CrazyEngineers
  • Lem0nHead
    Lem0nHead

    MemberMay 9, 2009

    voltage comparator

    hello
    I'm trying to auto-discharge a capacitor when it's charged using LM311 (voltage comparator), but although it worked on a protoboard while I was testing with a 12V PSU it's not working on the real thing
    any ideas on what may be wrong? schematic:
    [​IMG]

    (I tried swaping pins 2/3)

    a quick explanation:
    since 120V (actually 120*sqrt(2) = ~170V peak) is too much for the LM311, I used a voltage divider both on the main source and capacitor
    on the main souce I used smaller resistance to give enough current to the LM311 (it needs at least 12mA)
    on the capacitor I used a smaller proportion on the resistors so that it triggers when it reaches 95~98% of the main voltage

    the problem is that output pin is always 0V
    Replies
Howdy guest!
Dear guest, you must be logged-in to participate on CrazyEngineers. We would love to have you as a member of our community. Consider creating an account or login.
Replies
  • Rifaa

    MemberMay 9, 2009

    Buddy , the Inverting input pin is at Supply potential, there is noway the NI will go above to trigger the out put. the way I see it the output switches state when NI input Voltage goes above the Inverting input. So - pin is the reference Voltage and + pin is the Sensing Input.

    The Reference should be in a value in which the sensing voltage can cross over. If the reference as at the supply potential, how could the sensing voltage will go above it in order for the comparator to detect a difference in the input.

    Change the circuit the way I recommend
    First use a appropriate zener diode to supply the Vcc to the LM 311, remember to use a filter cap to suppress high frequencies.
    Secondly Use the potential divider as a reference and sensing so that the Input voltages does not go above the rated voltages provided in the data sheet of the comparator.
    One more thing u should be concerned is not to short circuit the capacitor to discharge, the way I see it the cap is quite a high value and the short circuit current can be too much for the SCR if it is not properly rated in the long run. Do use a low value resistor as a bleeder.
    post back if in doubt
    Are you sure? This action cannot be undone.
    Cancel
  • Lem0nHead

    MemberMay 9, 2009

    hello
    thanks for your answer
    I have an explanation for that

    I'm using a 120V AC though a diode bridge, so it becomes 120V DC with about 170V peaks
    I use a voltage divider with 9090ohm/660ohm (weird, but that's what I have easily available at this time for the prototype)
    that means I'm getting V=8.12V with i=13.2mA plugged on LM311 supply and reference pin

    the other voltage divider is from the capacitor, which can get charged up to the peak voltage (170V)
    the resistors used on the divider are 90909ohm/7333ohm which means the output voltage is 8.95V even at 120V (but since it can goes up to 170V, it may easily become at least 12V), so it definitely should get higher than the supply voltage



    Rifaa
    Buddy , the Inverting input pin is at Supply potential, there is noway the NI will go above to trigger the out put. the way I see it the output switches state when NI input Voltage goes above the Inverting input. So - pin is the reference Voltage and + pin is the Sensing Input.

    The Reference should be in a value in which the sensing voltage can cross over. If the reference as at the supply potential, how could the sensing voltage will go above it in order for the comparator to detect a difference in the input.

    Change the circuit the way I recommend
    First use a appropriate zener diode to supply the Vcc to the LM 311, remember to use a filter cap to suppress high frequencies.
    Secondly Use the potential divider as a reference and sensing so that the Input voltages does not go above the rated voltages provided in the data sheet of the comparator.
    One more thing u should be concerned is not to short circuit the capacitor to discharge, the way I see it the cap is quite a high value and the short circuit current can be too much for the SCR if it is not properly rated in the long run. Do use a low value resistor as a bleeder.
    post back if in doubt
    Are you sure? This action cannot be undone.
    Cancel
  • Rifaa

    MemberMay 9, 2009

    U say that u prototyped the board and it worked at 120V
    Are you sure? This action cannot be undone.
    Cancel
  • Lem0nHead

    MemberMay 9, 2009

    what actually worked was that:

    [​IMG]
    #-Link-Snipped-#

    (I later changed the voltmeter by a LED and it lighted correctly when it should)
    Are you sure? This action cannot be undone.
    Cancel
  • Rifaa

    MemberMay 9, 2009

    The wiring is way wrong in your practical one buddy
    Do u have msn ID. if so add my ID and I can guide you now.
    Are you sure? This action cannot be undone.
    Cancel
Home Channels Search Login Register