Announcement

Collapse
No announcement yet.

"Dick Smith" 55" LED backlight issue - model GE6880

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    "Dick Smith" 55" LED backlight issue - model GE6880

    Firstly this TV appears to have been made by Shenzen MTC (think that's the place) and its just re-badged as a "store brand" so it may be identical to other cheap brand TVs out there. Certainly looks like it has "generic" circuit boards inside.

    The TV is only ~4 months old but Dick Smith has gone into receivership and aren't honouring warranties otherwise I'd just go down that route

    Now the issue:

    Upon being plugged in there is a very brief flash of light from the back light. The panel itself still works as images appear dimly using a torch.

    based on this I'm assuming the fault is either in the back light assembly itself or in the PS board, because the panel itself still works and the tv responds to remote control etc.


    Power board is : TV5001-ZC02-01 it appears to be a generic 50" Power supply board. schematic available at: http://www.go-gddq.com/upload/2015-0...0618093118.pdf

    Schematic appears accurate except a few resistor values, generally feedback/setting pins on the ICs (I suspect these are because its driving a 55" screen/BLU not the 50" its intended for)


    power board output/input voltages are :
    12V: 12.6
    5V: 5.4
    5V stb: 5.4
    PSON: can't remember if 3 or 5. but it clearly turns the board on.
    BK/ON: ~3V
    ADJ: ~3V (does this and BK/ON seem right? not sure if they are meant to be closer to 3.3 or 5v)


    The back light unit itself consists of 2 separate circuits, each has 3 strips of 16 LEDs (48 total for each circuit, 96 in the entire assembly) as far as I can tell the LEDs within each circuit are all in series.

    I've tested each LED with the "diode test" on my DMM and every LED illuminates dimly. Also on power up it appears that every LED flashes.

    The PSU has 4 places for LED connectors, but only 2 have sockets soldered on, with each connected via a 0 ohm resistor to one of the disconnected sockets, this matches a possibility in the schematic so I assume it hasn't deviated too far from the design here.


    measured the voltages at startup and the LED+ pins each jumped to ~115V (coinciding with the flash) measured between the pin and ground then dropped (while writing this I realized that the LED+ pins are a common rail so of course they will be the same. Will have to measure between the two LED pins rather than to ground as that may help isolate the problem). My DMM is just a cheap one and probably doesn't have the fastest response time, so may well have gone higher.



    I cant see any burnt out chips or components. no fuses are blown. the TV screen keeps working and none of the voltage outputs turn off/change after the backlight does.

    Anyone able to help me pinpoint where/what it could be? Or even give me somewhere to start troubleshooting.

    I've attached some images of the PS board (assuming this is the most likely fault given what I've found so far). sorry for the poor quality, best I could do with my Phone camera. If better pictures are needed I'll try to borrow a decent camera.








    Attached Files

    #2
    Re: "Dick Smith" 55" LED backlight issue - model GE6880

    The circuit does not match (the circuit shows small TO-92 PWM controlled transistors, so other sections of the circuits may not match the board either), you have two LED driver circuits as you can tell by the two SMD PWM Controlled power MOSFETs close to the two LEDs connectors.
    The circuit will shut down if Over Voltage or Over current occur.
    What DCV do you get if the LED connectors are not in place?
    The LED may show OK when tested with meter but may fail when full Voltage/Current is applied.
    Last edited by budm; 04-29-2016, 09:30 AM.
    Never stop learning
    Basic LCD TV and Monitor troubleshooting guides.
    http://www.badcaps.net/forum/showthr...956#post305956

    Voltage Regulator (LDO) testing:
    http://www.badcaps.net/forum/showthr...999#post300999

    Inverter testing using old CFL:
    http://www.badcaps.net/forum/showthr...er+testing+cfl

    Tear down pictures : Hit the ">" Show Albums and stories" on the left side
    http://s807.photobucket.com/user/budm/library/

    TV Factory reset codes listing:
    http://www.badcaps.net/forum/showthread.php?t=24809

    Comment


      #3
      Re: "Dick Smith" 55" LED backlight issue - model GE6880

      yours way to test LEDs sorry to say wrong, they will illuminates dimly with Diode test that correct, but you need to test them at opposite way test as Diode, as they had Ziner Diode build in at Opposite. to forward currents to next LED, this Ziner become short contact or Open. so yours DMM Diode Test will beep on if one are short. or Zero if Open
      Attached Files
      Last edited by Diah; 04-29-2016, 09:58 AM.

