A different CAS typology with very low THD - 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 14th February 2011, 01:22 PM   #1
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
Default A different CAS typology with very low THD

I wanted to share what I've been working on for the past year or so. I been following the "The very best amplifier I've ever heard" (TVBAIEH) thread and it got me thinking. The best amp I've heard is the Allison variant that several folks hear helped with. When I say helped with, they came up with the scheme, I built and tested it. I want to acknowledge Keantoken, PaulBysouth and KenPeter who all helped immeasurably. You can read about the development of this amp on the following: (this configuration starts at post #308) Simulation Analysis of several unique Allison-based output stages.

I've compared this amp against a triple emitter follower and a nice 50watt Rotel amp. I can't claim to have heard all the world's great amps, but, this one is just a bit smoother than the others. I really enjoy music that features the human voice, from opera, to Frank Sinatra and everything in between. This amp helps those voices shine! I know these opinions are subjective, so, that's what you get...

To test the amp in simulation; I've compared it to Bob Cordell's Triple emitter follower Figure 3.10 on page 65 of his book. Using Bob's model files and Thd setup, I've put together the Figure 3.10 as a baseline and used the front end of this amp to graft the modified Allison on as the CAS.

I then ran each in simulation at various input voltages to get a picture of THD.

A note on the modified Allison, it runs in class A up to about 16watts into 8 ohms in this configuration, but, it can be easily adjusted by changing out one resistor.

Got to go, a bit more later.
Attached Images
File Type: png Triple_emitter_follower_fig_3.10.png (50.3 KB, 1066 views)
File Type: png Allison_with_cordell_front_end.png (57.2 KB, 1056 views)
File Type: png Allison_vs_Triple_emitter_follower.png (40.6 KB, 1003 views)
  Reply With Quote
Old 14th February 2011, 02:09 PM   #2
Telstar is offline Telstar  Italy
diyAudio Member
 
Join Date: Dec 2007
Location: Italy
Can you run other simulations?
I'd like to see HD2 to 9 at 10khz, 20khz and 50khz (at just 1W would be fine).

Besides, I'm more interested in your listening opinions than the sims. Which which amplifiers (or better, output stages connected to the same frontend) did you compare it?
__________________
"The total harmonic distortion is not a measure of the degree of distastefulness to the listener and it is recommended that its use should be discontinued." D. Masa, 1938
  Reply With Quote
Old 14th February 2011, 03:41 PM   #3
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
Quote:
Originally Posted by Telstar View Post
Can you run other simulations?
I'd like to see HD2 to 9 at 10khz, 20khz and 50khz (at just 1W would be fine).

Besides, I'm more interested in your listening opinions than the sims. Which which amplifiers (or better, output stages connected to the same frontend) did you compare it?
Yes, I'll run more simulations.

I have been using an LME49811 for the front end on all my test amps. It's on a separate board that plugs into the CAS board, so that I can swap out different typologies easily. In all I have built 7 or 8 different CAS typologies - some minor variations on others. I compared it to a pure Class A Allison, a double emitter follower and a triple emitter follower. The guys from National say that the 49811 contains pre-drivers, so that a triple isn't required, but, who knows? In comparison to the double emitter follower with the same type of driver and output transistors, re resistors, etc. the modified Allison made the voices more natural, more open and less strident. It's tough to describe these things, the double emitter follower sounded great, the modified Allison just sounded better in comparison over many hours of listening.

Ken
  Reply With Quote
Old 14th February 2011, 05:02 PM   #4
Telstar is offline Telstar  Italy
diyAudio Member
 
Join Date: Dec 2007
Location: Italy
Oh, so you are the guy that Anthony talked me about.

