I've recently recapped my Silver SP5050/SA5050 combo, which is almost the same as the HK725/770. There's 2 things that I haven't been able to work out though that I'd appreciate help on.
I'm using a known good CD player as the audio source.
1. Every time I turn the amp(s) on and play music, there is crackly distortion for about a minute, both channels impacted. This is not caused by pots being turned, and it goes away I guess once something warms up. But if I play music for 20 minutes so everything is warm, turn it off and then turn back on straight away, the same distortion is there for a minute. I've replaced all the electrolytics plus a couple of transistors, but wondering what that behaviour could be a symptom of?
2. The left channel for the "A" speakers doesn't work. The LED power display comes on, showing a signal for both channels, and the "B" speakers work fine. As far as I can see, there is one relay for A and another relay for B, so on that basis I guess there is some sort of fault on the actual speaker terminals or the board they mount on. Is that correct?
I'm using a known good CD player as the audio source.
1. Every time I turn the amp(s) on and play music, there is crackly distortion for about a minute, both channels impacted. This is not caused by pots being turned, and it goes away I guess once something warms up. But if I play music for 20 minutes so everything is warm, turn it off and then turn back on straight away, the same distortion is there for a minute. I've replaced all the electrolytics plus a couple of transistors, but wondering what that behaviour could be a symptom of?
2. The left channel for the "A" speakers doesn't work. The LED power display comes on, showing a signal for both channels, and the "B" speakers work fine. As far as I can see, there is one relay for A and another relay for B, so on that basis I guess there is some sort of fault on the actual speaker terminals or the board they mount on. Is that correct?
I've answered my own questions.
The first issue was the pre-amp, seems there is a relay that was commonly used back then and which commonly causes trouble like this. It's a Fujitsu 221D024, I can't easily find a replacement so will try cleaning it and see what happens.
The second issue was on the speaker terminal board, which is double sided. Dry joint on the component side, was actually a bit tricky to fix, but works now.
I now have another problem with the amp cutting out when hot - just for a second or two, then comes back, but it is not an issue at low volumes. Not quite sure how to troubleshoot that and haven't isolated it to the pre-amp or power amp yet.
The first issue was the pre-amp, seems there is a relay that was commonly used back then and which commonly causes trouble like this. It's a Fujitsu 221D024, I can't easily find a replacement so will try cleaning it and see what happens.
The second issue was on the speaker terminal board, which is double sided. Dry joint on the component side, was actually a bit tricky to fix, but works now.
I now have another problem with the amp cutting out when hot - just for a second or two, then comes back, but it is not an issue at low volumes. Not quite sure how to troubleshoot that and haven't isolated it to the pre-amp or power amp yet.
Ok, so some more testing done, and it's the power amp that is cutting out. When i turn the volume up, after just a few minutes, the relays will starting clicking off for a few seconds, then back on. If I turn the volume down it will happily play with no issues.
Hard to be certain, but it also sounds slightly distorted. I've set the bias to a total of 24mV across both test points, there was conflicting info in these threads, wondering if that could be part of the issue?
Hard to be certain, but it also sounds slightly distorted. I've set the bias to a total of 24mV across both test points, there was conflicting info in these threads, wondering if that could be part of the issue?
From the web, I've attached a service manual for benefit of members.
This design has flat gain to DC, so it will be sensitive to offset trim. I surmise the problem channel is more erratic compared to the better channel? Do you have a scope to better observe fast variations? You observe the 70mV output offset with no input to the PA?
Good luck!
This design has flat gain to DC, so it will be sensitive to offset trim. I surmise the problem channel is more erratic compared to the better channel? Do you have a scope to better observe fast variations? You observe the 70mV output offset with no input to the PA?
Good luck!
Yes. I set the offset to correct values a while back and then went chasing some issues with the pre-amp. With that solved, the only remaining problem is that both channels will regularly cut out for a second or two at high volumes.
At low volume it seems ok. The relays are switching in and out so I assumed it is something to do with the protection circuit and had a closer look at the offset.
I only have a multimeter, but on one channel it is a lot less stable and moves up and down over time. The trimpot is also very sensitive, moving it by 5 degrees can make the offset change by 200mV.
So I will replace the trimpot, but wondering if there is anything else I should be looking at.
At low volume it seems ok. The relays are switching in and out so I assumed it is something to do with the protection circuit and had a closer look at the offset.
I only have a multimeter, but on one channel it is a lot less stable and moves up and down over time. The trimpot is also very sensitive, moving it by 5 degrees can make the offset change by 200mV.
So I will replace the trimpot, but wondering if there is anything else I should be looking at.
From the schematic, the amp seems to have a DC gain of about 18.4 V/V. I estimate the the offset pot should produce about +/- 72mV at full rotations at the wiper terminal. In turn, this should yield about 2.65V peak-to-peak adjustment range observed at amp output. I would think that 200mV adjustment would correspond to roughly 8% of the adjustment range of the trimmer. Are the 20V Zeners (D402, D404) regulating properly? Do both channels behave similarly?
Another interesting clue would look at DC across R406. It will reveal net base bias current drawn by Q402 and Q406. Ideally it will be 0mV. Compare with the other channel for insight. If adjacent range is badly centered, the base currents of Q402 and Q406 may be badly mismatched. Let us know what you observe in each channel.
Another interesting clue would look at DC across R406. It will reveal net base bias current drawn by Q402 and Q406. Ideally it will be 0mV. Compare with the other channel for insight. If adjacent range is badly centered, the base currents of Q402 and Q406 may be badly mismatched. Let us know what you observe in each channel.
