New Class A, Super-A, Non-Switching : need a revival ? - Page 10 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

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 April 2012, 03:28 AM   #91
diyAudio Member
 
Join Date: Nov 2007
Location: Dallas
R2 is neither a CCS nor a bootstrap,
therefore current in Q1 is variable.
VBE of Q1 not held constant.
Does it matter? I dunno...

You have eliminated Q1 VBE resistor,
that was responsible for bleeding a
constant reserve current that serves
to assure output transistors stay on.

A misbehaving load could turn off a
fast recovering Schottky, but would
be unable to shut off either of the
slow recovering output transistors.

-----

OK I see it now.
The top Q2 (why do you have 2 Q2's?)
conducts R1 to U1 as a reserve current.
The bottom Q2 conducts R2 as a reserve.

Last edited by kenpeter; 29th April 2012 at 03:34 AM.
  Reply With Quote
Old 29th April 2012, 05:39 AM   #92
diyAudio Member
 
Join Date: Sep 2007
Send a message via Yahoo to linuxguru
Quote:
Originally Posted by kenpeter View Post
R2 is neither a CCS nor a bootstrap,
therefore current in Q1 is variable.
VBE of Q1 not held constant.
Does it matter? I dunno...
I probably should go with a bootstrap or JFET instead of R1. The simulated harmonics became nicer with just a plain R1 - from memory, H2 increased.

Quote:
You have eliminated Q1 VBE resistor,
that was responsible for bleeding a
constant reserve current that serves
to assure output transistors stay on.
I merged that resistor with another that was intended to bias the driving opamp into Class-A. It works, but maybe the value needs experimenting with.


Quote:
The top Q2 (why do you have 2 Q2's?)
Sorry, the lower one should be Q3.

I goofed while relabeling the Qs before exporting the schematic (they tend to start with numbers like Q11, Q13, etc. because I sometimes cut them down from some larger circuit like a discrete opamp, and I relabel them left-to-right, top-to-bottom for convenience. LTSpice catches these duplicates, but only if a simulation run is attempted.)

BTW, check out my LF03 discrete opamp schematic at your convenience - I just built the first prototype a couple of days ago, and it does sound good with the present Class-A push-pull output stage.

However, it could potentially give similar performance at lower quiescent current with your non-switching Class-AB stage.

LF03 discrete opamp

I'll post the prototype parts list shortly.
  Reply With Quote
Old 10th May 2013, 08:46 PM   #93
diyAudio Member
 
tiefbassuebertr's Avatar
 
Join Date: Jul 2005
Location: D-55629 Schwarzerden
Quote:
Originally Posted by steph_tsf View Post
At fisrt glance, the Ben Duncan books looks quite general. Digging into the individual chapters, the big and positive surprise is to realize how precise and accurate Ben Duncan is. The appendixes are unvaluable, plenty of references, plenty of historical info. Many many thanks for pointing this great source of info !

Page 133, about the non-switching amplifiers, it reads :
"This involved a kind of positive feedback, and carrying the risk of blowing-up the output stage ! It also relies on the switching of small diodes, which ameliorated and displaced, rather than overcoming the switching problem".
I guess Ben Duncan didn't wanted to write explicitely that the Technics SynchroBias, marketed as "New Class A", was displacing the problem instead of solving it. The Technics "New Class A" is thus a little bit apart. It should be clear and remembered that JVC "Super-A", Pioneer "Non Switching" and Sansui "Tanaka Non-Switching" are not relying on diodes to be switched.

It is amazing to see the LT1166 biasing integrated circuit, being ignored. Were there commercial power amplifiers using a LT1166 biasing integrated circuit ?

Like Margan, what's worrying me is the behaviour of the output stage when driving a complex load like a loudspeaker. Some optimizations like the ones discussed here may prove unable to cope with capacitive, inductive, or varying impedance loads, especially on non-periodic, non-static, non-symetric input signals like music is. In this context, is there a LTspice standard test gig I can use, for better knowing the behaviour of the output stage ?

