Lm317 as a current regulator

Discussion in 'General Electronics Chat' started by jody, Mar 14, 2014.

  1. jody

    Thread Starter Member

    Nov 12, 2012
    39
    0
    HI
    sorry if this is a bit basic ive got a mental block on understanding the method of operation so working it out bit by bit sowhen an lm317 is used as a current regulator am i correct that the output voltage is the input voltage minus the volt drop across the lm317 from input to output.

    jody
     
  2. crutschow

    Expert

    Mar 14, 2008
    12,977
    3,221
    Yes. And if it's configured as a current regulator its output voltage will vary with the load (of course also changing the drop across the LM317). This keeps the current constant through the load, even if the resistance of the load changes.
     
  3. #12

    Expert

    Nov 30, 2010
    16,248
    6,745
    When you use a 317 for a current regulator, the regulating resistor always has 1.25 volts across it (if there is a load). The job of the 317 chip is to use up all the voltage the load isn't using. Therefore, the voltage across the chip will vary quite a lot, depending on the load.
     
  4. jody

    Thread Starter Member

    Nov 12, 2012
    39
    0
    thank you. i was trying to visulize the current flow and voltage levels. because its a simple concept most sites i found assume you will get it straight away but im one that likes to see a breakdown of the operation step by step so it can sink in my brain.
     
  5. #12

    Expert

    Nov 30, 2010
    16,248
    6,745
    That's OK. We have both types here :D
     
  6. jody

    Thread Starter Member

    Nov 12, 2012
    39
    0
    one more question
    do i have to match the input voltage to the required output voltage when using it as a current regulator. so if i had say a 15v input i would get a 12v output even if i only had a component that used 5 volts as the load.
     
  7. #12

    Expert

    Nov 30, 2010
    16,248
    6,745
    Read post #3 again.
     
  8. chipwitch

    Member

    Mar 29, 2013
    48
    4
    Jody, would it help if you draw your load as a resistor into the basic schematic along with all the grounds? Then you can see clearly that the load makes up the other half of the voltage divider used by the adjust pin.

    The attachment shows the basic circuit for a voltage regulator. Notice that it's essentially the same as the current limiter circuit with the major exception that your load goes where R2 is. So, the current will always be the same. The MAXIMUM voltage output to the load will be Vin - R1 - dropout voltage. But, if your load say shorts out, then the regulator will drop the voltage to whatever is necessary to limit the current to what you've established in R1. Ohm's law always wins. I = 1.25V/R1. Calculate all your voltages based on that I.

    You will need a voltage regulator circuit to limit voltage to your load. Insert that circuit between the current limiter and your load.
     
    #12 likes this.
  9. jody

    Thread Starter Member

    Nov 12, 2012
    39
    0
    thanks
    i think ive got it now. i was confusing the voltage regulation side and the current side thinking when current was regulated also the voltage was at the same time.
    back to basics for me :mad:
     
  10. crutschow

    Expert

    Mar 14, 2008
    12,977
    3,221
    Yes, you need a voltage regulator as well as a current limiter if you want a fixed voltage to your load but also limit the circuit current to some value. In that case you would put the LM317 configured as a current limiter in series with an LM317 configured as a voltage regulator driving the load (say for 5V if you have a 5V load). Now the minimum input voltage would be the desired regulated output voltage plus about 6V since each LM317 needs a minimum of 3V across it for proper operation.

    To reduce the 3V drop across the current limit circuit you can use a discrete design such as BJT transistor-MOSFET current limiter. These can operate at normal currents with less than a volt drop.
     
    Last edited: Mar 16, 2014
Loading...