Yamaha G50-112 Schematic Needed

[rocknroller]

If there are 0V across it, then the relay is off.
As you mentioned there is no DC at the output transistors, and pin 2 of IC110 is 0V, so there is nothing wrong at the output, no signal telling IC110 to turn off relay.
So that leaves the IC110 and it's circuitry. Pin 5 I would have expected more negative, but I could be wrong. Check R208 and associated solder connections.

R208 measures low resistance when in circuit. Shows 3.8K which seems low for a 4.7K item.
It seems to have a good connection on both ends, but I touched both ends up anyway. Definitely has -15V on the one side and -0.76 on the other at chip pin 5.

Of course during my measurements now both .5a fuses went and I need to get some.

[g1]

Suggest you also replace the 4 caps Enzo mentioned (C169,170,171,172).

[Enzo]

It is possible, but in my experience, that IC is almost never bad.


It is now old, but when I say check the voltage on R196 and C169, the idea is that C169 was what charges up, but R196 is in parallel with it. SO they have the same voltage, and it will be far more convenient to measure it on the resistor than on the cap. Many times we need to measure at some inaccessible point, so we just look for something else connected to that point as a handy spot.

[rocknroller]

Suggest you also replace the 4 caps Enzo mentioned (C169,170,171,172).

Pulled all 4 caps. 169-171 measure to spec. 172 is a 47uf measuring 58uf and the bottom looks a little bulged. Couldn't see that until I pulled it off the board. RS had my replacement fuses but not the cap so I'll have to order. Really wish there were a real electronics store in my area!

Given I don't have replacements for the others but they measure to spec, would you still recommend replacing or just put back in circuit?

Also - does the failure of that C172 cap makes sense - it does appear to tie to that relay in a fashion.

[rocknroller]

Suggest you also replace the 4 caps Enzo mentioned (C169,170,171,172).

Replaced all 4 caps around IC110 - no fix.


Here the pin voltages I see

pin1 -0.62
pin2-pin4 0
pin5 -0.75
pin6 1.22
Pin7 0
pin8 1.26
pin9 3

Across R196 as before -17.3
D112 diode shows 22.5v on both sides as before. Not clear why this would be the case. Wouldn't that mean either the diode is shorted or the relay RY101 itself is? I don't see anything else feeding that except the IC110 chip which clearly isn't putting out that voltage. My old emco-35 meter has a diode check, but I no longer have the manual for it and not sure what the expected readings should be when good/bad/shorted I was either getting an out of range symbol or practically zero, depending on which way the leads were

Also of note - R197 100K resistor that goes to pin 1 of IC110 seems WAY off. Measuring 39K. Candidate for replacement? Possible cause of IC110 not functioning correctly?

[Enzo]

D112 diode shows 22.5v on both sides as before. Not clear why this would be the case. Wouldn't that mean either the diode is shorted or the relay RY101 itself is?

No it means no current is flowing through the relay/diode. Ohm's Law, no current, no voltage drop. But don't assume I am right, A shorted diode would also do that, just seems less likely to me. Measure resistance across the diode, it will either measure shorted or yoou will measure the resistance of the relay coil. I have no idea what that is, maybe 100-200 ohms?

IC110 completely controls the relay

R197. resistors do not drop to a third of their value, unless they have turned to conductive charcoal. Your 100k resistor has parallel circuitry around it causing the reading to be low. As with any part, if you doubt it, disconnect one end of it from the circuit THEN measure it. I will bet my lunch money it will read 100k that way.


OK, so you have 22v at the relay. Isn't that pretty much the voltage at C165 up by IC111? In other words, you have the supply voltage on the relay. Now look at IC110. Pin 6 drives the relay, and has 1v on it. Between the two is 390 ohm R201. How is that resistor? If it is intact, it has 21v across it, which would be over 1 watt dissipation. Is it getting hot? Is it open?

[Enzo]

D112 diode shows 22.5v on both sides as before. Not clear why this would be the case. Wouldn't that mean either the diode is shorted or the relay RY101 itself is?

No it means no current is flowing through the relay/diode. Ohm's Law, no current, no voltage drop. But don't assume I am right, A shorted diode would also do that, just seems less likely to me. Measure resistance across the diode, it will either measure shorted or yoou will measure the resistance of the relay coil. I have no idea what that is, maybe 100-200 ohms?

IC110 completely controls the relay

R197. resistors do not drop to a third of their value, unless they have turned to conductive charcoal. Your 100k resistor has parallel circuitry around it causing the reading to be low. As with any part, if you doubt it, disconnect one end of it from the circuit THEN measure it. I will bet my lunch money it will read 100k that way.


