All power amps “good ones” sound the same.

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

pucho812

Well-known member
Joined
Oct 4, 2004
Messages
14,830
Location
third stone from the sun
According to things I am reading today
All amplifiers “ the good ones”, all sound the same.
That their purpose is to be the straight wire with gain.
While I do like and agree with  the last part about wire with gain, I am wondering what constitutes “the good ones” and how one can make such a claim? 
 
pucho812 said:
According to things I am reading today
All amplifiers “ the good ones”, all sound the same.
That their purpose is to be the straight wire with gain.
While I do like and agree with  the last part about wire with gain, I am wondering what constitutes “the good ones” and how one can make such a claim?
It depends very much what category of amp. For years I used to think that nothing could beat a Crown 5002VZ. It used to be in the heavyweight category, until the norm became 10-20kW in a 2U box.
There are audible differences in this category. There's bound to be, when the box can put out 10+kW whereas it takes its power from a 117V 25A outlet. Every designer has their own compromise as to solving this seemingly impossible equation.
But I reckon you're not talking about this category.
Typical amps used in studios use power supplies that allow continuous operation at max power. One of the big points of difference is how they handle complex loads.
Years ago there was a comparison of  power amps used in conjunction with real speakers.
One of these speakers was so hard to drive that only one amp could drive it without restraint, the Crown PSA2.
A Sony amp, that had an amazing rise time of 0.35us (yes, 1MHz BW!) could only deliver about one third of its max power on them. Some amps failed miserably at one-tenth their max power!
Although the Sony was unfaultable on typical tests, done with a resistive dummy load, guess which one sounded best on these speakers!
Even the usual capacitive load test could not hint to this difference.
The big difference was that he Sony, as nearly all the others used the usual reentrant protection that samples the emitter currents and adds it with a dose of Vce. The Crown was the 1st commercial amp to use the analog modeller that simulates thermal dissipation and secondary breakdown.
Most class-D (switch-mode) amps are capable of driving very complex loads without undue distortion, but the intrinsic performance is not always excellent. The usual tests would allow assessing performance adequately.
Now, another category is audiophools, who think perfection lies in a Single Ended 211 in class A, with 0.5% THD at half-power.
 
well category would be  average amplifier in a studio driving  the studio speakers.

The comment comes from discussion in what is "high-fidelity"  and being told that a "good amplifier" does not have a sound.  To which I would disagree, and while ultimately we want as close to an original sonic just louder (wire with gain) in an amplifier we do not have that. That ultimately in practice the same everything(settings, volume, speakers and wires) but switching out the amps to different ones will sounds different.  That things like class A amps have more character and are not efficient but class D amps which are extremely efficient lack character.

and that is when I was met with all "good amps sound the same..."
I was then told how if an Amp colors the sonic, it is not "high-fidelity" to which I would say fidelity is in the eye of the beholder.  ;D
 
abbey road d enfer said:
Now, another category is audiophools, who think perfection lies in a Single Ended 211 in class A, with 0.5% THD at half-power.
I'll defend the audiophools.

But who can hear 0.5% of mostly second harmonic distortion? I know I can't unless I'm listening to a sine wave. Assuming the amp has low enough output impedance and reasonably flat frequency response it'll be good enough. I would say when listening to music at home or in a studio any amp that's good enough will sound the same.
 
I'm not a power amp guy but I would am highly skeptical that power amps really sound that different at all as long as they have vaguely the same specs. The reasons is simple: the speaker and room have a MUCH greater impact on sound because speakers are highly imperfect devices and reflections in the room cause all sorts of things to happen to the frequency response just depending on where your standing.
 
pucho812 said:
According to things I am reading today
All amplifiers “ the good ones”, all sound the same.
That their purpose is to be the straight wire with gain.
While I do like and agree with  the last part about wire with gain, I am wondering what constitutes “the good ones” and how one can make such a claim?

This is a pretty old topic, and back several decades ago there were easily audible differences between better and cheaper amps.

It is generally a poor use of time to argue with people about what they claim to hear on the internets. It can't be proved without extensive testing (double-blind with statistical significance, yadda yadda).

Audio amplifiers have benefitted from modern technology to the point where they are pretty much canned off the shelf solutions, like mic preamps are now. I have been out of the trenches for a couple decades but suspect the value class D amps offered by several major players probably use the same chip sets inside.