Cheers,
Steph
Today I contact LT Technology in this matter because I want to know any Commercial Audio amplifier devices with LT1166 inside.
  Reply With Quote
Old 10th May 2013, 09:18 PM   #94
mlloyd1 is offline mlloyd1  United States
diyAudio Member
 
mlloyd1's Avatar
 
Join Date: Feb 2001
Location: Northern Iliinois
good luck with that.
it is unusual for device suppliers to say what products are using their specific devices.

mlloyd1

Quote:
Originally Posted by tiefbassuebertr View Post
Today I contact LT Technology in this matter because I want to know any Commercial Audio amplifier devices with LT1166 inside.
  Reply With Quote
Old 11th May 2013, 02:28 PM   #95
diyAudio Member
 
Join Date: Sep 2006
Quote:
Originally Posted by tiefbassuebertr View Post
Today I contact LT Technology in this matter because I want to know any Commercial Audio amplifier devices with LT1166 inside.
I don't know of any commercial amplifier manufacturer using the LT1166, but I could definitely be wrong. I have always thought it a great part when properly used. Part of the problem is that it is intended for use with MOSFET output stages, and will not work (straightforwardly) with all MOSFETs, due to its need for bias voltage that is a function of gate threshold voltage. Moreover, the designs shown in the datasheet generally are not the best. I show some ways to use the LT1166 in my book.

Cheers,
Bob
  Reply With Quote
Old 22nd February 2016, 11:32 AM   #96
arnyk is offline arnyk  United States
Banned
 
Join Date: Jan 2014
Location: Grosse Pte Wds MI
Quote:
Originally Posted by steph_tsf View Post
hello,

would there be an intellectual property infringement if a company sells three small PCBs, or carrierboards, about 1.5 by 1.0 inch tall, fitted with SMDs, ready to use, providing each one a specific Class AB biaising scheme :

- "New Class A" bias module (see Technics folklore - synchro bias)
- "Super-A" bias module (see JVC folklore)
- "Non Switching" bias module (see Pioneer folklore)
(1) If the technology is published - shows up in a published professional paper.

(2) If the patent has run out of its usual 18 year run.

(Pick one)

Then the the IP is in the public domain according to US law.

The status of IP that is revealed in a service manual seems less clear.

If you can obtain the technical information without signing a NDA, and there is not a currently-running patent, then you are on seemingly solid ground for treating it as if it were in the public domain,.

Big caveat - I'm not a lawyer. When in doubt, ask one.

If there is a violation, then you might get a cease and desist letter which you can honor or not, at your own risk. Until you get the letter...

BTW if there was an issue with infringing IP, simply publishing it on a public forum might get you into all of the trouble you might be in.
  Reply With Quote
Old 14th October 2016, 08:20 AM   #97
Banned
 
Join Date: Oct 2016
Quote:
Originally Posted by kenpeter View Post
Far as I know: The easiest way to assure operation in this class is to cross
deliberately underbiased Schottkys. Thus forcing them to operate upon the
square law curve. You then use this law to re-curve your power transistors
for a super-smooth B crossing! Plus the reserve current always bled by R1.

It doesn't matter that Schottkys operate in Class B and turn off. There is
no significant tail current, and they recover fast. The slow transistors are
always conducting the reserve current, and thus in Class A at all times.

It doesn't matter the exact quiescent current of an underbiased Schottky
B crossing. Its always more than zero, and less than would blow anything.
Will probably drift a bit with temperature. You can bond Q3 Q4 to heatsink
if you want to assure hot drift always floats deeper into B rather than AB...
It probably isn't necessary.

