Complementary input stages with current mirrors - Page 2 - 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 28th November 2008, 10:31 PM   #11
jcx is online now jcx  United States
diyAudio Member
 
Join Date: Feb 2003
Location: ..
as I was saying, the default linking scheme is a little rough - if the thread title were automatically included it would help...

Unstable VAS current in amp from Slone book
  Reply With Quote
Old 1st December 2008, 12:09 AM   #12
jgedde is offline jgedde  United States
diyAudio Member
 
jgedde's Avatar
 
Join Date: Dec 2006
OK. I think I've got it! Distortion at 1 kHz = 0.000056%, distortion at 20 kHz = 0.00012% into 4 ohms. At least that what PSPICE fourier analysis says....

Here's the mystery... In an attempt to define the output of the complementary diff-amps, I added a circuit to servo the supply voltage to one diff-amp to maintain a desired output standing voltage thus defining the VAS standing current (See the current mirror on the upper input stage). It seems the circuit only needs to be on one of the two complementary input stages - the other one follows. If I add another circuit to the so that both stages are constrained, then the distortion increases by an order of magnitude. ?????

I developed the circuit below with the servo because I found that forcing current mirrors by coercing the output to be some fixed voltage causes an imbalance in the diffamp that negates much of what's gained by use of current mirrors in the complementary input stages.

So, I need advice on the following:
1) Is it expected that the non-constrained input stage output always track the constained one or is this a simulation behavior that won't be even remotely true in real life?

2) How should I compensate this beast?

3) Am I ready to break out the soldering iron?
Attached Images
File Type: jpg myamp.jpg (57.4 KB, 733 views)
  Reply With Quote
Old 1st December 2008, 03:37 AM   #13
GK is offline GK  Australia
Account disabled at member's request
 
Join Date: Jan 2006
Quote:
Originally posted by jgedde
OK. I think I've got it! Distortion at 1 kHz = 0.000056%, distortion at 20 kHz = 0.00012% into 4 ohms. At least that what PSPICE fourier analysis says....

Here's the mystery... In an attempt to define the output of the complementary diff-amps, I added a circuit to servo the supply voltage to one diff-amp to maintain a desired output standing voltage thus defining the VAS standing current (See the current mirror on the upper input stage). It seems the circuit only needs to be on one of the two complementary input stages - the other one follows. If I add another circuit to the so that both stages are constrained, then the distortion increases by an order of magnitude. ?????

I developed the circuit below with the servo because I found that forcing current mirrors by coercing the output to be some fixed voltage causes an imbalance in the diffamp that negates much of what's gained by use of current mirrors in the complementary input stages.

So, I need advice on the following:
1) Is it expected that the non-constrained input stage output always track the constained one or is this a simulation behavior that won't be even remotely true in real life?

2) How should I compensate this beast?

3) Am I ready to break out the soldering iron?

Hi

You are correct that only one LTP/VAS side will need to be servoed (the other side will be steered to match by the gnfb loop).

However your proposed servo doesn't make sense to me.

Attached below is a rather quick mspaint rendition of what I came up with when making my 12W amp. I then abandoned the idea because even without a servo and with 1k load resistors to "constrain" (gain limit - they do not cause any imbalance) the LTP's instead the input stage / VAS distortion was still not dominant over the class A output stage.

Cheers,
Glen
Attached Images
File Type: gif servo.gif (10.7 KB, 721 views)
  Reply With Quote
Old 1st December 2008, 11:35 AM   #14
jgedde is offline jgedde  United States
diyAudio Member
 
jgedde's Avatar
 
Join Date: Dec 2006
I whipped up a quick simplified version of the mirror "servo" in LTSpice. Here it is...

Basically Q2 sets the current through R1 based on the voltage between the CM output and the rail. Q1 is an emitter follower that sets the supply voltage so as to make the output a diode drop above the reference voltage at the junction of the zener and the resistor. Since the output of the current mirror naturally wants to be at a higher voltage, we only need to servo it down, not up in voltage. That's why this works.

