Symmetrical folded cascode. - 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 14th October 2004, 06:23 PM   #101
jcarr is offline jcarr  United States
diyAudio Member
 
Join Date: Jul 2002
Location: Tokyo, Japan
MikeB: Interesting results. I've been working on a power amp project and have been experimenting with a variety of topologies. I now have two final candidates, which are quite similar in their overall approach - non-complementary JFET input with BJT cascode, a variant of a CFP (Sziklai) output stage, output equipped with zobel but no output resistor/inductor network, DC coupling, and global NFB. The power supply is a low-noise switching-mode design.

The prime topoligical differences between the candidates occurs in the sections between the input and output - one version uses a VAS with cascode, and the other uses a folded cascode (with active current sources rather than resistors). Both versions are completely stable into all loads that I have tested them with. Both are pretty good, but on the whole, I find the folded cascode preferable. The top end sounds cleaner and more pure, and everything about the sound is more solid, including spatial imaging. OTOH, I may still choose the cascoded VAS, as the results are quite acceptable, and it is simpler and appears to be somewhat more bulletproof when abused.

Maybe the choice of output stage (Darlington or Sziklai) favors one intermediate-stage topology over another.

regards, jonathan carr
__________________
http://www.lyraconnoisseur.com/, http://www.lyraaudio.com
  Reply With Quote
Old 14th October 2004, 06:34 PM   #102
The one and only
 
Nelson Pass's Avatar
 
Join Date: Mar 2001
Quote:
Originally posted by jcarr
Maybe the choice of output stage (Darlington or Sziklai) favors one intermediate-stage topology over another.
The folded cascode tends to have less gain and wider bandwidth
and thus is usually more frequency stable in feedback loops.
I am of the opinion that the Sziklai are more sensitive to stability
issues, so you can easily imagine complementary (no pun
intended) qualities. On the other hand, folded cascode works
great with output followers too, as in the X600.
  Reply With Quote
Old 15th October 2004, 01:48 AM   #103
diyAudio Moderator Emeritus
 
lumanauw's Avatar
 
Join Date: Oct 2002
Location: Bandung
Send a message via Yahoo to lumanauw
Hi, Mike,

Quote:
and the other uses a folded cascode (with active current sources rather than resistors).
Have you tried my suggestion for your FC here ?http://www.diyaudio.com/forums/showt...060#post482060 Suggestion no.3

Maybe putting CCS instead of resistor in your FC will change your final result one more time
  Reply With Quote
Old 15th October 2004, 02:09 AM   #104
thanh is offline thanh  Viet Nam
diyAudio Member
 
thanh's Avatar
 
Join Date: Jan 2004
Location: ho chi minh city
Send a message via Yahoo to thanh
Quote:
OTOH, I may still choose the cascoded VAS, as the results are quite acceptable, and it is simpler and appears to be somewhat more bulletproof when abused
yes , or I can use mr evil's topo
__________________
Justice for Victims of Agent Orange
http://www.petitiononline.com/AOVN/Thank all of you!
  Reply With Quote
Old 15th October 2004, 02:15 AM   #105
thanh is offline thanh  Viet Nam
diyAudio Member
 
thanh's Avatar
 
Join Date: Jan 2004
Location: ho chi minh city
Send a message via Yahoo to thanh
Quote:
as the cfp introduces more delay.
yes !
mike! this is my old vas .I have just removed it yesterday
Attached Images
File Type: png myoldvas.png (5.5 KB, 516 views)
__________________
Justice for Victims of Agent Orange
http://www.petitiononline.com/AOVN/Thank all of you!
  Reply With Quote
Old 15th October 2004, 09:27 AM   #106
MikeB is offline MikeB  Germany
diyAudio Member
 
MikeB's Avatar
 
Join Date: Apr 2004
Location: GŁtersloh
Hi !

My experience was, that my cascoded vas has the same quality for
top end (trebles & co), but shows the better overall sound (cleaner).

To NP:
It's only sim'd, but with similar gains the cascoded vas showed higher
OL-bandwidth and less distortions. Maybe i have wrong configuration
for my FC ?

To lumanauw:
If i remember correct, for symetrical designs the ccs didn't work good,
biasing was not defined. (Similar problem to symetric currentmirror)

To thanh:
I had not really better results with cfp in vas, seems that cfp is most
interesting for diffamps ?


I don't think that folded cascode is bad, but in my particular case it
was outperformed by the cascoded vas.

