SX-880 basket case, with lots of issues. Wiil need some advice

[ccheath]

So, you're saying when you turn up the volume with NOTHING going in, it'll still shoot up? If so, with power on and selector to AUX, check and see if there's any residual voltages at the AUX jack. I have experienced similar to this here: http://www.audiokarma.org/forums/showthread.php?t=283423 But, you may or may not have that issue. You might just have leaky caps. Time to track down the culprit.

 

[Uncle Bambi]

So, you're saying when you turn up the volume with NOTHING going in, it'll still shoot up? If so, with power on and selector to AUX, check and see if there's any residual voltages at the AUX jack. I have experienced similar to this here: http://www.audiokarma.org/forums/showthread.php?t=283423 But, you may or may not have that issue. You might just have leaky caps. Time to track down the culprit.

Yes - if I turn up the volume with no load or with an 8 ohm resistor load, it shoots up into the hundreds of mV and then sends the meter out of range.

I measured .5v at the positive side of the AUX jack, to chassis ground.

Wanna hear something funny? I measured DC offset again, and watched as it dropped to just about 0mV. Turned the volume knob, and now it's acting normal. Now it's sitting at 0mV - .2mV without a load, and turning the volume knob has no effect. Now there is no voltage on the AUX jack, either.

I re-connected one speaker, and with the volume off it is dead silent again and increases normally in volume as I turn the knob.

Still no front bezel lights. EDIT/UPDATE: Resolved. fuse holder is intermittently breaking contact.

Something is unstable......

EDIT/UPDATE: Still running normally, after 1 hour. ARGHHHHHHH!!!!!!

 

[SolutionRoom]

I am pretty sure you know this, but I would make sure that the DC Offset (Balance) and the Idle Current is good per the manual (and stable), BEFORE doing any hooking up of anything to the speaker terms or touching the volume pot. The measuring at the speaker terminals with a 1Khz tone is usually for adjusting the output meters. That would be an ancillary task.

If you can get these adjustments close with it on the DBT without any bright glowing, then take it off the DBT or change bulb wattages, and try it again, ensuring that you set the VR 5,6,7,8 pots to the starting positions each time you change DBT setups, or remove it entirely.

I would not connect anything to the speaker terms unless I could get the DC Offset and Idle set first. The fact that you are on a DBT could be causing unpredictable behavior among circuits due to the lower power. And you risk damaging the output circuits/trannies.

I would then check all voltages at the terminal points to ensure they are correct and stable. Once all that is in order, then I would commence with speaker hookup and volume attempts. Otherwise, the smoky garage could become a reality.

 

[SolutionRoom]

Check FU1, it's a 2A fuse that is in the path of the 3 lamps that go to the dial. Also, it would not be entirely inconceivable for all 3 lamps to be burned out. They commonly go bad.

 

[Uncle Bambi]

I am pretty sure you know this, but I would make sure that the DC Offset (Balance) and the Idle Current is good per the manual (and stable), BEFORE doing any hooking up of anything to the speaker terms or touching the volume pot.

Well.... that's the thing. I checked DC Offset without load and saw the wild swinging, and then watched it settle down and behave. Right now it is on the DBT with no load, and pretty steady at -0.1mV, with occasional swings down to -2.mV.

I did put some cheap speakers on it at one point before it went nuts, but they are off now. I'm watching it, and it is behaving - how the hell am I going to catch what's causing the intermittent wild variation?

I would then check all voltages at the terminal points to ensure they are correct and stable. Once all that is in order, then I would commence with speaker hookup and volume attempts. Otherwise, the smoky garage could become a reality.

I'll check them a few more times, but here's what I see right now:

pin 2 -38.7
pin 3 38.7
pin 9 38.8
pin 11 -38-7
pin 15 39
pin 17 -38.8
pin 37 13.2
pin 38

Ok - so I just went and re-did the DC Offset and Idle Current, this time from the test points to be sure. Offset on both sides was near 30, and has now been dialed down to near 0mV on both sides. Idle Current was set on both sides to near dead-on 30mv - 1st on the DBT, then again without for both.

While I've taken the time to type this, I'm monitoring IC on the right channel, and it has crept up to 33.6mV No load, AUX selected, volume down, other controls set flat.

The front panel lights/fuse issue has been resolved - the fuse clip on the transformer side is breaking contact with the wire wrap pin that terminates next to it. If I jumper the pin and the fuse holder clip, I get lights.

