Connect with us

Dim output on one color on LED PAR

Discussion in 'LEDs and Optoelectronics' started by partyanimallighting, Jul 8, 2019.

  1. partyanimallighting


    Oct 22, 2012
    Well, first let me say it's been quite a while since I posted and the change up of the website took me by surprise and it took me a while to find out how to add a new thread but here I am. I have some LED PAR's in my lighting rig and one of them recently developed a problem. The units are 18 x 10 Watt Indoor LED PAR's operating on 120VAC and powered via a 24VDC 6.5A power supply, very basic, power supply to mainboard and display PCB feeding the LED diode plate. On the particular unit, all colors are working but the RED output is very dim compared to the other colors (Green, Blue, White). I've tried the diode plate on another PCB and the RED is fine so it's definitely a mainboard problem. So, on the problematic mainboard, I've so far changed the capacitor and the resistor (R220), and everything's fine so I'm lost. I know that the capacitor could cause the problem but it's fine. Any assistance to increase my knowledge would be greatly appreciated. I've attached a mirrored image of the PCB (front and back). Thanks!!!! LED PAR Complete.jpg MAINBOARD FRONT.jpg MAINBOARD MIRRORED.JPG MAINBOARD REAR.jpg
  2. davenn

    davenn Moderator

    Sep 5, 2009

    Which capacitor ?

    Did you do that other soldering ?
    It looks like that cap to the right of the left hand IC is no longer connected at one end
    And the IC may have shorts between pins that need to be checked out

  3. partyanimallighting


    Oct 22, 2012
    Yes. The soldering is poor but I'm getting a little long in the tooth and the components are tiny soooo.......those poor solder points are the parts I swapped out from an old mainboard just messing around looking for the cause of the problem. When the problem is resolved, I'll neaten up everything. I promise. As for the missing cap, it's missing on all the other mainboards so it's not a functional component that accidentally broke off. Oh, and I checked the solder points and there aren't any shorts between the pins.
  4. partyanimallighting


    Oct 22, 2012
    Yellow circles denote parts that were swapped out. PARTS SWAPPED OUT.jpg
  5. davenn

    davenn Moderator

    Sep 5, 2009

    Ok so you need to fix up the, ummmm, not too good soldering. Note what I said in the last post :)
  6. partyanimallighting


    Oct 22, 2012
    Cleaned up the solder points a bit. Which cap were you referring to?

  7. 73's de Edd

    73's de Edd

    Aug 21, 2015
    Sir partyanimallighting . . . . . .

    In none of the photos was I able to read the ID's of the mini flat pack I.C.s that feed drive to the 4 POWER XSTRS ? FETS.
    So, you have ascertained that your worked on (over) circuitry is for the RED drive , and that the PAR substitution isolated it as not being a LED fault. ?
    If being my situation, I would next isolate the final LED power drive circuitry by going to the 8 pin IC that they want to board designate as 240K .

    Each of the four units being used, has its #4 pin connected to the base/gate of its companion power semiconductor.

    I would use an Exacto knife with a #11 blade to open an adequate width cut in the foil to open the drive path of your "RED" stage and also on the stage just next to / above it.
    Then get some KYNAR wire wrap `30 gauge wire and make 4 jumpers and pre tin all leads and then solder tack them into connection to cross jump the drive outs from one unit, going to the other companion POWER semiconductors base/gate.

    Then if upon powering up and testing, the RED is still deficient . . . . . but now . . . . since its now receiving its input from another driver . . . . . you have one Ecap, one large inductor, one power diode and one emitter /source resistor to suspect in the RED power /output section.

    Feed back, what are my involved components mentioned . . . . ID numbers . . . .the 8 pin I.C. and the semiconductor, power devices, also the lower corners switching regulator LM series ? full part numbering.

    73's de Edd . . . . .

