Connect with us

LG W2246 Caps/Inverter?

Discussion in 'Troubleshooting and Repair' started by mwkenna, Oct 17, 2012.

  1. mwkenna

    mwkenna

    9
    0
    Oct 17, 2012
    Hi guys

    My LCD monitor recently started to seemingly power itself off after a few seconds after turning it on but the power light would stay illuminated, after some research it seems as though it could potentially be failed caps (although there are no visible signs).

    I have seen that there is a "flashlight trick" where you shine a flashlight into the monitor after it "goes black", I am assuming that this will test if the backlight is down only. I have performed this test but can't see any image on the screen - does this mean that the caps are likely not a candidate?

    If it helps, sometimes I notice a high pitched screech coming from the corner of the screen just before it goes off - not every time though.

    Thanks for any help offered!

    Mark.
     
  2. Jotto

    Jotto

    120
    0
    Aug 24, 2012
    You have a unbalanced load at the lamps, it uses a royer oscillator circuit or something similar. I have had caps that are bad on the inverter, but most of the time it doesn't really affect operation of the unit.

    I would check the transformers and lamp drive chips, replace the caps if you feel more comfortable doing that to eliminate that from the problem. They are cheap and usually there isn't a bunch of them. Usually you can use the caps for a ground point and test the lamp drive chips for shorts, they almost always short out.

    The flashlight trick will work if you have a place on the panel that will allow light on the tft screen. I use a amp meter to prove that. You can actually disconnect the tft panel and see a different current draw, tells you its working without having to use a flashlight. Or you can take the panel apart and usually see the picture (not recommended).

    Would also take a good look at the power supply, if your going to work on it might as well work on the high failure parts.
     
  3. mwkenna

    mwkenna

    9
    0
    Oct 17, 2012
    Hi Jotto

    Thanks for the info. I have replacement caps which I'm going to try tonight. If this fails I will indeed check the chips for shorts.

    One thing that I did notice was that the specified current output was different from what I saw on the multi-meter. The connector to the "VGA connector PCB" specifies 12V but I was getting more like 17 - does this shed some light?

    Also, if you could let me know how I would test the chips for shorts that would be great (dont have much experience with this level of electronics). I can post pictures if required?

    Thanks for the help!
    Mark.
     
  4. Jotto

    Jotto

    120
    0
    Aug 24, 2012
    Usually they are 8 pin chips. One side is the input and feedback, the other side all 4 pins are on input of the transformer. I usually just find a capacitor and hook a clip lead there and then to my meter (the ground side). I would assume you have a digital meter, put the meter in diode test then test all the pins on the input side of the chip, not the one connected to the transformer.

    This is where having knowledge of using a meter will help you. Look up the datasheet for the chips in question. Sometimes these will blow a hole in the chip making it easy to locate, sometimes it looks just fine but is still bad. Just check each one for a short.

    Testing the transformers is a little different, you can still use a meter, but its not a very good test. I use my scope as a function generator to test these, again not a great test but much better then a meter. If you have a scope and would like to try this I will post again on how to use it for testing the transformers.
     
  5. mwkenna

    mwkenna

    9
    0
    Oct 17, 2012
    I have replaced the caps and the problem remains. I am unsure of the location of the lamp drive chip(s) so I will post a photo of the board in question tomorrow. There are 3 distinct boards as far as I can tell:

    1- The thin and long board attached to the screen itself with ribbon connectors all the way along.
    2- The power board which contains the caps, transformer, power input connector
    3- A small board housing the VGA connector with a ribbon connector that goes to (1)

    Please forgive my ignorance however I am keen to learn.

    Thanks!
    Mark.
     
  6. Jotto

    Jotto

    120
    0
    Aug 24, 2012
    1. Inverter board, cables on side should connect to the lamps
    2. Power supply
    3. A/D board analog to digital

    Cable going from A/D board to inverter will be your Vcc, brightness, and switched voltage.
     
  7. mwkenna

    mwkenna

    9
    0
    Oct 17, 2012
    As promised I have taken some photos of the boards. Please can you indicate on the photos what to look at next. Note that I have now replaced the 16v caps on the power board but the problem persists.

