My New VAS Topology - 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 4th July 2013, 06:00 AM   #31
diyAudio Member
 
Join Date: Sep 2011
Here's the updated schematic

Here's how the overdrive protection / clamps work:

Zener diode D8 provides a starting voltage for the VAS bias such that the current is about 1mA above the target current we want. The bias action is now to engage Q15 just enough to shunt the zener into dropping bias voltage enough to reach the desired VAS current.

This change was needed so that when overdriven with a positive signal, the bias wouldn't try to pull down the bias voltage all the way leaving both legs without current and putting it into an indeterminate state. Now, however, if the VAS current reduces during overdrive, the bias shunt Q15 won't be invoked; It doesn't matter whether there's a positive or negative overdrive; the VAS current would go down as one device closes while the other opens up. The actual current is set by the closing device.

Now that the "maximum" VAS current is defined this way (target + ~1mA), this acts as protection too, and now the shunt clamps Q12/D5 and Q14/D9 will shunt excessive current from the LTP to the supplies on overdrive while preventing the VAS input from allowing too much current in the opening leg. By limiting this current, there will be current left in the closing leg still; the VAS thus will never be without current no matter how severe the input stage would be overdriven.

It works amazingly well I've did my obligatory THD 200K check and with minor tweaks to the inputstage, does 0.000% on the instrument still, that's below 0.0005%.

Is the actual schematic too hard to grasp? Is it too out of this world?
Attached Images
File Type: png VAS2.png (66.9 KB, 435 views)
__________________
Negative Feedback: The Need for Speed
  Reply With Quote
Old 4th July 2013, 02:44 PM   #32
diyAudio Member
 
Join Date: Jan 2013
Here is the basic concept of the vas. Q38,Q39 can be driven by either a diamond buffer or just conventional CFB input. In here lies the difference with your circuit. Patented by Grosser in 1991. A diamond buffer is much more linear than LTP so they didnt bother including more complexity to improve performance as they could obtain -70 db THD at 2 MHz.

I built a small 50w power amp around the circuit, works no problem although I couldnt find any benefit to what I hear compared to just a conventional CFB design with much lower complexity.
Attached Images
File Type: jpg Untitled34.jpg (88.6 KB, 428 views)
  Reply With Quote
Old 4th July 2013, 04:13 PM   #33
diyAudio Member
 
Join Date: Sep 2011
Yeah it looks similar but it's definitely different. In the VAS you show, the mirror outputs provide rails for the output bias transistors Q57 Q58 to 'glide' along as their bases are moved by Q54 and Q55's collector node, which seems to be the actual voltage output, rather than the mirror outputs are. It's a clever output stage I have to say. I'm sure the inputs of Q38 and Q39 can't be conncted - they need to have a voltage bias applied. But then again, I may be wrong in my understanding

You say a diamond buffer is more lineair contrary to an LTP. Are you suggesting the VAS you show has a diamond buffer as its inputstage, not shown?

I designed around a single LTP, contrary to a double symmetrical LTP to avoid having two control signal paths (error signals). One LTP gives me one push-pull output. This would allow the use of the 15pF compensation cap between the output and the VAS input having made it a single inverting input by tying the bases together of the VAS input devices. By implementing the signal paths this way is what made the THD figures drop dramatically. To have a push-pull VAS that has just one input. Another benefit is that I didn't need to hunt for P JFETs and try to obtain complementairy pairs having to match them.