    <-------- All of my information . . . . . just went data way -------->

  8. partyanimallighting


    Oct 22, 2012
    Hi 73's de Edd, glad to hear from you and I recall you've helped me in the past so thanks in advance once again. I've learnt quite a bit since you last assisted me but I still have so much to learn. I have ascertained that it is not an LED problem as the plate works fine on another PCB so it's definitely the main driver PCB. When I was "troubleshooting", I decided to swap out the "mini flat pack IC" on the RED path with the one from the GREEN path and when I tested it, it was still dim and the GREEN was fine so it's not that IC (I far). Following your instructions, I understand now that each of the "mini flat pack IC's" drive an individual POWER XSISTOR / FET via the #4 pin. So far so good. Then I got I want to be absolutely sure that I understand your instructions. Do I have to cut the foil to open the drive path or can I just lift the output leg on the RED power xsistor/fet from the board? Where should I do this? From pin#4 to the FET? After this, I'm definitely not sure what I have to do. I'm supposed to repeat the process on the next stage (GREEN) and then solder 4 individual jumper wires onto each of the #4 pins on each of the mini flat pack IC's and switch around and test ie: the BLUE output on FULL to the RED path and, if the RED lights dimly, it's the few components remaining in the output path that are the likely problem. I've tried my best with the feedback on the component numbers with two magnifying glasses!! Please be patient with me. It's a learning process.

    Attached Files:

  9. partyanimallighting


    Oct 22, 2012
    Any feedback??
  10. 73's de Edd

    73's de Edd

    Aug 21, 2015
    Sir partyanimallighting . . . . .

    Why 'soitanly . . . .here is a run down of what is being seen, using your 9 initial and 3 updated photos.
    On the first set of photos . . . #3 . . . on jpeg-45635 . . . . serves me best.
    I am seeing an octet of 50 ohm loading / matching resistors, relating to the 16 bare header pins at the top right of the board . They are associated with the in / out interfacing of the LED 7 seg display and control pots on the offboard communications board of the very first photo, showing it being plugged in, along with the LED PAR board.

    In its proximity, I also see the central / edge located WHITE 3 pin molex that mates with the 75176 line commnications chip that would permit 2 way serial digital interfacing into the STS u/p chip.
    I see that you are not using that input, but the first one that I mentioned .

    Top left corner, I see the top electret microphone element , with its black cloth windscreen.
    ALSO I see a LM358 to its right, which would relate to analog audio processing into the u/p, so that this unit would also have the option of audio pick up and modulating the LED displays to have a "LIGHT ORGAN" display capability, modulating in unison with the picked up music.

    There is being a BLUE 500K trim pot that is too high to be related to the to be mentione SMPS below , so its rotor must route up to the 1ufd block ceramic cap for pot wiper noise filtering and then to the 51K and into the LM358 for mike sensitivity adjustment for the "LIGHT ORGAN" mode of use.

    On the left just below the mike we see the component cluster of a switch mode dedicated power supply with its associated ringing inductor , rectifier and E-capacitor for a dedicated Minus 5 volt power supply.
    On my initial photo sightings, I could not make out the variant of that LM25XX chip, as for its being variable or fixed value voltage output . . . . now its seen as being minus 5.

    Traveling across the board to the right, we encounter the six sections within a flat pack 74H*C04. It appears to be the interfacing between the systems STC608AD-5V *** u/p and the four POWER LED drivers circuitry.
    Now I need YOUR eyes for inspection and some info retrieval.
    Look at the area between the left side of the u/p and the 7404 and you will see :

    At the very top, you see a silver via, to dissapear to the other side of the board for its foils completion path.

    Then a foil path that runs to the left to get out of any further viewing, by being blocked by the white 10K silk screen symbolization.

    The third foil seems to take a hook down and connect right into pin 1 of the 7404, which is a valid inverter input, as is being confirmed by the chip pinout, being placed below.


    The next foil goes right into a silver via to complete its path on the other board side.

    Then there two final foil paths at the bottom which route underneth the 7404, and I am unable to see.
    The output ports of the u/p should go into a pin 1-3-5-9-11 or 13 of the 7404.