Just like people still believe there are significant audible differences with speaker wire, amplifiers have almost become as neutral/transparent as wire.

When pondering such questions (like do amps sound the same?) we need to be alert to influences from the entire audio path. When listening this not only includes the loudspeakers but the room, not to mention source material.

It used to make me crazy back in the 80s when I would submit an extremely accurate phono preamp to a magazine reviewers who would then listen to it through his home system comprising a phono cartridge and loudspeakers that were easily orders of magnitude less accurate, not to mention who knows what kind of listening room acoustics. I even received night and day good/bad reviews when the exact same preamp was submitted to two different magazine reviewers.

The complete amplifiers sound alike thesis says that competent modern amplifiers will all sound the same when operated in their linear region. i.e. not current limited by weird speaker loads, clipped, or slew limited. Over the decades we have seen some problematic passive loudspeaker crossovers where some speaker designer over-tweaked the crossover hoping to improve the speaker performance, ignoring the problematic load it presented to the amplifier.

JR

 
 
B&W speakers are hard to drive. I’m not a big fan of B&W but many are. I have heard one and maybe two amps that can make those speakers sound good to my ear.
 
there has been a disturbing trend as of late that involves the elimination of vacuum tubes and large transformers from the power amp.

how this started, or who started it, we need to know.

that way proper law enforcement agencies can be contacted so that due process can take place which will hopefully result in jail time for the main offenders.
 
I'm surprised transformers don't get used more in transistor amps. I'm guessing between voltage limits and class D amps, they never had a chance.
 
CJ said:
there has been a disturbing trend as of late that involves the elimination of vacuum tubes and large transformers from the power amp.

how this started, or who started it, we need to know.

that way proper law enforcement agencies can be contacted so that due process can take place which will hopefully result in jail time for the main offenders.
eliminating costs that don't improve the sound, makes a better world for all of us.

JR
 
squarewave said:
I'm surprised transformers don't get used more in transistor amps. I'm guessing between voltage limits and class D amps, they never had a chance.
The last time I saw transformers used inside a transistor amp was back in the 60s... (a knight kit IIRC). As I recall the design looked like it was lifted from an older tube design but with transistors dropped in place of the tubes.

Transformers are expensive and not cost effective when solid state devices can perform the manipulations much more cheaply (and better).

JR 
 
JohnRoberts said:
Transformers are expensive and not cost effective when solid state devices can perform the manipulations much more cheaply (and better).
Are we interested in mass producing cheap stuff here? If I want a cheap amp that does the job, I'll just get some boring chip amp class D thing.

And it's not obvious to me that a simple transistor amp is "better" than a transistor transformer amp.

A transformer gives you builtin speaker protection. Yeah the transformer might get torched but it's more likely that the speakers are more (possibly much more) expensive than the transformer.

But perhaps more important the circuit can be much simpler. One of the primary challenges with power amps is trying to precisely control really high currents. Driving 4 ohm is practically a short. But with a transformer you're circuit only has to handle 50 ohms or whatever the primary is which is a LOT less stress on the circuit.

It could be smaller because you're not dissipating as much heat since it's high voltage vs high current. So your heat sink doesn't have to be huge.

It's also less likely to burst into flames and burn your house down which is usually not desirable.

With good availability of high voltage supplies and transistors, a transistor transformer amp is very doable at higher wattage. A 24V / 8 ohms = 3A * 24V = 36W RMS. So to do the same with a 50 ohm transistor transformer would be 60V / 50 ohms = 1.2A * 60 = 36W RMS. With 100V you can make a 100W amp.

Are there problems with this? Does lower inductance make for bad low frequency response? Maybe a mild negative impedance converter could solve that?
 
Actually the famous Mac MC-2300 transistor  amp has transformers, as well as a very high end high powered Sansui stereo amp whose model number escapes me, but it is only an auto transformer for Z taps.
 
nearly all  high power McIntosh power amplifiers are fitted with output autoformers, including the MC-2300;
dates back to circa 1967 with the introduction of the MC-2100 family.

legendary Sansui is the BA-5000, autoformer fitted.

though rather dated, Altec power amplifiers including the 1594 family used both interstage and output transformers.

a more modern application of output iron is the Bogen HTA-250a, easily dismissed, but imho, sounds excellent.
 

