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 20th October 2013, 04:22 AM   #121
fab is offline fab  Canada
diyAudio Member
 
fab's Avatar
 
Join Date: Mar 2004
Location: Canada
The SKA does not have a VAS. As MIIB I believe that a filter can be put in the input stage to reduce power supply variation injection. My Hiraga has a capacitor bank comparable to a powerful class A/B amp without issue and the Hiraga is a class A.

I am not saying that only CFA amps have good soundstage. I have heard VFA amps with very good soundstage too. But soundstage seems a common characteristic of CFA Amps I (and others) have heard. I am glad to hear that the SKA in VFA has a huge soundstage since it could show great potential it is not a question of choosing between VFA or CFA but to build both!

Fab

Last edited by fab; 20th October 2013 at 04:26 AM.
  Reply With Quote
Old 20th October 2013, 12:30 PM   #122
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Blog Entries: 2
Interesting comments about common drain output. Perhaps there is a lot about the sound that is related to this aspect of the topology. The SKA has very good reputation for sound. The Hiraga also. The Pass F5 too. And Rod Elliot's P3A.

In my way of thinking though, the SKA does have a VAS, it is the output stage ! - crazy when you think about it. But in the case of the SKA you can't put a filter on the power rails for the input stage because the PSRR is not limited by the input stage, it is a function of the output stage too, that common drain thing again.

I don't know if I have the time to build both CFA and VFA version - they need a good pcb to do it properly and the topologies would be hard to accommodate on one pcb.
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig.
  Reply With Quote
Old 20th October 2013, 02:13 PM   #123
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Blog Entries: 2
Attached is my simulation file for the VFA and VFA versions. I've tried to set them up so that they are fairly close in operating points but also practical.

I do like the CFA. However, I am getting 6dB worse distortion for 3rd and 5th harmonics. It looks as if the LTP is more linear.

If this isn't what you'd expect - where is my 'error' ?
Attached Files
File Type: asc TGM7-II-CFA.asc (29.5 KB, 32 views)
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig.
  Reply With Quote
Old 20th October 2013, 02:35 PM   #124
MiiB is online now MiiB  Denmark
diyAudio Member
 
Join Date: May 2005
Location: Denmark
In the CFA input you are loosing most of the input bias current through the feedback network, I would separate AC and DC and AC couple the feedback.
  Reply With Quote
Old 20th October 2013, 03:54 PM   #125
fab is offline fab  Canada
diyAudio Member
 
fab's Avatar
 
Join Date: Mar 2004
Location: Canada
Quote:
Originally Posted by Bigun View Post
Interesting comments about common drain output. Perhaps there is a lot about the sound that is related to this aspect of the topology. The SKA has very good reputation for sound. The Hiraga also. The Pass F5 too. And Rod Elliot's P3A.

In my way of thinking though, the SKA does have a VAS, it is the output stage ! - crazy when you think about it. But in the case of the SKA you can't put a filter on the power rails for the input stage because the PSRR is not limited by the input stage, it is a function of the output stage too, that common drain thing again.

I don't know if I have the time to build both CFA and VFA version - they need a good pcb to do it properly and the topologies would be hard to accommodate on one pcb.
If you consider the output stage of SKA as a VAS then the output stage is the same for the VFA or CFA and they( output stage) suffer the same supply variation. Thus filtering the input stage would improve PSSR of the CFA but surely not as good as the VFA. You mean common source thing....

Yes 2 different pcbs for VFA and CFA would be more appropriate but trying to be as close as possible for the common circuit part....

Fab
  Reply With Quote
Old 20th October 2013, 05:26 PM   #126
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Blog Entries: 2
Quote:
Originally Posted by MiiB View Post
In the CFA input you are loosing most of the input bias current through the feedback network, I would separate AC and DC and AC couple the feedback.
I assume you mean the CCS current is being lost not input bias current - the input bias is just what is needed for the bases of the input devices. The CCS current we can adjust, we can throw it away if we want. The only problem is that as we crank it up we have to use more robust transistors for the CCS. Thing is, just putting in caps to make it an AC coupled feedback doesn't help the distortion. I did figure out it is simply a case of lower OLG. The CFA front end I have used loses gain which I can recover if I further reduce the impedance of the feedback network. In this case I go with the AC coupled approach to avoid high CCS currents. I can get back the gain depending how far I want to reduce the feedback impedance. It directly affects stability though, less impedance makes it faster.

