THD & Loop Gain Enhancer - 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 29th June 2013, 05:53 PM   #1
Zero D is offline Zero D  United Kingdom
diyAudio Member
 
Join Date: Aug 2009
Lightbulb THD & Loop Gain Enhancer

The following is applicable to both Power Amps & OpAmps.

I've been tinkering for a number of weeks with an "Enhancer" circuit added to an Amp, which improves both the THD & Loop Gain specs. I'd be interested in hearing your comments & suggestions etc, for where it "might" be improved etc.

Harry Dymond
has been kind enough to assist me in various ways & by simulating the circuit & establishing that it improves the Loop Gain. He also provided me with several links http://www.analog.com/static/importe...tes/AN-107.pdf

I also observed greater bandwith with my Enhancer circuit, as in AN-107.

Figure 1 in AN-107

That arrangement seems different to mine.

Figure 3 in AN-107 - SOC = Gain peaking.

I noticed that in my bode plots too, i was able to reduce/flatten the peaking by juggling the values of the capacitors in parallel with the feedback resistors.

ODNF or no GNFB power amp I hadn't looked at it until yesterday. There do "appear" to be similar things going on, but i don't understand it all !

Looking at my circuit -

OP27 With Enhancer = 0.001% @ 1kHz

OP27 Without Enhancer = 0.003% @ 1kHz

Interestingly with a much higher bandwith OpAmp the distortion is a LOT worse, & the bandwith is about 10 times less ? No other components were changed, or Anything else !

OP37 With Enhancer = 0.159% @ 1kHz

OP37 Without Enhancer = 0.049% @ 1kHz
Attached Images
File Type: png Enhancer.png (6.6 KB, 333 views)
File Type: png THDRC-1.png (22.8 KB, 329 views)
File Type: png LG-HD.png (50.4 KB, 325 views)
File Type: png bode1.png (5.5 KB, 315 views)
File Type: png bode2.png (5.5 KB, 314 views)
  Reply With Quote
Old 29th June 2013, 08:56 PM   #2
diyAudio Member
 
Join Date: Feb 2013
Location: CRO
Send a message via Skype™ to overall feedback
Quote:
Originally Posted by Zero D View Post

Interestingly with a much higher bandwith OpAmp the distortion is a LOT worse, & the bandwith is about 10 times less ? No other components were changed, or Anything else !
In my circuit it is all the same, lm6181 and lf411 . At least simulations shows that. I made circuit with lf411 and still can not believe how it sounds. Few days after without change, but I have to try with better op amp for experiment. It is audio band we are interest for, not ultrasonic.
  Reply With Quote
Old 29th June 2013, 11:38 PM   #3
mcd99uk is offline mcd99uk  United Kingdom
diyAudio Member
 
Join Date: Mar 2008
Location: UK
Looks interesting. Extra gain without too much added phase. Maybe a good way to increase the amount of feedback and hence a reduction in THD.
  Reply With Quote
Old 30th June 2013, 12:40 AM   #4
jcx is offline jcx  United States
diyAudio Member
 
Join Date: Feb 2003
Location: ..
10 dB? - why bother - I show some composite op amps circuits with lots more audio band added gain:

High loop Gain Composite Op Amp Circuits

the 1st few manage lots of added gain inside the global loop

post #10 shows a sim of Harlod Black's 1st feed forward error correction method, in the "current dumping" form, applied to one Bob Cordell's example amps with slow output Q


I would worry about one of your schematics - not near enough current gain to drive 8 Ohm nominal loudspeaker load - added voltage gain can't help there

Last edited by jcx; 30th June 2013 at 12:44 AM.
  Reply With Quote
Old 30th June 2013, 03:49 AM   #5
Bonsai is offline Bonsai  Taiwan
diyAudio Member
 
Bonsai's Avatar
 
Join Date: Jul 2003
Location: Taipei, Taiwan
I see a lot of proposals to reduce THD - some are quite complex (Stochino is one example) and Walkers current dumping another, requiring some closely attached components. I've put my bets on what I call AFEC. It was pointed out a few weeks ago, Hitachi used a similar scheme about 25 years ago but I never took off for some reason. AFEC offers about 20 dB of distortion reduction at 20 kHz, and you get DC servo offset reduction as a by product. Most importantly, it's really simple!

Augmented Feedback Error Correction (AFEC)
__________________
bonsai
Amplifier Design and Construction for MUSIC! http://hifisonix.com/
  Reply With Quote
Old 30th June 2013, 10:07 AM   #6
mcd99uk is offline mcd99uk  United Kingdom
diyAudio Member
 
Join Date: Mar 2008
Location: UK
[QUOTE=Bonsai;3544209 I've put my bets on what I call AFEC.
[/QUOTE]

Me too.
[Off topic]
There are some possible small developments that could be used. Eg. peaking on square waves could be reduced by adding a small cap across the feedback resistor and maybe making use of an op amp with a shut down pin to switch the AFEC off during clipping via an opto used as a baker clamp instead of a diode.
[/Off topic]
  Reply With Quote
Old 30th June 2013, 10:30 AM   #7
Bonsai is offline Bonsai  Taiwan
diyAudio Member
 
Bonsai's Avatar
 
Join Date: Jul 2003
Location: Taipei, Taiwan
Some good ideas there! I will need to try them out.

