Clarification on ClassA linearity - 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 22nd January 2013, 07:18 PM   #1
alvise is offline alvise  Italy
diyAudio Member
 
Join Date: Jan 2013
Default Clarification on ClassA linearity

Dear forum's people, I just registered in the forum and I've already a question for the experts

I've read the theory of the class A amplifiers (let's consider the configuration with 2 base voltage divider, a collector resistance and a emitter resistance); in particular I've read that a very good linearity for this topology is guaranteed when the signal excursion at the base terminal is much lower than 26mV (because of the non linear dependency of the current of the base on the base voltage - the diode I/V curve). When more class A stages are cascaded, of course this criterion is not honored anymore: in fact the output of the first stage is largely greater than 26 mV. So the question is: when I see at least 2 class A cascaded stages AC coupled (topology needed to obtain a large voltage gain) I know that they introduce some amount of distorsion. Then, why the class A amplifiers are considered by many people (let's do not take into account the power efficiency issue) the best configuration in term of sound quality ? Are the Lin topology, for example, far worse than cascaded class A stages ?

many thanks
  Reply With Quote
Old 22nd January 2013, 08:15 PM   #2
diyAudio Member
 
Join Date: Jan 2011
Location: Fosser, Aurskog-Holand, Akershus, Norway.
Shortform answer.

A class A stage, never mind if it is in a preamp og in a power amp, is biased so the transistors are in the most linear region for transfering the signal input on the base of the transistor to the collector/emitter.
There is no particular signal level for the stage to obtain a linear transition but the 0,5 to 1,5V Base Emitter voltage needed for the transistor to "open".
A true Class A stage never have a signal level allowing the voltage between Base and Emitter to enter the non-linear transition area.
This is what makes the Class A stages more "musical" than Class B or more presise A/B where one of the output transistors is allowed to be completely turned off during the signal trace on the output. This introduces a possibility for non-linearity, best known as Crossover Distortion.

However a good Class A/B stage will have a bias set in a area where this represents no or little problem. Hence the name: A/B A bit Class a a bit Class B
But still: Class A/B works in class B most of the time. There is only when the signal-level is near 0V the stage actually is working as a Class A.
This level can possibly be 26mV, but also a bit away from this.
All depends on how high the bias actually is set.

As for the sonic quality, I have heard A/B-amplifiers with as good sonic quality as even very good engineered Class A-amplifiers. In blind tests that is.
When performing a "open" test there is always some who can hear the crossover, the shut off of the output device not working at any time. I can't.

Wether Linn topology is superior to cascaded Class A i will leave to others to write about. I haven't really investigated that.
__________________
Sooner or later you end up with TANDBERG
  Reply With Quote
Old 22nd January 2013, 08:16 PM   #3
effebi is offline effebi  Italy
diyAudio Member
 
effebi's Avatar
 
Join Date: Nov 2010
Location: Milan
Benvenuto/a Alvise!
About your question, I believe most than all is a matter of taste. You can make an extremely good od bad amplifier either in class A or class B.
The non linearity of the base char. can be solved with feedback.
One of the main problems that you have to fight with a class B is crossover distortion, that is not present in class A. However there are many circuits to avoid that, or to render it unnoticeable. And at the end, again feedback takes care of the rest.
  Reply With Quote
Old 22nd January 2013, 08:23 PM   #4
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
All BJT amplifiers need lots of feedback to improve linearity and to keep base-emitter signal voltages sufficiently small. The feedback will normally be some mixture of local and global feedback.
  Reply With Quote
Old 22nd January 2013, 08:38 PM   #5
GoatGuy is offline GoatGuy  United States
diyAudio Member
 
Join Date: Dec 2012
Location: SF Bay Area
Hi all,

Its a fair question, the Original Post. ... "How to determine the optimum bias for a single transistor stage, with Re (emitter resistor), Rc (collector), and input-to-Rb1-to-base, and base-to-Rb2-to-ground?"

The answer so far "well, it depends on the voltages and transistor, and desired input impedance" are all correct.

Hfe (small signal current gain) in a bipolar transistor varies from around 75 to over 500 for modern transistors. The "higher the better" does not hold - often such high Hfe transistors are more noisy, all in all. Say though a very average 200.

The first stage no matter what should be very light in gain. Its job is to amplify the signal a bit, but preserve the high-impedance input characteristic (if that's the design goal). Working with a first-stage gain of 10 dB (3.1x, voltage-to-voltage), and a nominal operating current of 10 milliamps (usually pretty standard for small-signal ends), then we need the output to be 3.1V if the input is 1V. Taking 2/3 of the supply voltage as the operating point of the collector (good rule of thumb), then the voltage drop of Rc is 1/3V divided by 10 ma. Now you have Rc.

