I’m already swapping out 10 capacitors, the original resistors have meaningful drift, and the power supply is a mess. From what I know, Bongiorno wired these amplifiers up because Hadley had shakey hands. Bongiorno was in his early 20s at the time and his work was sloppy on all the copies of this amplifier I’ve seen. Or maybe Hadley wired the PS. The boards are beautifully done on the second and nicer copy of this amplifier I have (the one I’m now working on). On the other 622 amp, I rebuilt the power supply and the difference was obvious as has been the case on my Marantz 8B, my McIntosh MC225, and one of my Fisher 50AZs.
Last edited:
The photo of the capacitor bank is impressive, but probably all Siemens caps are still fine.
It would be interesting to see pictures of the resistors, in my experience parts like this
rarely fail, only particular brands.
No recommendation from me to swap the Zener diodes, but your readings in post 39 are
correct.
It would be interesting to see pictures of the resistors, in my experience parts like this
rarely fail, only particular brands.
No recommendation from me to swap the Zener diodes, but your readings in post 39 are
correct.
Hi hadley,
I'll agree with as_audio completely on this. The zener diodes should be fine and have settled into their long term characteristics. If it were mine, I would leave the zener diodes alone.
Main filter capacitors have a lot of electrolyte inside them, and as long as the seals are good they ought to be okay. One thing you can do is with the amplifier running, no signal and warmed up, connect an oscilloscope across the filter capacitors on AC coupling. Start vertical range high and switch down for each measurement. You want your trigger coupling to be on "line". Look at the leading edge of the sawtooth. If it is slightly rounded the capacitor is absolutely perfect. If you see a sharp edge, it is still just fine. When capacitors degrade, you will see a "pip" appear on the leading edge of the sawtooth, becoming worse as the part degrades. You can see this long before an ESR meter shows a problem. Another sensitive measurement is the dielectric absorption, but you have to measure the capacitor at the frequencies it operates at. So for me, I use 120 Hz and 1 KHz.
Hope that helps.
-Chris
I'll agree with as_audio completely on this. The zener diodes should be fine and have settled into their long term characteristics. If it were mine, I would leave the zener diodes alone.
Main filter capacitors have a lot of electrolyte inside them, and as long as the seals are good they ought to be okay. One thing you can do is with the amplifier running, no signal and warmed up, connect an oscilloscope across the filter capacitors on AC coupling. Start vertical range high and switch down for each measurement. You want your trigger coupling to be on "line". Look at the leading edge of the sawtooth. If it is slightly rounded the capacitor is absolutely perfect. If you see a sharp edge, it is still just fine. When capacitors degrade, you will see a "pip" appear on the leading edge of the sawtooth, becoming worse as the part degrades. You can see this long before an ESR meter shows a problem. Another sensitive measurement is the dielectric absorption, but you have to measure the capacitor at the frequencies it operates at. So for me, I use 120 Hz and 1 KHz.
Hope that helps.
-Chris
Those capacitors all tested fine, but the sealing had bulges.
I left the zeners in on the side of the power supply I was working on. If I need to replace them now I know what they are.
One of the 1k resistors in this portion of the power supply was 20% off from its spec value so at least I didn't replace these resistors for nothing though I doubt the 20% would've caused much of an issue.
I will try that, thank you. the filter caps are riveted in and I would prefer not to replace them. Hopefully they're still good.
I've selected my differential pair dual transistors. I think I'm into audiophoolery territory:
Data sheet for those who don't know what these are:
Data sheet for those who don't know what these are:
Last edited:
The metal can was blank, but I'm fairly certain it was an early Fairchild unit. Six pins, same layout as the mat12.
It would be nice to read about your sonic impressions of the amp, knowing the
speakers used and preferably in comparison with better known amplifier models.
speakers used and preferably in comparison with better known amplifier models.
Original filter caps are hilariously bad so I’ll pull them today. The amplifier sounds incredible but the power consumption at lower frequencies spikes unacceptably. I give more impressions once I’ve done the swap.
I've run into a new issue: the power supply voltages are pretty far off from their specification.
Below measurements:
I took these voltages at a few different potential ground points and they did not change, so I am going to assume they're accurate. This is an issue because I can't zero out the voltage across the outputs.
Let me try to explain this another way: since this is a differential amplifier and I've already zeroed out the DC offset, I get 0mV with probes across the outputs for each channel. This is what the 200ohm trimmer potentiometer adjusts. Mine is basically in the middle of its range. That's good.
HOWEVER, if I ground one probe and measure voltage at each input, I'm getting something like 13V. I can only reach about 5V is if I turn the 25K trimmer pot to the appropriate extreme (what would you call this, DC Balance?). I think I should be at zero to balance out the voltages across the amplifier. So, these zeners need to go (or at least 2 of 3).
With that being said, why such a large gap between the DC voltage at the 10ohm resistor vs. the output voltages? Based on my simulations, dropping from -67.8VDC to -51VDC puts a lot of strain on the zener. Maybe this is why they chose two zeners adding up to 58VDC?
My question is this: is there a certain relationship I need to preserve between the -30, -36.8, and -51vdc supplies? Any idea why they chose those numbers?
Below measurements:
I took these voltages at a few different potential ground points and they did not change, so I am going to assume they're accurate. This is an issue because I can't zero out the voltage across the outputs.
Let me try to explain this another way: since this is a differential amplifier and I've already zeroed out the DC offset, I get 0mV with probes across the outputs for each channel. This is what the 200ohm trimmer potentiometer adjusts. Mine is basically in the middle of its range. That's good.
HOWEVER, if I ground one probe and measure voltage at each input, I'm getting something like 13V. I can only reach about 5V is if I turn the 25K trimmer pot to the appropriate extreme (what would you call this, DC Balance?). I think I should be at zero to balance out the voltages across the amplifier. So, these zeners need to go (or at least 2 of 3).
With that being said, why such a large gap between the DC voltage at the 10ohm resistor vs. the output voltages? Based on my simulations, dropping from -67.8VDC to -51VDC puts a lot of strain on the zener. Maybe this is why they chose two zeners adding up to 58VDC?
My question is this: is there a certain relationship I need to preserve between the -30, -36.8, and -51vdc supplies? Any idea why they chose those numbers?
Since -51V is basically unloaded by the amp it may be ok -
no ripple here ?
Check D7 with diode tester (dmm in semiconductor range).
no ripple here ?
Check D7 with diode tester (dmm in semiconductor range).
-51 V comes off a zener diode, so it isn't okay. The zener may have a bad connection, failing open would be extremely rare.
Suggested alignment procedure :
- confirm that all parts of amp and supply are connected properly
- avoid grounding any output when using a scope etc, this is a bridged output
- adjust speaker output pins to 0 V dc with respect to each other using probes on output with R11
- adjust one (either) speaker output to 0 V with respect to ground with R15.
- confirm that all parts of amp and supply are connected properly
- avoid grounding any output when using a scope etc, this is a bridged output
- adjust speaker output pins to 0 V dc with respect to each other using probes on output with R11
- adjust one (either) speaker output to 0 V with respect to ground with R15.
Last edited:
I should clarify. When they built this amplifier, they deviated from the schematic and used a 36V zener in series with another 22V zener. so they were going for -58v. Given the other values also being a bit higher than the schematic, I could still theoretically barely balance out the amplifier at -58v using the 25k trimmer pot, but not -62.8V. Is there any advantage to running these more negative voltages or should I just return the amplifier to the schematic?
The reading 67.8V is not correct, it must be negative.
Check dc at the connection R41, R42, R45, C13 also.
Check dc at the connection R41, R42, R45, C13 also.