Hanns-G Hi221DPB (Hannstar HSG1041) monitor: 2sec2black
Finally King Robot allowed me to create this topic. It's a brave new world, they used to say?
Hello everybody, I have a problem. As the title says, I've been hit by the "two seconds to black" train. Thing is, this monitor of mine seems to have been a great one because there's no discussions on it on the web, including here. No schematics either (or I'm too dumb to search properly). And yes, it's been working fine for the two-three years I've had it, bought used. Manufactured July 2009. Two days ago I woke up, started it, connected to the web, started mail application and went to the toilet; when I came back (couple minutes or so) it was blank, power still on as if it had the blank screensaver I always use but it wasn't that. Bad things always happen behind my back, always - if only I were there to see any unusual symptom, anything.
To the point. First thing I wanted to ask (nicely, for that matter) is if anyone here happens to have a schematics for this particular model and would be willing to share it. I did find a service manual for an earlier model (HW191/HSG1033) but as my luck goes precisely the part I'm interested in is completely different so posting that schematics here to work on it would be (almost) useless. I'm in the process of reverse-engineering the actual schematics from the power+inverter board but it's a tedious work so I'd be glad if anyone could spare this waste of time.
That apart, what I've been able to gather from testing and measuring around is that one of the inverters doesn't light up the lamps, while the other does. Tested with a spare pair of lamps from another LCD panel, in different configurations. The bad inverter pertains to the top lamps, presumably more stressed due to rising heat.
From measuring it appears that both transformers are good (resistance is similar on their output windings: 471? upper vs 477? lower; primaries are all about 0.3? each) and so are the double MOSFETs (AP9971GD) measured comparatively and individually (out of scale on 200? for DS, GD, GS).
These MOSFETS are driven directly by an OZ9938GN, outputs 1 and 15 going both to upper pair and lower pair so a defect in this driver could be excluded.
I've also checked as good as I could the solderings in the area and resoldered where unsure. Replaced all three capacitors on the main 12V rail that feeds the inverters although they didn't seem defective (not bulged, capacitance relatively OK measured comparatively, low ESR - all measured with an old analog multimeter).
What I found though is a blown diode that doesn't measure anything in any sense, even on the 2k? scale. This diode feeds the overvoltage protection pin 7 of the OZ9938GN driver with positive voltage from the 5V rail while being put to ground when the first lamp of each pair stops working. Considering the schematic I've been able to reverse engineer until now I find it odd for this diode to blow up.
Problem is, the diode has no marking whatsoever, only a white stripe on a black body, marking the cathode. I've attached a picture of it. (Sorry for the bad quality of the pictures, I don't have a modern camera and my hand is shaking.) There are three other identical diodes on the board and they seem to be alright: one putting pin 10 (ENA) of the driver to ground when in Off state and the other two feeding negative impulses directly from lamp connectors CN1 and CN3 precisely to the faulty diode.
What I'd like to know is if anyone knows exactly what model could this be and what could I use for a replacement. In the HSG1033 schematic a similar role is played by an LL4148WP (1N4148?) so I was thinking I could patch a regular (non-SMD) 1N4148 instead since I got plenty of those brand new.
However, logic says there has to be a primary defect somewhere else that led to such unusual defect as a blown diode and a faulty inverter. And by visually inspecting the board there is a corner where a resistor and a capacitor have overheated (as seen in picture). The capacitor is a 10pF/3kV parallel to the upper lamps' transformer output, the resistor is returning the feedback from first lamp of the upper pair. The capacitor seems alright (not shorted, since trafo output is 471?, however the 6? difference between outputs may rise a question), while the resistor measures 1.5M? according to the color codes and comparatively to the one on the other channel.
After finishing the reverse engineering for the faulty inverter I can tell there's nothing specific to that channel that could lock it up while leaving the other work (for the two seconds before blank, that is). MOSFETS are alright, transformer output seems alright, RC filters in the primary seem OK (R=33?, C not shorted). So, what is the problem with the upper inverter, actually? Again, logic points to the transformer (which could have been fried by a bad lamp). And if it's that, I'm f?udged - can't afford either, assuming the trafo can still be found, new.
What say you? Is it possible to be only a lazy pair of lamps that don't get to light up before the protection shutdown? Should I mount an 1N4148 and try a power up or could a prolonged use in a possibly faulty state produce more severe damage?
Additionally, does anyone have any experience with this model/power+inverter board, does anyone know the manufacturer ('LF' logo, dunno who can be)? Original schematics anyone?
Please chime in if you will. Thank you for reading that far.
Read responses in badcaps.net