John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Jump to Latest
Esperado

Esperado

Banned
Joined 2011
esperado.fr
RNMarsh said:
Which is better and why?
Click to expand...
Again, the question with response you know...;)
Guess it is the computer one ?
I let the others free to provide the technical response, but at least, one argument: it is not audio "hype" and expensive accordingly ;-)
dadod said:
Waly, you did not have courage to comment that amp (to many zeros and poles ) and now you are sarcastic again.
Click to expand...
Ask a viper to be vegetarian !
 
Last edited:
B

bcarso

diyAudio Member RIP
Joined 2005
billshurv said:
Outside the usual suspects on this forum not sure anyone is published decent measurements or even understanding what they should be!
Click to expand...
JA has gotten the notion from somewhere that the most demanding test for noise of an amplifier or preamp is to short the input and run at maximum gain. I have no idea where this comes from---the only situation I can envision would be where the input stage has a first resistor going to a summing node, so you get the voltage noise of the stage amplified as much as possible, and of course the resistor's thermal noise. But a more likely input is just the initial active device's control electrode---a bipolar's base, or a JFET's gate, or some paralleled combination of such.

If a bipolar, the base current noise is being ignored with such a grounded input. For JFETs the parallel or current noise is small.

With MM, or high-inductance MC (rare, and an example the Ortofon X5-MC) the parallel noise is important, as is the thermal noise of the cartridge damping resistor. So for phono preamps we need to assess this noise, as well as the series or voltage noise. It should be sufficient to measure with an open input and a shorted input, and determine what the parallel noise contribution is for a given cartridge, knowing the latter's inductance and resistance. The preamp would have to be stable with an open input.

Out of this we should be able to accurately estimate the overall signal-to-noise ratio of the system, including the thermal noise of the cartridge, the latter which one hopes will dominate. Of course surface noise adds a bunch, but I don't think excuses poor performance of the electronics.

Another approach would be to synthesize a cartridge of some roughly representative L and R, and to make life easier use a synthetic resistor for the bulk of the R, with much less than thermal noise.
 
gpapag

gpapag

Member
Joined 2002
Paid Member
You are welcome Brad.
IMO, these are the essential points

It (unfiltered white-noise) is essentially a test of slew-rate limiting, similar to the Holman square-wave test butmuch more difficult for a preamplifier to pass. In spite of our doubts as to its validity, I con-tinued to conduct the test with as many preamplifiers as I could to see if my subjective impressionscontinued to correlate with the test outcome.
…
Clipping Measurements: An Alternative to White Noise
Abbott Lahti reasoned that because the white-noise test was essentially a severe measure ofslew rate, or speed, we could effectively resolve the level necessary for a preamplifier to passthe white-noise test by simply measuring the relative clipping levels at 1 kHz, 10 kHz, and 50 kHz.This test would be quantitative, while the white-noise test is qualitative.
We observed (see column 6 of Table 1) that if the 50-kHz clipping level is at least 50% of the1-kHz level, the preamp sounds excellent. However, if the 50-kHz clipping level is 30% of the 1-kHz level or less, the unit sounds veiled. We therefore postulated that somewhere between theseextremes is the acceptable minimum slew rate. With all the units we have tested, the subjectivecorrelation has been 100%.
Click to expand...

George
 
billshurv

billshurv

Member
Joined 2014
Paid Member
I was just pondering this myself. We know that overload on ticks and pops is an issue, and most designers try to get as much overload margin in as possible. The clever ones also look carefully at overload recovery. With the fact that MC cartridges often have an ultrasonic resonant peak HF clipping could be the issue.

As JC measured all this back then I am sure he can tell me if I am on the right track or completely missed the point :)
 
john curl

john curl

Member
Joined 2003
Actually, I evaluate with what I hear. IF I had followed that ABX test, I would never have made the JC-80, Vendetta Research, CTC Blowtorch, or any of the Parasound products. Perhaps I should have stuck with medical electronics. '-)
It was kind of boring, but it was a pretty good living, better than audio. No, I am happy with my decision to stay in audio design. With ABX tests discouraging the competing designers, more for me!
 
3

31697B

Disabled Account
Joined 2012
1. Certainly, ABX and DBLT tests can be helpful. However, if no one ever did an abx or a dblt test ever for audio evaluation....... there will be plenty of wrong conclusions. But, I would not go so far as to say every single subjective evaluation will be wrong. Surely, there will also be plenty which are correct.

2. Take the LP.... we just have covered pages on it... but only maybe .01% of what there is to know about it has been brought up......the many cal/adjustments cart tips and all...... did every one of the changes and developments go thru abx or dblt? No. What then?



THx-RNMarsh
 
B

bcarso

diyAudio Member RIP
Joined 2005
In any event the white noise test is fascinating---although somewhat difficult to understand how that many preamps were so severely limited in slew rate and/or overload margin at high frequencies. It does lead me to want to stick with a design in which both SR and OL margin are ludicrously excessive by traditional measures.
 
Status
Not open for further replies.
Top Bottom