LM317 as current limiter?

Discussion in 'The Projects Forum' started by fireofenergy, Aug 12, 2013.

  1. fireofenergy

    Thread Starter New Member

    Dec 17, 2011
    20
    0
    Hi,
    I want to charge up a led light from a 12v battery and am using a dc buck converter (from ebay) that is rated at 3A max to convert the voltage down (for a li-ion). The converter will fry (especially since it wants to increase the amps in ratio to the lowered voltage).

    Is the LM317 designed to be connected to car batteries and such, as long as the resistor limits the current to below its rated? Or will it fry, too?

    Thanks.
     
  2. wayneh

    Expert

    Sep 9, 2010
    12,148
    3,058
    You can use an LM317 in a constant current regulator configuration. Fig 11 in the TI data sheet, if I recall.
     
  3. LDC3

    Active Member

    Apr 27, 2013
    920
    160
    How much current flowing depends on the load. If the load is 1000Ω, then a 12V source will provide only 12mA, even if it is capable of providing 10A. If you want to provide a lower voltage (than 12V) to the LED, you could just use the LM317 (provided the current is under 1.5A and you dissipate the heat).

    Tell us the specs of the LED (better yet, the whole project) so we can figure out what would be a good design for powering it.
     
  4. fireofenergy

    Thread Starter New Member

    Dec 17, 2011
    20
    0
    I'm using an adjustable converter from ebay to convert the volts from a car battery down to 3.6v for a lifepo4 battery. I already connected the converter to a 18v, 0.6A solar panel (10 watts) and the thing got hot, since it converted the .6A to like 2A (or whatever after inefficiency). So, the next "step" is to charge an old car battery from the little solar panel and connect the dc converter to that, but obviously with some kind of current limiter.
    I thought about the resistor idea, but don't like the fact that it wastes exactly half of the juice. If it's 13v and I put a 100 ohm in line, then it would be 130mA, but that is also what is wasted as heat in the resistor, isn't it?
    So, I assume that the lm317 would be far more efficient... IF it can be attached directly to the battery. From there, (if it can be connected directly to the battery) the dc converter is dialed in to just less than 3.6v for the lifepo4 batteries. And I would set the lm317 to about 250mA which the converter should convert to over about .6A or so (I'm figuring losses from voltage drop of LM317 and converter.
    Thanks
     
  5. wayneh

    Expert

    Sep 9, 2010
    12,148
    3,058
    I'm lost. I need a schematic. An LM317 is basically a smart resistor, so no more and no less efficient.
     
  6. John P

    AAC Fanatic!

    Oct 14, 2008
    1,634
    224
    To answer exactly the question you asked--no, an LM317 will not "fry" under any conditions that are within its specifications. It cannot be damaged by a short circuit or overheating. If you reach its current limit, it simply won't pass any more current, and if it heats up, it limits current flow to a level that holds the temperature at a safe level. In that second case, the better the heat sink you place it on, the more current it will allow, up to 1.5A.

    But if you're running an LM317T (TO-220 package) off a 12V battery, your maximum output power is going to be around 16 Watts. That's enough to do some damage to whatever device is being operated after the LM317. It won't prevent anything except itself from frying!
     
  7. fireofenergy

    Thread Starter New Member

    Dec 17, 2011
    20
    0
    John, Thanks.
    I'll get one and see how this little set up works. Wondering about the efficiency...

    wayneh, If it is no more efficient than a resistor, then I would just use the resistor.
    But I'm going to check it out anyways (after I buy it).

    Thanks!
     
  8. fireofenergy

    Thread Starter New Member

    Dec 17, 2011
    20
    0
    Guess I'm confused...
    After looking at the lm117 pdf,
    http://www.ti.com/lit/ds/symlink/lm117.pdf
    figure 47, I see that the resistor is set just like a normal resistor without the IC because it says V-ref/R1.
    But is "ref" the same as input voltage? And then, to confuse me further, it "links" ref to ".8ohm < R1 <120 ohm"... I'm trying to find what it means...
     
  9. fireofenergy

    Thread Starter New Member

    Dec 17, 2011
    20
    0
    Oh, think "ref" means in the 1.25v range...
     
  10. fireofenergy

    Thread Starter New Member

    Dec 17, 2011
    20
    0
    Guess I'm confused...
    After looking at the lm117 pdf,
    http://www.ti.com/lit/ds/symlink/lm117.pdf
    figure 47, I see that the resistor is set just like a normal resistor without the IC because it says V-ref/R1.
    But is "ref" the same as input voltage? And then, to confuse me further, it "links" ref to ".8ohm < R1 <120 ohm"... I'm trying to find what it means...


    Oh, think "ref" means in the 1.25v range...

    Sorry about all the double posts, just trying to edit, and delete and it won't.
     
  11. LDC3

    Active Member

    Apr 27, 2013
    920
    160
    From the datasheet.
    Vref is an internal voltage reference that the IC uses to set the output voltage.
     
  12. fireofenergy

    Thread Starter New Member

    Dec 17, 2011
    20
    0
    Right on, thanks!
     
  13. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
Loading...