Quote:
Originally Posted by fab View Post
If you consider the output stage of SKA as a VAS then the output stage is the same for the VFA or CFA and they( output stage) suffer the same supply variation. Thus filtering the input stage would improve PSSR of the CFA but surely not as good as the VFA.
It doesn't work - you can't put RC filters between the input and output stage in this topology - if you play with it in spice you'll see that the PSRR can even get worse.
Attached Files
File Type: asc TGM7-II-CFA(2).asc (42.7 KB, 23 views)
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig.
  Reply With Quote
Old 20th October 2013, 07:45 PM   #127
fab is offline fab  Canada
diyAudio Member
 
fab's Avatar
 
Join Date: Mar 2004
Location: Canada
Quote:
Originally Posted by Bigun View Post
...
It doesn't work - you can't put RC filters between the input and output stage in this topology - if you play with it in spice you'll see that the PSRR can even get worse.
I do not use spice but Microcap. Can you do a PDF of your circuit.
Where do you put the filter?

Fab
  Reply With Quote
Old 20th October 2013, 08:34 PM   #128
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Blog Entries: 2
OK - here's an image of a simplified circuit to show where I tried adding RC filters - of course you add them in pairs to both + and - rail. I show 3 locations. None of them work to improve PSRR, and two locations actually make it worse.

Here's why: the input signal comes in from stage left, it's referenced to ground. It is fed to the bases of the input transistors.

All transistors are essentially a voltage controlled device, a differential signal between base-emitter controls current through collector-emitter.

The emitters of the input devices are also referenced to ground through the feedback network. So the input signal modulates the current through the input devices.

This modulated current flows through the collector resistors to generate a voltage across these resistors. This amplified voltage is the difference in potential between the collector of the input transistor and the other side of the collector load, which is the power rail.

So the input pair convert the signal reference from ground over to the power rails. The power rails have essentially become the a.c. 'ground' for the signal at this point in the circuit.

Ignoring the emitter follower buffer for now, the differential signal across the input pair collector loads is applied to the gate-source of the output FETs. The output FET source is also referenced to the power rail. This input signal to the FET modulates the current through the FET and hence through the Drain and into the load. So finally, the signal to once again referenced to ground through the load (speaker).

So where the power rail is acting as a.c. ground for the signal between the input pair and the FETs you can't put a filter in it. If you put an RC filter in the supply rail at this point then any rail noise from the power supply will produce a voltage drop across the R of the filter. This voltage drop will add to the signal a.c. 'ground' between the input pair and FETs. The FETs don't know the difference between the real signal and a voltage drop across the R of a filter. So the rail noise is injected into the output. The PSRR gets worse with a filter.

An RC filter to isolate the current sources doesn't help either since the current sources are already well decoupled from the supply rails. It doesn't cause any harm here though.

Keep going Fab - let's not give up, let's find a way to make this a workable solution or decide we can live with the PSRR as it is ?
Attached Images
File Type: jpg TGM7-II-CFA(2) temp.jpg (200.2 KB, 328 views)
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig.

Last edited by Bigun; 20th October 2013 at 08:39 PM.
  Reply With Quote
Old 21st October 2013, 06:17 AM   #129
MiiB is online now MiiB  Denmark
diyAudio Member
 
Join Date: May 2005
Location: Denmark
If you want to get lower distortion and better control over the idle in the output stage i will suggest to mirror the signal into Q4 and Q3. (but what to do with the capacito feed forward...?

Another option could be to make a CFP input, this will lower distortion app 20-30 dB
  Reply With Quote
Old 21st October 2013, 12:36 PM   #130
fab is offline fab  Canada
diyAudio Member
 
fab's Avatar
 
Join Date: Mar 2004
Location: Canada
Quote:
Originally Posted by Bigun View Post
?..

Keep going Fab - let's not give up, let's find a way to make this a workable solution or decide we can live with the PSRR as it is ?
We need R=0 and very big cap

I do not have a big problem with the actual PSSR...

Fab
  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
Wanted SKA amplifier kit Greg Ball Dufus Swap Meet 7 29th July 2011 10:14 PM
Greg Ball (Amplifier Guru)'s SKA GB300 completed PCB for sale HiFiNutNut Swap Meet 4 23rd March 2011 07:14 AM
Greg Ball SKA 300D amplifier nycavsr2000 Swap Meet 4 1st February 2010 06:12 AM
Greg Ball SKA / contruction & application notes stoeffle Solid State 5 2nd April 2006 01:18 PM


New To Site? Need Help?

All times are GMT. The time now is 05:22 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