 

[SolutionRoom]

Well.... that's the thing. I checked DC Offset without load and saw the wild swinging, and then watched it settle down and behave. Right now it is on the DBT with no load, and pretty steady at -0.1mV, with occasional swings down to -2.mV.

I did put some cheap speakers on it at one point before it went nuts, but they are off now. I'm watching it, and it is behaving - how the hell am I going to catch what's causing the intermittent wild variation?



I'll check them a few more times, but here's what I see right now:

pin 2 -38.7
pin 3 38.7
pin 9 38.8
pin 11 -38-7
pin 15 39
pin 17 -38.8
pin 37 13.2
pin 38

Ok - so I just went and re-did the DC Offset and Idle Current, this time from the test points to be sure. Offset on both sides was near 30, and has now been dialed down to near 0mV on both sides. Idle Current was set on both sides to near dead-on 30mv - 1st on the DBT, then again without for both.

While I've taken the time to type this, I'm monitoring IC on the right channel, and it has crept up to 33.6mV No load, AUX selected, volume down, other controls set flat.

The front panel lights/fuse issue has been resolved - the fuse clip on the transformer side is breaking contact with the wire wrap pin that terminates next to it. If I jumper the pin and the fuse holder clip, I get lights.

Sounds like it might be doing pretty well now. It is common for the IC to climb as the components warm up. In fact, it could take from 30 mins to close to an hour for them to really settle into their operating temperature. In my experience. I'm no expert, but I did re-work a SX-3900, and with the non-switching architecture, it was a picky bastard to get dialed in and stable, and was unforgiving if things were allowed to get out of control. I think with the 880 you should have a bit more forgiveness to having things exact in the Offset and IC realms. If things are intermittent still after having found a loose connection causing the display panel lamp outages, then you will have to do some critical investigative work on the various boards for maybe a loose (cold solder?) connection. Do the shine a bright flashlight trick from one side of the PCB and look for light on the opposite side while gently moving components to check (with power off of course). Look for heat damage around transistors where solder joints could be disconnected. I had to shotgun several boards on my SX-3900 to find some sneaky issues/bad components, but it paid off. It certainly is better than it was when you began the journey:yes:

 

[Uncle Bambi]

Sounds like it might be doing pretty well now. It is common for the IC to climb as the components warm up. In fact, it could take from 30 mins to close to an hour for them to really settle into their operating temperature. In my experience. I'm no expert, but I did re-work a SX-3900, and with the non-switching architecture, it was a picky bastard to get dialed in and stable, and was unforgiving if things were allowed to get out of control. I think with the 880 you should have a bit more forgiveness to having things exact in the Offset and IC realms. If things are intermittent still after having found a loose connection causing the display panel lamp outages, then you will have to do some critical investigative work on the various boards for maybe a loose (cold solder?) connection. Do the shine a bright flashlight trick from one side of the PCB and look for light on the opposite side while gently moving components to check (with power off of course). Look for heat damage around transistors where solder joints could be disconnected. I had to shotgun several boards on my SX-3900 to find some sneaky issues/bad components, but it paid off. It certainly is better than it was when you began the journey:yes:

Thanks for all the tips. I'm not convinced that this one is stable, so I'm going to monitor the IC tomorrow without load for a while and see what happens.

The power regulator transistors, especially the 14v one, have obvious heat distress where they mount on the PCB. I re-flowed them all, and removed/added fresh solder on the 14v regulator. I have replacements in stock for them all, so maybe I'll do some r/r and modify the heats-inks while I'm at it.

Once I'm convinced that it is good, I'll probably do a re-cap.

The one standout symptom that really puzzled me was the fairly loud volume when the volume knob was turned all the way down. Could out of spec DC Offset and/or IC have caused that? If not, I probably still have issues to deal with.

 

[Watthour]

The high volume at pot zero position is not likely a manifestation of final amp bias voltages. Something has to get through the preamp stages (signal) to the final amp for the final amp to modulate its output.

It stands to reason that the dim bulb would show a higher current draw with the amp running a signal at a substantial level. Connect a dim bulb to a known good amplifier and observe how the lamp changes with increasing volume levels to the point where the input voltage is clipped enough to drop the protection relay.

