Samsung 172X - OZ980S?

[pedro]

Just had a Samsung 172X lobbed onto my bench, already opened up by the owner. One second of backlight then blank - wow, now where have I seen that before? <sigh>.

These use an external DC brick supply, and all surface mount stuff so no swags of through-hole electros to look at. No visible signs of distress anywhere.

The backlight controller appears to be OZ980S. Recall seeing links to a data sheet for these or a similar one - can anyone point me there?

There's also a micro-pot near the OZ980S which just *may* control the reference/threshold. Hoping against hope that the owner wasn't a compulsive pot-twiddler.

[endevite]

Re: Samsung 172X - OZ980S?

Just had a Samsung 172X lobbed onto my bench, already opened up by the owner. One second of backlight then blank - wow, now where have I seen that before? <sigh>.

These use an external DC brick supply, and all surface mount stuff so no swags of through-hole electros to look at. No visible signs of distress anywhere.

The backlight controller appears to be OZ980S. Recall seeing links to a data sheet for these or a similar one - can anyone point me there?

There's also a micro-pot near the OZ980S which just *may* control the reference/threshold. Hoping against hope that the owner wasn't a compulsive pot-twiddler.

Could you look again and see if that is perhaps a oz960? oz980 is single chip Smart Battery Selector equipped with Battery Crisis Management and Green Conditioning.

While the oz960 is ccfl control.

Here is the datasheet on the oz960. (it was either the 225bw or 226bw thread that I posted the oz964 and oz9938 datasheets.

[pedro]

Re: Samsung 172X - OZ980S?

When I saw the "OZ" I expected a 960, but it looks very much like an"8". If it is a "6" then it is the worst smear I have seen in a loooong time. But certainly from your info a 980 wouldn't be there.

Already had the data sheet for the 960, and find the O2MICRO website rterrible - no partnum search that I could find.

Tnx

[pedro]

Re: Samsung 172X - OZ980S?

This one gets more intriguing (i.e. exhausting) the further I go. Symptoms: Apply +14VG (brick) power, no activity. Press POWER bezel button, green LED lights. After about 2 secs, quick flash of backlight then LED and backlight gone. Repeats if POWER pressed again.

All the supplies on the main board are OK. When the POWER is pressed, the micro blips the BL_EN and the +5V_PANEL, but seems determined to do that only for a flash. Unlike many designs, the backlight controller chip doesn't get a sustained Vcc. There is a big loop, and the micro decides to shut down the Vcc to the OZ960 so it is hard to form a view regarding the Vref, and the circuitry doesn't lend itself to overriding the micro's drive.

Normally this wouldn't be a drama with a more conventional arrangement, but being SMD makes it that bit harder to just lift a leg and drive a line to see more than a transient. I have forced the BL_EN by grounding the base of Q101 (see attachment) and the BL lights up fine, but haven't been able to check the OZ960 pin voltages due to the lack of hands and the tiny SMD work area.

The service manual doesn't have the schematic of the inverter board, but from the attached part of the mainboard schematic showing the interface to the inverter board there is no feedback to the micro via this route. It isn't obvious how/why the micro decided to pull the plug IF it is due to an inverter issue.

Any suggestions?

[PlainBill]

Re: Samsung 172X - OZ980S?

This one gets more intriguing (i.e. exhausting) the further I go. Symptoms: Apply +14VG (brick) power, no activity. Press POWER bezel button, green LED lights. After about 2 secs, quick flash of backlight then LED and backlight gone. Repeats if POWER pressed again.

All the supplies on the main board are OK. When the POWER is pressed, the micro blips the BL_EN and the +5V_PANEL, but seems determined to do that only for a flash. Unlike many designs, the backlight controller chip doesn't get a sustained Vcc. There is a big loop, and the micro decides to shut down the Vcc to the OZ960 so it is hard to form a view regarding the Vref, and the circuitry doesn't lend itself to overriding the micro's drive.

Normally this wouldn't be a drama with a more conventional arrangement, but being SMD makes it that bit harder to just lift a leg and drive a line to see more than a transient. I have forced the BL_EN by grounding the base of Q101 (see attachment) and the BL lights up fine, but haven't been able to check the OZ960 pin voltages due to the lack of hands and the tiny SMD work area.

The service manual doesn't have the schematic of the inverter board, but from the attached part of the mainboard schematic showing the interface to the inverter board there is no feedback to the micro via this route. It isn't obvious how/why the micro decided to pull the plug IF it is due to an inverter issue.

Any suggestions?

Two suggestions. Both are long shots. If you haven't already done so, try it with the monitor hooked up to a computer.

Second, check the driver transistors on the inverter for shorts, or isolate the power to the inverter from the logic board.. The only other possibility I can think of is that when the inverter is turned on it overloads the power supply, the power drops, and the micro resets, turning off the inverter.

PlainBill

[pedro]

Re: Samsung 172X - OZ980S?

Two suggestions. Both are long shots. If you haven't already done so, try it with the monitor hooked up to a computer.

Second, check the driver transistors on the inverter for shorts, or isolate the power to the inverter from the logic board.. The only other possibility I can think of is that when the inverter is turned on it overloads the power supply, the power drops, and the micro resets, turning off the inverter.

PlainBill

Yes, I've tried it with a video signal, same symptoms. These things are a PITA to test as you need to have base/stand plugged in to the main board to be able to feed Dc and video in, and it is precarious to try and see the front of the screen.

If I force BL_EN the backlight runs fine, so no shorted transistors or faults else the OZ960 would dummy-spit. That's the strange bit, the inverter board functions check out OK but the main board shuts down the BL_EN line. In a through-hole unit I'd be more focussing on the power supply area. Have done ESR checks on all the SMD electros and all check out OK, reg volts are all good.

I agree that a sensed (real or not) O/L condition might be causing the micro to intervene in the orderly startup, but haven't spotted any signs of that occurring. I've monitored the 14v in from the brick on the DMM and there is only 20mV droop when the ON button is actioned, so I doubt the brick is going into O/L. The 5V switcher shifts about 10mV when I force BL_EN. As you would note from the schematic bit I posted, the inverter runs off the 14v supply and the only relevance of the +5v is for the BL_EN logic signal and the brightness mod line's driver.

When the BL_EN is off, the OZ960 Vcc is isolated from the 14v rail by a series switch. Forcing BL_EN actually turns on this switch and fires up the 14v to the OZ960 as well as passing through an enable to the chip.

Thanks for hanging in with me on this one.