So...I've replaced both the trimpots and readjusted the offset. For both channels it's now a lot more stable, I left it with no load for about 10 minutes and each side was drifting across a range of about 10mV.
I then connected a source and played some music and it was fine up to a comfortably loud level, but as soon as I turned it up further it started cutting out again. Checking the offset at this stage, both channels were about 40mV, and still drifting.
I adjusted the bias, decided to try going up to 23mV for each leg rather than 23mV total, but it still cuts out. I think it sounds a bit better though?
One channel is definitely not as balanced. When I was setting bias initially, left was 11.5+11.5, while right was 10.5+12.5. Across R406 is 5mV on left and ~20mV on right. The 2SA1111/2SC2591 pairs have been replaced on both sides. Not sure what else to look at, or how to check the zeners - are they regulators so measure volts across as well?
I then connected a source and played some music and it was fine up to a comfortably loud level, but as soon as I turned it up further it started cutting out again. Checking the offset at this stage, both channels were about 40mV, and still drifting.
I adjusted the bias, decided to try going up to 23mV for each leg rather than 23mV total, but it still cuts out. I think it sounds a bit better though?
One channel is definitely not as balanced. When I was setting bias initially, left was 11.5+11.5, while right was 10.5+12.5. Across R406 is 5mV on left and ~20mV on right. The 2SA1111/2SC2591 pairs have been replaced on both sides. Not sure what else to look at, or how to check the zeners - are they regulators so measure volts across as well?
Yes, D402 and D404 are shunt regulators and should present about 20 VDC across each. They are probably OK, but if they were open, they would exacerbate offset drift and make the DC trim much more "twitchy."
There's a large number of possible culprits. The problem could lie within the PA itself, or it might be a defect in the protection circuitry. Since the problem seems to be provoked by increasing volume, my suspicion is a protection problem--- but that's only a guess. The protection circuit is shared by both channels, and looks at both DC offset and current drawn; this makes diagnostic isolation more difficult.
I assume there's no evidence of mechanical intermittence? Do you hear any noisy, crackling sounds that might be associated with leakage currents, especially in the right channel where you observed 20 mV across R406? This elevated voltage is suggestive of low hfe (or leakage) in the differential pairs. Can you associate DC output drift with drifting voltage across R406? If you note the polarity of the drop across R406, we should be able to focus suspicion on Q402 vs. Q406. This may or may not be the responsible defect.
I can suggest further diagnostic isolation experiments, but they will become more tedious.
P.S. Please be safe! And I advise covering all but tip of you meter probes with heatshrink tubing--- a slip of a probe can wreak havoc.
Good luck!
There's a large number of possible culprits. The problem could lie within the PA itself, or it might be a defect in the protection circuitry. Since the problem seems to be provoked by increasing volume, my suspicion is a protection problem--- but that's only a guess. The protection circuit is shared by both channels, and looks at both DC offset and current drawn; this makes diagnostic isolation more difficult.
I assume there's no evidence of mechanical intermittence? Do you hear any noisy, crackling sounds that might be associated with leakage currents, especially in the right channel where you observed 20 mV across R406? This elevated voltage is suggestive of low hfe (or leakage) in the differential pairs. Can you associate DC output drift with drifting voltage across R406? If you note the polarity of the drop across R406, we should be able to focus suspicion on Q402 vs. Q406. This may or may not be the responsible defect.
I can suggest further diagnostic isolation experiments, but they will become more tedious.
P.S. Please be safe! And I advise covering all but tip of you meter probes with heatshrink tubing--- a slip of a probe can wreak havoc.
Good luck!
Last edited:
Thanks for the info.
The zeners are all reading ~19V, and there's no noise/crackles after I cleaned the relays in the pre-amp (SP5050). No obvious mechanical issues, I have pressed the boards in different places.
It's the right channel, so R405, I double checked and it's actually 40mV drop across it. Coincidentally (?) the drop is the same across R405 and R403, compared to 4mV across R404/406
R407 has 16.5v across it while R408 has 17.2V
R409 has 19.2v across it while R410 has 19.1v
It's hard to get measurements of other areas because I put in a larger heatsink which obscures most of the board now....I think I might just leave as is, it plays loudly enough and I have a Nakamichi AV-10/DVD-10S alongside that I can use in emergencies
The zeners are all reading ~19V, and there's no noise/crackles after I cleaned the relays in the pre-amp (SP5050). No obvious mechanical issues, I have pressed the boards in different places.
It's the right channel, so R405, I double checked and it's actually 40mV drop across it. Coincidentally (?) the drop is the same across R405 and R403, compared to 4mV across R404/406
R407 has 16.5v across it while R408 has 17.2V
R409 has 19.2v across it while R410 has 19.1v
It's hard to get measurements of other areas because I put in a larger heatsink which obscures most of the board now....I think I might just leave as is, it plays loudly enough and I have a Nakamichi AV-10/DVD-10S alongside that I can use in emergencies
Just a last thought, since the DC offset is drifting for both channels, that would imply the issue is on a common circuit, ie on the power output display board. I had read these could have heat problems so I pulled it out and had a good look - no obvious issues there. It's hard to get at so can't really measure anything whilst active.
** Is a DC reading of up to 50mV ok for the speakers? ** I'm assuming that if it gets close to damage levels the relay will kick in so I'm ok to keep using it as is.
** Is a DC reading of up to 50mV ok for the speakers? ** I'm assuming that if it gets close to damage levels the relay will kick in so I'm ok to keep using it as is.