John Curl's Blowtorch preamplifier part III

Status
Not open for further replies.
Jump to Latest
#12,825
RNMarsh said:
I asked the designer of the BenchMark DAC what was major problem for affecting the sound and he said it was the analog portion of the circuitry.
Click to expand...

I think Benchmark considers the clocks, clock regeneration, and jitter reduction functions all to be part of analog. In ESS Sabre dacs the power for that circuitry inside the dac chips is also considered analog, such as VCCA, with the 'A' referring to analog. In addition, there is AVCC, the analog output reference. In other words, it isn't just the output stage.
 
Last edited:
#12,826
Wow, so Benchmark considers clocks, etc as part of the ANALOG portion of the digital conversion, this makes sense. This is where I am at a loss. I can and have designed first class Class A, all fet, I to V converters, but I don't know how to improve the clocks, etc to any great extent. I already have a Sabre 9038 evaluation board, BUT I have to accept their selection of clocks, etc, and this does not seem good enough to take on the big boys of digital out there.
 
#12,828
john curl said:
T, I am not going to argue with you or Richard about the advantages of digital. Also, IF I had to buy a quality loudspeaker today, I would probably buy a JBL similar to Richard's, but I am a 'victim' of being given a speaker too expensive to refuse, yet not always as happy with it, as I might wish to be. My suspicion is the TIME response of the expensive speaker. PMA disputes me on this, so it is him that I am disputing, not you or Richard.
Now, I personally think that complementary jfets (preferably Toshiba) are the ONLY way to go in designing amps, preamps, D-A analog electronics, etc, so should we discuss ONLY designs with these devices dominant? '-)
Click to expand...

I'm sorry but that's PMA and me. I know you're hearing something, but it isn't sourced from what you think (time alignment).
 
#12,830
RE time alignment. I was just reading up on microphone calibration and ran across this tidbit (B&K Free Field Reciprocity
It should be added that the software used with the system accounts for the position of the
acoustic center of the microphones. The center position is a function of frequency. For low
frequencies and for the axial sound incidence, the center for LS1 microphones is about 9 mm in
front of the diaphragm. For increasing frequency it moves closer to the diaphragm and is at 8 – 10
kHz at the diaphragm itself. It continues to move with frequency and is a few millimeters behind
the diaphragm at higher frequencies. For LS2 microphones the corresponding numbers are 4.5
mm and 20 – 22 kHz. Data for the center position is implemented in the software, but the
positions are also estimated by the system itself, if measurements are made at three or more
distances. For reasons of transparency, the software stores measurement files, extended
measurement files, acoustic center files and result files in linear and logarithmic frequency steps.
The linear frequency steps are equal to the measurement steps. Results for logarithmic steps are
determined by linear interpolation with a resolution of 1/12-octave
Click to expand...

The shift of the acoustic center should not be a surprise but its not to be completely ignored and seems to relate to the diameter of the microphones. LS1 being 1" and LS2 being 1/2". One more correction to dial in when testing tweeters. And a source of a small group delay. It would be interesting to measure those parameters for recording microphones and see how they compare to speakers. They would also apply to Mangers measurements.
 
#12,831
Markw4 said:
I think Benchmark considers the clocks, clock regeneration, and jitter reduction functions all to be part of analog. In ESS Sabre dacs the power for that circuitry inside the dac chips is also considered analog, such as VCCA, with the 'A' referring to analog. In addition, there is AVCC, the analog output reference. In other words, it isn't just the output stage.
Click to expand...


I remember a discussion 28 years ago with Bob Stuart when I presented that argument and he was very clear that, whilst the analog side had been an issue in the 80s that, in his mind all those issues were solved. He did go on to win a lot of awards with his digital products. Of course he is making a lot more money selling snakeoil licensing now. Clever guy!
 
#12,832
Markw4 said:
I think Benchmark considers the clocks, clock regeneration, and jitter reduction functions all to be part of analog. In ESS Sabre dacs the power for that circuitry inside the dac chips is also considered analog, such as VCCA, with the 'A' referring to analog. In addition, there is AVCC, the analog output reference. In other words, it isn't just the output stage.
Click to expand...
Yes. And analog+digital power supplies.
That is exactly what some people improve in their Behringer DCX2496, plus the output stage, to transform a very cheap device ... in a 'high-end'* one ?

*Smiley here.
 
#12,835
john curl said:
Wow, so Benchmark considers clocks, etc as part of the ANALOG portion of the digital conversion, this makes sense. This is where I am at a loss. I can and have designed first class Class A, all fet, I to V converters, but I don't know how to improve the clocks, etc to any great extent. I already have a Sabre 9038 evaluation board, BUT I have to accept their selection of clocks, etc, and this does not seem good enough to take on the big boys of digital out there.
Click to expand...
Jocko has selection of crystals/oscillators (the first step to good digital sound) down pat, I can hunt down his contact details if you like.


Dan.
 
Last edited:
#12,840
billshurv said:
Historically vinyl formulations were different between plants and a guarded secret, so unlikely you will find that out. I'd call around hipster new boutique plants and ask where they get their pellets from then call them.
The real question is why?
Click to expand...
I find that different vinyl and especially bakelite formulations in themselves sound different.....I am interested to know the impurities in commercial carbon black as used in vinyl record production.



john curl said:
Thanks Max, I already have talked to Jocko about oscillators. This is a problem, because he has to hand select them from many. Not the best approach for production.
Click to expand...
Yes, I understand that he goes to quite some trouble to select the good ones by his criterion......and yes unsuitable for production.
It would be worthwhile to get a preferred oscillator from him and try it on your evaluation board or other designs to get an understanding of oscillator signatures.
There is other method that works especially well, PM me.

Dan.
 
Status
Not open for further replies.
Top Bottom