Vintage amplifiers that could challenge or approach current state of the art amplifiers

[Frank Dernie]

It seems to me that a fully differential amp has the potential to be much more expensive, due to almost double the number of components, and possibly less linear than single ended if the 2 halves of the circuit are not perfectly matched.
The only benefit I can think of is balanced connections which make it easier to connect (though I have never had an insoluble problem with single ended connections in over 50 years - not no problems just none I couldn't fix)

 

[DonH56]

A distinguishing feature of these differential designs is that the feedback is cross coupled. In a 'bridged' amp this is not the case.

Makes sense, thanks. Though you could do that in a bridged design, albeit putting a switch in the feedback path is usually a Bad Idea.

Apart from having a marketing story to tell, what advantage does a fully differential design give? Having lower rails is only important on portable equipment or something powered off a wall-wart, so of no practical benefit for a power amp. So, if that's the case that many modern power amps are fully differential, why?

S.

True differential circuits (and to some extent various quasi-differential circuits) offer about 3 dB higher SNR than single-ended circuits, ideally cancel even-order distortion terms, and provide greater rejection to common-mode noise from things like power supplies, induced EM, coupled RFI/EMI, and so forth. They are not generally ground-referenced so can also help break ground loops, and provide the ability to shield signals without having shield (noise) current in the signal path. Those are the main things I can think of off-hand.

See https://www.audiosciencereview.com/...d-balanced-unbalanced-and-all-that-jazz.1352/ for a very basic overview.

HTH - Don

 

[sergeauckland]

It seems to me that a fully differential amp has the potential to be much more expensive, due to almost double the number of components, and possibly less linear than single ended if the 2 halves of the circuit are not perfectly matched.
The only benefit I can think of is balanced connections which make it easier to connect (though I have never had an insoluble problem with single ended connections in over 50 years - not no problems just none I couldn't fix)

That's pretty much my view, I just don't see the point. It doesn't even make balanced inputs any easier as any common or garden opamp has a differential input, so trivial to make it balanced to the outside world.

But then there's so much these days I just don't see the point of. Social media?

S.

 

[DonH56]

Most op-amps have a differential input stage but move to a single-ended in the second gain stage and output. They are not fully-differential amplifiers (though there are a few that are).

 

[MakeMineVinyl]

.....ideally cancel even-order distortion terms....

I wouldn't consider that an advantage. The ideal is no distortion, but I'd rather keep the even harmonics and reject the odd ones.

 

[oldsysop]

Pioneer, Kenwood "trash" and "dreadful sounding things

 

[MakeMineVinyl]

A muted but sweet top end........OKAY.........

 

[sergeauckland]

Makes sense, thanks. Though you could do that in a bridged design, albeit putting a switch in the feedback path is usually a Bad Idea.



True differential circuits (and to some extent various quasi-differential circuits) offer about 3 dB higher SNR than single-ended circuits, ideally cancel even-order distortion terms, and provide greater rejection to common-mode noise from things like power supplies, induced EM, coupled RFI/EMI, and so forth. They are not generally ground-referenced so can also help break ground loops, and provide the ability to shield signals without having shield (noise) current in the signal path. Those are the main things I can think of off-hand.

See https://www.audiosciencereview.com/...d-balanced-unbalanced-and-all-that-jazz.1352/ for a very basic overview.

HTH - Don

Thanks for that. Although, given how good some conventional amplifiers from the 1980s were (are still), isn't it just like the perceived need for ever lower SINADs in DACs, an intellectual exercise for designers to see how low they can go?

I would need other reasons to buy a true differential amplifier over another 'normal' amplifier than even lower distortion and noise.

S.

 

[sergeauckland]

Most op-amps have a differential input stage but move to a single-ended in the second gain stage and output. They are not fully-differential amplifiers (though there are a few that are).

Yes, but my point was that the input can be fully balanced differential with no more effort than a single ended input. What the output is, I really don't care, nor do the loudspeakers.

S.

 

[Sgt. Ear Ache]

Pioneer, Kenwood "trash" and "dreadful sounding things

lol, something tells me ol' Paul would much prefer people buy new amps rather than vintage ones.

 

[KaiserSoze]

OK, now I get the title "Audio Science Review". It has to be measured and not heard. But when it can be measured, such as the interference from a power supply, you claim it cannot be heard, right? Well, go figure.

That's pretty much correct, and it is not the least bit silly or inconsistent or whatever you think. If someone thinks they are hearing something that isn't capable of being measured, it is all but certain that they are imagining something. Double-blind testing protocol can be used to determine whether they are in fact hearing something. On the other side of the coin, just because something is measurable doesn't necessarily mean that it is audible. There is nothing the least bit incongruous about this. The fact that you obviously think it is incongruous is a clear indication that you prefer not to think rationally. What reason do you have to think that everything that is measurable ought to be audible? None. What reason do you have to think that people don't ever imagine that they are hearing something that they aren't actually hearing? None. Double-blind testing demonstrates conclusively that there are artifacts that are detectable electronically and that no one can hear. Double-blind testing demonstrates conclusively that people often claim they hear things that they don't actually hear.

