John Curl's Blowtorch preamplifier part II

[SY]

Originally Posted by MikeBettinger

I don't have the bandwidth right now to put much time into this but think about how a very complex piece of music would appear in the compression discussion. And then think about what is involved in a piece of electronics that would allow it to keep all of the aspects of that recording sorted out, pure and clean sounding without the more dynamic aspects of the music stepping on the more subtle details... The signal still looks compressed in Jan's snapshots, but one system might allow you to hear a pin drop clearly with all hell breaking loose around it while another would turn it into sonic mush. Which measurement would point this out?

Before you consider "which measurement," you have to establish that the phenomenon is real (from the standpoint of the electronics chain- transducers are a different pot of piscines) and under what circumstances it happens, if any. That's the only way you can rationally propose a measurement that correlates to audibility. Otherwise, you can spend all day coming up with more and more exotic measurements which give no particular insight into actual sound quality but generate lots of pretty graphs.

[QSerraTico_Tico]

Originally Posted by RNMarsh

General rule for DIY'ers: A poor psrr amp can be offset by a really good, low noise, low z regulated power supply. The worse the power supply, the greater the amp needs a higher psrr. Usually with small signals the regulated power supply is common. But in power amps, for example, it isnt as common (rare?) to have low noise, regulated voltage and the need for higher psrr is more important.

nonsense

[SY]

Not nonsense, but IMO, backward. A high PSRR design is far less critical of its power supply (no "brute force" solution needed).

[RNMarsh]

Originally Posted by SY

Not nonsense, but IMO, backward. A high PSRR design is far less critical of its power supply (no "brute force" solution needed).

Agreed.

Also a low psrr would be more susceptable to the power supply quality. Its those words again :-)

-RNM

[RNMarsh]

Originally Posted by QSerraTico_Tico

nonsense

explain.

[john curl]

Richard is right. Both Charles Hansen and I use very low power supply rejection topologies. We have to make up for this by designing custom, super low noise/distortion power supply buffers to take the nominal DC to a really low noise DC.

[RNMarsh]

Originally Posted by MikeBettinger

And then think about what is involved in a piece of electronics that would allow it to keep all of the aspects of that recording sorted out, pure and clean sounding without the more dynamic aspects of the music stepping on the more subtle details... The signal still looks compressed in Jan's snapshots, but one system might allow you to hear a pin drop clearly with all hell breaking loose around it while another would turn it into sonic mush. Which measurement would point this out?

Mike

Real music has that quality. At Carnegy Hall, you can clearly hear a single tiny bell being struck from the furthest balcony while the orch plays. Its quit amazing to hear this but HiFi needs to be as good [without being recorded with a 'spot' mic on the bell.]

[Joshua_G]

Originally Posted by RNMarsh

General rule for DIY'ers: A poor psrr amp can be offset by a really good, low noise, low z regulated power supply. The worse the power supply, the greater the amp needs a higher psrr. Usually with small signals the regulated power supply is common. But in power amps, for example, it isnt as common (rare?) to have low noise, regulated voltage and the need for higher psrr is more important.

Indeed.
It seems that for small signal amps, it is easier and less costly to make a power supply, regulated, very quiet and with low Z out, than to make an exceptionally good amp with very good PSRR.
This is the path that John Curl and some others chose.

[Joshua_G]

Originally Posted by SY

Before you consider "which measurement," you have to establish that the phenomenon is real (from the standpoint of the electronics chain- transducers are a different pot of piscines) and under what circumstances it happens, if any. That's the only way you can rationally propose a measurement that correlates to audibility. Otherwise, you can spend all day coming up with more and more exotic measurements which give no particular insight into actual sound quality but generate lots of pretty graphs.

It may be misleading to ignore widely testified phenomena, only on the ground that presently we don't see the engineering causes that may lie behind those phenomena.

[MikeBettinger]

Originally Posted by SY

Before you consider "which measurement," you have to establish that the phenomenon is real (from the standpoint of the electronics chain- transducers are a different pot of piscines) and under what circumstances it happens, if any. That's the only way you can rationally propose a measurement that correlates to audibility. Otherwise, you can spend all day coming up with more and more exotic measurements which give no particular insight into actual sound quality but generate lots of pretty graphs.

What I was highlighting is a components ability to sort out the subilties in the music regardless of the complexities or dynamics. This is what allows a system to sound musical, or to my ear right. We've all heard components that sound great, for example, on simple music, or ones that need to be turned up to clear up, or still others that breeze through the most complex music with ease. Why is this?

Franks comments touched on this ability and I felt they warranted more discussion. Especially if the underlying focus here is to understand what differentiates quality audio after the fundamental quality engineering practices are put into a design.

From a measurement standpoint I was more or less just pondering... Wondering what sort of test would sort this out.

Mike