HANNSG Hi221 -- 2 secs to black feat. CCFL flicker & "burn" marks on power board

[ttmi100x]

Greetings everyone,


I have a HANNSG Hi221 monitor that exhibits two seconds to black issues (backlight goes off, display is still on as confirmed by shining a flashlight on it) with some additional oddities...


The monitor failed on me with a standard "2 seconds to black" some time ago prompting me to initially replace most of the electrolytic capacitors (some of them were bulging/leaking).
Things kinda sorta worked for a night, then the problem returned along with some flickering and "buzzing" noises.

Upon closer inspection (not sure how I missed it the first time?) I noticed a strong, dark discoloration (the attached pictures don't quite do it justice) in close proximity to one of the CCFL connectors, mostly under one of the resistors.
The resistor is actually so discolored that its color bands are no longer discernible, though looking at its "brother" near the other pair of CCFL connectors it would seem that it's orange-green-green-gold, i.e. 3.5M ohms(?).
Does that seem about right? Frankly with the resistor itself being some shade of green, the colors are rather hard to make out in general. Perhaps someone else in here can make more of it than I can.

I tested the CCFL connectors one by one as per budm's guide using a lamp I had laying around. This led my search back to to the connector that is nearest to the "burn marks", with the test lamp exhibiting the same flickering and "buzzing" noise as the monitor's own CCFL. As such, I assume that my CCFLs are fine and the problem resides somewhere near that one connector/ "burn mark" area.
There is a video of me testing the connector in question here - please let me know if you have issues loading the video/accessing the page; I can upload it somewhere else if need be.


This is where I am at a loss now however, due to my lack of knowledge and equipment - multimeter, soldering iron, and some spare small bits from other projects are about all I have at my disposal.


I haven't swapped or tested the blue ceramic/disk capacitor near the CCFL connector yet, so that's something I suppose I could look into, although it does not seem to exhibit any outward signs of damage, or so I think at least?

I could also pull out the resistor and check it since it sits above the biggest discoloration in that area of the board, but from what I've read a burnt out resistor would only be a symptom of something else being broken, rather than the source of the problem itself - or was I fed incorrect information in that regard?

The transformer is a mystery in itself to me for the same reason(s), but rather than rambling on about that I am attaching the results of me going through the "A guide on how to troubleshoot 2 seconds to black" and what I noticed while doing so.


---


1) Bad caps - as I said I swapped most of the electrolytic ones with the exception of the "big" mains filter cap and the blue ceramic disk ones.

2) Shorted transistor - couldn't spot one on my board, then again maybe my lack of knowledge had me looking in the wrong spots.

3) Open fuse - measured the picofuse closest to the affected connector at .012 on the 2k Ohm range on my meter. Worth mentioning that I sometimes get erratic results regardless of range; multimeter cables seem fine and battery seems to have enough juice. Could this just be related to the fuse still being connected to the board along with other components?

4) Bad transformer - this is where I probably ran into the biggest knowledge gaps.
Figured out the meaning of primary/secondary via google (i.e. the coils), and went through the pins as indicated... still feeling a little on the unsure side of things
I have 6 pins on one side (primary?) that gave me a reading of around .012 / .013 in the 2k range, while the 2 pins on the other side (secondaries?) gave me around .472 in the same range.
Again some erratic results at points, notably the primaries sometimes showing .002 / .004.
That being said both the transformer on the side with the "burn marks" and the one on the "good" side showed the same readings... Could this just be another symptom of those components still being soldered onto the board and other parts causing interference?

5) Bad CCFL - as per the info above, I believe my CCFLs to be okay, as the test lamp's symptoms matched those of the only CCFL that was exhibiting issues to begin with (see the video of my test linked up above).

6) Bad Diode - none around the affected area as far as I can tell.

7) Bad/cold/poor solder joints - two of the caps I replaced were admittedly soldered a bit "wildly" as I was figuring things out, however upon tracing their paths with my multimeter the connections do seem to be fine. Other than those, I'm not aware of any problem solder joints, nothing visible to me anyway.

8) Other bad components: "Voltage regulators, mosfets, other ICs, resistors can also be bad and need to measured to verify proper operation." - I'm out of my depth here so I have nothing to offer in this section as a whole at the time of writing this.


---


With regard to all of the above components, if there's anything else you'd like me to measure or go back to, feel free to let me know.

