The simple example of a 4Q circuit will work well, essentially as shown, if the right fets are used. However, getting these fets is difficult, so we often cascode. Sometimes, we both cascode and buffer the output with a mosfet complementary follower.
Still, the 'basic' circuit has a lot of design merit, even though it is somewhat impractical.
It was first used in the JC-80, about 35 years ago, pretty much as shown.
Still, the 'basic' circuit has a lot of design merit, even though it is somewhat impractical.
It was first used in the JC-80, about 35 years ago, pretty much as shown.
Last edited:
The 20ps -200ps are the correct numbers for audibility... High freq to low, respect.
...
-Richard
please show the listening tests that found these numbers
I'm very certain you can't because you once again don't read, recall, or present context - those numbers are calculated as you have been told many times now
tube reference
There is a shorter softcover (and oddly more expensive!) book by the same authors, Introduction to Electron Tubes, Van Nostrand, 1969.
My experience with some of the super tubes is that they have huge device-to-device variations, probably owing to the difficulty of holding grid-cathode spacing tolerances. My guess is they wind the grid wires around a cathode with a coating, and then dissolve the coating, but that's just conjecture.
But with triodes that are more consistent, some of them are remarkably good over a reasonable operating range in holding mu constant, so if the anode is very lightly loaded the gain is constant and the distortion is low, for common-cathode stages. Of course the Miller multiplier will be high and thus the effective capacitance at the grid.
The best book I've seen, from the last days of really innovative tube design: Gewartowski and Watson, Principles of Electron Tubes, Van Nostrand, 1965. Both Bell Labs guys.
There is a shorter softcover (and oddly more expensive!) book by the same authors, Introduction to Electron Tubes, Van Nostrand, 1969.
My experience with some of the super tubes is that they have huge device-to-device variations, probably owing to the difficulty of holding grid-cathode spacing tolerances. My guess is they wind the grid wires around a cathode with a coating, and then dissolve the coating, but that's just conjecture.
But with triodes that are more consistent, some of them are remarkably good over a reasonable operating range in holding mu constant, so if the anode is very lightly loaded the gain is constant and the distortion is low, for common-cathode stages. Of course the Miller multiplier will be high and thus the effective capacitance at the grid.
I haven't seen this one before, it looks really good. A free source: http://frank.pocnet.net/other/docs/GewartowskiWatson1965PrinciplesOfElectronTubes.pdf
Frank's web site has a lot of other good stuff, too. http://www.tubedata.info/
Last edited:
Use what ever numbers please you. And, what authoritative source do you believe in? Or have you done your own expert listening tests?
View attachment jitter-5.pdf
I intend to do my own listening tests on a variety of jitter related issues and interfacing. That way, as I've done with distortion, I can set my own standards as to what I can hear as relevant. Then it wont matter what you or any other authority says is or isnt audible. I'll find out for myself, thank you.
But, I am also interested in the differences found between CD and HD down loaded files, in particular.
THx-RNMarsh
Last edited:
You've made it clear for a long time that you are "above the law" as to audibility tests. I don't get it talented savants in other fields love to demonstrate their skills, I remember a guy who would solicit large integers from two different members of the audience and immediately give the product.
Last edited:
The late Richard Heyser used to say: "If two people can hear some audio difference, then it is real." A far cry from the double blind fanatics who attack people who openly listen to differences. Trust me, Richard Heyser too, was hassled by the same double blinders up till he was on his death bed. Same bunch, just slightly different topics at the time.
Skills? What skills? I have no skills. Everyone knows that.
I know some great card tricks, though. Like the one where you pick a card and remember it... show it to another.... put it back in the deck and shuffle it a dozen times or more and I pick the card you chose. Is that a skill?
-RM
I know some great card tricks, though. Like the one where you pick a card and remember it... show it to another.... put it back in the deck and shuffle it a dozen times or more and I pick the card you chose. Is that a skill?
-RM
Last edited:
Pretty easy to hear differences, it's the "why" that brings in the messiness ...
