STR-DE845/945 - Standby CKT help

"427Cobraman" wrote in message news: snipped-for-privacy@p10g2000cwp.googlegroups.com...

Very first thing is to make sure that no front panel buttons are being accidentally pushed by slightly wrong assembly of the front panel. Make sure that all buttons move, and click.

Assuming that there is no attempt by the main power relay to close, when you try to turn it on, you can try forcing it, by temporarily shorting the collector to the emitter on the relay driver transistor. DO NOT have any speakers connected to it at this time. If unit then appears to come up ok, check all of its display and operation functions. If anything here seems not to be working correctly, chances are it's a system control problem, possibly related to the work you already did on the front pcb. If all is still looking good, check across ALL speaker outputs for DC. If there is any, then check power supplies to output stages carefully. Some of these big amps monitor the power supplies, and if anything is wrong, block operation, although that's only a potential issue if the main power relay does close momentarily. If everything is still looking ok, you can give it some input, and see if the outputs look normal on all channels, on a 'scope. If this looks good, you can risk some speakers, and then carefully recheck all functions. If there are any noted problems, they should give you a clue as to where to look. If it all appears to be working ok, then the only circuitry that you then need to concentrate on, is the relay drive, back to the system control micro, but check the circuit around the system control micro carefully, for any pins that may have a power up blocking function. With the standby supply present, and the system control idling, you should be able to verify that there are no incorrect inputs to the micro.

Arfa

Could just be the micro isn't getting any power. A regulator might have died, a crystal might be intermittent, and I'm thinking there were a couple

Mark Z.

OK, I was wrong about the resistors - guess I was thinking of a different model.

Check Q1205 on the digital board and make sure it's putting out around 6 volts or so - and check D1211- it could be open or have been breaking down under load - which kind of seems like your problem. Make note of whether the micro is getting hot, or is stone cold. Could be an important clue.

Mark Z.

Thanks for the tips so far. I'll check those points out. I thought about shorting the C-E junction as mentioned but was worried about damaging things. I'll give it a try. Incidentally, the DE945 actually DOES power up and function fine if using the remote (silly me, was using the wrong on/off button). The 845 and 945 share most of the same parts, so if all else fails, I may be able to shotgun this (last, last resort). Will keep this updated. Thanks a million...

Alex

OK. Here's an update. I tried the front panel board from the 945 in the 845 and it worked great. I checked continuity on all of the switches on the bad board and none are shorted. I'm guessing a new board is in order. What is a good source for sony parts? I've used mostly pac parts but they don't seem to show this part, let alone many sony replacement parts.

Alex

If you've got the skill to have undertaken the repair in the first place, and have managed to successfuly transplant the board from the '9 to the '8, then I wouldn't give up quite yet. You say that it initially worked, so there's a good chance that there's not a lot wrong. As the other poster said, now you know for sure that the problem is in the system control, check the clock generator crystal for activity with a 'scope, check that the 5v actually reaching the Vcc pin on the micro is accurate and clean, and check that there is activity on the switch matrix lines, just to make sure that there is no leakage across any of them, even though no switches appear to be actually closed. Have you checked this one with the remote control ? You might need to use the remote which matches the board ie from the '9, unless you know for sure that both remotes work both units correctly.

Arfa

You're really going to have to troubleshoot it down further than just the board. First of all the board may very well not be available, and if it is, it's going to be expensive (over 100.00 maybe closer to 150.00 - I'll check.)

Mark Z.

Thanks for the encouragement boost. I would normally shoot deeper, but I lack the SMD solder/desolder tools I'd need, especially the micros on there. I've done chip resistors and caps, etc. with a standard iron on occasion, though. I did see that crystal there and will check it out as well as the switch matrices. The remote is actually from the 845. ...the adventure continues...

There is a guy over here who has created " Beer's Maxim " ( his name is Beer ). It basically states that " The more pins an IC has, the less likely it is to be faulty ". Over the years, I've found this to be a pretty good thing to keep in mind. On many occasions, I've condemned a micro, because it's easy to do. With the best will in the world, none of us mere mortals, no matter how experienced in the world of repairs, TRULY understands the black magic, sorcery and esoteric shenanigans, that go on inside a microcontroller IC, so when one doesn't work, we always condemn it. On the occasions when I have actually replaced such an IC - and you can do it perfectly successfully with ordinary workshop soldering equipment - it has cured the problem on probably fewer than 20% of the occasions. Almost always, unless there is a definite reason for the IC to be bad, such as lightning damage, screwed internal software ( people trying to hack DVD region protection etc ) or your 'scope probe having slipped ( !! ), the problem turns out to be something simple, elsewhere. Often, an input to the micro is wrong - even the /reset line. If you've got the time and inclination, I would recommend checking all of the pins on the micro on the working board, and noting the conditions on them, and then comparing to the bad board.

Also, as you previously did work to this board, involving soldering, check carefully, all over the board, with a strong magnifier, just in case a whisker of solder has flown off the end of the iron, and lodged across somewhere. Also check that there are no bent pins on any connectors, or damaged foil fingers on the ends of any flexiprints going to the board. It's amazing how often secondary faults, are caused by something you've done yourself ...

Arfa

Well, all switch matrices checked out as well as supply voltages and oscillators. ESR'd all electrolytics. The only thing that was off is the LED drive circuits 1,2, and 4 on IC102. On the good board, these are all at 4.7v when unit is powered off, regardless of state when in operation. On the bad board, all are at zero. I wouldn't think these would affect general operation, but who knows. I've checked, and a couple other eyes have checked for cracks, solder bridges and bent pins...everything looks good. I'm at wit's end. I'm sure I'll get motivated to dig around some more after a nice cold brew. Link to diagram:

Hmmm. That's interesting. The anodes of the LEDs are connected to B+ via their 150R limiter resistors. The cathodes are connected back to pins 1,2,4 on the micro. On the good board, in standby, the pins - ie the LED cathodes - are all at very nearly B+, which is exactly as you would expect. In standby, the LEDs will want to be off, so the driver transistors, internal to those pins, will be turned off. Now look at the situation with the bad board. The voltage at the three IC pins can only be low as a result of two conditions. First, the internal driver transistors are on. BUT, if this is the case, then the LEDs must be alight, as we know that there should be B+ on their anodes, even when the unit is in standby. The second condition which would lead to there being no voltage on the IC pins, AND the LEDs not alight, is B+ missing at the back side of the 150R limiter resistors.

It's hard to see how this could be the case, if all of the other voltages on the micro are correct. Never-the-less, it would be my best guess, as being the case. Although you can be pretty sure from your measurements so far, that B+ itself is established and correct, I'd give reasonable odds that it's not everywhere it should be. I think that the missing voltages at the LED drive pins are a big clue. My next move would be to check that B+ is present at the junction point of the three limiter resistors and C120 and pin 1 of IC105. Keep us posted on your findings.

Arfa

Well, it looks like B+ is not available on the ckt with the good board until unit is powered up, yet there is 4.7 volts at the pins on the IC while in standby. I've pulled all but 3 hairs out of my head trying to shoot as well as understand how this circuit is supposed to work. I also went and double-checked all voltages coming into the card via ribbon cable and everything checks out. *sigh* Incidentally, the owner said he is willing to pay for a replacement card if all else fails. Luckily he's a very patient neighbor! It's $95 bucks for the durned thing. Again, I appreciate all of the guidance and encouragement!

Alex