      Comment


        #4
        Re: "Dick Smith" 55" LED backlight issue - model GE6880

        Originally posted by budm View Post
        The circuit does not match (the circuit shows small TO-92 PWM controlled transistors, so other sections of the circuits may not match the board either)
        I assume you are referring to the 4 transistors with their bases connected to the "baseX" pins of the OB3354 controller? I hadn't noticed that the package doesn't match whats specified, but they are still EN13003 transistors as stated in the schematic. As I said there does indeed seem to be some changes over the schematic, but I haven't seen any glaring major ones yet. there might be substituted components and slight value changes but that is to be expected seeing as the circuit as designed was for a 50" backlight! Every part of the circuit I have looked at/traced thus far has matched the schematic in terms of structure.


        The circuit will shut down if Over Voltage or Over current occur.
        Is there any way of knowing which failure (if it is one of these) has occurred and which chip is responsible? I'd probably be looking at the OVP, SCP and Fault pins of the OB3354 controller? unfortunately I cant find a datasheet on this chip so I don't know what these pins expect.


        What DCV do you get if the LED connectors are not in place?
        without the LEDs plugged in:
        con5 has 114~120V
        Con2 has 118~122V

        With LEDs plugged in:
        con5 has 112~114V
        con2 has 102~112V
        (all DC, all between pins on the connector)

        while measuring this I noticed that the peak voltage comes after the flash from the LEDs, so I suspect that there is a input lag in my DMM. so the peak V may well have been higher.

        The LED may show OK when tested with meter but may fail when full Voltage/Current is applied.
        I was a bit afraid of that, is there any way (without specialist equipment preferably) for me to verify if the LEDs are good/bad? I saw in another thread someone had set up a camera with a long exposure time and was looking for a dimmer LED that may indicate specific LED failure, would something like that be a good idea?



        you have two LED driver circuits as you can tell by the two SMD PWM Controlled power MOSFETs close to the two LEDs connectors.
        I cant see what SMD power mosfets you are referring to. From what I can see there are only 2 SMD MOSFETS on the board, both are on the output voltage rails to the mainboard (5V and 12V).

        I've attached one of the previous pictures with the connectors marked just so that we're on the same page. Also I noticed in the picture of the top of the board that the LED connectors don't show up easily, because it seems that when I pull the connectors out it pulls the plastic socket off the pins with the plug!




        yours way to test LEDs sorry to say wrong, they will illuminates dimly with Diode test that correct, but you need to test them at opposite way test as Diode, as they had Ziner Diode build in at Opposite. to forward currents to next LED, this Ziner become short contact or Open. so yours DMM Diode Test will beep on if one are short. or Zero if Open
        I'm confused about this statement. It is my understanding that the current goes through the LED itself to the next LED in the chain (they're all in a series configuration)? Wouldn't the Zener diode be there to protect against reverse biasing the LED?
        Attached Files

        Comment


          #5
          Re: "Dick Smith" 55" LED backlight issue - model GE6880

          What I'd do especially bec the set is so new is check all the boards look very closely and check for bad connections.I bet you find bad connections that maybe causing that issue.Believe it not I've seen either new or fairly new sets have all kinds of bad connections.For example I bought a 40 inch Samsung 3D as is set that was fairly new they said the set goes off and comes on by itself.I figured it's a bad connection I actually found several cold soldering spots and bad soldering connections. Just an added note Samsungs are also known to have a bad surface mounted capacitor on the main board that causes the same type of issue.That set has worked fine ever since.Note make sure you check all the plug in wire connections try unplugging them and plugging them back in again and check their soldering connections.
          Last edited by Electronic Handyman; 04-29-2016, 10:16 PM.