The only real solution I've found is to eliminate every gremlin that one can think of, and when no differences can be heard that matter then one is in a good zone. And to make the differences obvious, use really "nasty" recordings where the inferior configuration is downright unpleasant to listen to, and the better reverses that perception.
OK participate in a refereed DBT where "you don't know".
EDIT - right now listening to the Italian JVC/RCA compilation of Ennio Morricone's spaghetti western music on LP, who needs CD or HD scary wonderful sound .
Last edited:
Maybe in the 20th century. (below) is a typical 21'st century setup - 7 !
Hubs with 2 or 4 ports apiece. I'm not even using them all.
I can plug everything USB I own into just one "enhanced" with no latency !
The "enhanced" - usb2.0 hubs with a older PC - before PCI-E ...
might just do 35-200mb/s. PCI-E bus USB can do the full 480mb/s !
12mb/s is the old PCI 1.0 (1995 PC's). Mice still default to this , even on
a super fast 2.0 interface.
I've tried every sample rate my PC toslink (or usb) will output .
An onboard 2013 PC audio codec will negotiate 44.1 -192Khz using
only <10% of the usb bandwidth ,on a cheap "pocket DAC".
OS
Attachments
If only for the negative tempco the Hitachi Laterals are Non'Pariel(sp?)
Happens they were truly designed for audio .... (as advertised)
edit: what the world needs is Magnetec 2sk213 .... TO220 1/2 amp 200V maybe even TO92
with attendant complement
Last edited:
This is really at the center of it all..... taking a position... the theme is one view or the other is the 'right' view to take.
I take the position that R.Heyser and M.Hawksford do.... people do hear things that are really, actually there. And, I am not worried that i might make 10 mistakes for every 'right' one.
I learn a lot from the process.
THx-RNMarsh
Hi John,
That is a nice circuit. Did you come up with it, or did someone else?
BTW, is that a common mode DC servo I see there? I had been scratching my head about what sets the common mode output level in the earlier simplified version of the circuit you showed.
Also, it looks like the common mode gain of the circuit you showed earlier is unity. This would imply that the CMRR of the circuit is just equal to the amount of the differential gain. Is that correct?
Cheers,
Bob
Of course he's dead now so you can put whatever words you want into his mouth without any benefit of context if he ever did say such a thing.
So he was in denial of his own humanity as well?
se
Bob, of course I designed this circuit, starting with a complementary differential bipolar version in 1969. Of course, over the years, first, we got jfets, then mosfets, and I upgraded to them. By 1980, the circuit shown was practical, and one version was put into the line amp of the JC-80.
I first met Richard Heyser in 1968 at an AES convention in Hollywood. He, Paul Klipsch, and I would get together for some Kentucky Whiskey that Paul brought. I stayed close friends with Richard Heyser, for almost 20 years, and I would often call him at home when I had an important, if not vexing question. Once it was 'What is true delay?' He was one of the most brilliant people I have ever associated myself with.
He would often 'stun' me with statements like. Negative feedback is a serious problem in audio design. 'That if 2 or more people heard an audio difference, then it is real.' etc
And he showed Bascom King and me his first TDS measurement system, more than 40 years ago.
Richard and I worked in the 70's to make the AES as good a society as possible, but I became disillusioned with the AES along with Walt Jung, when we were at upper level committee meetings and realized it was more like a boys club, rather than a professional organization.
It was a real loss to lose Richard Heyser, and later, Michael Gerzon, two people I made peace between, and two of the smartest people I ever had the privilege of knowing.
He would often 'stun' me with statements like. Negative feedback is a serious problem in audio design. 'That if 2 or more people heard an audio difference, then it is real.' etc
And he showed Bascom King and me his first TDS measurement system, more than 40 years ago.
Richard and I worked in the 70's to make the AES as good a society as possible, but I became disillusioned with the AES along with Walt Jung, when we were at upper level committee meetings and realized it was more like a boys club, rather than a professional organization.
It was a real loss to lose Richard Heyser, and later, Michael Gerzon, two people I made peace between, and two of the smartest people I ever had the privilege of knowing.
Attachments
- Status
- Not open for further replies.