Greetings fellow forumites,
I have a 1974 Marantz 1120 on my bench. It's in really good nick considering it's 40 years old! Still working on both channels and no noisy transistors.
Had low audio level on one channel when first powered up. DC offset is over 200mv on that channel and I see it's been repaired before - about 20 years ago according to the owner. It's entirely possible the DC offset (and the bias for that matter) were not adjusted or checked after the repair. Both channels seem to run at about the same temperature so I'm not too worried about it.
This quasi-complimentary design uses a pair of SJ2519 as the output transistors, apparently they were custom made for Marantz by Motorola. (Pete B, I think I have you to thank for that info) Anyway I have yet to find a datasheet for them but in another thread it would seem they are 160v 16a devices. At the last repair someone fitted 3055s - 100v 15a devices. I was rather surprised to see this but they've lasted 20 years. I'd be interested in what you more knowledgeable guys think about this. The circuit diagram shows the power-amps supply rails as 45-0-45. I was considering fitting MJ15003 devices - 140v 20a. On the other hand I wonder if it's not best to let sleeping dogs lie? Thing is, I have to remove them to change the caps on that board....see the dilemma?
Questions regarding cap replacement.
The service manual states that all electrolytics used have tolerances of +50% -20%.
How do you guys feel about upping the values on replacements? I'm referring to the power supply of course. I wouldn't change any values in amplifier, coupling or timing circuits.
Current values are 220uF, 150uF, 680uF off the one rectifier and 8000uF off the other. How about 330, 220 and 1000? Or 820 if 1000 is pushing it for the last one? And how about 15,000 to 22,000uF for the main psu caps? I can get those values in the same screw terminal-can size as the originals. It would be plug and play.
I usually measure new caps before i fit them and they often fail to reach the stated value, for example, a brand new 220uF Rubycon may only measure 200uF. Interestingly, a lot of these 40 year old caps still measure in excess of their stated values.
I think back then caps were a lot more expensive and way bigger so it's possible that these were considerations when the values were first chosen.
I just don't want to recap this thing and give it back with less 'effective' capacitance in the circuit then it had when it came in. Sure they'll be brand new and I'm sure it will work fine - but right now I have a choice.
My concerns are that the caps off the first rectifier are 'cascaded' - sorry if that's the wrong terminology. The first pair of 220uf are fed through 10 ohm resistors, the second pair of 220uf through 100 ohm following the preceding pair. Then come the 150uF through 470 ohm. The 680uF are off the regulated 22v rails. Bigger values mean more inrush current from a cold start. I wouldn't want to overstress anything, however the proposed % increase is relatively small.
I'd value any comments. Thanks in advance.
Chris
I have a 1974 Marantz 1120 on my bench. It's in really good nick considering it's 40 years old! Still working on both channels and no noisy transistors.
Had low audio level on one channel when first powered up. DC offset is over 200mv on that channel and I see it's been repaired before - about 20 years ago according to the owner. It's entirely possible the DC offset (and the bias for that matter) were not adjusted or checked after the repair. Both channels seem to run at about the same temperature so I'm not too worried about it.
This quasi-complimentary design uses a pair of SJ2519 as the output transistors, apparently they were custom made for Marantz by Motorola. (Pete B, I think I have you to thank for that info) Anyway I have yet to find a datasheet for them but in another thread it would seem they are 160v 16a devices. At the last repair someone fitted 3055s - 100v 15a devices. I was rather surprised to see this but they've lasted 20 years. I'd be interested in what you more knowledgeable guys think about this. The circuit diagram shows the power-amps supply rails as 45-0-45. I was considering fitting MJ15003 devices - 140v 20a. On the other hand I wonder if it's not best to let sleeping dogs lie? Thing is, I have to remove them to change the caps on that board....see the dilemma?
Questions regarding cap replacement.
The service manual states that all electrolytics used have tolerances of +50% -20%.
How do you guys feel about upping the values on replacements? I'm referring to the power supply of course. I wouldn't change any values in amplifier, coupling or timing circuits.
Current values are 220uF, 150uF, 680uF off the one rectifier and 8000uF off the other. How about 330, 220 and 1000? Or 820 if 1000 is pushing it for the last one? And how about 15,000 to 22,000uF for the main psu caps? I can get those values in the same screw terminal-can size as the originals. It would be plug and play.
I usually measure new caps before i fit them and they often fail to reach the stated value, for example, a brand new 220uF Rubycon may only measure 200uF. Interestingly, a lot of these 40 year old caps still measure in excess of their stated values.
I think back then caps were a lot more expensive and way bigger so it's possible that these were considerations when the values were first chosen.
I just don't want to recap this thing and give it back with less 'effective' capacitance in the circuit then it had when it came in. Sure they'll be brand new and I'm sure it will work fine - but right now I have a choice.
My concerns are that the caps off the first rectifier are 'cascaded' - sorry if that's the wrong terminology. The first pair of 220uf are fed through 10 ohm resistors, the second pair of 220uf through 100 ohm following the preceding pair. Then come the 150uF through 470 ohm. The 680uF are off the regulated 22v rails. Bigger values mean more inrush current from a cold start. I wouldn't want to overstress anything, however the proposed % increase is relatively small.
I'd value any comments. Thanks in advance.
Chris
in a few words
you may beef up anything behind regulators and after diodes ... keep all feedback , decoupling and similar function capacitors at the same level of uf increase temp voltage and quality ...
Small range 0.47 Uf to 2.2 uf Dc blocking electrolytics may replaced with MKT grade capacitors only if you can lay your hands in 63V versions with pitch 5mm and expected small physical size
Inrush current might be an issue think of that versus beefing up
""Regulation"" through series resistors and capacitors might be easily replaced with regulators 78-317 style ...more quiet more accurate ...
20 + years old 2N3055 is a dead slow device bit more than 2 mhz I bet you that original transistor was easily 30 that is 10 times up .If Marantz thought so then there is got to be a reason for that ...
Some info on the procedure can be found here :
http://www.diyaudio.com/forums/solid-state/136261-vintage-amplifier-repair-upgrade-manual.html
Kind regards
Sakis
you may beef up anything behind regulators and after diodes ... keep all feedback , decoupling and similar function capacitors at the same level of uf increase temp voltage and quality ...
Small range 0.47 Uf to 2.2 uf Dc blocking electrolytics may replaced with MKT grade capacitors only if you can lay your hands in 63V versions with pitch 5mm and expected small physical size
Inrush current might be an issue think of that versus beefing up
""Regulation"" through series resistors and capacitors might be easily replaced with regulators 78-317 style ...more quiet more accurate ...
20 + years old 2N3055 is a dead slow device bit more than 2 mhz I bet you that original transistor was easily 30 that is 10 times up .If Marantz thought so then there is got to be a reason for that ...
Some info on the procedure can be found here :
http://www.diyaudio.com/forums/solid-state/136261-vintage-amplifier-repair-upgrade-manual.html
Kind regards
Sakis
- Status
- Not open for further replies.