I'd be suspect of circuit grounding and connections, preamp transistors out of their bias voltages (resistors and solder connections), a signal cap shorting, the volume pots, and similar failures. The frustrating part will be getting the amp to repeat its symptoms. Intermittents are not fun sometimes.

 

[Uncle Bambi]

The high volume at pot zero position is not likely a manifestation of final amp bias voltages. Something has to get through the preamp stages (signal) to the final amp for the final amp to modulate its output.

It stands to reason that the dim bulb would show a higher current draw with the amp running a signal at a substantial level. Connect a dim bulb to a known good amplifier and observe how the lamp changes with increasing volume levels to the point where the input voltage is clipped enough to drop the protection relay.

I'd be suspect of circuit grounding and connections, preamp transistors out of their bias voltages (resistors and solder connections), a signal cap shorting, the volume pots, and similar failures. The frustrating part will be getting the amp to repeat its symptoms. Intermittents are not fun sometimes.

Would it be a good or bad idea to shotgun the pre-amp transistors, and do a complete recap? I wouldn't learn anything about where the exact fault was, but I'm not sure how else to proceed. I have the voltage regulator transistors on-hand, but that would be throwing a dart, eh?

Right now I'm going to fire it up cold, and just listen to it at low levels to see if it repeats. It seemed to do it fairly quickly after the first fire-up, but maybe it's heat related?


As an aside - the darn thing has a really nice tuner..... either that, or it is in top alignment.

 

[ccheath]

There are no preamp transistors. They're all IC's unless you're talking about the the main amp. I'd try disconnecting c229 and c230 and turn on low filter. That should completely disconnect the preamp from the main. Then you can see if your main amp is stable.

I wouldn't try shotgun approach just yet. I'd just simply replace coupling caps first but I'd start replacing c227 and c229 for one channel. c228 and c230 for the other channel. Both pairs are on the same circuit when low filter is off but they're different values so watch for that.

 

[Uncle Bambi]

There are no preamp transistors. They're all IC's unless you're talking about the the main amp. I'd try disconnecting c229 and c230 and turn on low filter. That should completely disconnect the preamp from the main. Then you can see if your main amp is stable.

I wouldn't try shotgun approach just yet. I'd just simply replace coupling caps first but I'd start replacing c227 and c229 for one channel. c228 and c230 for the other channel. Both pairs are on the same circuit when low filter is off but they're different values so watch for that.

I've been listening to a local FM station at low volume for the last 1-1/2 hours, and it has been perfect so far. IC is rock solid and sitting at 29mV. Offset is at around 4-6mV. So, it is stable....... right now, anyway.

If I need to replace those cap pairs I'll either have to place an order or make do with what's in the drawer. Looking them up and visually verifying them on the board:

C229, C230 ceanl - .22uf 50v
C227, C228 ceanl - 2.2uf 50v

I must have a hundred extra capacitors laying around here from previous projects, but I don't have those. :sigh:

Here's what I have on hand:

For C229, C230
.1uf 50v poly film caps
.22uf 10v poly film caps
.33uf 50v Nichicon UKT caps (general purpose)
.47uf 50v Nichicon UPM caps (low impedance)

C227, C228
1uf 50v Nichicon UKL (low leakage)
1uf 63v poly film caps
3.3uf 50v Nichicon UPW caps (general purpose)


It's Memorial Day and RS is closed. I'm done for now, unless you guys see something in the above list that will be "good enough".

 

[Watthour]

I was talking about the pre-drivers (like Q15, 17, 19, the duals 5-8, etc.) in the final AF amp stage. Pardon my ignorance, but I'm not a dedicated audio guy. I deal mostly with servoamplifiers, RF power, and industrial electronics, where we call about anything upstream of the final outputs a "pre-amp" stage. I understand the semantics can vary, as do the applications of the devices.

Regardless, those transistors must be either passing signal from upstream or pulling up or down the following amp stage(s). Half-splitting the circuit should isolate it faster. The PITA is apparently getting it to repeat so it can be diagnosed. If you have a signal tracer handy if/when it happens again you might want to verify signals at C217/218 (upstream of the tone controls) to see if the level is askew at those points. I believe that was the thought of ccheath in using the low filter to alter the signal route upstream of the final amp board.

Another random thought is that VR5 and/or VR6 might be intermittent. Those are the DC Balance adjustments. If one of those goes intermittently open the zero point could go through the roof, causing your apparent high current load.

