Help with current limiting

Discussion in 'General Electronics Chat' started by 1dimwit, Jun 2, 2012.

  1. 1dimwit

    Thread Starter New Member

    Jun 2, 2012
    I have this circuit...

    D1 = 1N4007
    R1 = 220R
    12V is supplied by a car battery - so there is plenty of amps available.

    The motor is a cheap air compressor - there is no spec for it on the box, but is supposed to be used with a car battery


    It seems to work OK simply switching on and staying on, or switching off, but if the motor stalls - or stop/starts many times/sec, then the TIP122 gets fried.

    I guess I need to reduce the current flow somehow, to keep it in spec for the TIP122

    I thought that the only way to do this was to put some sort of current limiter in circuit, but googling about I found a discussion that seemed to imply that I could simply change R1. Is this right ? Up circuit from R1 is a 74HC595.

    or do I have to put in a current limiter ? I've got a current limiter circuit that looks like this ...


    If I could limit the current to say 5A - (the max for a TIP122) how do I go about calculating the heatsink size ?

  2. crutschow


    Mar 14, 2008
    If you limit the current to the motor then it may not start and could burn out. Just use a larger BJT transistor or a power N type MOSFET.

    Changing R1 is not an effective way to limit the current.

    Why does the compressor motor stop/starts many times/sec? :confused:

    You have to know the motor current to calculate the heat sink you need. Can you measure it with a multimeter?
  3. kubeek

    AAC Fanatic!

    Sep 20, 2005
    Especialy when the transistor cannot bear the load to begin with.
    You could theoretically limit the current like that, but this is not going to solve the transistor burning, since the stalled motor is basically a short, you need the transistor to be able to dissipate 12V*Imax. If you have 5A, you need to dissipate 60W of heat. The tiny TIP can dissipate 65W with infinite heatsink (imagine cooling with liquid nitrogen). Not a very bright idea to run it close to absolute maximum limits.

    I think you should use some hefty mosfet like IRFP240 or basically any N mosfet from IRFP line. Then you don´t need no current limiting since the transistor should be able to easily bear the stall current for a while.
    1dimwit likes this.
  4. 1dimwit

    Thread Starter New Member

    Jun 2, 2012
    Well it's a long story. It is computer controlled. Compressor actuates a pneumatic ram. There's feedback via a linear pot into a ADC. What with all the mechanical and digital tolerances, sometimes the computer decides the ram needs to go a little further, sometimes a little less, and the compressor starts/stops for each movement.

    I could fix that but that wouldn't solve a motor stall.

    I need a permanent solution.

  5. #12


    Nov 30, 2010
    Bottom line: Do not try to limit the current. Get a bigger transistor, or two, so they can survive the needs of the motor and feed it properly.
  6. vladtess


    Jan 5, 2011
    Use a cap with PWM to avoid spikes.
  7. crutschow


    Mar 14, 2008
    If you measure the resistance of the motor when unpowered you can use that to calculate the stall or starting current. From that you can pick a transistor with sufficient rating (25-50%% higher than the stall current). That should solve you failure problem.
  8. raviypujar

    New Member

    May 20, 2012
    I suggest you use two TIP122 transistors in darlington configuration so that the current drawn by the motor is divided between them .

    - Ravi
    Last edited by a moderator: Jun 17, 2012
  9. Ron H

    AAC Fanatic!

    Apr 14, 2005
    A low Rds(on) power MOSFET will not get as hot as Darlingtons.
    Last edited by a moderator: Jun 17, 2012
  10. 1dimwit

    Thread Starter New Member

    Jun 2, 2012
    OK. Thanks for all the help and suggestions on this. I'll try a bigger MOSFET like the IRFP240, see how that goes.
  11. SgtWookie


    Jul 17, 2007
    The IRFP240 is not a logic-level MOSFET:
    See that the Rds(on) specifications are only given when Vgs=10v.
    Don't look at the threshold voltage; that just tells you what voltage you need to be below in order to turn it off.

    Try something like the IRLU8721:
    These have a low Rds(on) even with Vgs=4.5v; a low gate charge (Qg=8.5nC), rather high drain current (65A max) and the Vdss being 30 exceeds your requirements.