Sound Quality Vs. Measurements

Status
Not open for further replies.
Pano, you only have 1/2 the story. Class A is not necessarily the most thermally stable operating condition. It can be shown that heavy Class AB-1 is as good or better, if you go back to the root equations. However, thermal stability, both transient and long term is really important, and real designers make sure that the THERMAL CAPACITANCE is very high by close mounting the output devices on good insulators (if necessary) properly torqued down and used with thermal grease on a substantial heatsink . This removes most of the short term gain changes due to temperature.
High input stage operating current, using only a part of the class A portion can also be helpful.

Interesting John , could you elaborate more on your previous comment about toroid transformer vs a super-quality EI or R core transformer.
 
Sure, SY. But there might be in typical tube amps.

Sure, like a cat does not need to run before jumping on a fence, tube amp does not need to have such low THD and DF to sound nice. However, my Pyramids have lots of feedbacks, stabilized voltages, and they show comparable with transistor amps DF and THD. I designed them to drive high-end line arrays, but as far as I know all prototypes I made drive Maggies and speakers with crossovers meant to be driven by SS amps.

I.e. you can design and build tube amps with low THD and high DF, but nobody actually needs it. Especially those who drive full-range high efficiency speakers. That's why you don't see so many tube amps with extremely good parameters that are not actually needed. Especially, when they lead to horrible transients on overload, in all amps, no matter made they on diamonds, or on steam engines.
 
Pano, you only have 1/2 the story. Class A is not necessarily the most thermally stable operating condition. It can be shown that heavy Class AB-1 is as good or better, if you go back to the root equations.

John; it is true only for biased in class A complementary followers. My amps that have bootstrapped source follower loaded on counter-modulated current source mean totally different story: power dissipation varies with variation of load impedance only, not with the signal.
 
Disabled Account
Joined 2004
This is a DIY site, right? People here do their own building because they are not happy with commercial products, in one way or another. So, if engineers do amps by the numbers or by quality of sound it's really something that does not interest me. I don't buy amps from other people - thanks God I'm free of that hassle. They must be doing something wrong anyway because I'm not an engineer and the amps that I build generally sound better to me. The only thing that I miss is the quality of the chassis. In that area I'm miles behind and that's what I most often admire - a well engineered chassis.

No wonder that people interested in these debates either build for other people or buy commercial amps.
 
I've always wondered if the real problem and perceived variability in amps performance assuming the usual very low THD is really in how the speakers perform with different amps, and whether it's for reasons not usually considered during amp design. Like, how do amps cope with back emf from speakers? Does anyone measure output distortion with 1kHz applied to the input with bursts of 100Hz also applied to the output (not the input) to muddle things up a bit and be a bit more like music with back emf from the speaker.

How much back emf is there typically, I don't know how to measure it, but I bet it's time delayed and has a non-flat frequency response. Even worse, passive crossovers behave totally different to normal with back emf applied, i.e. in reverse, creating sharp resonances as the speaker sees a parallel high Q LC network due to the amp looking much like a short circuit at the output. I know I've kept 20 ohm approx resistors across each drive unit in my much modified 3-way speakers because I can hear these muddled high Q effects reduced dramatically, with transparency much increased.
 
I've always wondered if the real problem and perceived variability in amps performance assuming the usual very low THD is really in how the speakers perform with different amps, and whether it's for reasons not usually considered during amp design. Like, how do amps cope with back emf from speakers? Does anyone measure output distortion with 1kHz applied to the input with bursts of 100Hz also applied to the output (not the input) to muddle things up a bit and be a bit more like music with back emf from the speaker.

How much back emf is there typically, I don't know how to measure it, but I bet it's time delayed and has a non-flat frequency response. Even worse, passive crossovers behave totally different to normal with back emf applied, i.e. in reverse, creating sharp resonances as the speaker sees a parallel high Q LC network due to the amp looking much like a short circuit at the output. I know I've kept 20 ohm approx resistors across each drive unit in my much modified 3-way speakers because I can hear these muddled high Q effects reduced dramatically, with transparency much increased.

It is possible to check how linear is output resistance of the amp applying current to output. Say, 100W of sine wave through 8 Ohm resistor, when the amp itself is at idle. FFT plot is quite revealing.

Long time ago when our laboratory designed thick film 100W ICs I did something similar, bridging a couple of ICs.
 
Now you are thinking, sbrads. Matti Otala addressed this problem back in 1977. It is called IIM. I am relatively sure that my IIM is as low or lower than virtually any feedback controlled power amp, due to the inherently low open loop output impedance that I design into my power amps. This lowers the potential for IIM. I let the natural output impedance absorb the reverse EMF before the feedback loop has to do anything significant. It 'might' also be another reason why triode amps always have sounded better than pentode amps.
 
It 'might' also be another reason why triode amps always have sounded better than pentode amps.

I know that it is true, without any doubt. When I apply local feedback across pentode output stages the effect is like with triodes with very low output resistance. However, it requires more power from drivers, but pentode LTP drives nicely pentodes with parallel feedback by voltage across them.
 
sbrads
You don't have to wonder. It's ugly. That would suggest a self powered design would have the advantage. If we could get a good amp designer to hook up with a good speaker designer.... The ones I have listened to have never really showed off the advantage. Saw a nifty Meridian system today. Way out of my price, but at least they seem to be thinking.
 
Isn't it a strange coincidence that the point beyond which amplifiers start to behave unacceptably, apparently, is at the upper limit of human hearing? The same electronic components that are used for processing signals in the many megaherz or gigaherz ranges (i.e. nothing to do with audio) can only just be configured to reproduce a complex waveform whose components are all within a bandwidth of 20kHz, or so it seems. It's a good job that the physics of air, and the laws of biology mean that our ears evolved the way they did.

Or is it the case that even if our ears ran out at 10Hz, people would still manage to make a living telling us that our amplifiers are rubbish, and that they can hear huge differences between valves and transistors, and that circuits without feedback are so much better than those with?

A BC547 TO92 Transistor behaves well, and is used in circuits at audio frequencies and RF

An important development that has already had a commercial sale cycle namely companding,especially in the form of DBX Type 1 ( David Blackmer ) needs to be reconsidered as commonplace in audio systems. David had many splendid ideas and was most interested in extending the band of perceived aural frequency much higher :)
David E. Blackmer - Wikipedia, the free encyclopedia
and article "The World Beyond 20Khz" http://www.drtmastering.com/blackmer.htm


Cheers / Chris
 
Last edited:
Status
Not open for further replies.