Connect with us

Inverter Repair... I could use some pointers from the experts.

Discussion in 'Electronic Repair' started by Michael Kennedy, Apr 19, 2006.

Scroll to continue with content
  1. I have a NEC Multisync LCD 1525M Monitor with a bad inverter (built by
    Potrans model# DA00812600). It had a burnt 22 pF 3kv ceramic capacitor. It
    is the cap just before wher the ccfl lamp hooks up. I replaced this cap with
    the closest value I could find at the local surplus dealer, a 24pF 1kv cap.
    I also replaced the .5a fuse with a regular fast blow and tried it out. It
    worked for a short while. I powered it off and back on and it blew the .5a
    fuse. I don't know if the original was a slow blow or not, but I can't get
    anything out the side that had the burnt cap.

    I can't find any shorts on the board and don't really know where to go from
    here.. Any poiters before I replace every component on the board?

    Here are some pictures.. I know the fuse i put on there looks really bad but
    I didn't have a small resistor sized fuse to go on there.
    Before
    http://mikescomputers.homeip.net/LCD/DSC00357.JPG
    http://mikescomputers.homeip.net/LCD/DSC00358.JPG
    After
    http://mikescomputers.homeip.net/LCD/DSC00365.JPG
    Whole board
    http://mikescomputers.homeip.net/LCD/DSC00361.JPG
     
  2. Dave D

    Dave D Guest

    This is a bit vague to me. Are you saying there are two CCFL tubes and the
    transformer has a seperate secondary and coupling cap for each lamp, and
    that one lamp worked for a while? If only one lamp lights that should tell
    you the inverter is working but there is a problem with either the relevent
    secondary winding (check the DC resistance matches the other) the continuity
    of the PCB (it looks very badly burnt) or the lamp itself. If the two lamps
    share a secondary winding then that narrows it down even further. Connect
    the other lamp to the 'working' output and see if it lights up.


    The PCB is a mess, and carbonised PCB will have a significant amount of
    conductivity to a HV high impedance source. Consider cutting away all the
    offending bits of the PCB and if necessary soldering the capacitor straight
    to the transformer leg. It won't look pretty bit it's better than the
    ramifications of presenting HT across burned PCB! It's possible the heat
    generated during the burning has damaged the transformer.
    It's proably not a short on the primary side, more likely HV leakage on the
    secondary side. The secondary of the inverter will knock out approaching
    IIRC around 90V or so into the right load, and several thousand volts off
    load. A relatively high resistance leakage on the secondary could easily
    draw enough extra supply current draw to pop the 500mA fuse.

    Don't worry about that. Get it working then worry about aesthetics.

    Dave
     
  3. I've tried ohming from the fuse to just before the bad capacitor and it ohms
    1347ohms on the bad side and 1341 ohms on the good side.. So it doesn't
    look like I have a short of any kind that I can see. Would this be a good
    assumption..

    Sorry for posting so much but I really want to learn about fixing
    electronics. This inverter seems like a simple enough project. I just need
    some help.

    I have an old 200khz oscillosocope that needs calibrating but it works.
    Would this be usefull at all for working on this? I wouldn't immagine it
    being usefull at all without some sort of wave generator.

    - Mike
     
  4. Dave

    Dave Guest

    The original was probably a slow blow at probably 1 or 1.25 amp (but I'm
    guessing). The ccfl's are probably 4 watt bulbs, 2 of them, total draw 8
    watts, inverter efficiency is around 75% which means the whole thing is
    drawing 10 watts when running correctly. With a 12 volt supply and you get
    just under an amp draw.

    If you have a scope, see what kind of a trace you get on either side of C9.
    If the transformer is open on the secondary, you'll get a clean 1/2 wave on
    either side of C9. If it's working, you'll get the 1/2 wave but it will be
    misshapen and may have some harmonics on it. The voltage on C9 side is low
    since you are on the primary. The inverter will usually work for about 2 -
    4 seconds with the bulbs unplugged. The square chip in the center is the
    controller and will shut the inverter down if it doesn't sense the bulbs
    light up. Here's a file that should give you a close approximation of your
    inverter: go to http://www.anachip.com/eng/supports/tech/appnote.php and
    take a look at ANP005

    The other thing you can do is to check continuity on the transformer. The
    primary windings and the feedback winding are very low but the secondary
    winding should be 100 to 400 ohms if I recall correctly.