Remember: We are talking about a deliberate underbias. No more than one
half the Schottky stack can ever be fully on at once. We strictly regulate
the voltage sum across the stack, so that could never happen. Runaway
just isn't a likely event, even if you did nothing to thermally compensate.
Whats a Schottky diode gonna dissipate in this application? Less than 2W.
TO-220 packages are common, so thermals are going nowhere! Hot power
devices only do the work, but are never trusted to make the decisions.

The smoothness of the crossing is assured by the quasi-complimentary
square laws, and not by a specific bias current that has to be controlled.
No attempt is made to define what actual small B crossing current might
be? The extra reserve current bleeder assures the power devices will be
operating well above the small crossing of the Schottkys.

I have been reading this post #6 in the first page, it's very interesting. It might be too old and Kenpeter might not even on the forum. I just have a question on this circuit.

From my understanding of the circuit on the left, the voltage drop across R1 is 2Vbe of Q1 and Q2 + 4 schottky diode voltage. The reserve current is through R1. When it swing negative, D1 and D2 are turned off. Q2 is kept on by the reserve current and follow Q1.

If I am right, the reserve current keeps Q1 on all the time, but if D1 and D2 turn off, you still have only Q1 driving the load alone. That is still class B by definition even thought Q2 remains on. How is this having advantage over just the simple ClassAB running into class B when driving low?

Thanks

Last edited by Audiocrase; 14th October 2016 at 08:34 AM.
  Reply With Quote
Old 15th October 2016, 03:47 AM   #98
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Location: Waterloo, ON or Herefordshire
Blog Entries: 2
There are many non-switching schemes that simply keep both devices conducting but without really contributing to the load and I don't see the point anymore. I used to think it would be a good thing so that cross-over distortion was gone, but you always end up with some other distortion. It seems to me, still naively perhaps, that if you want simple circuits without feedback you need proper Class A and a superb power supply. Otherwise just use Class AB with plenty of feedback and it will have superb performance when designed properly that the case for non-switching AB or full Class A becomes difficult to support. Perhaps one day Class D will reach the point where Class AB no longer makes sense either - at the high end it may be close already but not at the cheap end.
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig.
  Reply With Quote
Old 15th October 2016, 06:42 AM   #99
Banned
 
Join Date: Oct 2016
Thanks for the reply Bigun.

I know you were on this thread a lot. Is that the reason this thread got dropped because there is always a catch of something and nobody manage to come up with a design that does not introduce extra distortion?

Maybe that's the reason most schematics of even high end amps don't use this at all and just go on optimizing the crossover distortion and higher loop gain. The only one seems to have success is Pass Lab Aleph current source.
  Reply With Quote
Old 15th October 2016, 06:46 AM   #100
AKSA is offline AKSA  Australia
diyAudio Member
 
Join Date: Sep 2001
Location: Melbourne, Australia
I agree emphatically with all of Bigun's points. I too have noticed that when you manage to reduce the switching issues, you find virtually no improvement in SQ. Whether there is clearly remanent crossover artefacts, OR, as Bigun suggests, the non-switching trick create other distortions, the same result - no improvement to SQ.

Feedback and careful operating points of output devices seems to fix the perceived crossover artefacts. I have found the biggest issues come from the extraction of the fb error, the voltage amplifier, and the drive stage for the output devices. These are the pre-eminent problems; solve them, and a Class AB with some global feedback sounds absolutely ethereal.

Ciao,

Hugh
__________________
Aspen Amplifiers P/L (Australia)
www.aksaonline.com
  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
margan : non switching class B sreten Solid State 16 22nd May 2011 05:43 PM
Revival of incredible LM3886 design udailey Group Buys 82 3rd January 2009 01:03 AM
New class A biasing (with non-switching class AB overflow) Steven Solid State 34 24th March 2007 07:40 PM
switching class d amp steven344 Class D 10 20th November 2005 04:22 AM
Non-switching amp same as Class A? mjarve Solid State 2 26th June 2004 08:43 PM


New To Site? Need Help?

All times are GMT. The time now is 11:30 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Copyright 1999-2016 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki