John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Jump to Latest
hhoyt

hhoyt

Member
Joined 2010
PMA

PMA

Member
Joined 2002
pmacura.cz/audiopage.html
Paid Member
Well, I went through Hirata's papers. His view is based on nature of musical signals as assymetric waveforms with zero DC component over period of time. He suggests several test waveforms that are assymetric, but positive and negative areas are equal (= zero DC spectral component). Then he tests several examples of non-linear transfer functions and examines how they affect spectrum of input test waveform. He finds most of "amplifier type" non-lineraities to affect lower frequencies of the original test spectrum and "loudspeaker type" non-linearity to affect higher frequencies of the original test spectrum. His papers bring interesting hints.

I would like to ask John Curl what where the results of his Hirata based tests, what did they reveal.
 
john curl

john curl

Member
Joined 2003
One 'unusual' result was with a standard 'op amp' type design, and attempting to add an AC balance control composed of a pot and a large capacitor.
The circuit was conventional: An improved model of the original JC-2 phono amp, composed of IC matched pair of n channel input devices, the second stage was a complementary quad of 2 beta matched npn and 2 matched pnp devices and a complementary jfet output. This design, even though using the best matching that was available, both Gm, beta, and thermally, still had some residual 2'nd harmonic. I thought that an AC balance control could be added that would ONLY effect the differential AC gain of the second stage, and balance out the distortion, leaving only 3'rd. The gain offset would be very small, with the primary gain provided by other resistors connected to the supply. A pot was added across the emitter terminals of the second stage pnp devices, with the wiper connected to a cap and then to the positive terminal. A small offset from center in the pot providing extra gain to one leg of the second stage, and slightly less gain to the other leg of the second stage.
In any case, from a harmonic, IM distortion point of view, it worked perfectly. HOWEVER, it changed the Hirata test considerably. So, either I could have lowest IM, or lowest Hirata, I had to take a choice. I can't remember whether having the AC gains balanced by exactly centering the pot made an improvement or whether removing the cap entirely from the wiper to the power supply was necessary, unfortunately. (It was 30 years ago, or so) but I concluded that it was a real problem, and I left out the AC control. Sorry, a schematic would make this easier to see what I was doing.
Still, it was interesting.
 
john curl

john curl

Member
Joined 2003
I would like to elaborate on my last input 'interesting' and why my reply was so short.
At first, I was going to say 'DA? No way!' or its equivalent, and then I realized that there WAS DA in the Tantalum cap that I used, and I was trapping myself from being OPEN to something, because I thought that I knew about DA, better than most, and it wasn't immediately obvious to me that this might be a cause. I caught myself doing what I often criticize in others, 'Concluding, without examination of the actual facts and process'. I then shortened my response to 'interesting'. Thanks again, PMA for bringing these points up.
 
Status
Not open for further replies.
Top Bottom