Class B w/o crossover distortion (1975) - Page 11 - 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 11th February 2010, 06:17 PM   #101
hahfran is offline hahfran  Germany
diyAudio Member
 
Join Date: Dec 2007
Quote:
Originally Posted by Elvee View Post
If the cap had 15V, this would mean that all the bias current goes into the string of predriver transistors, with nothing left for Q2.
It would be a hard-switching class B.
Provided the current source Q7 drives 15 mA then
at the collector of Q7 the voltage must be -15 v and at the base of Q5 the voltage is 0 v.
  Reply With Quote
Old 11th February 2010, 08:00 PM   #102
hahfran is offline hahfran  Germany
diyAudio Member
 
Join Date: Dec 2007
Ok doing the sim I confirm your findings seems the value of C is critical
  Reply With Quote
Old 11th February 2010, 08:41 PM   #103
hahfran is offline hahfran  Germany
diyAudio Member
 
Join Date: Dec 2007
With this clamp already proposed by Hugh the thd is down to 0.3% at 15 v in 1 kHz
It can be reduced further but depends on property of power BJT.
The odd harmonics are one order of magnitude lower than the even. I guess that sounds good.
Attached Images
File Type: jpg vishm.jpg (85.3 KB, 441 views)

Last edited by hahfran; 11th February 2010 at 08:54 PM.
  Reply With Quote
Old 11th February 2010, 09:00 PM   #104
AKSA is offline AKSA  Australia
diyAudio Member
 
Join Date: Sep 2001
Location: Melbourne, Australia
Hi Hahfran,

Quote:
The odd harmonics are one order of magnitude lower than the even. I guess that sounds good.
Yes, generally my experience too. The best profile, according to Jean Hiraga twenty years ago, is the monotonic decreasing profile, where successive harmonic peaks form a straight line down to about -140dB.

BTW, is Q6 in your schemat NPN (it may be an error?)

Hugh
__________________
Aspen Amplifiers P/L (Australia)
www.aksaonline.com
  Reply With Quote
Old 11th February 2010, 09:58 PM   #105
hahfran is offline hahfran  Germany
diyAudio Member
 
Join Date: Dec 2007
Not quite...I've been testing the reliability of LTSpice fourier computation and the schematic shown was such a test....thd significantly improved over the correct circuit...
at least according to Spice ...
Yes Hiraga's rule has been confirmed in numerous listening tests.
  Reply With Quote
Old 11th February 2010, 10:58 PM   #106
AKSA is offline AKSA  Australia
diyAudio Member
 
Join Date: Sep 2001
Location: Melbourne, Australia
Hahfran,

For Q6 to pass a current into Q7's base, its base must be 0.6V lower than its emitter, not higher, and since the 68R resistor is connected to the emitter of the output device Q7, no base current can flow. If LTSpice is working with this connection, there is a bug.....

I really think Q6 should be a 2N5551 (and in fact I'd change it to a beefier device like a 2SC4793, MJE340 or MJE15030).

To properly simulate the outsized BE junction on Q7, diode D2 might be something capable of 1A. I was never sure about the cap; 12nF seems about right.

Cheers,

Hugh
__________________
Aspen Amplifiers P/L (Australia)
www.aksaonline.com
  Reply With Quote
Old 12th February 2010, 07:46 AM   #107
hahfran is offline hahfran  Germany
diyAudio Member
 
Join Date: Dec 2007
Quote:
Originally Posted by AKSA View Post
Hahfran,

For Q6 to pass a current into Q7's base, its base must be 0.6V lower than its emitter, not higher, and since the 68R resistor is connected to the emitter of the output device Q7, no base current can flow. If LTSpice is working with this connection, there is a bug.....

I really think Q6 should be a 2N5551 (and in fact I'd change it to a beefier device like a 2SC4793, MJE340 or MJE15030).

To properly simulate the outsized BE junction on Q7, diode D2 might be something capable of 1A. I was never sure about the cap; 12nF seems about right.