One thing, if you use an opamp with a shut down pin, the turn on and turn off need to be very fast and clean. Another option here is to simply reduce the AFEC gain during clipping. This could be acheived for example, by shunting the AFEC feedback loop with a low value resistor - again, response time in order to avoid any funny artifacts is the key. If you are building anything, I's be interested to see your results. I am some months away from my new power amp (CFA 400 Watts) but that will incorporate AFEC which I will be able to switch in and out.
__________________
bonsai
Amplifier Design and Construction for MUSIC! http://hifisonix.com/

Last edited by Bonsai; 30th June 2013 at 10:36 AM.
  Reply With Quote
Old 30th June 2013, 10:38 AM   #8
Zero D is offline Zero D  United Kingdom
diyAudio Member
 
Join Date: Aug 2009
Quote:
Originally Posted by overall feedback

It is audio band we are interest for, not ultrasonic
Sure, but having a high internal bandwith with very low THD, is a good starting point.

@ mcd99uk

Thanks

Quote:
Originally Posted by jcx

10 dB? - why bother
Well that's just for starters. The circuit i've shown is one that sims & works as described.

The 8.2 Ohm load was just there as one option. With for eg 32 Ohms, THD @ 1kHz is unmeasurable.

But it's NOT really about this Amp, it's about the Enhancer circuit which could be configured to work with other Amps & PreAmps etc. The components around U2 & connected to U1, are only provisional. The main purpose is for people to experiment with other values, & sim it themselves, to discover what's possible etc, & post their results

I'll take a look at the link, thanks.

@ Bonsai

No doubt your AFEC is another valuable contribution in the grand scheme of things, & i wish you All the best with it My design is still in it's early stages, & isn't perfected yet. Hopefully with input from others who are more skilled in the art, it "might" be

***************

I've realised why the OP37 THD result was so poor. I forgot it requires a gain of 5 or more. Changing it improves things substantially.

U1 = OP37 & U2 = OP27 - C4 = 33pf & C8 = 10pf

THD @ 1kHz = 0.001% or less

The bandwith looks pretty good to me, & that's with those output trannies !
Attached Images
File Type: png Bode-OP37.png (5.6 KB, 56 views)
File Type: png Phase-OP37.png (4.5 KB, 24 views)
  Reply With Quote
Old 30th June 2013, 10:55 AM   #9
diyAudio Member
 
Join Date: May 2005
Location: Bristol, UK
Er, what exactly is the difference between Zero D's circuit and AFEC? It's exactly the same principle as far as I can see.

Bonsai, I've had a very brief look at your AFEC document (as always, thank you for documenting your work so thoroughly and sharing it freely ) and I don't trust your loop gain simulation. You should see a difference between AFEC on and AFEC off. The circuit is not feed-forward, so where is exactly is the improvement in THD coming from if not by an increase in loop gain?

Is figure 5 the circuit you used for loop gain simulation? The output of the AFEC circuit (after R74) is labelled "Vfb", but I can't see any other node in the circuit labelled similarly, so where is the AFEC signal going?
  Reply With Quote
Old 30th June 2013, 12:35 PM   #10
Bonsai is offline Bonsai  Taiwan
diyAudio Member
 
Bonsai's Avatar
 
Join Date: Jul 2003
Location: Taipei, Taiwan
Harry Its not feed forward - as I note in the document, it augments the normal feedback loop, so yes, it does work by increasing loop gain, but not by brute force. Most of the error correction (read: distortion reduction) in a normal amp is provided by the main loop. Any residual error (distortion) can be seen as the main loop simply running out of steam. AFEC corrects for that by comparing the output signal with the input and injecting an error correction signal into the feedback node. This is in marked contrast to feed forward schemes, or approaches that accomplish low distortion through very high loop gain.

As I noted in my write up, some more work is required to flesh out the design. Your comments about the loop gain plots are acknowleged - thanks for that input (hell, its difficult to avoid a pun here, no matter what I write!).

Maybe you'd like to take a bash at the loop gain equations which would help to formalize the approach.
__________________
bonsai
Amplifier Design and Construction for MUSIC! http://hifisonix.com/

Last edited by Bonsai; 30th June 2013 at 12:37 PM.
  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
High loop Gain Composite Op Amp Circuits jcx Solid State 25 25th May 2014 12:54 AM
Throwing away open loop gain with an opamp near or at unity gain okapi Pass Labs 3 25th December 2009 01:23 PM
hi-loop gain amp mfc Solid State 49 10th January 2009 06:10 AM
Gain Blockwith Variable Loop Gain Bonsai Solid State 17 11th July 2008 02:40 PM
Gain loop formula fscarpa58 Everything Else 2 16th January 2004 11:19 AM


New To Site? Need Help?

All times are GMT. The time now is 02:13 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