Since I have no clue on how to proceed from here, other than swapping the ceramic cap near the CCFL connector, or checking the resistor, I decided to post my issue on here before going any further to get some pointers or at least check whether I'm heading in the right direction to begin with.


Any and all help is most appreciated, since I've been erring in the dark for a bit now with this monitor.
Thanks!

[CapLeaker]

As for the resistor, measure it up with your DMM. To me is brown black black gold = 10 ohm.
something was working hard. If you get already erratic measurements with your DMM checking out something simple like a fuse and you got trouble, then either your battery, your leads, or your DMM are screwed. You need to stop right here until you got your DMM situation sorted out. Other than that you are going to drive yourself nuts with weird measurements.

Compare the two backlight transformers with you DMM in a more smaller range ohm setting so you get more resolution primaries and secondaries. If one winding is out more like 2 percent than the other, the transformer is toast and the protection circuit kicks in. You could disable that protection circuit, but for that you really got to know what you are doing and I don't recommend it, unless one is experienced.
Cold solder joints do not have to be visible, most are, some are not.
How do you figure your CCFLs are o.k? Are they black on their ends? When the monitor worked, did they ran on full tilt backlight and maybe had already a pinkish hue on anything white? What about the ccfl connector and the wire attached?

However, your DMM needs to be working 100% first.

[ttmi100x]

Thank you for your reply, it's much appreciated!

I'm gonna look into inspecting the DMM /replacing it as soon as I'm out of the nightshift loop and can actually get stuff done during the day, after that I'll look into the transformers as you suggested.
Battery is fine, and leads seemed fine but I'll double check.

On the subject of cold solder joints; these should be found if I continuity test, right?

My assumption on the CCFLs being okay came from the fact that both the monitor's lamp (only on that one specific connector) as well as the completely unrelated (and perfectly functional) lamp I used for the testing (as seen in the video I linked in my original post) exhibited the same flickering and buzzing. As such I went by the logic that the issue wasn't stemming from the lamps but rather component(s) on the board.
I suppose the lamps could've been damaged by whatever is going on on the board, which is the point you're trying to make here I'm guessing?
On first inspection I noticed no visible damage on the CCFL, but I'll give it a closer look once I'm free.

Finally, in regards to full tilt, discolorations, or changes in hue; there was no such thing prior to the monitor failing. The monitor started flickering, then the backlight went completely.
This is when I swapped the leaky/bulging caps.
After that, as mentioned, the monitor ran for like a night, then the buzzing started, along with brief flickering, and the backlight dying once again.
I will check the connector and wires again as suggested.

Thanks again for your input, suggestions and help!

[stj]

check the small blue capacitors in series with the tubes,
the resistor is for monitoring current - check it's resistance to ground and compare to the other one - incase something it connects with is bad.
whats the part number on the 16pin chip under the transformer?

one other thing, if the panel is made by philips you may have bad soldering where the wires connect to the ends of the tubes!

[ttmi100x]

Thanks for the reply!

check the small blue capacitors in series with the tubes,
the resistor is for monitoring current - check it's resistance to ground and compare to the other one - incase something it connects with is bad.

Will do as soon as I can verify that my meter is showing proper readings.
Also thanks for clarifying the role of the resistor, that definitely had me wondering.

one other thing, if the panel is made by philips you may have bad soldering where the wires connect to the ends of the tubes!

Back of the panel says it's an INNOLUX panel, but I'll be sure to check; I was gonna give that another look anyway after what CapLeaker suggested in his reply.

Edit:

whats the part number on the 16pin chip under the transformer?

Just to make sure we're talking about the same thing - are we talking about the chip in the attached picture? Just asking because it has 8pins, and I don't see anything else with 16 pins in that vicinity.
In any case, the part number would be " 9971GD 930966".
If you were talking about something else, please let me know and I'll have another look.

Thanks again for the help!

[stj]

the 16pin chip is on the other side of the board - that 8pin chip is the mosfets

[ttmi100x]

the 16pin chip is on the other side of the board - that 8pin chip is the mosfets

My bad, this one, right?

The number is rather hard to read, partly because of the surface, partly because of the component age (I'm guessing) had to grab a magnifying glass and blast it with light from the right angle.
Anyway, as far as I can tell it goes as follows:

OZ9938GN (The "O" could be a zero but I'm leaning more towards O as it does look slightly wider than the zeros)
A86710 20
0924B