resistor dependent pwm circuit

Discussion in 'The Projects Forum' started by vdubin, Sep 22, 2009.

  1. vdubin

    Thread Starter New Member

    Sep 22, 2009
    7
    0
    I am looking for a circuit diagram for a pwm circuit that is controlled by a variable resistor. what I am looking to do is control a peltier cooler by a temperature sensing resistor. the voltage source I have is automotive 12v-15v dc.

    thanks guys.
     
  2. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    You could adapt the circuit shown here - you would use your variable (sensing) resistor in combination with a fixed resistance of suitable value to replace the control potentiometer in the circuit shown in the link.

    http://www.allaboutcircuits.com/vol_6/chpt_6/9.html
     
  3. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    Then there is this one...

    555 PWM Oscillator

    [​IMG]

    There should be a 10Ω resistor between R1 and pin 3

    There there is this one...

    [​IMG]

    If you use a comparitor such as a LM339 there should be a 10KΩ resistor to Vcc.
     
  4. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    I've done these, you just need thermistor and one comparator as in a standard on/off temperature controller, but then you add feedback so it oscillates around the comparator setpoint. Perfect PWM that changes duty to keep the temperature steady.

    The addition of a small cap from comparator input to ground lets you tune the frequency to (say 4kHz) fast enough to save the peltier from thermal shock and slow enough to reduce switching losses.
     
  5. vdubin

    Thread Starter New Member

    Sep 22, 2009
    7
    0
    thanks guys. I will use one of these, i'll probably try all 3 just for fun.
     
  6. vdubin

    Thread Starter New Member

    Sep 22, 2009
    7
    0
    RB: I am building this circuit and you mentioned a cap on the input of the comparator to make the switching freq around 4k. what size cap would you recommend? sorry, I can assemble and fix circuit, but I don't know enough to design them.
    [​IMG]
     
  7. Dx3

    Member

    Jun 19, 2010
    87
    7
    It calculates to 6.89 nanofarads
    or you can change the 33k resistor to 2.27k
     
    vdubin likes this.
  8. vdubin

    Thread Starter New Member

    Sep 22, 2009
    7
    0
    6.89 nano = .007 micro unless my math is wrong.
    probably ceramic disc cap my best bet.
     
    Last edited: Jul 10, 2010
  9. Dx3

    Member

    Jun 19, 2010
    87
    7
    True but... that size cap probably doesn't exist. That's why I included the resistor size. the formula is: Time = 1.1RC
    Time is 1/frequency
     
  10. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    The 1458 is an old and slow opamp that can't source much current, and you are trying to use it to drive a power transistor. That's just not going to work very well.
     
  11. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    I recognize this circuit from the AAC experiments volume. It would work, but a resistor in line with the diodes should be there. These experiments are verified (as in they have been built), but many of them aren't exactly practical. I'm not excepting my entries in this.
     
  12. vdubin

    Thread Starter New Member

    Sep 22, 2009
    7
    0
    bill, the circuit is the one you recommended from the experiments section.

    I am trying to drive a power transister that can carry a 10A load at a frequency that will not damage my peltiers I am trying to control.
    if the 1458is old, slow and cannot handle current, could you recommend one that could?..

    I am going to be driving two large peltier coolers here
    there are actually 2 peltiers in each run in parallel.
    ideally the circuit would have 2 temp sensors to compare and use the resistant differance to control the peltiers.
    check input temp, if temp increasing then increase cooling, check output temp for goal, if at goal then maintain if not increase cooling. however, as long as I have a temp sensor I can make the circuit work by monitoring the over all output temps and adjust the cooler to compensate. I do know that I will have to wire in a bypass to give the peltiers a full power on ocassion for fast cooling as necessary.

    thanks guys.
     
    Last edited: Jul 10, 2010
  13. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    If you are using a 12VDC power supply you could use MOSFETs. I'm working on a new section for my article, you can read about how to use them in Chapter 10 in my work in progress area, LEDs, 555s, Flashers, and Light Chasers (forum edition). You will need to loose the op amps though. A 555 will drive a MOSFET well, put a 100Ω resistor between the output of the 555 and the gate.

    MOSFETs require almost no current on the gate past a single surge when switching. A BJT requires some current on the BE to keep them turned on. For large current applications this can be a major deficit.
     
Loading...