    That charred board may end up being your biggest liability since the voltage
    on that side of things normally runs around 600vac @ 60Khz or so. It might
    just be easier to find a cheap monitor with a cracked LCD on eBay and take
    the boards out of it.

    Dave
     
  5. First off thanks for the detailed reply!
    Let me try to clear up the vagueness.. It has 2 transformers one on each end
    of the pcb and each has a coupling cap. It also has 2 500ma fuses. The fuse
    which goes to the non working side keeps blowing. I had both lamps powered
    up and working for a short while, but they were a bit dim.(may be ccfl
    tubes) Then I shut it off and powered it back on and the fuse blew on the
    "bad" side. I havn't been able to get the "bad" side to light again. It just
    blows the fuse imedately. I've ohmed from the fuse to the transformer's
    output and it is within 6 ohms of the working side.
    I've cut away all the burnt PCB and soldered the cap directly to the lamp
    socket. I tried that and it still is blowing fuses so I guess thats not it.

    The transformer might be the problem. It has a resistance of 668 ohms vs 931
    ohms on the good one. I'm not sure which is the primary and secondary but
    I'm guessing the primary is the one which ohms close to zero. Check my
    picture to make sure I'm ohming them in the correct place.
    http://mikescomputers.homeip.net/DSC00367.JPG


    - Mike
     
  6. Dave D

    Dave D Guest

    The secondary of the transformer should be isolated from the primary side
    and therefore the fuse, so such measurements are irrelevent. Try checking
    the switching transistors for shorts, as the fuse is blowing I'd start
    there. I think you'll also find the transformer has a shorted winding. Take
    both transformers out of circuit and measure the windings, they should have
    very similar readings between pins. If not, you likely have a bad one.
    Yes, it will tell you if there is drive to the transformer and therefore
    confirm whether or not the primary side of the circuit is working.
    The inverter is an oscillator and generates its own waveform to drive the
    transformer, you don't need a 'wave generator'.

    Dave
     
  7. I beleive I have found a shorted zener diode at ZD1 it ohms 119 both ways.
    ZD2 on the opposite side of the board ohms 343 foward and over 700 reverse.
    I guess that is why I'm blowing fuses. Now to get a chart on how to read
    zener diodes. :)

    - Mike
     
  8. I took the transformers off and they both ohm very similarly... I think they
    are both good, but it is late and I'll give it a more careful check
    tomorrow.

    - Mike
     
  9. Okay.. I found where I had a short.. My output cap was shorting to
    something. I cut the lead off shorter and I no longer have a short and have
    stopped blowing fuses... Yey! :)

    I think I may have the wrong value zener. The inverter comes on for a split
    second and turns back off. Both Lamps are working now... well at least they
    are comming on for a split second..

    What now? This is the zener I replaced with a 1N5237b738 (supposedly a 8.2v
    zener)
    http://mikescomputers.homeip.net/LCD/zd2.jpg
    Someone told me that this was a 8v2 zener in another post titled SMT Diode
    Identification

    - Mike
     
  10. Dave

    Dave Guest

    Does it come on long enough for you to take some voltage measurements on the
    good side and start comparing with the bad side? Also some scope traces on
    the primary of the transformer would help. That should help you figure out
    if the replacement zener is correct.

    Another place to start is with the feedback circuit. Make sure the proper
    voltage is coming back to the controller chip. If you trace the low side of
    the ccfl plug (not the line with the output cap) you'll see that the low
    side hits a few small diodes, resistors and caps. Eventually it will end up
    at one pin of the controller chip. That line has to have a certain voltage
    on it to tell the controller chip to stay on. If the voltage back to that
    line is not right the controller shuts the inverter down after a second or
    two.

    Dave
     
  11. Could my problem be related to having a 24pF cap instead of the 22pF cap
    that it is supposed to have.
    The 24pF is only rated for 1kv also..

    My local surplus dealer didn't have any high voltage 22pF caps.. I'm
    guessing I'm going to have to order that part.

