John Curl's Blowtorch preamplifier part II

[jan.didden]

Originally Posted by ThorstenL

Hi,

Sorry to be persistent, but either it is:

"All amplifiers that show no appreciable difference between input and output signals also sound the same"

[snip]Ciao T

My definition of 'no appreciable difference between input and output signals' was: there is no audible difference between input and output signal. Logically, amplifiers with that attribute therefore sound the same. How could it be otherwise?

jan didden

[ThorstenL]

Jan,

Originally Posted by janneman

My definition of 'no appreciable difference between input and output signals' was: there is no audible difference between input and output signal.

How do we determine "no audible difference between input and output signal" for an amplifier? It is doable for cables, possibly even for preamp's and maybe DAC's, but amplifiers?

What is your "reference" for the "input signal" that allows audible comparison to allows to conclude "no audible difference" in a manner that is fair and sensible?

Originally Posted by janneman

Logically, amplifiers with that attribute therefore sound the same. How could it be otherwise?

Yes, they would be the same. However you have a major issue to explain how you can quantify said attributes?

Ciao T

[jan.didden]

Originally Posted by ThorstenL

[snip]
Yes, they would be the same. However you have a major issue to explain how you can quantify said attributes?

Ciao T

Yes that's a major pita. But, as I said, engineers are good in solving clearly defined problems . This one is like bootstrapping.

An idea: how about using a 'good' headphone, switching between input and output (suitably attenuated), if the source can handle the load? The headphone doesn't need to be ruler flat or something because you listen for differences. Even an amplifed headphone could work, as headphone amps are probably an order of magnitude more accurate than power amps.

Question: since we only listen to minute differences, we need a method to show minute differences, not necessarily show the differences exactly as they are. So, even a power amp could be used. Take the example that you have an amp that has 1% THD, and you want to know if that is audible. You take another power amp, with that same 1% THD, and you switch it between the input and the attenuated output of the DUT. How important is that 1% THD of the amp you use for listening? Is it important at all, or irrelevant?

What does the gathered intelligence out there feel about that?

jan didden

[Pano]

I've often used headphones thru an attenuation network and dummy load to listen for differences. It's not a bad technique. But the dummy load in my case was always purely resistive, so that's not ideal. And I do miss some stuff on headphones that I hear with speakers - spacial stuff, mostly.

[ThorstenL]

Jan,

Originally Posted by janneman

Yes that's a major pita. But, as I said, engineers are good in solving clearly defined problems . This one is like bootstrapping.

An idea: how about using a 'good' headphone, switching between input and output (suitably attenuated), if the source can handle the load?

Hold on, you presented your assertion as observed fact. I can think of a few possible solution, however this is not what I asked, what I asked is what test setup YOU used to establish "no audible difference between input and output" for "best amplifiers"?

I really need to know the experimental setup you used to establish your results to know if I should give credence to the claims you made.

Ciao T

[jan.didden]

Originally Posted by Pano

I've often used headphones thru an attenuation network and dummy load to listen for differences. It's not a bad technique. But the dummy load in my case was always purely resistive, so that's not ideal. And I do miss some stuff on headphones that I hear with speakers - spacial stuff, mostly.

Yes ideally you'd use a real-world load, a speaker or a dummy speaker load.
About the spacial things, do you think we could have the situation that two amps have a different spacial rendering, yet sound the same on the headphone?

jan didden

[Pano]

I tired to build a dummy load out of a speaker that I could not hear, but that didn't work. Not sure that a passive L,R,C network would be as good a dummy - but it should be better than just a resistor..

Yes, I think that two amps could have different spacial renderings and "I" wouldn't hear it on headphones. But that's me and my ears.

[SY]

There's one test I use which no loudspeaker has passed- but it consistently works on headphones. "For a Thousand Mothers" on Jethro Tull's Stand Up. The main section of the song (in E) ends, then a reprise in D begins. In the beginning of the reprise, there's a little drum and cymbal figure- with excellent headphones, you can make out a sound like a champagne cork in the second part of that figure, right as Clive Bunker moves from cymbals to tom. Never been able to hear that on loudspeakers.

[simon7000]

Ok,

My space is big enough I can load two amplifiers with loudspeakers and look at the difference. When I tried this after balancing the output level, several issues show up.

The gain of the amplifiers does change a bit as they heat up and cool down. Even using two channels of the same stereo amplifier, there are artifacts that show up. How do you judge these?

I can't imagine two different amplifiers not having even more artifacts than two of what should be the same.

That is how I got to the resistor vs the loudspeaker load test. The amplifiers I am familiar with do not drive resistors and loudspeakers the same. It is waaay to early for me to present the results of this bit of playing around.

[Pano]

Originally Posted by SY

you can make out a sound like a champagne cork in the second part of that figure, right as Clive Bunker moves from cymbals to tom.

I just hear a tiny "ticK" there. No cork.