 

[KaiserSoze]

To be honest, with me it is not about the weight or dimensions, but I am feeling a little troubled just reading the words 'switching power supply'. In my personal experience such technology has very little in common with an audiophile listening experience. For example, I had quite a bid of noise on the power grid despite running a dedicated power cord. It turned out that 50% of this came from the DAC's switching power supply. I replaced it with a 100 VA linear power supply and guess what... Switching power supplies might be good for shipping costs and for the energy bill, but they tend to mess up the sound.

Some of what you wrote is so untenable that I think I must not be correctly interpreting.

I had quite a bid of noise on the power grid despite running a dedicated power cord.

What do you mean exactly by "dedicated power cord", and why did you think it would help with suppressing noise on the main power feed?

... I am feeling a little troubled just reading the words 'switching power supply'. ...Switching power supplies might be good for shipping costs and for the energy bill, but they tend to mess up the sound.

Say what? Are you not claiming in essence that switching power supplies introduce a type of noise that is incapable of being suppressed with filtering? It certainly seems that this is what you are saying. I read it several times and I can't see any way to avoid this interpretation. Yes, I am quite certain that this is what you are saying, that no sort of filtering can possibly suppress the noise introduced by a switching power supply. Do you know of any theoretical reason for why this would be? Do you have any guess as to why it would be?

 

[Digital Mastering System]

Are you old enough to remember Southwest Technical Products?

I built a bunch in the 70's and 80's. Still have the preamp and the 30W ".01". Use the power amp as my test bench speaker amp.

 

[Martin]

This was my dream amp in the early 1980's:

I'd love to hear one of their new 2HP-D amps:

Martin

 

[DonH56]

Thanks for that. Although, given how good some conventional amplifiers from the 1980s were (are still), isn't it just like the perceived need for ever lower SINADs in DACs, an intellectual exercise for designers to see how low they can go?

I would need other reasons to buy a true differential amplifier over another 'normal' amplifier than even lower distortion and noise.

S.

Yeah, got that. I could argue that with all the transistors you can put on an IC, why not go differential, and get that extra bit of noise and power supply immunity?

Yes, but my point was that the input can be fully balanced differential with no more effort than a single ended input. What the output is, I really don't care, nor do the loudspeakers.

S.

Without a differential output it is harder to match input impedances when feedback is applied since you only have one polarity at the output. For non-ideal op-amps the (+) and (-) inputs are different in-circuit. So you don't get "true" differential inputs from a typical (single) op-amp. The math works all the way to the speakers.

I'm a fan of differential operation and that is pretty much what I have always designed in my day job. For audio, especially consumer audio, the biggest benefit is the ability to break a ground loop, followed by common-mode noise rejection (CMRR). But those are not of benefit to all, and the improved dynamic range is not a big deal given where most designs are these days (though to get there DAC outputs are almost universally differential with conversion taking place in the buffer circuit after the DAC, still inside the box audiophiles call a "DAC").

For me, the biggest benefit in my home system is the ability to run my back subs on a separate circuit without ground loop noise. When I was doing live sound and recording, differential mic circuits were the only way to keep the noise down. Thus my past history in audio and other analog circuit design predisposes, or biases, me toward differential but it is not something I normally pay much attention to when buying a piece of audio gear. If two have similar features and performance for the price I'll get the differential one, why not? If you are convinced differential costs more and is of no benefit, not much point in discussing it further. Pros and cons...

 

[Robin L]

This was my dream amp in the early 1980's:

I'd love to hear one of their new 2HP-D amps:

Martin

Had one of SAE's integrated amps, crapped out on me some years back. I suspect the capacitors were starting to bail. Seriously, not that great.

 

[KaiserSoze]

Pioneer, Kenwood "trash" and "dreadful sounding things

Oh no, not another one of these guys on YouTube. They come out of the woodwork.

 

[MakeMineVinyl]

This was my dream amp in the early 1980's:

I'd love to hear one of their new 2HP-D amps:

Martin

the 2HP is a great amp.

 

[DonH56]

I wouldn't consider that an advantage. The ideal is no distortion, but I'd rather keep the even harmonics and reject the odd ones.

Yah, but there's no easy way to do that (ideal, or reject odd order terms). I prefer to eliminate as much distortion as I can. You consider having more distortion an advantage?

 

[DonH56]

It seems to me that a fully differential amp has the potential to be much more expensive, due to almost double the number of components, and possibly less linear than single ended if the 2 halves of the circuit are not perfectly matched.
The only benefit I can think of is balanced connections which make it easier to connect (though I have never had an insoluble problem with single ended connections in over 50 years - not no problems just none I couldn't fix)

Very much depends upon the implementation. Virtually all DACs have differential outputs, and a dif pair on an IC is cheap and easy and works much, much better than a single-ended amplifier circuit. The extra circuitry and power to linearize and bias (stably) the single-ended cell often outweighs the cost of going differential. Most input stages are differential; output stages are often (usually) not, but they are wrapped in a feedback loop and so benefit from the previous cells' performance.

I have had a lot of problems with single-ended connections but based on the mood here I must be the only one... But again unless I have a specific need I have never really paid attention to whether my home gear is differential or not.