Mike
  Reply With Quote
Old 15th October 2004, 01:22 PM   #107
jcarr is offline jcarr  United States
diyAudio Member
 
Join Date: Jul 2002
Location: Tokyo, Japan
Hi Nelson:

I agree that folded cascodes are usually more stable (and require less compensation) inside feedback loops. However, it has not been my experience that this is entirely due to less OL gain.

Comparing a normal VAS vs. a folded cascode circuit - both matched for similar OL gain (around 90dB in my case), my findings are that the folded cascode requires less compensation (and a less complex compensation scheme) to stabilize.

Using the folded cascode, without any local degeneration in the input stage, and compensation applied only to the inputs of the output stage, I can apply 50-some dB of global feedback before my ears tell me that I'm overdoing things. If I were only looking for stability on the scope, I would certainly be able to go higher. I don't recall the exact figures for the VAS version off the top of my head, but assuming a similarly simple compensation scheme, I think it would be around 8dB less global NFB before I started noticing audible artifacts.

In my experience, the appropriate phase compensation (scheme and amount) as well as amount of global NFB will be affected by the circuit structure and pcb layout, so definitely YMWV.

regards, jonathan carr
__________________
http://www.lyraconnoisseur.com/, http://www.lyraaudio.com
  Reply With Quote
Old 15th October 2004, 02:41 PM   #108
diyAudio Moderator Emeritus
 
lumanauw's Avatar
 
Join Date: Oct 2002
Location: Bandung
Send a message via Yahoo to lumanauw
Quote:
before my ears tell me that I'm overdoing things.
Quote:
less global NFB before I started noticing audible artifacts.
Can you share something. What's the clue of this point? Is it audible, or we should also use help of meters.

A good designer once write that there is a "sweet spot" for an amp, that is the right OL gain with the right Feedback value. That means we cannot blindly forced to use low OL gain or high OL gain. They just need the right value with the right feedback factor, not certain low value or certain high value.

Sometimes I have problem with "not enjoyable" sound from an audio amplifier. I estimated the frequencies is about 1K,2K---5K. The high midrange region (not trebles yet).
It can be eliminated by lowering the OL gain, with the same Closed Loop gain.

I dont know what causes this "too much" midhigh. Is my modification of lowering OL gain is the right cure for this? It seems works.

Is this one of the clue for finding this "sweet spot"?
  Reply With Quote
Old 15th October 2004, 02:44 PM   #109
Mr Evil is offline Mr Evil  United Kingdom
diyAudio Member
 
Join Date: Aug 2004
Location: Behind you
As far as compromises between bandwidth/gain and other characteristics go, I can't really offer any concrete opinions. Once distortion has been brought down to reasonabl levels, I honestly cannot be sure whether any differences I percieve are real or imagined. Measurements aren't much use either, since there it's not obvious which characteristics are to be favoured; Should distortion be lowered further, or is it beneficial to minimize phase shift at higher frequencies? My instincts say that the answer probably lies between the two extremes.

Anyway, back to output stages: Perhaps one with a low input impedance would work well. It would match better to the high output impedance of the cascode.
__________________
https://mrevil.asvachin.eu/
  Reply With Quote
Old 15th October 2004, 03:04 PM   #110
MikeB is offline MikeB  Germany
diyAudio Member
 
MikeB's Avatar
 
Join Date: Apr 2004
Location: GŁtersloh
Quote:
Originally posted by lumanauw


Sometimes I have problem with "not enjoyable" sound from an audio amplifier. I estimated the frequencies is about 1K,2K---5K. The high midrange region (not trebles yet).
It can be eliminated by lowering the OL gain, with the same Closed Loop gain.

I dont know what causes this "too much" midhigh. Is my modification of lowering OL gain is the right cure for this? It seems works.

Hi lumanauw !
This could mean that the cornerfreq of your ol-bandwidth was
perhaps ~3khz. So lowering the gain simply improved ol-bandwidth,
moving the cornerfreq to a level no longer audible.
The even worse thing is, you mentioned frequencies are those
most sensible to human ear...

Also possible is that you did not properly adjust feedbackcaps,
this make an amp sound too bright, as closedloopgain gets modified
by phaseshifts between input and feedback. Typically you use a small
cap (~22pF) in paralell to the voltagedivider in feedback to compensate this.

Mike
  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
folded cascode the_tool Chip Amps 1 27th June 2006 09:53 PM
New mixed-device-type symmetrical folded cascode amp Mr Evil Solid State 31 15th August 2005 11:37 AM
folded cascode puzzling drakonis Pass Labs 6 30th March 2002 09:04 PM


New To Site? Need Help?

All times are GMT. The time now is 01:03 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