          Comment


            #6
            Re: "Dick Smith" 55" LED backlight issue - model GE6880

            OK, I traced out he wrong connection for the LED Cathode return lines, it uses two Transistors/or MOSFETs (need P/N) for each string of the LED, see the marking in the pictures.
            BTW please check the DIMMING CONTROL (adj) line pin 3 of the OB3354 to see if it drops out after the LED drop out.
            The OVP is done by sensing the Vdrops on the R257 (1K).
            I cannot find the spec for the ON3354 tey so I do not know what the FAULT pin 9 is used for, from the look of it, it looks to be an output for latch shutdown (Q115, 114) of the P-ON signal. the PR should normally be HIGH.
            BTW, the anti-parallel Zener in the LED is for ESD protection.
            OSRAM came up with built-in protection:
            http://www.ledsmagazine.com/articles...itigation.html
            Attached Files
            Last edited by budm; 04-29-2016, 10:52 PM.
            Never stop learning
            Basic LCD TV and Monitor troubleshooting guides.
            http://www.badcaps.net/forum/showthr...956#post305956

            Voltage Regulator (LDO) testing:
            http://www.badcaps.net/forum/showthr...999#post300999

            Inverter testing using old CFL:
            http://www.badcaps.net/forum/showthr...er+testing+cfl

            Tear down pictures : Hit the ">" Show Albums and stories" on the left side
            http://s807.photobucket.com/user/budm/library/

            TV Factory reset codes listing:
            http://www.badcaps.net/forum/showthread.php?t=24809

            Comment


              #7
              Re: "Dick Smith" 55" LED backlight issue - model GE6880

              I already had a look and can't Identify any recognizably bad solder joints or loose connections, so short of re-soldering every joint on the board/s I think that's the end of that path of investigation. It may well be a bad joint in the end, but without being visually identifiable the diagnostic approach really remains the same from here as far as I can see.

              Comment


                #8
                Re: "Dick Smith" 55" LED backlight issue - model GE6880

                Look at the ADJ signal to see if it is present when the backlights drops out or you can disconnect the ADJ line from feeding the board and see if the LEDs will stay on.
                Also check the DCV between GND and the Cathode return to see what you get.
                Last edited by budm; 04-29-2016, 11:03 PM.
                Never stop learning
                Basic LCD TV and Monitor troubleshooting guides.
                http://www.badcaps.net/forum/showthr...956#post305956

                Voltage Regulator (LDO) testing:
                http://www.badcaps.net/forum/showthr...999#post300999

                Inverter testing using old CFL:
                http://www.badcaps.net/forum/showthr...er+testing+cfl

                Tear down pictures : Hit the ">" Show Albums and stories" on the left side
                http://s807.photobucket.com/user/budm/library/

                TV Factory reset codes listing:
                http://www.badcaps.net/forum/showthread.php?t=24809

                Comment


                  #9
                  Re: "Dick Smith" 55" LED backlight issue - model GE6880

                  Originally posted by budm View Post
                  OK, I traced out he wrong connection for the LED Cathode return lines, it uses two Transistors/or MOSFETs (need P/N) for each string of the LED, see the marking in the pictures.
                  BTW please check the DIMMING CONTROL (adj) line pin 3 of the OB3354 to see if it drops out after the LED drop out.
                  The OVP is done by sensing the Vdrops on the R257 (1K).
                  I cannot find the spec for the ON3354 tey so I do not know what the FAULT pin 9 is used for, it looks to be an output for latch shutdown of the P-ON signal. the PR should normally be HIGH.
                  BTW, the anti-parallel Zener in the LED is for ESD protection.
                  The part numbers are EN13003, they appear to be a NPN transistors. I pulled all 4 out and did a transistor test (just the basic one that pops up when you google "NPN transistor test") and they all appear to check out OK (as far as that test can diagnose). That part of the circuit you traced matches whats in the schematic with a few differences like only 4 diodes not the 8 in the schematics and the jumper 0Ohm resistors, but again it is functionally the same, and seeing as the set worked for a few months I cant really see that wrong components or slightly different setup is the reason it isn't working (it may well be the ROOT cause of the failure especially if they cheaped out when they redesigned it for 55" not 50" and didn't use correctly rated components for that change!)


                  the ADJ line starts at 0V then goes up to ~3V approximately coinciding with the flash (oh how I wish I had better test equipment than a multi meter!) and remains at ~3V.

                  the FAULT pin goes up to 5V almost immediately after being plugged in and stays there.

                  So that really leaves the OVP or SCP correct?





