• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Capacitor coupled power amps

ConradH

ConradH

Lunatic Member
While rebuilding a Marantz receiver (2230) and having previously rebuilt a Sansui 2000x, I wondered about sound, economics and conventional wisdom.

Both receivers sound really great. Both receivers have single supplies and capacitively coupled power amps, not dual supplies and direct output coupling. I had always thought the former design was done purely to save money, and the more desirable design was always a full complimentary dual supply design.

Back when these were built, big can capacitors were quite cheap, unlike the inflated prices we have today. If you use one big cap for the power supply, and 2 caps to couple the outputs, that's 3 caps. But, you got the savings of not using power PNP output transistors, though I'm not sure how great those savings were. I do know PNP performance back then wasn't what it is today.

With a dual supply complimentary design you just have 2 power supply caps. The price and performance of PNP devices is very good today, so that should be the more economical design.

So...

1) How much of the "sound" of these receivers might be due to the "inferior" cap coupled design?

2) Is there any real reason to build a single supply cap coupled amp these days?

3) Why do you think they chose to go single supply back then?
 
Register to hide this ad
Some well-regarded designers are sure that capacitors in the signal path do have a sonic contribution. John Curl said he designed the DC output servo circuit into the Halo line of amps specifically so he could eliminate all caps from the signal path. I will also note that Mr. Pass designed the J2 without coupling caps, although that may have been, at least in part, in keeping with the "simpler is better" motif.
 
grateful said:
The fully direct coupled design is considered a superior design from the engineering perspective.
Click to expand...

That's certainly true and I wouldn't previously have considered building anything else, but what prompted this discussion is that the receivers listed, amongst many others, are widely praised as having unusually good sound quality. There are a huge number of "better" lower distortion full complimentary circuits that outperform them, yet are not widely praised for their musicality.

I'm of the belief that all amps sound the same, but only with a long list of caveats- operating within their limits, with low distortion, flat response and low output impedance. Plus a few other things. Those circuits above don't really meet that criteria, so I can accept that they may influence the sound in a positive way. Or, maybe it's a case of the lie repeated many times that becomes the new truth?
 
Many quasi complementary designs have coloration to them. Perhaps it is harmonic distortion or slightly rolled off treble. Many like it I do not.
 
A quasi amp can have very low distortion and flat response. it depends on the design. Both comp and quasi can be dual or single supply. I'm not sure if that is in question, but it seems the OP is thinking quasi is for single supply. This was often the case in the early days. With minor mods it is possible to turn about any amp into single supply, cap coupled or vise versa. One benefit of cap coupled is that if the amp lets go, your speakers don't get DC through them.
 
Last edited:
All true John, but the designs the op is talking about are the single rail cap coupled ones. In general they have lower slew rates, more limited bandwidth and higher distortion. As to whether that has anything to do with their sound quality, I don't have a clue. But I do know there is science to this and much of it is measureable.
 
IMO, it's all measurable! It's just that I hear over and over how much people like the sound of their Sansui and Marantz receivers, that in hard number terms aren't all that great. I've measured my Sansui 2000x and know it has some minor colorations in both the tone board and amp.

I was thinking single supply as I don't think there have been many dual supply amps combined with capacitor coupled outputs, though one can certainly build that if the mood strikes.

The reasons that people like or dislike various audio products is a subject I find quite fascinating, though I doubt we'll ever come up with predictive rules on the subject.
 
ConradH said:
IMO, it's all measurable! It's just that I hear over and over how much people like the sound of their Sansui and Marantz receivers, that in hard number terms aren't all that great. I've measured my Sansui 2000x and know it has some minor colorations in both the tone board and amp.

I was thinking single supply as I don't think there have been many dual supply amps combined with capacitor coupled outputs, though one can certainly build that if the mood strikes.

The reasons that people like or dislike various audio products is a subject I find quite fascinating, though I doubt we'll ever come up with predictive rules on the subject.
Click to expand...

Once you have dual supplies the output coupling cap is more or less unnecessary, in addition to being large and costly, especially if the amp is expected to have a -3dB point at a few Hz.
 
ConradH said:
The reasons that people like or dislike various audio products is a subject I find quite fascinating, though I doubt we'll ever come up with predictive rules on the subject.
Click to expand...

Absolutely. One guy hates it, another guy loves it. Ask two other people and you may get the opposite response. Amps, speakers, media, whatever! So subjective.

There was a double blind test performed that demonstrated how listeners couldn't detect harmonic distortion in music until over a certain level. 2.5% I think was given. Many of the power amps in early SS receivers had THD figures that are nothing to write home about, yet perhaps still too low to be audible to the owner.

My audio philosophy is simple. If it sounds good, it is good.
 
fpitas said:
Once you have dual supplies the output coupling cap is more or less unnecessary, in addition to being large and costly, especially if the amp is expected to have a -3dB point at a few Hz.
Click to expand...

I suppose I could say that output transformers on a solid state amp are quite unnecessary, but that might offend followers of McIntosh. :yes:
 
ConradH said:
I suppose I could say that output transformers on a solid state amp are quite unnecessary, but that might offend followers of McIntosh. :yes:
Click to expand...

Yes, it might. IMHO that kind of design is getting into the arena of special effects.
 
McIntosh SS amps don't have output transformers they have autoformers. There is no primary simply a tapped secondary. This allows the output transistors to be designed to operate optimally into the 2.5 ohm load of the coil. This keeps them in the soa at all times and allows high power with lower voltage rails.
 
IMO, an autotransformer is a transformer and so is a tapped inductor. Regardless of semantics, there's definitely some impedance transforming device at the back end of the thing. They certainly had a good reason for using it, though IMO the reasons become less valid as semiconductor devices improved over the years.
 
ConradH said:
IMO, an autotransformer is a transformer and so is a tapped inductor. Regardless of semantics, there's definitely some impedance transforming device at the back end of the thing. They certainly had a good reason for using it, though IMO the reasons become less valid as semiconductor devices improved over the years.
Click to expand...

I agree an autoformer is a type of transformer, and shares many of the vices and virtues of a transformer. Their reasons for keeping it were probably that users expected that sound.
 
The claim was made that McIntosh ss amps have output transformers. That is not correct. There is no inductive coupling. There is not the limited bandwith that transformers entail.
 
There is no inductive coupling the output transistors are still diirectly connected to the load. They are more similar to the output inductors used in virtually all ss amps. They also shunt dc to ground which prevents speaker burnout if the ouput section fails.
 
You must log in or register to reply here.
Back
Top Bottom