Likewise Re is determined easily by "setting" it to a few volts - 2 to 5 is just fine. Whatever that setting is (that you make, and I'd choose 3V) then again, 3V divided by 10 ma is the resistor. Bypass cap? Sure... but it decreases linearity.

OK, next is to set quiescent (no signal) base bias current. Easy! 10ma / 200 Hfe = 50 microamps! Now we're swinging... The base can be abstracted as Ve (emitter voltage) + 0.7v, and if there were just a resistor from V+ to base for bias, it would be (V+ minus (0.7 + Ve)) / 50 microamp. As any good transistor design text will tell you, this isn't very good in practice because of the variability of transistors. It is much better to give them a "synthetic bias supply" using a pair of resistors from V+ to GND, which behaves like a power-supply having perhaps a +2V over Vb divided voltage point. At that point, figuring out the necessary series resistance from base to input jack to provide the required input impedance matching ... is all that's left.

GoatGuy
  Reply With Quote
Old 22nd January 2013, 08:50 PM   #6
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
?
  Reply With Quote
Old 22nd January 2013, 08:53 PM   #7
sreten is offline sreten  United Kingdom
diyAudio Member
 
Join Date: Nov 2003
Location: Brighton UK
Quote:
Originally Posted by alvise View Post
Dear forum's people, I just registered in the forum and I've already a question for the experts

I've read the theory of the class A amplifiers (let's consider the configuration with 2 base voltage divider, a collector resistance and a emitter resistance); in particular I've read that a very good linearity for this topology is guaranteed when the signal excursion at the base terminal is much lower than 26mV (because of the non linear dependency of the current of the base on the base voltage - the diode I/V curve). When more class A stages are cascaded, of course this criterion is not honored anymore: in fact the output of the first stage is largely greater than 26 mV. So the question is: when I see at least 2 class A cascaded stages AC coupled (topology needed to obtain a large voltage gain) I know that they introduce some amount of distorsion. Then, why the class A amplifiers are considered by many people (let's do not take into account the power efficiency issue) the best configuration in term of sound quality ? Are the Lin topology, for example, far worse than cascaded class A stages ?

many thanks
Hi,

Nothing in the above makes much sense to me, so its impossible
to sensibly discuss class A in any real way in the above terminology.

The Lin topology is for a complete power amplifier, has its own
history, and not much to do with any issues related to class A.

I've no idea where you have got the 26mV idea from,
but whatever it is you have misunderstood it, AFAICT.

rgds, sreten.

26mV is related to the variation of Vbe with temperature.
__________________
There is nothing so practical as a really good theory - Ludwig Boltzmann
When your only tool is a hammer, every problem looks like a nail - Abraham Maslow

Last edited by sreten; 22nd January 2013 at 09:10 PM.
  Reply With Quote
Old 22nd January 2013, 09:09 PM   #8
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
I assume the OP has read that if you feed a BJT an XmV peak signal you get X% 2nd harmonic distortion, so he is puzzled how amplifiers can handle larger signals. I also assume he has just seen the classic textbook common-emitter stage, which almost never appears in real circuits since about 1965.
  Reply With Quote
Old 22nd January 2013, 09:24 PM   #9
sreten is offline sreten  United Kingdom
diyAudio Member
 
Join Date: Nov 2003
Location: Brighton UK
Hi,

Well one things for sure so far, nothing has been clarified.

rgds, sreten,
__________________
There is nothing so practical as a really good theory - Ludwig Boltzmann
When your only tool is a hammer, every problem looks like a nail - Abraham Maslow
  Reply With Quote
Old 23rd January 2013, 07:21 PM   #10
diyAudio Member
 
Join Date: Apr 2010
Location: Coffs Harbour
You appear to misunderstand the (small) difference between most practical class A and AB amplifiers.
I have the impression you believe that they are A or AB throughout, when normally both types operate
in class A up to the output stage. Perhaps I misunderstand what you are saying also but this type of
operation is true of most linear audio amplifiers, buffers and op-amps.

They can even be identical in every respect apart from just the bias current applied to the output stage.

If you are referring to text that perhaps is confusing, please post a reference, link etc if it is available in
English, as we may be able to assist.
__________________
regards
  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
Linearity rsumperl Tubes / Valves 15 3rd December 2010 06:45 PM
NFB Clarification cbj591 Tubes / Valves 9 17th July 2010 02:17 PM
An attempt for clarification. Lumba Ogir Solid State 1 4th March 2009 07:46 PM
Linearity maxpou Solid State 42 8th February 2008 10:44 PM
improve the linearity E007 Solid State 1 3rd May 2004 09:23 AM


New To Site? Need Help?

All times are GMT. The time now is 11:21 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