The most linear stage for large voltage swings - Lender's Circuit

Jump to Latest
Z

Zero D

Member
Joined 2009
@ tiefbassuebertr

This "might" be the guy ?

United States Patent Application 20090309659

Published Applications Database Search Results: <SRCH> in <BANR>.

Micro-Coaxial Lines for Active Hybrid-Monolithic Circuits," N. Ehsan, E. Cullens, K. Vanhille, D. Frey, S. Rondineau, R. Actis, S. Jessup, R. Lender Jr., A. Immorlica, D. Nair, D. Filipovic', Z. Popovic', IEEE International Microwave Symposium Digest, 2009, pp. 465-468, Boston, MA, June 2009

University of Colorado - Microwave and RF Research Group - Publications
Click to expand...

You could try & contact him via those addresses !
 
djk

djk

R.I.P
Joined 2001
"There is no evidence anywhere that this is true. "

That you are aware of.

I would have to think that the local feedback, accomplished with no extra parts, would have to improve the linearity.

Do you have any measurements to back up your doubt?

Could it be improved on, given a different topology and more parts? Maybe, maybe not. Tests and measurements might help, but will it sound any better?

Active current sources for loads in the front end improve the open loop gain, and while that reduces the distortion measurement, it may sound worse. Is this progress? Is this worth increasing the complexity and parts count?
 
M

michaelkiwanuka

Banned
Joined 2010
djk said:
Could it be improved on, given a different topology and more parts? Maybe, maybe not. Tests and measurements might help, but will it sound any better?

Active current sources for loads in the front end improve the open loop gain, and while that reduces the distortion measurement, it may sound worse. Is this progress? Is this worth increasing the complexity and parts count?
Click to expand...

From my perspective good amplifier design is about maximising forward path linearity while maximising forward path gain, and, simultaneously, minimising forward path poles.

Folded (or complementary) cascode arrangements like those of Borbely and Lender are effectively single gain-stage designs which do not maximise forward path gain or improve forward path linearity anywhere near enough to justify their use.

The best compromise is obtained by using a two stage design where local feedback can be used to enhance forward path linearity without excessively reducing the forward path gain necessary for adequate loop gain when major loop feedback is applied.

Moreover, the two gain stage design generates two dominant poles in the forward path that can be split by means of minor loop feedback (Miller compensation, TPC, "TMC") about the second stage to stablise the major feedback loop.
 
djk

djk

R.I.P
Joined 2001
".of course active current sources add to distortion. "

I have actually seen measurements showing the distortion decreasing by an order of magnitude, yet listening tends to show the opposite.

"Folded (or complementary) cascode arrangements like those of Borbely and Lender are effectively single gain-stage designs"

Are we talking about the same topology?

I think not.
 
Last edited:
StevenOH

StevenOH

Member
Joined 2004
Maybe, reading EB article in a thorough manner will help to uncover the
"mystery" of Lender design.:)
 

Attachments

  • Borbely 60W Mosfet - Lender.PNG
    Borbely 60W Mosfet - Lender.PNG
    113.1 KB · Views: 902
tiefbassuebertr

tiefbassuebertr

Member
Joined 2005
tiefbassuebertragung.de
tiefbassuebertr said:
Thank you for this advices. here the right link:
1) 404 | University of Essex
If don't goes open please copy this in the URL address field and add the correct characters at the beginning and at the end as a substitute for the wrong
2) http_www_essex.ac.uk/csee/research/audio_lab/malcolmspubdocs/J10%20Enhanced%20cascode_pdf
then the URL will no longer be mutilated further through this forum
Click to expand...
URL is death. but this link goes open:
https://web.archive.org/web/2013012..._lab/malcolmspubdocs/J10 Enhanced cascode.pdf
 
Last edited:
tiefbassuebertr

tiefbassuebertr

Member
Joined 2005
tiefbassuebertragung.de

Attachments

  • J10 Enhanced cascode.pdf
    733 KB · Views: 118
tiefbassuebertr

tiefbassuebertr

Member
Joined 2005
tiefbassuebertragung.de
Bonsai said:
"by the thread title you'd be better off pursuing cascodes:

REDUCTION OF TRANSISTOR SLOPE DISTORTION IN LARGE SIGNAL AMPLIFIERS, M.O.J. Hawksford, JAES, vol.36, no.4, pp.213-222, April 1988

http://www.essex.ac.uk/csee/research... cascode.pdf "


This is the VAS technique I was refering to in my earlier post on this thread.

Using the simple standard current amplifier topology front end, but modding the VAS stage to use Hawksfords technique, I consistently get THD (simmed) below 500ppb. Between the VAS and the load, I'm using a diamond buffer. Hawksfords VAS is really amazing in my view. This weekend, I plan to mod my power current power amp VAS to Hawksford (I was supposed to do it a few weeks ago, but have not had time).
Click to expand...
URL is dead, I have upload the pdf in my previous post.
 
You must log in or register to reply here.
Top Bottom