The added complexity is mostly because of the auto bias feature I included. I wanted that to compensate for mirror errors, temperature drift and component differences. The prototype has no matched transistors, I just arbitrary grabbed them as needed

I'll have to try a diamond buffer input and see how that goes
__________________
Negative Feedback: The Need for Speed
  Reply With Quote
Old 4th July 2013, 09:45 PM   #34
jony is offline jony  Poland
diyAudio Member
 
Join Date: Jun 2004
Location: wroclaw
Quote:
Originally Posted by manso View Post
Patented by Grosser in 1991.
Can you post the patent number?
  Reply With Quote
Old 4th July 2013, 11:42 PM   #35
dimitri is offline dimitri  United States
diyAudio Member
 
dimitri's Avatar
 
Join Date: Oct 2002
Location: retired
Send a message via ICQ to dimitri
>>Grosser in 1991
Can you please mention pat #? Thanks
  Reply With Quote
Old 5th July 2013, 05:40 AM   #36
diyAudio Member
 
Join Date: Sep 2011
There, I made a different version without the autobias feature. I wanted to see how much it would reduce part count. I havev to admit though that at HF (100K) THD performance is worse with the bias circuit. It's not much and the performance is still miles and miles better than the fully summetrical amp I made. But without the bias circuit I was able to do 250KHz full power and still have 0.000% (> 100dB) THD. At 2MHz it was down to 76dB.

Without the bias, the VAS current will be subject to inaccuracies no more than your average run off the milll VAS. I turned the current mirrors into Wilson mirrors to ensure the mirror currents are as accurate as possible.

Attached you'll find the schematic without the bias feature. The current is now coarsely set by D8. It is equal to (5.6V - 4 x Vbe) / (R14 + R15).

This week I'll implement this current schematic on the plugboard and see if it still works as intended in reallife. But I have to order a bunch of various zener diodes; without them I can't build and test. Updates soon!
Attached Images
File Type: png VAS-NoBias.png (79.4 KB, 354 views)
__________________
Negative Feedback: The Need for Speed
  Reply With Quote
Old 5th July 2013, 09:51 AM   #37
diyAudio Member
 
Join Date: Jan 2013
Quote:
Originally Posted by jony View Post
Can you post the patent number?
Sorry gentlemen, I dont have the number. I just kept relevant schematics from the patents I deemed excellent or innovative. Its not too difficult to find though, use google patent search and search for Grosser. He has quite a few, many of AD current range of very high performance opamp designs are his doing.
  Reply With Quote
Old 5th July 2013, 10:35 AM   #38
diyAudio Member
 
Join Date: Apr 2010
Location: Bath, UK
Interesting about the Patent, however I seem to recall a circuit very similar to the latest version posted by MagicBox from the 1970's. without the FET input stage, which was used for a high-power Amplifier design here in the UK at that time. Might be some 'Prior Art' in this?
  Reply With Quote
Old 5th July 2013, 11:09 AM   #39
dimitri is offline dimitri  United States
diyAudio Member
 
dimitri's Avatar
 
Join Date: Oct 2002
Location: retired
Send a message via ICQ to dimitri
4,970,470 DC-coupled transimpedance amplifier
  Reply With Quote
Old 5th July 2013, 11:27 AM   #40
diyAudio Member
 
Join Date: Sep 2011
Quote:
Originally Posted by audiomik View Post
Interesting about the Patent, however I seem to recall a circuit very similar to the latest version posted by MagicBox from the 1970's. without the FET input stage, which was used for a high-power Amplifier design here in the UK at that time. Might be some 'Prior Art' in this?
Yeah the circuit could operate without input stage. It's just that the bases need to be bias'ed around ground. I've found though that with the additional gain of an input stage the combination (input + VAS) yields much lower distortion. It however doesn't do anything to bandwith which is limited mainly by Cmiller of the VAS output.
__________________
Negative Feedback: The Need for Speed
  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
Differential VAS topology danny92 Solid State 13 11th June 2013 06:03 PM
Bob Cordell's Complementary IPS-VAS topology danny92 Solid State 30 20th May 2013 05:15 PM
Differential VAS topology Bob Cordell based danny92 Solid State 0 6th May 2013 04:41 AM
A superior VAS clamp and VAS current limit for the Blameless. GK Solid State 128 19th October 2011 03:03 PM
low VAS sealed better than high VAS? matos81 Subwoofers 3 7th April 2006 11:28 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