John Curl's Blowtorch preamplifier part III

Status
Not open for further replies.
Nice reference though he cites even earlier work. If you looked hard enough you could probably find some references from WWII radar work in combining correlated signals with amplifiers having uncorrelated noise. The basic principle of paralleling n amplifiers to get square root of n noise is very old (I'm referring to this basic principle not any of your other work).
http://www.nj7p.org/Manuals/PDFs/Books/MIT-Radiation-Lab-Series-V18-Radar-Engineering.pdf

Chapter 12 has this plus references to earlier work. Or, maybe the details are in Chapter 13.

Going through it all makes my head hurt. Since you're retired, you probably have the time to wade through this.
 
Member
Joined 2002
Paid Member
I had a lady friend who was in a small amateur ensemble and she told me how strange the scoring was for stuff like this.

Experimental music notation resources - Process - lines
There is a common misconception – almost omnipresent in the domain of contemporary academic music - that there is a complete and reciprocal match between a written score and its sounding equivalent. This goes to show how far today’s culture is alienated from the living phenomenon of music and its practice!

What you discover, when you actually make music, is that you can find everything in
the score except the music itself. Mahler, of course, had already said this.

George
 
Hi Richard,
Yes, I know and accept that. I'm still doubtful that Ampex couldn't use that design as John created it. Any lab needing to amplify weak signals would need low noise, and a DC servo to keep that amplifier from drifting into clipping.

I don't see they would be interested in a DC servo at that time. Tape
machine electronics had -tons- of coupling caps. I'm doubtful eliminating one
of them would have been a priority and DC operating points in high gain
front ends are not hard to control.

I think that John might have discovered prior art if he didn't know about it, and was the first to apply it to audio. However, I am also somewhat interested about the fact that John's development was not known to his employer and that this work was exactly what he was supposed to do for Ampex at the time. They could have bumped Studer off the top machine rank and had a different future had they been able to access John's development.

-Chris

When it comes to sonics, early Studer and Ampex machines are two
flavours and it's more personal preference as to what is subjectively better.

Studers real strength were their transport and tape handling = superb
Swiss engineering. I still laugh when I see people upgrading their old Ampex
static tape guides with roller bearings - why would you design it that way in
the first place?

T
 
I have known people who can play all the right notes in the right order at the right time, as specified in the score, yet somehow no music emerges from the instrument.

Going back to oscillators, I seem to recall reading an article about the difficulty in getting two crystal oscillators on near enough the same frequency (possibly nominally the same frequency) to act independently when they have to share a box or PSU. I can't remember the context. Eventually they will synchronise.
 
I have known people who can play all the right notes in the right order at the right time, as specified in the score, yet somehow no music emerges from the instrument.

A score does not contain exact timing information. One of the limitations of music notation. As a result of the limitations, we don't know exactly how classical music was intended by the composer to be played.
 
I have known people who can play all the right notes in the right order at the right time, as specified in the score, yet somehow no music emerges from the instrument.

Going back to oscillators, I seem to recall reading an article about the difficulty in getting two crystal oscillators on near enough the same frequency (possibly nominally the same frequency) to act independently when they have to share a box or PSU. I can't remember the context. Eventually they will synchronise.

Probably injection locking via supply or incidental radiation. A neat experiment would be to put one of Victors oscillators (tweeked to be as close to 1k as possible) on the same rail as a crystal controlled 1k oscillator.
 
This is all nonsense, guys. First of all, it was 50 years ago! You know, 1968. The head of the Audio Division allowed me to do some advanced research for future audio tape recorder development, besides my main task of making a practical phase-locked capstan motor servo that cost about as much as the AC motor it replaced.
My work on low noise was part of this effort, and I paralleled both bipolar transistors and jfets on reproduce input stages to prove the concept. I left the audio department in early 1969, and moved to the Research Division, because I had made contacts there that were forward thinking like I was. I was hated in the audio department, because I was pushy, and forward thinking. It only made them mad.
When it came to the complementary differential input stage power amp, I originally designed it for myself, of course, but when an opening appeared in the audio department to improve a small monitor amp that was to be used with a portable tape recorder, I gave it to the guy in charge. He tried it, but our mutual supervisor said that it was too complex, and to discard it. They used 4-6 active transistors in those days to make a small amp.
Later, in 1969, in the Research Dept, I was assigned to make a balanced bridge 50V-50A current output motor drive power amp, and I built it also with a complementary differential input stage. They also found that it was too complex and hard to understand its operation when I left Ampex to go back into hi fi at a custom audio shop, so they dropped it too. Today, I am taking that same balanced bridge basic power amp concept to make a 1000W into 4 ohm, battleship. It will take awhile.
 
Last edited:
Member
Joined 2004
Paid Member
Absolutely..The print can be finer than the vc wire, it can be matched to number of turns.

jn

Easier and cheaper to wind a 38 AWG coil on top of the existing coil. That would be .004" thick. You can go smaller but the wire gets more expensive and fragile. Since there is no current the gauge is not an issue.

Making a cylinder flex circuit would be quite expensive if you can even figure out how.

You could make an aluminium former in a spiral with maybe 4 turns. The output would be low but it may work. Connecting leadout wires would be a bitch.
 
Studers real strength were their transport and tape handling = superb
Swiss engineering. I still laugh when I see people upgrading their old Ampex
static tape guides with roller bearings - why would you design it that way in
the first place?

T

And the transport was more important to the sound than the electronics. I remember that if you could afford it you bought the Studer A80 24 track, if you couldnt you bought the MCI. The Studers electronics where also better. More head room ( +28dbm output) lower output impedance ( 30 ohms) easier to align.
 
A score does not contain exact timing information. One of the limitations of music notation. As a result of the limitations, we don't know exactly how classical music was intended by the composer to be played.

Hopefully it was passed generation to generation by musicians and conductors as in pre written literature, the Odessy for example. Sure it wont be exact but maybe close enough.
 
Again, this is mostly nonsense. It was best to throw away the Record-Reproduce electronics of either the Studer or the Ampex and start from scratch. That is what I did 3 times. The Studer transport was significantly better than the earlier Ampex transports, and I just stuck with Studer to make upgrades.
 

Attachments

  • photo 3.jpg
    photo 3.jpg
    197.5 KB · Views: 248
Last edited:
Member
Joined 2002
Paid Member
I have known people who can play all the right notes in the right order at the right time, as specified in the score, yet somehow no music emerges from the instrument.
It takes many hours, days and weeks to learn to play a simple musical piece. If you record the process you understand.
And then by the years, when you revisit, other changes take place.

A score does not contain exact timing information. One of the limitations of music notation. As a result of the limitations, we don't know exactly how classical music was intended by the composer to be played.

I have asked myself a lot of times how is it possible music compositions lasting 2 - 3 hours to be performed by different conductors/orchestras, with their tempo to be acoustically perceived as different and in the end they all end at the same time but 2-3 minutes apart.
Performance interpretation is what makes listening to music fascinating .
And timing is only one of the attributes that in a few cases is not provided with exact information on the score, although timing is the one attribute that is provided with the most exact information in most cases.

I don't think your fellow mod (keninkr) was that impressed with some of my favorite George Crumb LP's.

Maybe it was the medium. You should have brought in a CD :D
YouTube

George
 

Attachments

  • DSCN4777.JPG
    DSCN4777.JPG
    644.9 KB · Views: 187
Status
Not open for further replies.