                  On second thoughts.. Is it possible the fault is in U103 (OB2263) or T103 or their associated circuitry? I originally measured LED+ to ground, and the LED+ is just the smoothed/filtered output of T103 Pins 9+10. so if the fault were the OB3354 (sorry think I mis-labeled this earlier as "ON") then LED+ should remain high? but seeing as this isn't the case is it likely that the OB2263 chip has gone into protect mode not the other one? OR have I just messed up my understanding of how this thing works somewhere

                  Comment


                    #10
                    Re: "Dick Smith" 55" LED backlight issue - model GE6880

                    Originally posted by budm View Post
                    Look at the ADJ signal to see if it is present when the backlights drops out or you can disconnect the ADJ line from feeding the board and see if the LEDs will stay on.
                    Also check the DCV between GND and the Cathode return to see what you get.
                    by "cathode return" where do you mean? as in the cathode pin of the LED or elsewhere on the circuit?

                    Comment


                      #11
                      Re: "Dick Smith" 55" LED backlight issue - model GE6880

                      So PWM stays OK, the PR (FAULT, I believe this pin goes LOW if the total LED current draw is above the limit if some LEDs shorted out, may be) signal stays HIGH other wise it would have kill the P-ON signal and shutdown the power supply PFC booster circuit and the rest of the switched supplies, so yes that leave the OVP signal, I do not know what the SCP is.
                      BTW, what DCV do you get on the Cathode return lines, they should be very close to each other. To fool the OVP, you can add another 1K resistor in parallel with the R257 just to see if the LEDs will stay on.
                      Last edited by budm; 04-30-2016, 12:02 AM.
                      Never stop learning
                      Basic LCD TV and Monitor troubleshooting guides.
                      http://www.badcaps.net/forum/showthr...956#post305956

                      Voltage Regulator (LDO) testing:
                      http://www.badcaps.net/forum/showthr...999#post300999

                      Inverter testing using old CFL:
                      http://www.badcaps.net/forum/showthr...er+testing+cfl

                      Tear down pictures : Hit the ">" Show Albums and stories" on the left side
                      http://s807.photobucket.com/user/budm/library/

                      TV Factory reset codes listing:
                      http://www.badcaps.net/forum/showthread.php?t=24809

                      Comment


                        #12
                        Re: "Dick Smith" 55" LED backlight issue - model GE6880

                        Originally posted by rockit711 View Post
                        by "cathode return" where do you mean? as in the cathode pin of the LED or elsewhere on the circuit?
                        The Cathode return are the pin for each LED connector that I marked which is connected to the Collector (center pin) of those transistors.
                        Never stop learning
                        Basic LCD TV and Monitor troubleshooting guides.
                        http://www.badcaps.net/forum/showthr...956#post305956

                        Voltage Regulator (LDO) testing:
                        http://www.badcaps.net/forum/showthr...999#post300999

                        Inverter testing using old CFL:
                        http://www.badcaps.net/forum/showthr...er+testing+cfl

                        Tear down pictures : Hit the ">" Show Albums and stories" on the left side
                        http://s807.photobucket.com/user/budm/library/

                        TV Factory reset codes listing:
                        http://www.badcaps.net/forum/showthread.php?t=24809

                        Comment


                          #13
                          Re: "Dick Smith" 55" LED backlight issue - model GE6880

                          OK, I just reread your post 9, are you saying that the L+ goes down or drain down to 0V or so after the backlights go off?
                          Never stop learning
                          Basic LCD TV and Monitor troubleshooting guides.
                          http://www.badcaps.net/forum/showthr...956#post305956

                          Voltage Regulator (LDO) testing:
                          http://www.badcaps.net/forum/showthr...999#post300999

                          Inverter testing using old CFL:
                          http://www.badcaps.net/forum/showthr...er+testing+cfl

                          Tear down pictures : Hit the ">" Show Albums and stories" on the left side
                          http://s807.photobucket.com/user/budm/library/

                          TV Factory reset codes listing:
                          http://www.badcaps.net/forum/showthread.php?t=24809

                          Comment


                            #14
                            Re: "Dick Smith" 55" LED backlight issue - model GE6880

                            OK, I found out what the FAULT pin 9 is for SHORTS OR OPEN LED STRING error out signal.
                            Never stop learning
                            Basic LCD TV and Monitor troubleshooting guides.
                            http://www.badcaps.net/forum/showthr...956#post305956

