Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Solid State Talk all about solid state amplification.

No-global-loop amplification
No-global-loop 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 2nd February 2018, 10:41 AM   #151
matze is offline matze  Europe
diyAudio Member
 
Join Date: Sep 2012
Quote:
Originally Posted by traderbam View Post
Comparison
You can set 1/B = TRX so the closed loop gain is independent of A but the feedback loop gain is ABX, the same as the conventional system.
The third last line on the sheet seems to agree with results from post #135 and #128.
  Reply With Quote
Old 2nd February 2018, 12:40 PM   #152
traderbam is offline traderbam  Canada
diyAudio Member
 
traderbam's Avatar
 
Join Date: Jan 2002
Location: Earth
Quote:
Originally Posted by vzaichenko View Post
OK, in my system X = 1, making the formulas even simpler.
Now, here's the point - overall gain in my system is equal to T.
It's independent of A - that's exactly the condition for ODNF to work properly.
Ok. ODNF => 1/B=TRX (X=1 and T= your G and R=5.1k)
Quote:
The error channel's loop characteristics, shown in the post #145, practically do not influence the amplifier's overall characteristics - they are strongly canceled-out after all.
If I ground the output of A - nothing happens except the distortion increase, as there's no (-error) injection anymore.
Neither the amplitude nor the phase responses change.
No.
By your own measurement, the op-amps apply 18dB of NFB at low f. This will cause the op-amps to try to control the output voltage and this certainly will affect the overall characteristics of the amp. Eg: try measuring the output resistance. I also notice in post #142 that the overall unity gain f has doubled or more with ODNF and the phase response slightly straightened.

It is tempting to think that the closer the output voltage tracks the input voltage the less feedback exists. But this isn't so! Feedback affects the sensitivity of the circuit to input/output difference regardless of what that difference is. The feedback is always "fully on".

The absence of change of the overall closed loop gain does not imply there is no feedback. The op-amps are simply using their local excess voltage gain to control the global closed loop gain to be the same as before. Because that's what you have told them to do.
__________________
Everything matters.

Last edited by traderbam; 2nd February 2018 at 01:01 PM.
  Reply With Quote
Old 2nd February 2018, 12:44 PM   #153
traderbam is offline traderbam  Canada
diyAudio Member
 
traderbam's Avatar
 
Join Date: Jan 2002
Location: Earth
Quote:
Originally Posted by matze View Post
The third last line on the sheet seems to agree with results from post #135 and #128.
Barring mistakes, the diagrams and the equations should agree.
__________________
Everything matters.
  Reply With Quote
Old 2nd February 2018, 01:04 PM   #154
vzaichenko is offline vzaichenko  Russian Federation
diyAudio Member
 
vzaichenko's Avatar
 
Join Date: Jan 2010
Location: Moscow, Russia
No-global-loop amplification
Arrow That's the way it works

In my particular case E = 18db, however, as 18db / (1+18db) is rather close to 1, it does not influence the overall gain / phase characteristics of the amplifier, as I mentioned earlier.

Once again, the dominant component in the formula is A, so the amplifier's gain, phase and other characteristics are defined by A.
Distortion D is a sum of Da and Dx - so all distortion components are corrected.
E only influences the accuracy of the error correction - the higher E means the lower Error at the output.

ODNF does not really change the "character" of the amplifier evenly decreasing the harmonics amplitude. I see it in my prototype measurements. That's exactly what I like it to do
Attached Images
File Type: jpg ODNF formula.JPG (77.3 KB, 1285 views)
__________________
If you can't explain it simply, you don't understand it well enough (c) Albert Einstein
http://vzaudio.com/
  Reply With Quote
Old 2nd February 2018, 01:16 PM   #155
matze is offline matze  Europe
diyAudio Member
 
Join Date: Sep 2012
Quote:
Originally Posted by vzaichenko View Post
Once again, the dominant component in the formula is A, so the amplifier's gain, phase and other characteristics are defined by A.
Distortion D is a sum of Da and Dx - so all distortion components are corrected.
E only influences the accuracy of the error correction - the higher E means the lower Error at the output.

ODNF does not really change the "character" of the amplifier evenly decreasing the harmonics amplitude. I see it in my prototype measurements. That's exactly what I like it to do
OK.