Same output transistors?
Now much NFB?
__________________
"The total harmonic distortion is not a measure of the degree of distastefulness to the listener and it is recommended that its use should be discontinued." D. Masa, 1938
  Reply With Quote
Old 14th February 2011, 06:13 PM   #5
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
Quote:
Originally Posted by Telstar View Post
Oh, so you are the guy that Anthony talked me about. Yup that would be me, hopefully he said nice things.Same output transistors? Same type onsemi 3281/1302's, thermal trak version for the double emitter follower configuration.Now much NFB?
  Reply With Quote
Old 14th February 2011, 06:21 PM   #6
Telstar is offline Telstar  Italy
diyAudio Member
 
Join Date: Dec 2007
Location: Italy
I meant how much NFB? Same in both cases?
__________________
"The total harmonic distortion is not a measure of the degree of distastefulness to the listener and it is recommended that its use should be discontinued." D. Masa, 1938
  Reply With Quote
Old 14th February 2011, 06:30 PM   #7
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
Quote:
Originally Posted by Telstar View Post
I meant how much NFB? Same in both cases?
As far as I know, though I believe that Keantoken mentioned that the Allison has a kind of local NFB loop. But the NFB is set up like the simulation. The gain setting resistors for both configuration.
  Reply With Quote
Old 15th February 2011, 01:55 AM   #8
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
Default hd2-9

The following are the HD2-10 results at 1k, 10k, 20k and 50k respectively and the FFT at 1k. All at approx. 1 watt output.
Attached Images
File Type: png HD2-9_1k.png (16.4 KB, 827 views)
File Type: png HD2-9_10k.png (18.0 KB, 774 views)
File Type: png HD2-9_20k.png (17.4 KB, 88 views)
File Type: png HD2-9_50k.png (17.3 KB, 92 views)
File Type: png FFT_at_1k_1watt.png (32.6 KB, 166 views)
  Reply With Quote
Old 15th February 2011, 02:14 AM   #9
diyAudio Member
 
Join Date: Feb 2008
Location: Melbourne
Klewis,
Referring to the second schematic in post #1 (ie the Allison circuit), if you add a resistor (of about 100 ohms) between the emitters of Q11 and Q18, then these output transistors will never turn off and the only switching in the circuit will be the two Schottky diodes. I doubt that you'll be able to measure any difference though.

Some other things you might like to try:
- Use irf610 and irf9610 Mosfets for the drivers (in place of Q3, Q6), or
- swap the output transistors (Q11 and Q18) for some big mosfets (irf240 and irf9240 or similar).
Don't forget the gate stopper resistors.

Paul Bysouth
  Reply With Quote
Old 15th February 2011, 04:30 AM   #10
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
Quote:
Originally Posted by PaulBysouth View Post
Klewis,
Referring to the second schematic in post #1 (ie the Allison circuit), if you add a resistor (of about 100 ohms) between the emitters of Q11 and Q18, then these output transistors will never turn off and the only switching in the circuit will be the two Schottky diodes. I doubt that you'll be able to measure any difference though.

Some other things you might like to try:
- Use irf610 and irf9610 Mosfets for the drivers (in place of Q3, Q6), or
- swap the output transistors (Q11 and Q18) for some big mosfets (irf240 and irf9240 or similar).
Don't forget the gate stopper resistors.

Paul Bysouth
Hi Paul,

Nice to hear from you again. Your help in the past has been great! I think you meant Q17 & Q18, I'll give this a try. The resistors that feed the emitters of these transistor are different values. I found that if you tune these so that the collector currents are equal, it drives the distortion down further. My implementation plan includes a pot at one of these resistors to balance the current, ultimately replaced by a fixed resistor.

I've actually been using mje15034/35's for the drivers, they are much more stable than the 2sc4793/a1837 used in the simulation. They also have a close match pnp to npn (keantoken's suggestion). What do you see as the are the benefits of the mosfets for drivers and outputs?

Ken
  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
Ultra low noise amp with JFET's - How to reduce the THD ?? toogool Solid State 9 21st May 2010 09:54 AM
Published/DIY ultra-low THD designs? leadbelly Tubes / Valves 10 2nd April 2007 03:41 AM
Look at this 42V/20A max PSU, and at this low THD oscillator. rephil Everything Else 3 6th August 2006 08:04 PM
low THD guitar preamp Antonix Tubes / Valves 4 29th June 2006 04:49 PM


New To Site? Need Help?

All times are GMT. The time now is 11:40 PM.


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