Thought an update was in order after having used and thoroughly enjoyed the amp for a few months now - finally going to recap the main power-amp board first, followed by the other sections. PSU already done, of course.
I still get that occasional very high pitched whistle from the SMPS, and it goes if I tap the case so it's obviously a vibrating component. Noticed on the SMPS board that there's a coil that isn't covered in any kind of resin like some of the other parts are, so I'll probably slip some heatshrink over that and see if that doesn't stop the occasional noise. I've noticed many modern SMPS's have heatshrink over the smaller coils as well so hopefully that should be a safe and simple fix.
Other than that it's been working absolutely fine everyday and sounds great. I still need to fix that stuck VU meter though but that's at the bottom of things to do.
Not sure what to do about removing wires to the power-amp board to allow easier access to desoldering caps etc. All the wires are wire-wrapped in place. Do these pull off, or do they still need to be desoldered? Or does one apply heat to the 'wrap' section whilst sliding it off the anchor point?
Cheers,
- John
I still get that occasional very high pitched whistle from the SMPS, and it goes if I tap the case so it's obviously a vibrating component. Noticed on the SMPS board that there's a coil that isn't covered in any kind of resin like some of the other parts are, so I'll probably slip some heatshrink over that and see if that doesn't stop the occasional noise. I've noticed many modern SMPS's have heatshrink over the smaller coils as well so hopefully that should be a safe and simple fix.
Other than that it's been working absolutely fine everyday and sounds great. I still need to fix that stuck VU meter though but that's at the bottom of things to do.
Not sure what to do about removing wires to the power-amp board to allow easier access to desoldering caps etc. All the wires are wire-wrapped in place. Do these pull off, or do they still need to be desoldered? Or does one apply heat to the 'wrap' section whilst sliding it off the anchor point?
Cheers,
- John
I'll try the heatshrink and see what happens - fingers crossed!
Whilst going through my parts bin for a couple of 1000uF 25V caps for the power amp PCB board (C310 & C360), I found a pair of Panasonic FMs, Rubycon ZLHs and Rubycon ZLJs.
Does anybody know which would be the best choice I'm out of my depth here:
http://www.rubycon.co.jp/en/catalog/e_pdfs/e_SERIES-CHART.pdf
www.panasonic.com/industrial/components/pdf/ABA0000CE108.pdf
Cheers,
- John
Last edited:
Hi John, really pleased the amp is still going strong... great job.
The caps. Its not a "demanding" application at that point in the circuit, certainly not with regard to ripple current and low E.S.R. etc. Usually the larger a cap is physically (for the same V and C) and the better. I would always fit 105C temp rated parts these days.
Its not demanding because its fed via 4.7k resistor from the main rails which severely limits any "rate of change" of voltage across the cap. The cap then goes on to feed an internal regulator in the IC.
The varnish trick Jaycee mentions is something I used to do a lot with line output transformer on TV's. You have to literally submerge the tranny for several minutes, then just hang it up to drip and dry. Sometimes successful, sometimes not but nothing lost.
The caps. Its not a "demanding" application at that point in the circuit, certainly not with regard to ripple current and low E.S.R. etc. Usually the larger a cap is physically (for the same V and C) and the better. I would always fit 105C temp rated parts these days.
Its not demanding because its fed via 4.7k resistor from the main rails which severely limits any "rate of change" of voltage across the cap. The cap then goes on to feed an internal regulator in the IC.
The varnish trick Jaycee mentions is something I used to do a lot with line output transformer on TV's. You have to literally submerge the tranny for several minutes, then just hang it up to drip and dry. Sometimes successful, sometimes not but nothing lost.
Greetings Mooly!
Well I have you to thank for this wonderful amp working at all
Thanks for the advice on those two caps - went with the ZLJs in the end. All other caps replaced on the power amp board now too. Took the opportunity to also clear off 30+ years of dust and grime from the PCB and components - lovely job. Also replaced the output transistor mica spacers and cleaned up all the old heatsink paste and used fresh silpads. At the very least it all looks nicer now
Onto setting the trim pots (also replaced with new Vishay types, set at middle positions for now).
