

    Thread Starter New Member

    Jan 26, 2011
    8
    0
    that would be great thanks
     
  4. Gundalf

    New Member

    Jun 22, 2010
    13
    0
    OK, I''l descrbe it as I don't have access to circuit drawing software at the point in time.

    Take a standard npn such as 2n3904 and connect the collector to base through a 1M0 or 2M0 resistor. This biases the base on very slightly and draws a small current, about 3uA. Transistor Hfe is typically 80 or so, and a collector current of about 0.2mA flows assuming the collector is held at 6V.

    A resistor of 33K will therefore take about 0.2mA at 6v.

    OK, so we have now a potential divider dividing the 12V line. One side is the 33K resistor, the other side is this self biased transistor.

    Now.. if you warm up the transistor you increase the current in the base and hence the amplified collector current. Hence more current flows and it's apparant resistance drops, vice versa if it's cooled. This arrangement is quite sensitive wrt to temperature and quite repeatable.

    Feed the mid point into one input of an op amp. Feed the other side of the opamp (I use TL082's usually) with preset mutlerterm pot and you have an adjustable temperature dependant comparator.

    Output then drives the base of BFY51 whcih can switch an amp of current, usually enough for all but the chunkiest of DC fans (the bfy may need heatsinking though - check the dissipation and spec sheets)

    HTH - I'll get the schematic up when I get chance.
     
  5. wayneh

    Expert

    Sep 9, 2010
    12,100
    3,036
    Take a look at the circuit I was using to control a cooler.
    http://forum.allaboutcircuits.com/showthread.php?t=42774

    Ignore the extra stuff below and to the right of the TEC cooler, and imagine the TEC is your fan.

    You'll see there is a LM35 temperature sensor IC to read 10mV per °C, a comparator to compare that voltage to a reference, and finally a MOSFET to switch the TEC on or off. This type of circuit, with your fan replacing the TEC, would be perfect for you. The MOSFET in particular is better at switching than a regular BJT.

    I later added an op-amp to amplify the LM35 voltage. This allowed me to reduce the dead-band hysteresis from ~1°C to ~0.05°C. Most applications wouldn't require that refinement.
    http://forum.allaboutcircuits.com/showthread.php?t=43408
    Note that the op-amp in the circuit posted there is what was causing my problem at the time - it couldn't sense near ground. Everything was fine once I replaced it.
     
    Last edited: Jan 26, 2011
Loading...