                            Voltage Regulator (LDO) testing:
                            http://www.badcaps.net/forum/showthr...999#post300999

                            Inverter testing using old CFL:
                            http://www.badcaps.net/forum/showthr...er+testing+cfl

                            Tear down pictures : Hit the ">" Show Albums and stories" on the left side
                            http://s807.photobucket.com/user/budm/library/

                            TV Factory reset codes listing:
                            http://www.badcaps.net/forum/showthread.php?t=24809

                            Comment


                              #15
                              Re: "Dick Smith" 55" LED backlight issue - model GE6880

                              Originally posted by budm View Post
                              OK, I just reread your post 9, are you saying that the L+ goes down or drain down to 0V or so after the backlights go off?
                              correct, LED+ (measured between the LED+ rail and ground) appears to drop down to 0V (not instantly, but maybe a second or 2, I'll time it if speed is important) after the backlights flash.

                              I also just measured the cathode return, it appears both start at 4.8V as soon as plugged in, then both drop to ~0V at/after the LEDs flash, then sit as ~0.2V both quite close to each other.



                              Also playing on my previous hunch I measured the FB pin of U103 (OB2263 chip), its the output of a feedback photo transistor (U203) and it stays at ~0V. There are some slight fluctuations, the highest I saw it hit over several power up cycles was 0.4V but at no point during my measurements did it get much above 0V for any amount of time. the datasheet for this IC (U103) is avaliable:

                              https://cdn.badcaps-static.com/pdfs/...bcd438bec6.pdf


                              I'm a bit confused about exactly what the FB pin does for this chip, but I suspect it isn't supposed to be at 0V the entire time! I might be mistaken, but it looks like it should be between .7V and 3.7V and be a factor in the PWM duty cycle for the switching MOSFET, in which case 0V may simply shut down this MOSFET permanantly?

                              Comment


                                #16
                                Re: "Dick Smith" 55" LED backlight issue - model GE6880

                                Originally posted by rockit711 View Post
                                I'm confused about this statement. It is my understanding that the current goes through the LED itself to the next LED in the chain (they're all in a series configuration)? Wouldn't the Zener diode be there to protect against reverse biasing the LED?
                                No , Zener Diodeare responsible to forward the currents to next LED.

                                you have one Shorted, please take look on this diagram to understand how most LED Backlight working. test the Zener of each you will find one shorted.

                                Translate German to Eng. Kurz Anschlß = Short contact, Licht Diode = Light Diode
                                Attached Files

                                Comment


                                  #17
                                  Re: "Dick Smith" 55" LED backlight issue - model GE6880

                                  If what you say is correct then I don't see how the LEDs work at all. Unless i really misunderstand something or am making a stupid mistake somewhere; in that configuration the LEDs are all wired in backwards! (Anodes to ground). So none of them would work like that and the current would flow through the zeners ONLY. Also its irrelevant if the zeners have gone open as the LEDs would still work, and if the zener has gone short circuit the LED wouldn't have illuminated under then diode test.

                                  Anyway, starting to suspect the u103 chip is the cause. So will do more poking around that chip and its associated circuitry tomorrow and hopefully figure something more out.

                                  Comment


                                    #18
                                    Re: "Dick Smith" 55" LED backlight issue - model GE6880

                                    You can get an idea of how OB3354 works from the following schematic.
                                    Attached Files
                                    Willing to help but I'm no expert.

                                    Comment


                                      #19
                                      Re: "Dick Smith" 55" LED backlight issue - model GE6880

                                      as you like rockit711.

                                      Comment


                                        #20
                                        Re: "Dick Smith" 55" LED backlight issue - model GE6880

                                        Originally posted by rockit711 View Post
                                        I'm a bit confused about exactly what the FB pin does for this chip, but I suspect it isn't supposed to be at 0V the entire time! I might be mistaken, but it looks like it should be between .7V and 3.7V and be a factor in the PWM duty cycle for the switching MOSFET, in which case 0V may simply shut down this MOSFET permanantly?

                                        Looking at the pdf the Opto is switched on by the secondary DC voltage so the transistor within the package would be switched on. So having 0V at the FB pin looks correct to me.
                                        Willing to help but I'm no expert.

                                        Comment

                                        Working...
                                        X