    Link to photos:
    https://www.dropbox.com/sh/ei1hrkif5ty7yss/pOKZJSSTfn

    I appreciate the help!
    Mark.
     
  8. mwkenna

    mwkenna

    9
    0
    Oct 17, 2012
  9. Jotto

    Jotto

    120
    0
    Aug 24, 2012
    I can't read the schematic, the dpi setting is too low when I print it out.

    Inverter and power supply is on the same board. You need to find the Vcc and switched voltage for turning the inverter on. Transformer seems to be good by what you have said. So that leaves a problem on the inverter or the lamps.

    Don't mess with the long board. That will not be your problem.
     
  10. mwkenna

    mwkenna

    9
    0
    Oct 17, 2012
    The lamps are illuminating and then switching off, the voltage going to both the lamps spikes around 3V and then drops back to zero after ~2 seconds.

    If I were to remove the lamp connectors (i.e. the screen is now totally disconnected) and re-measure the voltage and find that the voltage is still spiking then dropping back to zero that would indicate that the lamps are good and the problem lays in the inverter right? In which case I could just replace the inverter?

    Thanks,
    Mark.
     
  11. mwkenna

    mwkenna

    9
    0
    Oct 17, 2012
    And with regards to the PDF, if you save it to the PC first, you can zoom right in on the schematics very clearly. For some reason there is no ZOOM functionality in DropBox's PDF viewer.

    Cheers,
    Mark.
     
  12. Jotto

    Jotto

    120
    0
    Aug 24, 2012
    I think you have your meter set to DC, and should be AC. Transformers block DC and pass AC.

    Yes you could buy another power supply/inverter. An indication the lamps are going bad is a red tinge to them. The lamps will turn on, but if one is drawing more then the other it will shut down.

    Also, make sure you hook the meter up before you power the unit, if you try to test it while power is on it will shut down.
     
  13. mwkenna

    mwkenna

    9
    0
    Oct 17, 2012
    Ah my bad, so AC current goes to the lamp connectors. I *may* be able to test with another screen as I have another monitor that's very similar... Might be worth a go.
     
  14. mwkenna

    mwkenna

    9
    0
    Oct 17, 2012
    Ok I have located what I think is the lamp driver chip. Can you please verify how I would test this chip:

    http://www.dz863.com/pinout-821027163-APM4568J/

    With a multimeter?

    I have tested using the diode setting on the multimeter but I dont think I am doing it correctly.

    Thanks,
    Mark.
     
  15. Jotto

    Jotto

    120
    0
    Aug 24, 2012
    Since you have another monitor, I would take an use the lamps from the other panel for testing. Size of the lamps don't matter for a short test, but if the screens are different sizes I wouldn't let it run longer then 1 minute. Don't remove the lamps from the other panel, just use the panel next to the other panel.

    Notice that there is a gate, drain and source. They are numbered 1 and 2, just use your meter in diode test. I don't have this particular chip, but you will be able to turn it on an off. It will check like a diode until you turn it on, then it will act as a short. Desolder all the drains on the one side of the chip, or if easier for you remove the complete chip before testing.

    I will give you a example for testing this component. I used a K2698.

    Meter in diode test mode.

    Black lead to Drain, Red lead to Source should test like a diode. Remove Red lead and touch the Gate, return to the Source it should now be shorted. Remove the Black lead from Drain and touch Gate and then back to Drain and it should test like a diode again. Test both on the chip and see if one is bad.

    If it turns out to be a lamp, I can provide you with the lamps, be careful with the rubber boots. They break easily from the rubber fusing to the lamp. I also have those depending on which one your require.
     
    Last edited: Oct 19, 2012
  16. davenn

    davenn Moderator

    13,294
    1,758
    Sep 5, 2009
    Mark WARNING .... you realise there's 1000+ Volts going to the lamps from the inverter
    not a Voltage you should be trying to measure with a meter unless its a damn good one and DONT be touching the probe handles..... they may not be good enough to insulate you at that voltage!


    unfortunately that service manual doesnt give any actual circuit info on the power upply or inverter The driver chips appear to be on the underside of the PSU board you showed in photo3 of that initial set of photos

    Dave
     
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

-