OK, so you have 22v at the relay. Isn't that pretty much the voltage at C165 up by IC111? In other words, you have the supply voltage on the relay. Now look at IC110. Pin 6 drives the relay, and has 1v on it. Between the two is 390 ohm R201. How is that resistor? If it is intact, it has 21v across it, which would be over 1 watt dissipation. Is it getting hot? Is it open?

[rocknroller]

No it means no current is flowing through the relay/diode. Ohm's Law, no current, no voltage drop. But don't assume I am right, A shorted diode would also do that, just seems less likely to me. Measure resistance across the diode, it will either measure shorted or yoou will measure the resistance of the relay coil. I have no idea what that is, maybe 100-200 ohms?

IC110 completely controls the relay

R197. resistors do not drop to a third of their value, unless they have turned to conductive charcoal. Your 100k resistor has parallel circuitry around it causing the reading to be low. As with any part, if you doubt it, disconnect one end of it from the circuit THEN measure it. I will bet my lunch money it will read 100k that way.


OK, so you have 22v at the relay. Isn't that pretty much the voltage at C165 up by IC111? In other words, you have the supply voltage on the relay. Now look at IC110. Pin 6 drives the relay, and has 1v on it. Between the two is 390 ohm R201. How is that resistor? If it is intact, it has 21v across it, which would be over 1 watt dissipation. Is it getting hot? Is it open?

D112 measures 1.2 ohms across, when in circuit. Getting it out would require removing the relay itself or large cap next to it. It's pretty inaccessible from the top!
Hopefully your lunch money is small - you were correct. I didn't consider the impact of the parallel circuitry. I'm certainly learning (in some cases re-learning) with every new post!
R201 is definitely hot - but measures 389 ohms which is pretty much right on the money.

[Enzo]

Then your shorted diode idea is probably right.

The 390 ohm resistor cannot get hot without current flowing, so that 1v on the output of the IC is apparently not because the R201 is open. I'd believe a shorted diode before I';d believe a shorted relay coil, but I have been wrong already today, so...

No matter what we find, that board has to come up to get at the solder side, so bite the bullet, pull it up and pull that diode and see if it is indeed shorted. A plain old vanilla 1N400x diode will work perfectly there. All I stock are 1N4007, but a 1N4003 or something would be fine.

Do not power up without the diode though.

[rocknroller]

Then your shorted diode idea is probably right.

The 390 ohm resistor cannot get hot without current flowing, so that 1v on the output of the IC is apparently not because the R201 is open. I'd believe a shorted diode before I';d believe a shorted relay coil, but I have been wrong already today, so...

No matter what we find, that board has to come up to get at the solder side, so bite the bullet, pull it up and pull that diode and see if it is indeed shorted. A plain old vanilla 1N400x diode will work perfectly there. All I stock are 1N4007, but a 1N4003 or something would be fine.

Do not power up without the diode though.

Got it off the board. I already had the solder side out as I needed that to do all the cicuit tracing/tests anyway. The real trick was keeping a tweezer heatsink on the diode (which I do remember from being a kid) just in case it was actually good and I had to put it back (put one back WILL be the real challenge this board!)
Anyway, measures 1.6 ohms out of circuit as well as it id before in circuit and my diode check feature on my meter shows 0 in both directions. On D111 I get .77 or something in one so I guess that confirms the diode is shot. The schematic shows this as a W03B but no luck on a quick google search so I'll take your advice and purchase a 1N4003 and install.

So assuming that's it (and I hate to even say that until it's really done done) - Suspected cause of failure? I played the amp for 15-20 minutes and all was fine, put a way for a week, move to another room and turn it on and diode is now dead? Just stroke of bad timing luck that the part took one too many minutes of use, or something else on the board/circuit that caused it?

[Enzo]

Yes, 1.6 ohms on any diode would be bad.


You will never know why that particular part failed. Everything works until it doesn't. Why does the light burn out in your table lamp today instead of tomorrow or yesterday?

[rocknroller]

Yes, 1.6 ohms on any diode would be bad.


You will never know why that particular part failed. Everything works until it doesn't. Why does the light burn out in your table lamp today instead of tomorrow or yesterday?

I get that. I guess I was just expecting that, if it would fail, it would be sometime during the first use as opposed to DOA on the next turn on. Just makes me curious if some surge type thing occurred at power down or at power up that might have killed it. But then again, I have had light bulbs die immediately when turning them on and they were fine the last time they were on....

