Current Limiting R Value

Discussion in 'General Electronics Chat' started by pntrbl, May 6, 2008.

  1. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    I'm working with an LM339 comparator and a 5.1V 1N4733 Zener on a 12V supply intending to use the the zener as a constant reference to compare against. Realizing there needs to be a resistor from the supply voltage to the zener/comparator junction, I've downloaded datasheets on both ,(way over my head on that :)), and googled the you know what outa zener diode circuits in an attempt to come up with the correct value.

    All of which confused me but good!

    The resistor will need to drop the 6.9V the zener doesn't need so all I need to figure out is how much current the zener and comparator combination need. 6.9 divided by the current should do it.

    On the 339 datasheet from National Semi I see Typical Input Bias Current is 25 nA. nA? Isn't that a little small? That's @5VDC 25C BTW and I'll be on a 12V supply.

    On the zener sheet from Surge Components other than the 49ma test current I haven't a clue .....

    Any help?

    SP
     
  2. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    You want to limit the current through the zener, but allow enough that the load its regulating gets a constant voltage. Assuming that a couple of mills current will be enough, it's easy to figure. 12 volts - 5.1 volts is 6.9 volts. Using R = E/I, we plug in 6.9 volts/.002 amps and get 345 ohms. Standard values are 330 and 360. Whichever is easiest to obtain will work. 1/4 watt size will be just fine.
     
  3. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    Thanx for the quick response beenthere. I'm surprised it's "safe" to assume a couple of mills will be enough, but I make mountains out of molehills all the time ... :D

    SP
     
  4. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Do you have any information to share that suggests that 2 mills won't be enough? Or you can use the same formula and calculate a resistor value that allows more current.
     
  5. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    You bought a "high current" 5.1V zener diode. It is spec'd at a current of 49mA and might not even work at only 2mA or its voltage will be only 4.2V.
    I would use a low current BZX79C5V1 (there are others) zener diode instead that is spec'd at a current of only 5mA. Then I would operate it at 5mA where it will be 5.1V.

    Yes, the input current of a comparator or opamp is very low. Some have FET inputs that draw no current.
     
  6. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    You slipped the decimal point. 6.9V/2mA=3.45k.
     
  7. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    Mea culpa - it was late.
     
  8. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    Thanx for the continued discussion gents. I don't know nuthin' from nuthin', but my breadboard arrived today .... so I'll be learning!

    As far as the 2 mils vs 49 on a 1N4733 all I know is the datasheet lists 49 as the test current. The rest is all about a buncha stuff that doesn't give me a clue. For that matter the only reason I have a 1N4733 is because that's what they had in stock at the local RadioShack.

    Less current for the same job seems like a good idea so I'll be hunting up a BZX79C. Good to know op amp input currents are so low too. I never even heard of a nano amp before .....

    Thanx again.

    SP
     
  9. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I had some 1N4733A's around here, so decided to test one of them.
    4.6v @ 2mA
    4.97v @ 10mA
    5v @ 12.3mA
    5.1v @ 27.8mA
    5.16v @ 50mA
    5.2v @ 70ma
    Your mileage may vary. Note that I was using uncalibrated DMM's that cost under $10 each if I'd paid full price. :rolleyes: However, you'd likely get a pretty stable voltage if you were running at least 20mA through it.
    R = 6.9v/0.02A = 345 Ohms. P = EI = 138mW, a 1/4W resistor would do.
    For 50mA:
    R = 6.9/0.05 = 138 Ohms. P = EI = 345mW, you'd need a 1/2 W resistor.

    I did my tests using three DMM's, an 1N4733A, a 150 Ohm 1 Watt carbon resistor, and a Lambda 160W 0.25-54V power supply.
     
  10. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    You da man Sarge! Good info.

    SP
     
  11. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    A few more things:
    1) You should use a small capacitor across the Zener to ground. Otherwise, you might get some noise on the input. A 0.1uF (100nF) ceramic or tantalum cap should do nicely.

    2) Don't forget that your LM339 comparator should have a bypass capacitor across it's supply pins.

    3) The LM339 has open collector outputs. It can sink current (up to 20mA), but it can't source it. Since you're using a 12v supply, use a 1.2k pull-up resistor on the output.

    4) If you're not using all of the comparators, you should ground the unused inputs. Otherwise, they may "float" and be an unexpected source of noise.
     
  12. pntrbl

    Thread Starter Active Member

    Apr 21, 2008
    123
    0
    Thanks to all the advice I received in this thread I was able to bread a 339 up and it worked right outa the box. No smoke. The output pops from next to nuthin' to full supply as soon as I cross the zener.

    This begs the question of what to do with it next, :p, but I'll start another thread over that.

    So Thanx again.

    SP
     
  13. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    My practice is to ground only one input of each unused comparator, leaving the other input to float high (the input transistors are PNPs). If that makes you nervous, you could tie the other input to Vcc, but I don't think it's necessary. This leaves the unused comparators switched hard to one state or the other, preventing internal oscillation or "rattling around". :)
     
Loading...