Bob Cordell Interview: Negative Feedback - Page 52 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

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 17th March 2007, 11:34 AM   #511
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Quote:
Originally posted by lumanauw
Hi, Estuart,Your schematic in post #497, how to determine the value of C1-C2-R1? Is there any paper/article about it?

Hi lumanauw,


The equivalent series capacitance of C1 and C2 should be equal to the original Cdom. The are no strict rules for choosing C2 and R1, but the impedance of network C2 & R1 should chosen such that it is not too much loaded by C1. So, if C2 is 5 to 10 times C1, itís OK. Regarding R1, see my previous post.

Cheers,
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het lichtÖ(H.M. van Randwijk)
  Reply With Quote
Old 17th March 2007, 12:17 PM   #512
diyAudio Moderator Emeritus
 
lumanauw's Avatar
 
Join Date: Oct 2002
Location: Bandung
Send a message via Yahoo to lumanauw
Hi, Estuart,

Thanks for the explenation

Your compensation scheme reminds me of a work of Lars Clausen.

It is different, it is about taking feedback point. Feedback point is taken parrarel between VAS/C6=1nf and output/R24=1k. Then this parrarel point is taken to differential.
Attached Images
File Type: gif ulcamp1.gif (73.3 KB, 1054 views)
  Reply With Quote
Old 17th March 2007, 12:44 PM   #513
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Quote:
Originally posted by mikeks
...while DPC can be persuaded to deliver ~60dB at the same frequency.
Hi Mike,

Now I see what you mean: TWO pole compensation(TPC), rather off topic, hence the confusion.
Indeed, linearity can be improved by increasing the bandwidth of the global feedback loop. By means of TPC however, is the worst thing you can do. I repeat myself, have look at post 466 and 467. The phase margin at 10Ö20kHz is terrible low. This does not imply that the system is unstable, but it causes a considerable amount of overshoot. So forget this miserable thingy.
Linearity can also be improved by increasing the loop bandwidth around the O/P stage (thatís what Iím doing), leaving the global feedback loop and step response unaffected.
Convince yourself by running a sim (I suppose you do have a decent simulator, otherwise end of discussion), one with R1 (in my diagram) tied to ground and one with R1 tied to the output. THD should be the same, step response differs.
BTW, use a VAS as shown on post 505, nothing else!

Regards,
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het lichtÖ(H.M. van Randwijk)
  Reply With Quote
Old 17th March 2007, 01:08 PM   #514
GK is offline GK  Australia
Account disabled at member's request
 
Join Date: Jan 2006
Quote:
Originally posted by estuart


Hi Mike,

. This does not imply that the system is unstable, but it causes a considerable amount of overshoot. So forget this miserable thingy.

Iíve been spicing TPC to see how good it is in comparison to other methods for a <1ppm THD-20 500W+ amplifier design Iíve almost finalised. I though it was great at first, examining the resultant slew rate. Then I did a bode plot into a reactive load. Looked like one of Evil Knievelís motorcycle ramps.........


Cheers,
Glen

EDIT: Oops. Terminology confusion. Not TPC, although I've seen people call it TPC. Not the TPC you are talking about, I think. Arrggh.
  Reply With Quote
Old 17th March 2007, 01:12 PM   #515
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Quote:
Originally posted by lumanauw
Hi, Estuart,
Your compensation scheme reminds me of a work of Lars Clausen.
Hi lumanauw,

Funny. Indeed, based on same principle, except that feedback is led to a different point. Probably doesnít work with BJTs in the O/P stage, but with MOSFETs itís OK.
Interested in an other example (which I simulated, but not build)?
Do I have papers on this subject? No, nothing, sorry.

Cheers,
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het lichtÖ(H.M. van Randwijk)
  Reply With Quote
Old 17th March 2007, 01:50 PM   #516
AndrewT is online now AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally posted by Graham Maynard
Using a resistor from output to Darlington VAS base can also help to reduce output stage crossover distortion and improve NFB response by flattening the open loop bandwidth.
Using an equal value resistor from other half of mirror to ground prevents start-up thump and preserves differential input balance.

This is a Sim demo cct only.
Hi Graham,
I note the use of a relatively low Iq value.
The improvement in output stage distortion reduction will probably be more evident when the stage is under-biased.
If the Iq is re-set to 70mA, will the benefits of the revised NFB be as marked?
Will the results be changed at all when a higher IQ is set?
__________________
regards Andrew T.
  Reply With Quote
Old 17th March 2007, 03:12 PM   #517
mikeks is offline mikeks  United Kingdom
Account Disabled
 
Join Date: Jun 2004
Location: Animal farm
Edmond, let's cut to the chase: what is major (global) loop transmission (gain) with your set up compared with ordinary Miller compensation?

http://www.diyaudio.com/forums/showt...75#post1160075
  Reply With Quote
Old 17th March 2007, 03:54 PM   #518
diyAudio Member
 
Join Date: Sep 2006
Quote:
Originally posted by estuart



Exactly, thatís the crux. Chapeaux!
See below for this simple but valuable trick.
Note: It doesnít work without a preceding beta enhancement transistor in the VAS stage. Probably, thatís why D. Self misunderstood the potential of this subtle modification.

Cheers,

Who originally came up with this form of compensation?

It seems to be a fairly nice and inexpensive trick. As described by others, it looks basically like Miller compensation taken from the output of the power stage, but then bypassed (transitioned) to use the more stable output of the VAS itself by C2 and R1 at a frequency on the order of 250 kHz. I think that this topology might benefit from connecting C2 to an emitter-follower (fast) buffered output of the VAS. This would isolate the VAS from currents flowing in R1. This emitter follower could just be the first guy in the output triple.

Double pole compensation has value, but it needs to be used carefully. There are not too many truly free lunches. However, careful selection of its component values allows one to exercise a tradeoff between minimum phase margin and amount of loop gain enhancement it delivers at 20 kHz.

A more generalized approach is to seek a loop gain rolloff on the order of a constant 9 dB/octave, which gives a phase margin on the order of about 45 degrees. This is just a compromize between the conservative 6 dB/octave and the deadly 12 dB/octave. Indeed, double pole compensation ("T" compensation as it was called at Bell Labs in the 1970's where we integrated it into op amps) can be thought of as just one form of approximation to 9 dB/octave compensation.

Bob
  Reply With Quote
Old 17th March 2007, 04:06 PM   #519
diyAudio Moderator Emeritus
 
lumanauw's Avatar
 
Join Date: Oct 2002
Location: Bandung
Send a message via Yahoo to lumanauw
Hi, Estuart,
Quote:
Interested in an other example (which I simulated, but not build)?
Off course

Please submit them

I have 1 question. What happens if I don't use buffer before VAS (omitting Q2 in your graph, the input differential pair directly feds base of Q3)?
  Reply With Quote
Old 17th March 2007, 04:08 PM   #520
mikeks is offline mikeks  United Kingdom
Account Disabled
 
Join Date: Jun 2004
Location: Animal farm
Quote:
Originally posted by Bob Cordell
It seems to be a fairly nice and inexpensive trick. As described by others, it looks basically like Miller compensation taken from the output of the power stage, but then bypassed (transitioned) to use the more stable output of the VAS itself by C2 and R1 at a frequency on the order of 250 kHz. I think that this topology might benefit from connecting C2 to an emitter-follower (fast) buffered output of the VAS. This would isolate the VAS from currents flowing in R1. This emitter follower could just be the first guy in the output triple.
Bob

Edmond's arrangement can only be said to constitute an "improvement" if the later is expressed as an increase in loop gain in the audio band without compromising stability.
  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 12:39 PM.


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