John Curl's Blowtorch preamplifier part II - Page 3334 - diyAudio
Go Back   Home > Forums > Member Areas > The Lounge

The Lounge A place to talk about almost anything but politics and religion.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 22nd January 2013, 06:29 PM   #33331
jcx is offline jcx  United States
diyAudio Member
 
Join Date: Feb 2003
Location: ..
Default please, john, the "poor me" and the arrogance really don't mix well

feedback only works well for a fairly linear open loop - can't magically bridge a deadband only try to push faster trough it - with limited success

but the when the amp is free of deadband, abrupt gain changes then feedback over a certain level does reduce all orders of "locally Lipschitz" nonlinearities where the feedback factor is large

and much of the pious "make the amp as linear as possible before applying feedback" is in reality just choosing how to distribute feedback between local, and larger loops up to the global feedback

once you have decided how many, what type and how heavily biased the gain elements are


op amp open loop linearity plots are rare, but appeared at least 3 decades ago in some (PMI) datasheets - Pease has written on this - published test circuits, plots - it really is text book (or app note) engineering today

if monolith op amp designers don't show you then measure it, if they choose too low bias current, too small output Q, then bias, buffer the output if you like the rest of the specs - if not pass it by

you could even look at less than 20 yr old op amps

Last edited by jcx; 22nd January 2013 at 06:44 PM.
  Reply With Quote
Old 22nd January 2013, 06:37 PM   #33332
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: cambridge ma
Default -200dB and no mains noise

Here's an interesting physics paper. Notice the noise graphs, no mains lines at all. http://physics.syr.edu/~bplourde/pap...-voltmeter.pdf
__________________
"The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important."
  Reply With Quote
Old 22nd January 2013, 06:39 PM   #33333
kamis is offline kamis  Serbia
diyAudio Member
 
Join Date: Mar 2006
Location: Obrenovac,Serbia
Quote:
Originally Posted by john curl View Post
For the record, the IC op amp that I use in place of the AD797 has to amplify over 30 dB or 40 times and has a final load of 800 ohms above 2KHz. Not so easy as some might guess.
It should and must only be LME 49990 Ic.
  Reply With Quote
Old 22nd January 2013, 06:43 PM   #33334
diyAudio Member
 
john curl's Avatar
 
Join Date: Jul 2003
Location: berkeley ca
Right you are!
  Reply With Quote
Old 22nd January 2013, 06:48 PM   #33335
gk7 is offline gk7
diyAudio Member
 
Join Date: Oct 2007
Location: Vienna, Austria
Quote:
Originally Posted by SY View Post
...apparently incorrect claim that it is what distinguishes ICs from discrete, relegating the former to "mid fi."
No one doubts that AD797 and OPA637 are excellent (and comparably expensive) parts but you won´t
find them in "mid fi" where the design decisions are made by bean counters.

Would be interesting to see the results for JRC4580 and similar dual opamps you
typically find in consumer gear. (But maybe PMA does not store such junk in his lab. ;-) )
  Reply With Quote
Old 22nd January 2013, 06:59 PM   #33336
diyAudio Member
 
Chris Hornbeck's Avatar
 
Join Date: May 2011
Location: Little Rock
Quote:
Originally Posted by jcx View Post
and much of the pious "make the amp as linear as possible before applying feedback" is in reality just choosing how to distribute feedback between local, and larger loops up to the global feedback
Another *possible* description (not testable AFAIK, so not even a hypothesis) is that intrinsic linearity, before even local feedback, is important. Sounds wacky at first blush, but does go a long way towards explaining such strange things as the ease with which good sound is obtained with vacuum valves, the SET phenomenon, folks' preference for FETs, etc. Maybe it's all self-delusion, but maybe there's some there there.

Thanks,
Chris
  Reply With Quote
Old 22nd January 2013, 07:02 PM   #33337
bcarso is offline bcarso  United States
diyAudio Member
 
Join Date: May 2005
Location: Canoga Park, California
Yes, Chris and jcx say sooth.

In a given small-enough region of operation, our assumptions are that a series expansion of some number of terms accurately describes the nature of the beast, and when feedback is applied it reduces distortion in a predictable way. Thus we are allowed to do the measurements with high noise gain and extrapolate to much lower gains. And the curve of growth of higher-order terms is likewise predictable as a function of level.
  Reply With Quote
Old 22nd January 2013, 07:11 PM   #33338
bcarso is offline bcarso  United States
diyAudio Member
 
Join Date: May 2005
Location: Canoga Park, California
Quote:
Originally Posted by Chris Hornbeck View Post
Another *possible* description (not testable AFAIK, so not even a hypothesis) is that intrinsic linearity, before even local feedback, is important. Sounds wacky at first blush, but does go a long way towards explaining such strange things as the ease with which good sound is obtained with vacuum valves, the SET phenomenon, folks' preference for FETs, etc. Maybe it's all self-delusion, but maybe there's some there there.

Thanks,
Chris
I'm sure you've seen the theoretical paper of Boyk and Sussman that's floating around, afaik never published in a journal although looking journal-ready to these eyes at least. Sussman is a technical heavy hitter, apparent from his publications and attested to by Scott W.

They deal with ideal models, with appropriate qualification, of FETs, bipolars, and tubes, and look at IM distortion terms arising from very simple structures with local feedback. Someone will have the ref link I'm sure, or I can dig it up.

But in essence, it's a kind of elaboration of the old Baxandall stuff about higher order terms of HD appearing with small amounts of feedback around nonlinear systems, even though the overall performance by some measures is enhanced.

I suspect the focus on IMD, versus HD, makes it a little more relevant to what we seek.
  Reply With Quote
Old 22nd January 2013, 07:12 PM   #33339
diyAudio Member
 
Chris Hornbeck's Avatar
 
Join Date: May 2011
Location: Little Rock
And we assume monotonicity.

Thanks,
Chris
  Reply With Quote
Old 22nd January 2013, 07:12 PM   #33340
PMA is offline PMA  Europe
Design engineer, consultant
diyAudio Member
 
PMA's Avatar
 
Join Date: Apr 2002
Location: Prague
Quote:
Originally Posted by gk7 View Post

Would be interesting to see the results for JRC4580 and similar dual opamps you
typically find in consumer gear. (But maybe PMA does not store such junk in his lab. ;-) )
No I do not . But the last page of my zip/pdf that was posted here several hours ago shows uA748. Or precisely MAA748, the Tesla equivalent of uA748, with Ccomp 10pF, which makes it faster than uA741 (equals Ccomp 30pF). If you want to laugh, here it goes again: (noisy and with THD increasing fast with frequency)
Attached Images
File Type: png uA748_gain40dB.PNG (52.7 KB, 149 views)
File Type: png uA748.PNG (65.8 KB, 149 views)
__________________
Pavel Macura
http://web.telecom.cz/macura/audiopage.html
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 01:04 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2