hooking up a LED lights wrong dont know how to fix

Discussion in 'General Electronics Chat' started by dingo95966, Oct 14, 2015.

  1. dingo95966

    Thread Starter New Member

    Oct 14, 2015
    So i am doing a project with my kid that was supposed to be easy that i screwed up i have 78 Diode i was suppose to hook them up in series 6 led in a row and 1 ohm resistor on a 12 v circuit.
    what i did was soldered them in parallel instead. so i have set of 2 led tided together and was going to daisy chain them. well i hooked up power to one set of 2 and blow the light thought it was in series so voltage was set for 4v that's when i fingered out what i did wrong. i don't want to un solder all the leds. leds are different in voltage i have 2v and 3 v Diode don't know what to do to fix the problem i created. any help would be appreciated
    Last edited: Oct 14, 2015
  2. Hypatia's Protege

    Distinguished Member

    Mar 1, 2015
    Clearly, you'll need to remove and replace the damaged LEDs... Instead of improvisation, I would suggest that you make what changes are necessary to wire the project as intended -- a little work now compares favorably to further 'surprises':cool:

    Best regards and good luck!:)
    Last edited: Oct 14, 2015
    blocco a spirale likes this.
  3. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    Discard everything you've done already, because it is clearly wrong, and create a design on paper that makes sense. It is not complicated; there will be a resistor and a small number of LEDs connected in series across a 12V supply. Obviously, don't solder all LEDs until you have established that your design works.

    A 1Ω resistor is too low a value for most LEDs.
    Last edited: Oct 14, 2015
    Hypatia's Protege likes this.
  4. AnalogKid

    Distinguished Member

    Aug 1, 2013
    There is nothing in your post about your experience base or skill set, so, I'm gonna start at the beginning. There are 10 rules to follow for a successful electronics project, and everyone on this forum either already knows them, or learns them the hard way:
    1. Draw a schematic.
    2. See #1.
    3. See #1.
    4. See #1.

    Drawing a schematic forces you to see your idea in a physical way without actually touching any parts. A schematic not only controls the design, it controls how you think about the design. It is the universal language of electronics and a huge time saver. You can go through dozens of posts with thousands of words trying to describe what you want to do and how you actually did it, or you can draw a quick sketch, shoot it with your phone, and post it. There is a chance that even if you had wired everything correctly your idea still would not have worked. Without a schematic, we'll never know.

    Last edited: Oct 14, 2015
  5. SgtWookie


    Jul 17, 2007
    It sounds like you used an LED "wizard" on another website; to call it a "wizard" does a gross disservice to new experimenters, as if the supply voltage is evenly divsible by the stated LED Vf (forward voltage) then the resistor value selected is 1 Ohm. This gives very poor current regulation through the LED string, as even minor fluctuations in supply voltage or LED Vf due to temperature changes can easily and quickly cause overcurrent in the LED string and failures. Subjecting LEDs to current in excess of their ratings not only reduces their lifespan, but permanently reduces their maximum brightness.

    How were you planning on powering the LEDs? If using a regulated power supply plugged into a wall outlet, the voltage will likely stay within a few percent of rated voltage up to around 80% of rated current output. If battery powered, your mileage will vary.

    Do you know the LEDs typical Vf at the rated current? And what is the rated current?
  6. dl324

    Distinguished Member

    Mar 30, 2015
    Nothing ever seems to be as easy as you thought it would be...
    How were you expecting this to work? LEDs these days have a forward voltage of over 2V. If you have no "headroom" in the supply, you can't control LED current.
    Make this a learning experience. Teach your child the correct way to undo this mistake and teach him/her that mistakes can happen if you don't have a plan before starting a project.
    You never wire LEDs in parallel without ballast resistors because there will be slight variations in forward voltage; even for LEDs of the same brightness category. LEDs hogging current may appear to be brighter than the others and that is generally objectionable.
  7. dingo95966