When setting DC balance to 0v per channel the manual doesn't explicitly say if a pair of speakers should be connected or not. However this appears to be the case in the diagrams. Should I use an 8-ohm power resistor per channel when setting the DC balance?
When setting the DC bias (22mV) per channel, the service manual states no source should be connected, but doesn't mention speakers. Assume these are not connected here?
Many thanks,
- John
Well I have you to thank for this wonderful amp working at all
Thanks for the advice on those two caps - went with the ZLJs in the end. All other caps replaced on the power amp board now too. Took the opportunity to also clear off 30+ years of dust and grime from the PCB and components - lovely job
. Also replaced the output transistor mica spacers and cleaned up all the old heatsink paste and used fresh silpads. At the very least it all looks nicer now Onto setting the trim pots (also replaced with new Vishay types, set at middle positions for now).
When setting DC balance to 0v per channel the manual doesn't explicitly say if a pair of speakers should be connected or not. However this appears to be the case in the diagrams. Should I use an 8-ohm power resistor per channel when setting the DC balance?
When setting the DC bias (22mV) per channel, the service manual states no source should be connected, but doesn't mention speakers. Assume these are not connected here?
Many thanks,
- John
Last edited:
It's easiest to say with no speakers and no source. And no dummy load. Reasons why...
The DC offset should be the same with and without speakers and adjusts the same with and without.
The DC bias current reading could be influenced by a load (speakers or resistor) if the DC offset were not exactly 0.000 volts. Thats because the offset would in itself cause a current to flow in the speaker and alter the 22 mv reading. So it's always with nothing connected.
The DC offset should be the same with and without speakers and adjusts the same with and without.
The DC bias current reading could be influenced by a load (speakers or resistor) if the DC offset were not exactly 0.000 volts. Thats because the offset would in itself cause a current to flow in the speaker and alter the 22 mv reading. So it's always with nothing connected.
Cheers for that Mooly.
However it does appear to indicate that an 8-ohm power resistor is needed when setting the DC balance to 0V. Have attached that part of the service manual to see what you think:
Even though the manual indicates setting 0v DC balance first, I guess it's safer to check the DC Bias first to ensure it's not excessively high before any other adjustments are made?
However it does appear to indicate that an 8-ohm power resistor is needed when setting the DC balance to 0V. Have attached that part of the service manual to see what you think:
An externally hosted image should be here but it was not working when we last tested it.
Even though the manual indicates setting 0v DC balance first, I guess it's safer to check the DC Bias first to ensure it's not excessively high before any other adjustments are made?
Last edited:
Hi John, you will find the DC offset is essentially the same with and without the speakers or a load. Try it
The bias should not interact with the offset. Once the offset is really down to zero or as near as you can get it then the bias will be correct whether adjusted with or without a load (speaker).
If the offset were say 1 volt or even something ridiculous like 8 volts if it were possible to swing it that far, then the bias would still be correct with no load. Connect a load and that 8 volts would push 1 amp through one or other of the outputs depending on the polarity of the offset. That would swamp the bias current reading.
On a practical level the bias will drift as the outputs get hot so its always best practice to set it finally when the amp is hot... after playing music loud for say 20 minutes. Amps vary in how well the bias tracks temperature.
The DC offset should be essentially stable with temperature to with a few millivolts drift.
The bias should not interact with the offset. Once the offset is really down to zero or as near as you can get it then the bias will be correct whether adjusted with or without a load (speaker).
If the offset were say 1 volt or even something ridiculous like 8 volts if it were possible to swing it that far, then the bias would still be correct with no load. Connect a load and that 8 volts would push 1 amp through one or other of the outputs depending on the polarity of the offset. That would swamp the bias current reading.
On a practical level the bias will drift as the outputs get hot so its always best practice to set it finally when the amp is hot... after playing music loud for say 20 minutes. Amps vary in how well the bias tracks temperature.
The DC offset should be essentially stable with temperature to with a few millivolts drift.
