"Amps all sound the same" (more or less)

[Rana]

I used to be quite vocally militant about all competently designed amplifiers (feeding typical speaker loads) all sounding the same.....until I started buying speakers that would actually show up the differences!

 

[Sue Pertwee-Tyr]

I've explained several times, sometimes in great detail, why it can't possibly make a difference and Martin Clark also provided an excellent technical explanation of precisely the same a few months back...
IIRC my most detailed explanation, with diagrams etc, was on AOS....

Yeah, sorry Jez, never done AoS.

 

[Arkless Electronics]

Dynamic bias controls the bias voltage so the output transistors never switch off, I believe there were a few, Nelson Pass has a patent on one and Sansui and Toshiba produced amplifiers using the technique. If you do bias a 100 watt amplifier into class A for the first watt you will end up with a standing dissipation of around a hundred watts which will require a big heat sink and long life high temp capacitors and a severe attack of the vapours in the finance department.

.... 50W for a 100WPC stereo amp but one hell of a lot to get 1WPC in class A yes!
As you'll know there are many benefits to class A beyond being non switching and so not producing crossover distortion and hence many of these non switching topologies are by no means the holy grail of a cool running class A amp! Krell's sustained plateau bias is maybe one of the more successful attempts but still gets hot etc... just less so than permanent class A

 

[Arkless Electronics]

Yeah, sorry Jez, never done AoS.

No problem Steve. You're one of the good guys FWIW you've persuaded me I was actually talking bollox several times in off topic threads

 

[Columbo]

And there's the famous test in 1978 proving that they do sound the same if driven within their limits...

Does this apply as much or less to modern amps?

 

[Columbo]

Excuse me picking out parts of your post, but I could not stop myself from commenting.
The valve set distortion is a common and understandable reaction to why a well made set amp may sound good compare to a 'neutral' class a or a/b solid state job,
I build both valve and solid state single ended amps, and in my view, both can sound a real step up from all of my push pull solid state amps it's not distortion that's making the difference its less cancellation of the audio signal. Push pull operation, simple power supplies poorly matched transistors and valves all come into play to cancel the original signal.
Most amplifiers fall apart when the simple test tones, that most amps are measured with are replaced by a far more taxing music signal.
I'm not pro valve or solid state, I just like what i think sounds the best.
For me it single ended operation.
Slightly more second harmonic but much less of all the other issues that make some of these so called neutral amplifiers

Excuse my utter ignorance... what's a single ended operation amp?

 

[Columbo]

I used to be quite vocally militant about all competently designed amplifiers (feeding typical speaker loads) all sounding the same.....until I started buying speakers that would actually show up the differences!

What sort of level of speaker are we talking to begin hearing significant differences?

 

[russel]

No problem Steve. You're one of the good guys FWIW you've persuaded me I was actually talking bollox several times in off topic threads

Isn’t against internet rules to admit you have been talking rubbish?, I think the last person to do that was banned in perpetuity and that was in 1996.

 

[Arkless Electronics]

Isn’t against internet rules to admit you have been talking rubbish?, I think the last person to do that was banned in perpetuity and that was in 1996.

Quite correct yes... but I usually only talk shite about politics, and professional politicians talk much worse shite about politics than I do!
Just today Jeremy Corbyn was voted the best prime minister we never had. I've said this myself weeks back in off topic... so either great minds think alike or fools rarely differ

 

[Steve 57]

Excuse my utter ignorance... what's a single ended operation amp?

Single ended is where the signal is amplified through a single transistor or valve, the music waveform stays in one piece. All single ended amps are class a.

most amplifiers are push pull where the positive and negative parts of the signal ( think in and out of the speaker cone )are split with one device amplifying the positive and the other amplifying the negative. The idea is that both sides match, and the waveform retains its original shape, but alot of things have to be right for that to happen, the advantage is that you can get alot more watts out of a push pull pair,

in push pull there are different classes, a ab b c.. in simple terms its just different spacing of the two halves of the signal wider apart to get more volume.. if you can understand that you will start to understand the quality relationship to different classes of amplifier
Hope that makes sense.

 

[Arkless Electronics]

Single ended is where the signal is amplified through a single transistor or valve, the music waveform stays in one piece.

most amplifiers are push pull where the positive and negative parts of the signal ( think in and out of the speaker cone )are split with one device amplifying the positive and the other amplifying the negative. The idea is that both sides match, and the waveform retains its original shape, but alot of things have to be right for that to happen, the advantage is that you can get alot more watts out of a push pull pair,

in push pull there are different classes, a ab b c.. in simple terms its just different spacings of the two halves of the signal wider apart to get more volume.. if you can understand that you will start to understand the quality relationship to different classes of amplifier
Hope that makes sense.