    Thread Starter New Member

    Oct 14, 2015
    OK i took your guys advice and scrap the project and starting over. I did have the plans drown out its just that i wired it wrong but i did use LED "wizard" sound like a bad place to get my calculation on ohm size.This is my scenario i wont to hook up a 12v system run both 2v and 3v led 20ma and i have 78 led. what would be the best way to hook them up series or parallel and how do account for different voltage size so if i do led in series 2v+2v+3v+3v=10v would the led burn out because the input is 12 v
  8. KeepItSimpleStupid

    Well-Known Member

    Mar 4, 2014
    It will be a number of strings in series/parallel. Your LED drops are a little course. Go here: https://learn.adafruit.com/all-about-leds/the-led-datasheet and note that there is a range of Vf. Using max Vf is what you should use. Sometime min Vf is needed too.

    Without some means of current limiting, the LEDs may burn out.

    Let's say you have red and green LEDs. There is a possibility that the brightness could be very different at the same current.
    So, I would take two LEDs in series (in reality all of your colors) first and compare the brightness at the designed current.

    e.g. R <= 12-(2+2)/20e-3 If that works and it's the "right" brightness, your good to go.

    If your numbers are right then R <= 2/20e-3 or 100 ohms.

    The wattage of the resistor needs to be greater than (20e-2)^2*100 for 20 mA and 100 ohms.

    Your 2V headroom might be too close, so I would repeat the calcs using Vf(min) and also Vfmax You will get different values of R. Picking the average is probably the way to go. The calculation may tell you you have to have increased headroom.

    This is too simple to blindly follow a LED calculator. It's also complicated.

    Why is it complicated?

    The eye has different sensitivities to wavelength. The LEDs are different colors. The voltage drops are different because of the colors.

    So, before you do any wiring you really need to determine if the blue/green/white/red or whatever LEDS are the same perceived brightness at the same current. So, put all of the colors together and test. The datasheet probably doesn't include the eye's sensitivity. Not sure.

    If that test passes, then fine tune the resistor to the desired brightness. If the circuit is battery powered, Vf combined with the battery discharge curve matters.

    You can Pulse Width Modulate to change the brightness. That's another topic. You can also add a high wattage resistor to lower the brightness. You can also use an IC specifically designed for this.

    Initial resistor size
    Final resistor size based on brightness/Power supply (e.g. Battery or regulated or unregulated 12 V power supply)

    Hopefully, you can mix and match to get the same brightness.

    Just my take.
  9. dl324

    Distinguished Member

    Mar 30, 2015
    Both. The first thing you should do is separate the LEDs according to luminous intensity at the planned current. I suggest you try some LEDs at 10mA to see if they're sufficiently bright at a lower current.

    The picture below shows luminous intensity data for 4 different LEDs; each with 3 different intensity groupings. The thing to note is that even within an intensity grouping, there is a range of brightness.

    You should have some way of limiting current. What is the purpose of the connected LEDs? Are they just for show? Do you want to group them by colors or intermix them?

    If you connect a string similar of LEDs in series, you can simplify your resistor calculations. Or you could get a little fancier and make a simple constant current source to drive each series string of LEDs and not have to worry about calculating current limit resistors for each string (you haven't stated whether the power source is a battery or a power supply). There was a post within the past 2-3 weeks about a constant current source that used 2 transistors and 2 resistors. A simpler one could replace one of the transistors with an LED. An example of the latter was posted a month or so ago.
    Last edited: Oct 14, 2015
  10. Bernard

    AAC Fanatic!

    Aug 7, 2008
    It is not much trouble to check all LEDs for Vfd. Last October checked 100 SMT LEDs & ended up in just two stacks, 2.8 & 2.9; averaged 3 sec/LED. Make up strings that come close to 9 V. In simple LED tester, blue LED Vf should be greater than LED under test, maybe 2 red LEDs in series for white or other LEDs .
    I use 75% LED V & 25% for resistor.
    LED Tester 00000.png
    Last edited: Oct 15, 2015