    - Mike
     
  12. I forgot to mention that it was the output cap.
     
  13. Check that your replacement capacitor is not shorted again. I repaired an
    inverter very similar to yours... even the failure mode was the same... the
    capacitor going to the CFL charred by itself and shorted. While it was
    shorted, the inverter would power on for half second, and then turn off by
    itself.

    Regarding your Diode question, maybe you will find this page interesting:

    http://www.marsport.demon.co.uk/smd/sod80.htm

    Towards the end of this page you will find a color chart for SMT diodes. If
    my monitor is rendering correctly the colors of the picture you posted, your
    diode has the following ring bands RED, BLACK, GREEN. Using the table
    provided, that would be a RLZ12B zener diode manufactured by Rohm.

    Here you have the datasheet for all Rohm diodes and their voltages:

    http://www.rohm.com/products/databook/di/pdf/rlz12b.pdf

    As you can see, you really need a 12V zener diode.

    I hope all this info helps you.

    Good luck!
     
  14. lsmartino

    lsmartino Guest

    Check that your replacement capacitor is not shorted again. I repaired
    an
    inverter very similar to yours... even the failure mode was the same...
    the
    capacitor going to the CFL charred by itself and shorted. While it was
    shorted, the inverter would power on for half second, and then turn off
    by
    itself.

    Regarding your Diode question, maybe you will find this page
    interesting:

    http://www.marsport.demon.co.uk/smd/sod80.htm

    Towards the end of this page you will find a color chart for SMT
    diodes. If
    my monitor is rendering correctly the colors of the picture you posted,
    your
    diode has the following ring bands RED, BLACK, GREEN. Using the table
    provided, that would be a RLZ12B zener diode manufactured by Rohm.

    Here you have the datasheet for all Rohm diodes and their voltages:

    http://www.rohm.com/products/databook/di/pdf/rlz12b.pdf

    As you can see, you really need a 12V zener diode, not an 8.2V one.

    I hope all this info helps you.

    Good luck!
     
  15. hmm.. Thanks.
    I found out how the diode got blown.. If you power the inverter on without a
    lamp it blows the diode.

    I tested the good diode and you're right it was a 12v zener diode.

    The odd thing is I had this working and put the monitor back togeather. As
    soon as I got it back togeather it stopped working and started powering on
    for only a second again.. err..

    I really think this could be related to my improper output capicitor that I
    used.. No one has commented on that. It had a 22pF cap and I replaced it
    with a 24pF and I also tried a 20pF.

    - Mike
     
  16. lsmartino

    lsmartino Guest

    I don´t think the 22pf versus 24pf will make any difference. Most caps
    have tolerances as high as 20%... so a cap marked as 22pf could have a
    capacitance as high as 26.4pf or as low as 15.6pf... and it´s value
    would be within tolerance. The circuit designer surely took that into
    account when the circuit was designed.

    What worries me is the 1kv rating of the substitute cap... try to get
    one of 2kv at least, and repeat the tests.

    Probably the 1kv cap is shorting under load.
     
  17. What worries me is the 1kv rating of the substitute cap... try to get
    one of 2kv at least, and repeat the tests.

    Probably the 1kv cap is shorting under load.

    hmm.. Cap shorting.. Well I'm not blowing fuses anymore but that does sound
    quite possible. I guess I'm going to have to order one since skycraft
    doesn't have any in stock.

    - Mike
     
  18. I think I fixed it!!! I think you were right about a capacitor shorting,
    just not the one we thought..

    Anyway I was at the surplus store today and bought 2 18pF 3kv caps and
    replaced both output caps on the inverter board. I replaced the one on the
    "bad" side of the board first to no avail.. I then thought why not replace
    the one on the "good" side and I did and it turned on..

    My only regret is ordering the 22pF caps from mouser that I no longer need..
    :) oh well you live you learn.. I'll classify this as an education
    expense.. It doesn't come anywhere what my tuition for school is and is way
    more fun..

    I probably couldn't have fixed this without the help of Dave, lsmartino, and
    Afra Dailey.. Thanks everybody..

    Just hopefully I havn't spoken too soon.. :)

    - Mike
     
  19. lsmartino

    lsmartino Guest

    Congratulations!

    You will see that the repair will hold over time.

    Good luck!
     
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

-