Class i and siblings - Page 4 - 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 21st December 2011, 08:13 PM   #31
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Default Not very promising.

Hi Damir,

I'm not surprised at all, as you are torturing the OPS. For two reasons you got ugly results:

1. Drivers for the top respectively the bottom OP devices consist of only a single EF stage instead of a push-pull stage. Regrettably, it's a bit difficult to implement the latter, as such arrangement will compromise the headroom of the bias circuit. (Therefore, I prefer vertical MOSFETs, with a much higher Vgs than the Vbe of a BJT). Sliding bias schemes are crying for separate push-pull drivers, as a speed-up capacitor (between the emitters) is forbidden. If you still insist to use BJTs, well, it has a prise.

2. Because of the limited speed of the internal loops inside AB2 (the ULGF is 'only' 3MHz) you must filter the transients of a square wave test signal drastically, preferable by means of a 2nd order Butterworth low pass filter with a Fc of around 250kHz. In this way the slew rate, or more precisely the rise and fall time is kept to a safe value. Now, it may look as if the slew rate is limited (to 50V/us or so), but actually, we are dealing here with a finite rise and fall time. Nothing wrong with that as long as it is determined by a filter and not by deficiencies of the amplifier itself.
BTW, in case of a 1st order filter, you have to chose Fc much lower. That's why I recommend a 2nd order filter.
(hope this helps)

Cheers,
E.
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)
  Reply With Quote
Old 21st December 2011, 08:21 PM   #32
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Quote:
Originally Posted by miralin View Post
Hi Edmond
Once again my poor English leads to possible misunderstanding. It is because the posts to this thread are the first attempts of English language conversation on technical issues for all my life. I’ll make another try some days later.
Cheers Mir
Hi Mir,

No problem and don't give up.
Drawing a schematic will also help to avoid confusion. Do you have a scanner? If so, just draw it with pencil and paper.

Cheers,
E.
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)
  Reply With Quote
Old 21st December 2011, 09:06 PM   #33
Waly is offline Waly  United Kingdom
diyAudio Member
 
Waly's Avatar
 
Join Date: Jan 2011
Location: London
Quote:
Originally Posted by Edmond Stuart View Post
(...) you must filter the transients of a square wave test signal drastically, preferable by means of a 2nd order Butterworth low pass filter with a Fc of around 250kHz. In this way the slew rate, or more precisely the rise and fall time is kept to a safe value. Now, it may look as if the slew rate is limited (to 50V/us or so), but actually, we are dealing here with a finite rise and fall time. Nothing wrong with that as long as it is determined by a filter and not by deficiencies of the amplifier itself(...)
To avoid the overshooting, I was wondering why "filter the transients of a square wave test signal drastically" is here acceptable, while for a two pole compensated amp it's "a fake solution": Bob Cordell's Power amplifier book
  Reply With Quote
Old 21st December 2011, 09:19 PM   #34
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Shut up Walter (alias Waly, alias YWN) and mind your own business.
If you don't know what you are talking about keep quiet
or go trolling elsewhere, please.
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)

Last edited by Edmond Stuart; 21st December 2011 at 09:30 PM.
  Reply With Quote
Old 22nd December 2011, 08:02 AM   #35
ontoaba is offline ontoaba  Indonesia
diyAudio Member
 
ontoaba's Avatar
 
Join Date: Nov 2009
Location: Kudus, Malang, Dieng
Hi Edmond,

Is that what transistor is replaceable with cheaper, I mean is it ok if I replace them with 3904-3906?

Damir VAS is saturated in squarewave test, and it is three stages so it need more time to recover. Ofcourse Audio amp won't eat any square waves, using filter will help.
  Reply With Quote
Old 22nd December 2011, 09:18 AM   #36
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Quote:
Originally Posted by ontoaba View Post
Hi Edmond,

Is that what transistor is replaceable with cheaper, I mean is it ok if I replace them with 3904-3906?
Hi Ontoaba,

Sure, that is, if you take the time and effort to (hand) select them for tight matching of Vbe, within a few mV.

As an aside (and joke), one might argue that, given the time to select the right trannies, AB2 doesn't deserve the name 'auto bias' any longer, as setting the bias by means of a trim pot takes far less time.