Attachments

  • xfmr.jpg
    xfmr.jpg
    128.2 KB · Views: 15
squarewave said:
Are there problems with this? Does lower inductance make for bad low frequency response? Maybe a mild negative impedance converter could solve that?
There is a limit to this scheme, when the necessary rail voltage become excessive.
A 500W 5ohm amplifier reqires +/-70V rails, so the output devices must be rated at min. 140V, which is OK. Many types are available.
Following your thinking, a transformer-based amp of 500W into 50 ohms would require +/-210V rails, and output devices rated at 420V, an entirely different sector of the semi-conductor area. Indeed, there are HV transistors, but they are generally optimized for different parameters than audio.
McIntosh use transformers, but the rails are not terribly different than conventional amps.
Also the transformers are used in a specific configuration where they are under 100% NFB, just to make sure performance is perfectly under control and does not depend too much on transformer variations.
 
pucho812 said:
That things like class A amps have more character and are not efficient but class D amps which are extremely efficient lack character.

and that is when I was met with all "good amps sound the same..."
I was then told how if an Amp colors the sonic, it is not "high-fidelity" to which I would say fidelity is in the eye of the beholder.  ;D

Fidelity is clear and simple, the definition of fidelity is being "faithful" to the sound source, so distortion is not fidelity. There has been several tests which show that people can't tell the difference between two amplifiers in a double blind test as long as the amps are not clipping or do not have something inherently wrong with them, a test that comes to mind is this http://matrixhifi.com/ENG_contenedor_ppec.htm in which they double blind tested a cheap Behringer amp vs a more expensive one, statistically, people couldn't tell the difference.

What does "character" even mean? its one of those terms audiphools love to use like "forward sounding" or "resolving".
 
user 37518 said:
Fidelity is clear and simple, the definition of fidelity is being "faithful" to the sound source, so distortion is not fidelity. There has been several tests which show that people can't tell the difference between two amplifiers in a double blind test as long as the amps are not clipping or do not have something inherently wrong with them, a test that comes to mind is this http://matrixhifi.com/ENG_contenedor_ppec.htm in which they double blind tested a cheap Behringer amp vs a more expensive one, statistically, people couldn't tell the difference.

What does "character" even mean? its one of those terms audiphools love to use like "forward sounding" or "resolving".
+1

Back when I was working at Peavey I would dread returning from some business trip and learning about some listening test my boss (the one with his name on all the buildings) had requested. These informal listening tests, when I was out of town would invariably be poorly controlled, making any results untrustworthy. I would sometimes have to do damage control.

Speaking of amplifier listening tests and I did plenty, I observed that uncritical listeners would often prefer the sound of an amplifier allowed to clip, compared to the same amplifier with a clean, fast, clip limiter. What they were hearing was the clipped amplifier playing louder, the old Hifi salesman's dirty trick #1, louder sounds better to customers who don't realize what is going on.

Sadly allowing the amplifiers to clip stressed the amplifiers, and more importantly stressed the loudspeakers more, part of Peavey's successful formula was protecting inexperienced customers from themselves.

JR
 
I could easily design a test where you couldn’t tell the difference between two amplifiers. I could also design a test where you easily could.
 
Gold said:
I could easily design a test where you couldn’t tell the difference between two amplifiers. I could also design a test where you easily could.
I am not sure of your point(s), and what "difference" your tests would reveal.

There are comparison strategies that are more revealing of small differences, like null tests. I recall one amplifier test I did last century where I loaded down both amplifiers I was comparing with typical speaker loads, and played them at full volume (loud) inside a sound proof room. Then I bridged a third loudspeaker across the the two amplifier hots playing in a quiet room. If the two amps are truly identical the null speaker will be dead silent, but even minor differences in either amplifier's phase response, etc will express as audio signal in the null output. Note: this test doesn't tell you which one is different only that they are not identical.

It is easy enough to not normalize playback volume to make amplifiers sound different, even when they aren't different (refer to dirty hifi trick #1).

JR 
 
abbey road d enfer said:
McIntosh use transformers, but the rails are not terribly different than conventional amps.
Also the transformers are used in a specific configuration where they are under 100% NFB, just to make sure performance is perfectly under control and does not depend too much on transformer variations.
not aware of any Mac output autoformer wrapped in feedback, not the 2300 nor the 2100.
 

Attachments

  • 2300.jpg
    2300.jpg
    167.7 KB · Views: 11

Latest posts

Back
Top