100 LED ladder - Is this right?

Discussion in 'The Projects Forum' started by polyploidy, Nov 11, 2012.

  1. polyploidy

    Thread Starter New Member

    Nov 11, 2012
    1
    0
    Hello, I'm trying to build something like this with 100 interlocked LEDs (is there an official name for a configuration like this?)
    [​IMG]

    I had great difficulty trying to plug this into a LED calculator, until it was brought to my attention that the ladder should wired like a singular LED. Does this sound right? Below is what I came up with with these 520LB7C LEDS. I also noticed with these is that the datasheet says the max forward current is 30mA, but the graph on page 4 says that the max forward current is 20mA when at 3.2 V. This is alright, right? Thanks! :)
    [​IMG]
     
  2. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    You connected 100 LEDs in parallel which is how light bulbs are connected, not LEDs.
    The datasheet shows that some LEDs have a forward voltage of 2.8V and others have a forward voltage of 3.4V or anywhere in between. Then LEDs with a low forward voltage will hog all the current and burn out. Then other remaining LEDs with a low forward voltage will burn out. They all might burn out soon.

    Flashlight (torch) manufacturers connect many identical LEDs in parallel because they buy millions of them. They test and group them into groups with exactly the same forward voltage. You could do that too if you buy millions of them.

    Since you do not know the actual forward voltage of each LED then each one needs to have its own resistor to limit the current. Calculate the resistor for 25mA when the forward voltage is the lowest, 2.8V:
    (6V - 2.8V)/25mA= 128 ohms. Use 130 ohms or 150 ohms. Then an LED that has a forward voltage of 3.4V will have a current of 17.3mA when the resistor is 150 ohms.

    You might be lucky and all your LEDs were made at the same time and are identical.
    Then your single resistor circuit will be fine.

    Your "wizard" forgot to tell you that the LEDs will get EXTREMELY hot because they are close together. The heating is 3.2V x 2A= 6.4W which is lot of concentrated heat.
    The LEDs in the center might melt.
     
  3. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Agreed. Many lights you see wired this way are test selected to work, but with a batch you have bought it is a crap shoot. You might win, but you will probably lose.

    LEDs, 555s, Flashers, and Light Chasers
     
  4. spinnaker

    AAC Fanatic!

    Oct 29, 2009
    4,887
    1,016
    Not to still the thread but what if you have a buck regulator to limit the current? Assuming a single buck regulator, you will still have the same issue as a single resiistor correct?
     
  5. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The problem is that the LED forward voltages are probably different. The LED or LEDs with the lowest voltage will light (but LEDs with higher voltages will not light) then too much current will flow and burn out the LEDs. Then there are fewer remaining working LEDs that have too much current so they also burn out. Then the few remaining LEDs also get too much current and burn out.

    A buck regulator will not help this problem.
     
  6. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    No, they don't parallel real well and will have varying brightness levels.
     
  7. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    You can test them for matching on a protoboard. Set up a 10V source and a 1K resistor in series and plug in each LED and read the diode drop voltage. If they match within a few mV, maybe OK to parallel.

    The temp problem is difficult and you would need a good sized PCB to suck the heat out of the diodes.
     
Loading...