[Enzo]

Yeah, it would be nice if things failed on a schedule, then we could plan for it. But they don'r... We all just naturally want everything to be because of something, it is wired into us. We all want to say "Oh, this broke because I left the amp on too long, or because I played too loud, or because lightning hit it." But even if we are watching, we usually don't see the failure. Imagine you are in a boat out in the river, the Niagara river. Out in the middle, you can paddle to shore easy enough. But there will be a point in the river, that you will be swept over the falls before you can make it to shore. At one minute, you have time to make it, the next minute you don't. The river looks the same either place, nothing around you changed, just... you won't make it.


What am I rambling about? A voltage surge can happen and weaken a part. Lightning strikes nearby and stresses your output transistors. Your amp bounced a little too hard on the trunk of your car crossing train tracks. Damage happened to something in the amp, but it still runs. Eventually the part goes ahead and fails, but it is 6 months after the stress that killed it. Now how can we look at that failure and think, "Oh yeah, that must have happened last April when we went over the train tracks." So like the river boat, your thing is doomed, but it just doesn;t show yet.

There is also what we just call "random component failure." There is nothing in this circuit that should cause that diode to fail. But it did anyway.

In computers there are "soft" memory errors. The tiny microscopic bits of silicon in the memory ICs are working just fine. Do you know what cosmic rays are? Tiny bits of atoms, they rain down on us. They have high energy and lightning speed, and they pass right through us. Every now and then one hits an atom of matter. Passing through you, it might hit an atom in a skin cell or whatever. You will never know it. But in a memory IC, that bit of atomic nucleus can hit a tiny bit of silicon that is holding a bit of memory, and alter its state. A 1 becomes a zero. And your computer crashes. Nothing wring with the IC, reboot and it is fine, but that cosmic ray affected it. What if a cosmic ray knocks an atom out of place in a component of your amp? Unless you have a cloud chamber handy, you will never know there are cosmic rays, yet there they are.


No I don't think a cosmic ray bit your diode, but it serves to show we can't always know why something fails.

[rocknroller]

Yeah, it would be nice if things failed on a schedule, then we could plan for it. But they don'r... We all just naturally want everything to be because of something, it is wired into us. We all want to say "Oh, this broke because I left the amp on too long, or because I played too loud, or because lightning hit it." But even if we are watching, we usually don't see the failure. Imagine you are in a boat out in the river, the Niagara river. Out in the middle, you can paddle to shore easy enough. But there will be a point in the river, that you will be swept over the falls before you can make it to shore. At one minute, you have time to make it, the next minute you don't. The river looks the same either place, nothing around you changed, just... you won't make it.


What am I rambling about? A voltage surge can happen and weaken a part. Lightning strikes nearby and stresses your output transistors. Your amp bounced a little too hard on the trunk of your car crossing train tracks. Damage happened to something in the amp, but it still runs. Eventually the part goes ahead and fails, but it is 6 months after the stress that killed it. Now how can we look at that failure and think, "Oh yeah, that must have happened last April when we went over the train tracks." So like the river boat, your thing is doomed, but it just doesn;t show yet.

There is also what we just call "random component failure." There is nothing in this circuit that should cause that diode to fail. But it did anyway.

In computers there are "soft" memory errors. The tiny microscopic bits of silicon in the memory ICs are working just fine. Do you know what cosmic rays are? Tiny bits of atoms, they rain down on us. They have high energy and lightning speed, and they pass right through us. Every now and then one hits an atom of matter. Passing through you, it might hit an atom in a skin cell or whatever. You will never know it. But in a memory IC, that bit of atomic nucleus can hit a tiny bit of silicon that is holding a bit of memory, and alter its state. A 1 becomes a zero. And your computer crashes. Nothing wring with the IC, reboot and it is fine, but that cosmic ray affected it. What if a cosmic ray knocks an atom out of place in a component of your amp? Unless you have a cloud chamber handy, you will never know there are cosmic rays, yet there they are.


No I don't think a cosmic ray bit your diode, but it serves to show we can't always know why something fails.

My replacement diode came to day. Interesting that the old diode had more of that Zener diode look to it (orange colored with the green bands) but this new one with the more classic diode style of black with white stripe) matches the cross reference. Put it in and everything now appears to work fine!! I'll test it out for a few days before reposting it just to make sure everything is good but it certainly seems healthy at the moment.

I want thank everyone here that responded - especially Enzo and DrGonzo. You have not only helped me resolve this but educated me in the process as well and i will certainly pay it forward! Thanks again!

Jim

[Enzo]

Diodes can come in glass packages, plastic packages and ceramic packages. I have 1N400x types in both glass and plastic in my drawer. Don;t read too much into it.