Electronics > Repair

LCD monitor repair - inverter issue

(1/3) > >>

dr_p:
Hello!


For the last 3-4 months I've been trying to repair a broken LCD monitor. It's more of an ego thing now  :scared: so I won't give up. I read a couple of books about LCD repairs, I looked at tens of schematics, video tutorials, I figured thing out but now I'm stuck and I need help.


Long story short the LCD is externally powered by a 3.3A 12V brick. It worked for 3-4years, then a PCB fuse in the LCD blew. Someone shorted it with a piece of wire and it kept on working for 2-3 months (?!). Then it died, putting the power supply in current limit.


Now comes the funny part - someone worked on the inverter board (separate board) and replaced all the SOT-23 and SOIC-8 parts with other (random) parts. What?! They're the same size.  |O |O |O So I received an LCD unit with (basically) missing unknown transistors in the on/off (control) circuit, missing unknown MOSFETS and a yet unknown fault.

I replaced the fuse with a 5x20mm holder, replaced all capacitors (even if ESR, capacitance and looks were ok). I isolated the problem on the inverter board(except the 5V regulator, it's ok and happily powering the logic board).

I studied schematics and read stuff, and learned of the existence of digital transistors, so it's clear why there were missing resistors footprints on the PCB. As I had no equivalent parts, I soldered in TO-92s and 1/4W resistors. I used BC557, BC547 and 10K resistors.



The startup circuit now works and supplies 6.8V to the controller (BIT3713) that seems to want to start, but goes into shutdown. I then soldered in a pair of IRF7105PbF as they seem to be fine for the job, however they don't seem to open properly at 5V, so they are getting hot. After more investigation I found a schematic that is almost correct, except for slightly different values of passives. It's using FSD8958A MOSFETs so I went ahead and bought some and soldered them in.



I have picture now, but the controller still shuts down in 1-2 seconds. I checked DC resistance of both secondaries, they are the same - 1.5K or 15K, don't remember - so probably the transformers are fine. I checked the 4 feedback points (with 10M impedance DMM) and they were 0.25V, 0.17V, 0.25V, 0.35V. I then replaced the 1K resistor to ground (for the 0.17V point) with 2K. It rose to 0.40V and the LCD stayed on for 1-2 minutes drawing 1.5-2.2A from my bench supply. Mind you it looks too bright on the top, though. But then it draws more than 3.5A and constantly sends the PSU in current limit and shutdown :bullshit: so the LCD keeps power cycling. In that brief moment that is on I have picture.

I put a scope on the 12V input, the 5V rail for the logic board, and the 6.8V to the controller IC, all are relatively clean. MOSFETS seem to be fine, they are not shorted, they still get the 0-5V square wave that was enough to fully open them, capacitors are ok.

I reverted to the 1K feedback resistor and the problem is still there and I don't know what else to do. Now that I think of it there's a slight chance I only checked one MOSFET, not both. It was really late and I just gave up for the day.

Can someone please help?

Edit: P.S. this is the full inside layout, FYI:

wraper:
Probably one of the CCFL tubes is bad. Try to swap connectors and measure feedback points again.

dr_p:
well now it's not working any more, except for 1-2 seconds. All the feedback points are over 1V now.  :-//

mariush:
You can buy inverter boards from eBay for under 10$ and know for sure the inverter is working properly.

For example, see this : http://www.ebay.com/itm/10-25V-Universal-LCD-Inverters-Mini-LCD-Inverter-Board-for-17-22-Widescreen-/390651669520?pt=US_Monitor_Replacement_Parts&hash=item5af4a76c10

The 1-2 second thing is usually a timeout... the controller tries to ignite a cfl lamp and if it notices the lamp doesn't work within certain parameters (too high current drawn from transformer or it just doesn't "ignite" etc) so you may want to check the cfl tubes... often it can be a simple case of loose wire to the end of the cfl tubes, they get desoldered from the heat sometimes.
They also go bad due to old age, lots of hours... if the cfl tubes are old the ends are black, dark brown or dark yellow.

You can make some improvized cfl tubes from simple cfl house bulbs ... just open up the plastic case, remove the electronics and connect wires directly to the glass tube wires. See this gallery  for some examples : http://s807.photobucket.com/user/budm/library/Inverter%20tester%20using%20old%20CFL?sort=3&page=1

wraper:
You can try to fool protection by disconnecting Isen pin from the circuit and connecting it to regulated voltage divider. Try something like 100K resistor to VDD and ~30K pot to GND. Then adjust until it stops shutting down so you can measure what happens in the circuit.

Navigation

[0] Message Index

[#] Next page

There was an error while thanking
Thanking...
Go to full version