Cheers,
Matthias
  Reply With Quote
Old 2nd February 2018, 01:19 PM   #156
vzaichenko is offline vzaichenko  Russian Federation
diyAudio Member
 
vzaichenko's Avatar
 
Join Date: Jan 2010
Location: Moscow, Russia
No-global-loop amplification
Quote:
Originally Posted by traderbam View Post
By your own measurement, the op-amps apply 18dB of NFB at low f.
No, because of the reason I mentioned - the "+" input of the opamp was grounded for that measurement . Connecting it to the input makes a big difference, making the real influence looking as 8 / (1 + 8) = 1db (18db = 8 times). This is what I also see in simulation if I connect the "+" input to the signal source.
__________________
If you can't explain it simply, you don't understand it well enough (c) Albert Einstein
http://vzaudio.com/
  Reply With Quote
Old 2nd February 2018, 01:22 PM   #157
traderbam is offline traderbam  Canada
diyAudio Member
 
traderbam's Avatar
 
Join Date: Jan 2002
Location: Earth
Quote:
Originally Posted by matze View Post
The third last line on the sheet seems to agree with results from post #135 and #128.
Yes #135 does except that I have treated the IPS as a transconductance amp and my equation includes R. I haven't checked #128.
__________________
Everything matters.
  Reply With Quote
Old 2nd February 2018, 01:30 PM   #158
traderbam is offline traderbam  Canada
diyAudio Member
 
traderbam's Avatar
 
Join Date: Jan 2002
Location: Earth
Quote:
Originally Posted by vzaichenko View Post
No, because of the reason I mentioned - the "+" input of the opamp was grounded for that measurement . Connecting it to the input makes a big difference, making the real influence looking as 8 / (1 + 8) = 1db (18db = 8 times). This is what I also see in simulation if I connect the "+" input to the signal source.
Re post #154 diagram
The transfer function is

Vo/Vi = [ (AR + K)X ] / ( 1 + KBX )

The feedback loop gain is KBX.
Can you estimate KBX for your circuit?

Re measuring feedback
You need to connect everything up in the sim as in the proper operation of the circuit. Then insert an ac voltage source (ac=1 dc=0) into the feedback loop. Then measure the ratio of the voltages either side of the voltage source. Do not apply any input signal.
__________________
Everything matters.
  Reply With Quote
Old 2nd February 2018, 01:48 PM   #159
traderbam is offline traderbam  Canada
diyAudio Member
 
traderbam's Avatar
 
Join Date: Jan 2002
Location: Earth
Looking at the schematic in post #49, a quick calculation (always a mistake!) gives me KB = 8.4 or 18dB. Do you get the same figure?
__________________
Everything matters.
  Reply With Quote
Old 2nd February 2018, 01:50 PM   #160
vzaichenko is offline vzaichenko  Russian Federation
diyAudio Member
 
vzaichenko's Avatar
 
Join Date: Jan 2010
Location: Moscow, Russia
No-global-loop amplification
You're trying to consider K as a gain stage. From the overall amplifier's point of view, it's not. It's a small helper, producing the difference between the input signal and attenuated output signal.

The transfer function:
Vo/Vi = AX

And then, we have additional component from the error channel, looking like:
KBX / (1 + KBX)

This additional component sets the rate of distortion cancellation. Nothing else.
Assuming, the amplitudes at the "+" and "-" inputs of the opamp are equal.
__________________
If you can't explain it simply, you don't understand it well enough (c) Albert Einstein
http://vzaudio.com/
  Reply With Quote

Reply


No-global-loop amplificationHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Closed-Loop Amplification with Driver Cone Position Fedback xrk971 Planars & Exotics 44 23rd July 2013 02:51 AM
No global feedback preamplifier smithomo Solid State 6 18th December 2011 03:31 PM
possible advantage to global??? mashaffer Tubes / Valves 20 28th December 2010 06:43 PM
GAS - Global Amplification Stage Stee Solid State 17 19th August 2008 04:58 AM
hum,ground loop,earth loop problem with your answers please frank2395 Pass Labs 2 17th May 2005 07:46 PM


New To Site? Need Help?

All times are GMT. The time now is 07:33 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio
Wiki