Thanks,
John
Attached Images
File Type: jpg mirror op set.jpg (22.9 KB, 606 views)
  Reply With Quote
Old 1st December 2008, 11:38 AM   #15
jgedde is offline jgedde  United States
diyAudio Member
 
jgedde's Avatar
 
Join Date: Dec 2006
Here is the ASC file...
Attached Files
File Type: txt mirror op set.asc.txt (1.7 KB, 30 views)
  Reply With Quote
Old 1st December 2008, 11:58 AM   #16
GK is offline GK  Australia
Account disabled at member's request
 
Join Date: Jan 2006
Quote:
Originally posted by jgedde
I whipped up a quick simplified version of the mirror "servo" in LTSpice. Here it is...

Basically Q2 sets the current through R1 based on the voltage between the CM output and the rail. Q1 is an emitter follower that sets the supply voltage so as to make the output a diode drop above the reference voltage at the junction of the zener and the resistor. Since the output of the current mirror naturally wants to be at a higher voltage, we only need to servo it down, not up in voltage. That's why this works.

Thanks,
John

Hi John

I can understand how the servo loop functions, but I can't see how it can work without seriously compromising the function of the current mirror - Q3 (in your subsequent attachment) will be running with minimal Vce. Basically, the voltage across R2 will not begin to drop until Q3 begins to get cut off.

BTW, I had to change my avatar because my new girlfriend was getting jealous, Iím currently fabricating a custom intake manifold to fit a pair of my 45DCOE Weber side draughtís to the Landrover engine and the 12W amp doesnít sound (which is as it should be for a HiFi amp, IMHO).

Cheers,
Glen
  Reply With Quote
Old 2nd December 2008, 01:12 AM   #17
diyAudio Member
 
LineSource's Avatar
 
Join Date: Sep 2003
Location: SiliconValley
Quote:
Originally posted by G.Kleinschmidt
I can understand how the servo loop functions, but I can't see how it can work without seriously compromising the function of the current mirror.
Glen

Glen,

What is the best DC servo solution you have seen that does not voltage-error sum at the DIFF input? For a symmetrical diff pair input topology, I have seen circuits which:
1) adjust current sources on input diff pair
2) adjust the voltages to the input cascode pair
3) adjust the the VAS current source pair


I am working with a bipolar version of Nelson Pass Symmetrical diff pair input Super-Sym topology. Symmetrical diff inputs driving + and - output circuits which drive load.
  Reply With Quote
Old 2nd December 2008, 03:16 AM   #18
GK is offline GK  Australia
Account disabled at member's request
 
Join Date: Jan 2006
Quote:
Originally posted by LineSource



Glen,

What is the best DC servo solution you have seen that does not voltage-error sum at the DIFF input? For a symmetrical diff pair input topology, I have seen circuits which:
1) adjust current sources on input diff pair
2) adjust the voltages to the input cascode pair
3) adjust the the VAS current source pair

Hi

I think the 'best' option would depend on the specific application. However not to confuse two different things, the servos we are talking about here are for defining/stabilising the VAS current in esoteric fully-symmetrical amps with lots of first stage (LTP) gain.
Such servos do not null out the offset voltage at the speaker terminals caused by Vbe/RE/etc tolerances of the input devices.

Cheers,
Glen
  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


Similar Threads
Thread Thread Starter Forum Replies Last Post
How critical is matching PNP/NPN in complementary stages? nelsonvandal Solid State 29 20th January 2009 08:36 PM
Q on PP & current mirrors JoshK Tubes / Valves 13 26th March 2007 05:16 PM
Current mirrors driving FETs Kevinbd Solid State 23 1st December 2006 12:50 AM
Current Mirrors and first post here vynuhl.addict Solid State 6 16th November 2006 08:43 PM
BCV61/62 monolithic current mirrors Mr Evil Parts 0 23rd February 2005 02:45 PM


New To Site? Need Help?

All times are GMT. The time now is 03:25 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