 

[Uncle Bambi]

Stupid thing has been running all day........

 

[SolutionRoom]

Stupid thing has been running all day........

Give her a little shake...

 

[ccheath]

I wouldn't think it's stupid thing.

Watthour have pointed out a very important thing such as VR5 and VR6 which reminds me of more important VR7 and VR8. 780 does not have them but 880 DOES. Those are BIAS adjustments. If those are intermittent, **POOF** goes your outputs. Do NOT try to wiggle them while running. If they go intermittent, you'll fry your outputs. I'd take them out and deoxit them and give them a good workout and follow up with Deoxit Faderlube. After that, use your meter set at OHM and while hooked to meter (out of circuit) and while watching the meter, turn the pot slowly so as if you see any intermittent activity. Also wiggle the leads and the turning cap on the pot to see if it's getting intermittent connection while still hooked to the meter. But, I'd just replace those since they are located in a dangerous part of the circuit (dangerous to the output transistors that is). Make note of the position that the pot is in because you'll want to set it back.

I had one bias pot gone bad in my SX-1250's left channel. The heatsink was heating up so fast that in 5 minute it was toward the hot side. The BIAS was supposed to be 100mv in SX-1250 but it was way above 300mv. Amazing my outputs didn't fry.

 

[Uncle Bambi]

Give her a little shake...

That's crazy talk, Marge......


Actually, I've done that. Been tapping it, too. So far it's been behaving. It'll probably wait until I button it up to self-destruct.

I'm going to go ahead and put a re-cap list together. If anyone wants to suggest possible transistor candidates for a preemptive strike replacement, I'm all ears.

 

[Uncle Bambi]

I wouldn't think it's stupid thing.

Watthour have pointed out a very important thing such as VR5 and VR6 which reminds me of more important VR7 and VR8. 780 does not have them but 880 DOES. Those are BIAS adjustments. If those are intermittent, **POOF** goes your outputs. Do NOT try to wiggle them while running. If they go intermittent, you'll fry your outputs. I'd take them out and deoxit them and give them a good workout and follow up with Deoxit Faderlube. After that, use your meter set at OHM and while hooked to meter (out of circuit) and while watching the meter, turn the pot slowly so as if you see any intermittent activity. Also wiggle the leads and the turning cap on the pot to see if it's getting intermittent connection while still hooked to the meter. But, I'd just replace those since they are located in a dangerous part of the circuit (dangerous to the output transistors that is). Make note of the position that the pot is in because you'll want to set it back.

Hmmm..... I did the bias adjustment right after the offset adjustment, and haven't had an issue since. Hmmm.........

Since I'm going to re-cap anyway, I don't mind buying replacements and changing them out. Any idea what the Mouser part # might be?

 

[ccheath]

Hmmm..... I did the bias adjustment right after the offset adjustment, and haven't had an issue since. Hmmm.........

Since I'm going to re-cap anyway, I don't mind buying replacements and changing them out. Any idea what the Mouser part # might be?

Ok, I just don't know the condition your receiver is in. You have in the beginning of this thread that there's alot of problems that need fix. That is what made me think. If you already have tampered with it with no issue, then disregard. However, I'd still replace them.

 

[SolutionRoom]

Hmmm..... I did the bias adjustment right after the offset adjustment, and haven't had an issue since. Hmmm.........

Since I'm going to re-cap anyway, I don't mind buying replacements and changing them out. Any idea what the Mouser part # might be?

Yeah. There are pots you can get that are not so "touchy" when adjusting them, having a smoother control of resistance. Not sure of the part number though, sorry. I would replace them. Might be Bournes??? Just off memory.

 

[Uncle Bambi]

Ok, I just don't know the condition your receiver is in. You have in the beginning of this thread that there's alot of problems that need fix. That is what made me think. If you already have tampered with it with no issue, then disregard. However, I'd still replace them.

Yes - I picked it up as a non-running unit. Problems I found/corrected:

1) missing wires for + and - voltage from the main caps to the big output transistors

2)two missing jfets (acting as current source)

3)missing volume pot

4)2A front panel fuse holder had a bad/open solder joint to the wire wrap pin that distributes the 8vAC to the front bezel lights

I did initially did the offset off the speaker lugs, but then switched to test point pins to get it done. I then did the IC. Since then, it's been a good boy.