    While the solely needed 4 output ports of the 7404 should be pin 2-4-6-8-10 or 12. The opted for / utilized 4 connections should go to the to the 4 vias that you see just in the foreground from the 7404 which has a quad resistor pack with its 4 side by side 1K resistors within it.
    Confirm if I don't then see the left resistor sneaking under the large E-cap to its left and ending up at the pin 5 input into the RED driver IC chip.
    The second resistor in that pack makes a quick route down to the pin 5 input into the GREEN driver IC chip.
    The third resistor in that pack makes a sneaky under and around the large E-cap to its right and down to the
    pin 5 input of the BLUE driver IC.
    The fourth resistor makes a vewy-vewy round about trip, eventually getting to the pin 5 input of the WHITE driver IC.

    NOW . . . they seem to have initially assigned the board markings as " 240K " cryptic shorthand for these 4 drivers board markings.
    I see no similarities in the ???? markings ???? on the cases of those IC's with that cryptic 240K, so I'm guessing that a production change altered the part # being finally used. . . . . AFTER PC bare boards were already fabricated.

    With my best enhancement if the most visibly marked IC case , I agree with 6 characters being present on the first line.
    One usually expects the second line typically being proprietary manufactiring data, akin to plant, shift, date code, etc.
    I cannot get /relate to your 6 digits identification of the unit.
    It looks more like a 5 or S, then a Q or O or zero, then agreeing on the possible two 9's , but the next item is not a vertical up and down slash as an I or 1 would be.
    Then the final charactewr looks like a skinny zero or a skinny stylized backwards D .
    Laser cut etching is universal for these markings . . . . except stamp pad bogus manufacturers.
    What I think has happened, is that the conformal coating has filled in the laser cavities flush and made the contrast then being absolutely minimal.

    OPTIMIZED IMAGE . . . . Mystery "DRIVER" I.C.


    Try HARD pressing a Q- tip wetted in denatured alcohol / or / Paint thinner / or / Lacquer thinner / or / Methy Ethyl Ketone / or / Xylene / or / Acetone / or / Naptha ( aka Ronson lighter fluid) / or / Brake cleaner / or / Fuel injector cleaner . . . . . . . types of more aggresive solvents and rub vigorously to see what takes the conformal coating off and leaves you with a thwen relatedly more contrasty and readible marking.

    No problems, with your now confirming the POWER FET's used as being N channel POWER FET's by INT RECT

    SUMMARY . . . . .

    Your swapping the driver iC's between the RED and GREEN with no changes confirmed both of them being OK . . . . if the GREEN then still operated O.K.
    So you can disregard my suggestion of lifting and cross swapping the driver output connections between the gates of the RED and GREEN power FETS. You have already effectively accompliahed that test.

    Now we have to sort out, if the problem is being the signal into the drivers from the u/p thru the 7404 stages
    OR if the output of the 4 POWER FETS to the associative 1 inductor, 1 E-cap and 1 diode sets that are used within the
    separate 4 stages to that WHITE Molex pin connector.

    What you might do now is power up the unit and monitor at the top of the resistors in the resistor packs four 1K resistors with a DVM set in AC voltage testing mode and swing that colors adjustment pot thru its range.
    E.G. . . . . . I'm expecting that if you monitor the top of the second 1K resistor for GREEN info, that you will get a measurable reading at MAX CW pot position and a decreasing one at MID pot position.
    Then step a resistor to the right for BLUE info and then for WHITE info and expect all of them to read similar and FINALLY move back to resistor 1 and see how its reading compares to the three normally functioning
    ones previously tested.
    If that last test differs, we work on that , if being the same as the three others, we concentrate on the power outputs 4 components.

    * Details

    * 74H*C04 recheck that IC and expect it to be a misread, and NOT to be a N, but a HC instead, at the speed this unit is being expected to work at.

    *** It's 8051 design based and this unit is clocking at 24 Mhz

    BTW . . . . .
    Very 1st photo . . . . very top light cluster at 12:00 position . . . . now drop down one position.
    ??????? What happened there, some GOOBER try to snuff out his cigarette butt , there ?
    Or do you have outdoor use and an insect nesting problem ?

    73's de Edd . . . . .

    I was born by Cesarean section . . . . . but not so you'd notice. . . . . It's just that when I leave a house, I go out through the window.

    Last edited: Jul 15, 2019 at 2:50 PM
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day