LED Circuits

Discussion in 'Feedback and Suggestions' started by Adjuster, Oct 20, 2011.

  1. Adjuster

    Thread Starter Well-Known Member

    Dec 26, 2010
    2,147
    300
    Would it be helpful to have some kind of Sticky or other warning about the danger of operating LEDs without current restriction?

    I have seen a number of instances of beginners' travails with LEDs recently, but this example involving a mains transformer and rechargeable batteries prompted me to put this in writing. http://forum.allaboutcircuits.com/attachment.php?attachmentid=35476&d=1319094779 We must hope that the OP for this article will get the necessary advice, but a standard reference would be useful.

    From time to time I have considered referring such enquirers to Bill Marsden's comprehensive notes,http://forum.allaboutcircuits.com/showthread.php?t=19075 but as a relative newcomer here myself was unsure if that would be the done thing. There is also an article in the forum book http://www.allaboutcircuits.com/vol_3/chpt_3/12.html#03294.png This is all very useful, but I wonder if it may also help to have a short simplified warning presentation, in language aimed at the sort of real beginner who is in most danger of making these mistakes. It may be better for them to grasp the basics first, before moving on to detailed information and an interesting selection of LED circuits.
     
  2. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    It might work, but if they don't read the other sources I suspect they won't read a sticky. Just an opinion (that I'm willing to change).

    I've had beginners argue with me about the subject, they want to treat them like light bulbs and that is all there is to it.
     
  3. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
    If they burn out some LEDs. Well then they will learn quicker. Standard low power LEDs as components are quite harmless. So I see no reason to be concern about using low power LEDs in any way. They are quite ideal trail error components.
     
  4. Adjuster

    Thread Starter Well-Known Member

    Dec 26, 2010
    2,147
    300
    Well, perhaps that's fair enough, it was just a thought. Possibly I was as much reacting to the fellow wanting to slam a rectifier output straight into a SLA battery, as to the LED issues. Obviously we cannot protect everyone who plans projects likely to go up in smoke, but there does seem to be a particular problem with LEDs. Getting folk to accept that they really are sharp stiff little things that need to have some padding round them can be an uphill struggle.

    The parallel issue is another annoyance. Lots of cr@ppy and a few fairly successful lamps are made that seem to break the rules on this. Some of them last surprisingly well, which makes it hard to convince junior not to do the same. Actually, I'd love to know how some do appear to get away with it, but that's another story. My guess is that it has to do with careful grading in factories dealing with huge lots of devices, by which token it is probably impracticable to do at home. It may also be difficult to repair such a unit if a single device fails. Last I heard, the likes of Mouser don't sell LEDs in 10mV forward voltage groups.

    Perhaps learning by experience is the best way, but I will probably find it hard to resist commenting next time I see one of these.
     
  5. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    How they get away with it is weak batteries, that have a large amount of internal resistance. The resistor is still there, it is just inside the battery.
     
  6. Adjuster

    Thread Starter Well-Known Member

    Dec 26, 2010
    2,147
    300
    That's clear enough, but I was thinking of how they get even vaguely similar brightness out of devices in parallel groups, even if there is some attempt at overall current control, with a resistor or otherwise.

    Some cheap-and-nasty torches have obviously inconsistent brightnesses, but others do a lot better than you might expect. I would guess at closer process control, binning for Vf, or maybe even devices with deliberately high internal resistances - another sort of hidden resistor.

    All of this is, of course, off the original topic.
     
  7. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    Yes, but it relates in that we are going to have to explain it to these folks if such a sticky is written.

    I suspect the actually match the Vf of the LEDs. It can be done easily enough with quantities of LEDs.
     
  8. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Actually, in a given batch of LEDs, perhaps 75% to 80% of them will be very close in Vf at a given If. There will be a few that will be quite above or below the median.

    It's really not that hard to match Vf's if you have a constant current source (LM317T or LM317L, a 62 Ohm resistor) a $3 Harbor Freight dmm and a 12v supply; it just takes some patience - but unless someone tells a newbie it needs to be done, they will have to do a LOT of reading to find out for themselves.

    Series strings are another matter. If you're running, say, 3 LEDs in a series string, and a bunch of those strings in parallel, the Vf's of the strings will tend to be pretty much the same due to the Monte Carlo distribution.
     
  9. kami lee

    New Member

    Apr 17, 2012
    1
    0


    i think this is a good idea for those people who want to get some information about it !
     
  10. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    Welcome to AAC.

    Did you notice the text books listed on top of this page? LEDs are included, I helped rewrite some of it myself on the LED section.

    If you didn't notice that, why would you notice a sticky?

    I have written a tutorial I usually introduce to the beginners. I started it specifically on LEDs, and it took a life of its own.

    LEDs, 555s, Flashers, and Light Chasers
     
  11. KJ6EAD

    Senior Member

    Apr 30, 2011
    1,425
    363
    I've experimented with Ningbo lights from Harbor Freight and concluded that the LEDs have high internal resistance. If you take the typical device with 24 5mm LEDs in parallel made to run from 3 AAA batteries and supply it with a 480mA current source, the temperature rise is alarming.
     
  12. #12

    Expert

    Nov 30, 2010
    16,261
    6,770
    Agreeing with Bill in post #2, some pesky beginners absoultely refuse to add a resistor and work for weeks wondering why their LEDs won't do what they want.

    Agreeing with post #4, LEDs are harmless enough to let the beginners use them for trial and error.

    Still, the basic concept of a "breakover" device compared to a (relatively) linear device is a lesson beginners need to learn. It's a kind of breaking the (Ohm's) law situation and thus requires some education.

    I vote for a "sticky" in spite of the fact that nobody will read it until we direct them to it. At least, that way, we will have something to direct them to instead of explaining it over and over as the weeks and months go by.
     
    KJ6EAD likes this.
Loading...