Cheers,

Hugh
Yes there is a bug... LTSpice does the dc setup perfectly and continues with the harmonics analysis. I have been using LTSPice "certified" models to be sure there is no .model error.
PSpice does not find a dc operating point.
I had been working on theory of semiconductors once and hence am very wary of dynamic simulations...one should always have a nice piece of artwork
for ex. a decorated flask filled with enough grains of salt on the desk when
working with sims....
  Reply With Quote
Old 12th February 2010, 08:10 AM   #108
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: Great City of Turnhout, Belgium
Blog Entries: 7
Quote:
Originally Posted by hahfran View Post
Yes there is a bug... LTSpice does the dc setup perfectly and continues with the harmonics analysis. I have been using LTSPice "certified" models to be sure there is no .model error.
PSpice does not find a dc operating point.
I had been working on theory of semiconductors once and hence am very wary of dynamic simulations...one should always have a nice piece of artwork
for ex. a decorated flask filled with enough grains of salt on the desk when
working with sims....
Some/most sims (not sure about LTspice) do small-signal ac analysis (which includes harmonic analysis IIRC) around a linearized bias point (that's why it is called small-signal analysis). This often works even with the DC conditions way wrong. I wouldn't call it a bug, it's the nature of the beast.

jd
__________________
If you don't change your beliefs, your life will be like this forever. Is that good news? - W. S. Maugham
Check out Linear Audio!
  Reply With Quote
Old 12th February 2010, 08:39 AM   #109
hahfran is offline hahfran  Germany
diyAudio Member
 
Join Date: Dec 2007
Yes the sims can hardly do more, in terms of large signal ac analysis. Large signal would work well with triodes where i_anode= const*Vgc^3/2 is a sound approximation ( except for the small range where i_anode is almost 0) ...alas those semiconductors are somehow
"quantum-mechanical"... electron states .... quantum wells ..non linear differential equations...none of that applies to tubes ..but could tubes operate in the vicinity of black holes or will all those electrons get sucked up by the black hole before they reach the plate? Fortunately none of those questions are valid if it is about a Keith Jarret solo concert these are the fundamental things
  Reply With Quote
Old 12th February 2010, 08:57 AM   #110
Elvee is offline Elvee  Belgium
diyAudio Member
 
Elvee's Avatar
 
Join Date: Sep 2006
Quote:
Originally Posted by hahfran View Post
With this clamp already proposed by Hugh the thd is down to 0.3% at 15 v in 1 kHz
It can be reduced further but depends on property of power BJT.
The odd harmonics are one order of magnitude lower than the even. I guess that sounds good.
I arrive at a much lower distortion figure under steady state conditions.
But you have to take care that the circuit has effectively reached its steady state before you make the measurement.
What I do is to run the circuit for a very long time (1s) with the intended signal level, and then, I look at final voltage across the cap and include .ic statements forcing that voltage at the next run.
If you rely on a .op, it wont work, because the bias level is dependent on the dynamic operating conditions.

With 15Vpk, the distortion is under 0.02%, almost pure 2nd harmonic.

But the circuit has a sweet spot regarding the source resistance: with 15K at the input, the figure is down to 0.003%.
Note that on your schematic, the cap value is 0.22 farad.

BTW, simulations are no evil, you have to understand what is going on and make sure it follows exactly the path you intend.
They are simply a tool, useful but dumb, and relying too much on default settings may give incoherent results. You have to make yourself the critical decisions.


AKSA,

Q6 cannot be a NPN: this would make the positive half-side inverting, and it would conflict with the non-inverting negative side.
The topology is in fact a circlotron with PNP composites at the output, and the Rush cascode as the phase shifter.
  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
low crossover-distortion class AB output stage topologies capslock Solid State 72 10th May 2013 08:48 PM
Exactly titles of old transistor data books from NEC, SONY and Fujitsu (1975 - 1985) tiefbassuebertr Solid State 5 12th December 2011 06:31 PM
Crossover distortion - be gone! Circlotron Solid State 4 24th August 2009 01:09 PM
Crossover distortion lumanauw Solid State 100 14th January 2005 01:58 AM
Crossover Distortion? Killjoy99 Multi-Way 8 30th March 2004 06:10 AM


New To Site? Need Help?

All times are GMT. The time now is 06:27 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