Thanks for that Mooly
Went ahead and set the DC balance to 0V (or as close to it as possible i.e. 0.03mV on one channel, 0.00 on the other!) OK.
However I cannot get 22mV on either channel. I get 17mV with the pot turned to max on the left channel, and 18mV on the right, again with the pot turned to max.
Is this indicative of a problem or do I need to let the amp warm up (as you mention above)?
Hope this isn't some new problem to hunt down - ARGHHHHH!
Went ahead and set the DC balance to 0V (or as close to it as possible i.e. 0.03mV on one channel, 0.00 on the other!) OK.
However I cannot get 22mV on either channel. I get 17mV with the pot turned to max on the left channel, and 18mV on the right, again with the pot turned to max.
Is this indicative of a problem or do I need to let the amp warm up (as you mention above)?
Hope this isn't some new problem to hunt down - ARGHHHHH!
Was just looking for the circuit for power amp lol, and see its on this page.
It won't be faulty as such. 22mv across 0.22 ohm is 100 milliamp which is a typical value for any class ab amp. Are those 0.22 ohms original and of the correct value. I'm sure they are.
All I can suggest is you see what happens as the amp warms and keep a check.
What you can try for interest. Turn the pots to give minimum voltage which is minimum current (coldest running). Listen to the amp like that at very low volume on speakers (not phones) and to a wide variety of music (strings are best). Can you hear the distortion as a "rough" sound ? Increase the bias by only a fraction and it should go.
I don't know why it won't go quite to what the manual says... but it won't be a fault if both channels are the same. Sometimes manuals are wrong, as a service tech I have seen that many times... so its a possiblity.
It won't be faulty as such. 22mv across 0.22 ohm is 100 milliamp which is a typical value for any class ab amp. Are those 0.22 ohms original and of the correct value. I'm sure they are.
All I can suggest is you see what happens as the amp warms and keep a check.
What you can try for interest. Turn the pots to give minimum voltage which is minimum current (coldest running). Listen to the amp like that at very low volume on speakers (not phones) and to a wide variety of music (strings are best). Can you hear the distortion as a "rough" sound ? Increase the bias by only a fraction and it should go.
I don't know why it won't go quite to what the manual says... but it won't be a fault if both channels are the same. Sometimes manuals are wrong, as a service tech I have seen that many times... so its a possiblity.
I'll do as you suggest Mooly and let it warm up then take the readings again - assume it's OK to leave the probes connected whilst playing music through the speakers, rather than having to scrabble to turn the amp off, attach the probes and take measurements in the space of a few seconds?
Judging by the cowboy repair work that was done to the SMPS, it wouldn't surprise me if something was changed on the power-amp PCB which might explain why I can only get 17mV max. instead of the expected 22mV. I know the output transistors were changed from the original Hitachi 2SD675 / 2SB655 pairs to OnSemi MJ21193 / MJ21194 so perhaps that might partially explain it?
Judging by the cowboy repair work that was done to the SMPS, it wouldn't surprise me if something was changed on the power-amp PCB which might explain why I can only get 17mV max. instead of the expected 22mV. I know the output transistors were changed from the original Hitachi 2SD675 / 2SB655 pairs to OnSemi MJ21193 / MJ21194 so perhaps that might partially explain it?
Last edited:
Its fine to leave probes connected, just turn the volume down to check the readings.
I couldn't remember without re-reading it all whether the amps had been worked on.
Transistor types won't make any difference. What I would do is double check the 0.22 ohms really are that and also check those 56 ohm and 100 ohm resistors nearby (in the grey area on circuit). Also check those 270 ohms also in a grey box. If any of those have been replaced with incorrect values it could alter the range of bias available.
I couldn't remember without re-reading it all whether the amps had been worked on.
Transistor types won't make any difference. What I would do is double check the 0.22 ohms really are that and also check those 56 ohm and 100 ohm resistors nearby (in the grey area on circuit). Also check those 270 ohms also in a grey box. If any of those have been replaced with incorrect values it could alter the range of bias available.