Quote:
Damir VAS is saturated in square wave test, and it is three stages so it need more time to recover. Of course Audio amp won't eat any square waves, using filter will help.
You're right, amps won't eat any square waves. Tortured with infinite fast transients any amp will be pushed beyond its limits.
In case of this amp I don't know yet what causes it to get 'berserk'. Saturation, lack of headroom of the bias circuit, missing Baker clamps? With vert. MOSFETs I got much higher slew rates. Maybe AB2 doesn't like BJT output devices anyhow. I have to figure that out.

Cheers,
E.
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)
  Reply With Quote
Old 22nd December 2011, 12:17 PM   #37
Waly is offline Waly  United Kingdom
diyAudio Member
 
Waly's Avatar
 
Join Date: Jan 2011
Location: London
Quote:
Originally Posted by Edmond Stuart View Post
(...)amps won't eat any square waves. Tortured with infinite fast transients any amp will be pushed beyond its limits.
As far as I can tell, from both theoretical and practical perspectives, this is incorrect. Both bandwidth and slew rate limitations are not good reasons for any amp to go "beserk", as long as signal levels are beyond clipping.

Care to support your statement?

The reason for the overshooting response of your autobias scheme is to me pretty obvious: the forward and the nonlinear reverse autobias feedback paths have local close coupled poles. And, to add insult to injury, the forward path is also nonlinear, as the forward pole is signal level dependent (being mostly defined by the mosfet Miller capacitance, strongly nonlinear when Vds approaches zero), hence a worsening in the step response at clipping onset.

Unfortunately, although it is possible to separate these poles (the LT1166 circuit mentioned in another thread is a good example, with the loop input filter, per the datasheet application example), but then most of the spectacular distortion reduction effect would be lost. A trade is required here, don't know how far it can be pushed. As it is though, I don't think the current step response is acceptable for audio applications.
  Reply With Quote
Old 22nd December 2011, 02:12 PM   #38
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Default push-pull drivers

Quote:
Originally Posted by dadod View Post
Hi Edmond,
I did some simulation with square wave, and it is not very encouraging. If I test with a bit higher input signal it vent berserk.
[snip]
Damir
Hi Damir,

As I explained elsewhere, in order to discharge the gates or bases of the OP devices, you really need separate push-pull drivers, as you can't use the ubiquitous speedup cap. Without them, of course the OPS 'went berserk' when subjected to an unconditioned square wave.

In the meantime I've figured out how to implement such drivers* (without compromising the headroom of the bias circuit). Now, the square wave response is much better. You still need a filter (anyhow), but I could increase the cutoff frequency to 500kHz with out any troubles. Without filter, hardly any overshoot (rather 'undershoot') and more seriously, an enormous cross-conduction. Needless to say to avoid that at any time.

Below the SW response at 50 KHz and 30Vpk. Blue = Vout; Red = Ie; Max. SR = 100V/us. Notice that there is still some cross-conduction: about 1A, though during a very short period of time: less than 1us.

* at the expense of 4 additional trannies, two for the drivers and two for the bias spreaders.

Cheers,
E.

PS: I'm sorry for the interference by Mr Troll. I've put him on my ignore list. Hope it helps.
Attached Images
File Type: png AB2-BJT-PP-SWR.png (8.5 KB, 264 views)
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)
  Reply With Quote
Old 22nd December 2011, 02:54 PM   #39
dadod is offline dadod  Croatia
diyAudio Member
 
dadod's Avatar
 
Join Date: Apr 2006
Location: Zagreb
Hi Edmond,
Could you show the circuit?
Damir
  Reply With Quote
Old 22nd December 2011, 03:18 PM   #40
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Of course, but first I like to do some more sims.

Cheers,
E.
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)
  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
Class-AB meets Class-D: Yamaha's EEEngine Topology - where are this Diy Projects? tiefbassuebertr Solid State 29 14th April 2014 04:12 PM
Collection of Class B topologies <100mA Idle and Sound closest by Class A tiefbassuebertr Solid State 37 27th July 2012 08:04 AM
Can a Class AB PP amp be said to be operating in Class A at low signal levels? ray_moth Tubes / Valves 19 23rd January 2009 07:52 PM
How about a round-up of Class A kit power amps, or collectable vintage class A? Brisso57 Solid State 4 14th February 2007 10:30 AM


New To Site? Need Help?

All times are GMT. The time now is 09:14 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