Whoa there! In push pull class A it also "stays in one piece" but with less asymmetry.

Daft analogy time again.... Where's the Ladybird book illustrator when you need them... Imagine lumberjacks, who are OK, the smell of pine.. and heatsinks

Are you sitting comfortably boys and girl? Then I'll begin. (that's showing yer age if you remember that 'on't wireless)

A two man large bow saw.

Single ended class A is like one man on his own doing both the pushing and the pulling... with a spring in place of the other bloke.. There's no gap where no ones doing anything but he is probably stronger pulling than pushing and as he gets to the end of his reach... plus the spring is helping in one direction but hindering in the other...
Until the "start sawing call" goes out the spring is holding the saw steadily in the middle against the strength of the bloke on the other end. This work he's continually putting in is the wasted energy, the heat, of class A.

Class B would be two blokes, one on each end, but as it gets nearly to half way one just lets go.. in both directions... there's a glitch in the movement as one stops pushing and the other starts pulling... or vice versa...
Until the "start sawing call" goes out the bow saw is resting on the plank whilst they have a cuppa. No work is being done, no heat, but they have to run back to the saw, late, and then the movement is strong but jerky.

Class A/B like above but each helps the other one out by continuing to push until it gets a little beyond the half way point and only gradually reducing the push whilst the other takes up the pull.
They stay in post and with a good grip on the saw and a light pressure but only just breaking a sweat.....

Class A push pull they both push in one direction and pull in the other for the full movement of the saw and as one reaches a point where their reach limits the strength they can put into it the other is reaching a point where they can pull at their strongest and this equalises the total into a constant smooth motion twixt saw and wood...
They both pull or push full strength against the other to keep saw still in the middle until "start sawing call" goes out but each is only using half the strength of the bloke working against the spring in the single ended case. Still lots of energy being wasted as heat but rather more efficient than pulling against the spring.

Mismatched output transistors can even be brought in to this as being one geezer goes to the gym more often than the other

 

[Rana]

What sort of level of speaker are we talking to begin hearing significant differences?

Hi Colombo, for me it started after I bought Decware ERR speakers a few year's back (I was firmly in the "choose your system around your speaker" camp in those days, and still am to a major extent), they revealed very clear differences in detail and presentation in amps such that I went through numerous brands and types and became a real amp swapper for a while. I ultimately decided that juicy SS amps worked best for my ears despite my Decware speakers being designed for flea-powered valves and that I listened at low/moderate volumes. I have since settled on Linx and AudioNet amps powering (even more) revealing MC Audio RL21s in both my systems.

 

[torstoi]

Sums it up quite well & after a long journey I have finally landed in more or less the exact same camp.
If with different hardware, but also juicy solid state as the momentary status quo.

And my take on the question..if all amps sound the same to you,
then your prob is in the speakers.

If you're lucky..

 

[mandryka]

Dynamic bias controls the bias voltage so the output transistors never switch off, I believe there were a few, Nelson Pass has a patent on one and Sansui and Toshiba produced amplifiers using the technique. If you do bias a 100 watt amplifier into class A for the first watt you will end up with a standing dissipation of around a hundred watts which will require a big heat sink and long life high temp capacitors and a severe attack of the vapours in the finance department.

Thanks. I found an internet discussion which said that Denon POA 6600 and 4400 implement dynamic bias - there are a couple for sale on eBay at the moment in fact, but I couldn’t see any comments on the sound or the quality of the amps themselves.

Did you make your own current dumping amps, or have you found a Quad amp you like? Or does someone else commercially use current dumping?

 

[gints]

What about even harmonics? Single ended class A has them and push pull does not. That is correct?

 

[Steve 57]

What about even harmonics? Single ended class A has them and push pull does not. That is correct?

Yes that correct, how much of a deal that is depends on a few things, single ended amps 2nd harmonic rises with output.
My speakers are very efficient (if commercial would be around 103db the pair) , normal volume would be running much less than a watt, as I turn it up to max volume there is no audible difference until I'm overdriving the amplifier.
There are good and bad examples in all amplifier appologies.

Jez, I think your comments echoed what I tried to explain in a simple as way as I could.
Your 1st paragraph was in my view not quite right ( or saying the same as me differently ? ) as the signal is still being split through 2 amplifying devices, in theory if these where perfectly matched the signal would remain completely as it was input.
The reality is they are generally not closely matched and and the result is some cancellation to the signal and asymmetry.
Cheers..

 

[PhilDick]

Is this dynamic biasing the same as what Arcam are doing in their Glass G amps to achieve Class A like operation?

The following is a quote from their doc about the A49:

The A49’s output stage includes a proprietary error correction circuit that modulates the modest standing currents in the
output stage and ensures a near-constant output impedance for peak currents of up to about +/- 4 amps, corresponding to well over
50W in to 8 ohms. The A49 thus behaves exactly like a classical class A amplifier up to this power level in terms of performance
but without the heat penalty.

https://www.arcam.co.uk/ugc/tor/p49/Technical Notes/The_technologies_behind_the_A49_101014_A.pdf

 

[Arkless Electronics]

Yes that correct, how much of a deal that is depends on a few things, single ended amps 2nd harmonic rises with output.
My speakers are very efficient (if commercial would be around 103db the pair) , normal volume would be running much less than a watt, as I turn it up to max volume there is no audible difference until I'm overdriving the amplifier.
There are good and bad examples in all amplifier appologies.

Jez, I think your comments echoed what I tried to explain in a simple as way as I could.
Your 1st paragraph was in my view not quite right ( or saying the same as me differently ? ) as the signal is still being split through 2 amplifying devices, in theory if these where perfectly matched the signal would remain completely as it was input.
The reality is they are generally not closely matched and and the result is some cancellation to the signal and asymmetry.
Cheers..

Class A push pull will still be more symmetrical than single ended even if the devices are not well matched though....
The very fact that they are in class A will tend to mean much better matching in practice than is usual for class A/B anyway as the devices never switch off and so there can be no mismatch of switch on voltage and slope around the switch on region.

2nd harmonic distortion will usually be MUCH more prevalent than any other and in a single ended output stage with say 3% distortion it would probably come down to say 0.5% or less simply by making the output stage push pull and without any other actions being taken! This is of course due to the asymmetry of a single ended class A stage, and in particular, in my "laymen s guide to amp classes" above, the fact the guy is pulling against the spring (as well as sawing the wood) in one direction but being helped by the spring in the other.

 

[Arkless Electronics]

Is this dynamic biasing the same as what Arcam are doing in their Glass G amps to achieve Class A like operation?

The following is a quote from their doc about the A49:

The A49’s output stage includes a proprietary error correction circuit that modulates the modest standing currents in the
output stage and ensures a near-constant output impedance for peak currents of up to about +/- 4 amps, corresponding to well over
50W in to 8 ohms. The A49 thus behaves exactly like a classical class A amplifier up to this power level in terms of performance
but without the heat penalty.

https://www.arcam.co.uk/ugc/tor/p49/Technical Notes/The_technologies_behind_the_A49_101014_A.pdf

All quite different. rather beyond the scope of the thread to go into detail on these ideas I'm afraid. It looks like a very nice amplifier BTW but unfortunately for Arcam the LM4702 IC (virtually like a gain clone power amp but powers external output transistors to get much higher powers) has bee obsoleted after only about 12-15 years of production...
The continuous takeovers and conglomerations of semiconductor companies is a terrible thing as in this case when Texas Instruments (TI) bought out National Semiconductors (NS) NS had recently brought out about the best performing and widest range of IC's for hi fi amplifiers ever seen and TI immediately made maybe 2/3rd's of the range obsolete!

 

[Jim Audiomisc]

I'm not for one moment suggesting that the differences are due to competence of design.... although that does vary considerably as well!
Unfortunately the reasons for the differences are completely and utterly unknown to science. We can speculate and speculate but just when we think we have hit the nail on the head we can often find a unit that breaks most of the "rules" we think are important and yet sounds great....

I'd suggest that the problem here is trying to consider the 'best' choice of amp *without considering the choice of speakers, etc*.

Thus an amp which is a perfect textbook example of "wire with gain" would sound worse than some other am whose deviations *suited* the choice of speakers, etc. e.g. having an impedance interaction due to non-zero output impedance that combined with the speaker impedance to give a changed response that the user prefer.

The amp designer can't win this battle... unless they also sell you the speakers and decide your room and taste in music. So you have to take these things into account. I'd suggest this is well known the 'science', but people keep persisting in not taking it into account and expect a "one amp to rule them all".

Works for gross problems, but not for interaction dependent ones.