F5 Cascoded - an alternative for my interest in the Balanced F5 - Page 3 - diyAudio
Go Back   Home > Forums > Amplifiers > Pass Labs

Pass Labs This forum is dedicated to Pass Labs discussion.

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 22nd January 2011, 06:46 PM   #21
NYCOne is offline NYCOne  United States
diyAudio Member
 
Join Date: Jul 2010
Blog Entries: 1
Juma,

Thanks for the PCB layout. For the rest of the thread: Juma designed the layout and I added the cascode bits.

I'm using the 3W 150s because I'm not sure whether I want 50 ohms at the end of the day or 75 ohms for feedback. With the 3W version, I have the choice. The crowding is easily solved by using two layers (I'll take a pic from the side to show this).



Quote:
Originally Posted by juma View Post
@NYCOne

I look at your PCB in post #1 and it's a bit crowded with 150R/3W feedback resistors.
I planned that PCB for 3x150R/2W feedback resistors since I can't buy locally 100R/3W ones.
You can use 2x100R/3W to avoid crowding.
  Reply With Quote
Old 22nd January 2011, 06:48 PM   #22
flg is offline flg  United States
diyAudio Member
 
flg's Avatar
 
Join Date: Jul 2005
Location: North East
O.K. I have some time. I don't know exactly where people are having trouble but, lets see if I can shed some light on the F5 case of input cascode modification. Basically we can read about the benifits elsewere, as in the previous sited papers authored by N.P. or Borbely etc. I will just try to squeeze in a cascode device as designed into the CViller boards I have (rev2 I guess). The schematic can be found here diyAudio - cviller go down to F5. I'm sure if you only want to know what time it is, you should skip to the next post.
We will be maintaining something very close to the original qurrent through input JFETs Q1,2 but, we will place the emitter of the selected Bipolars Q101,102 to the drain of the JFETs through a second hole provided on the PCB and marked "casc" I believe. No change in the JFETs 10 ohm source Rs is necessary.
Pictured in most of these cascoded F5 schematics is a voltage divider from each rail to Gnd or maybe from rail to rail(R101,102,103,104 in CVillers board). I think in almost all the schematics I've seen, the voltage divider is setting up 1/2 of the rail voltage as a refrence for the cascode. That might be O.K. but, I seem to remember some of these "experts" saying these JFETs work the best at around 8-10Vds so thats where I would put my ref voltage. The drain of the input JFETs will see about .6V less than the refrence voltage you set at the base of Q101 and 102. It is a personal choice based not only on where you want the JFETs to operate but also what the rail voltage is. In the case of a 2X25VRMS xfrmer giving about 32-33V rails I would want my voltage divider to provide about 25% of the rail voltage to the base of the Cascode xistor. So where CViller and N.P. and others have 2 equal value restors in the R101,102,103,104 position you might get 16.5V at the base of Q101,102. Me though, I think I would use more like a 12k for R101,103 and a 4k for R102,104 giving me about 8V at the base of Q101,102. A reasonable limit for the JFETs Vds is probably around 20V due to the power dissapation it will inccur with 5mA through it. I think any more than about 100mW is starting to push it a little.
I think what I have described would be just fine for the 32-33V rail situation without any other changes than adjusting P1,P2 as normal. Hopefully I'm getting this discussion rolling a little. I have undoubtedly overlooked things, dare I say due to my intelegence but, I'm open for further questions, ideas, etc...
__________________
"It was the perfect high end audio product: Exotic, inefficient, expensive, unavailable, and toxic." N.P.
  Reply With Quote
Old 22nd January 2011, 07:00 PM   #23
NYCOne is offline NYCOne  United States
diyAudio Member
 
Join Date: Jul 2010
Blog Entries: 1
I'll give you an example:

On the Balanced thread I asked about NTC thermistors. I was trying to decide what to buy. I was concerned about the steady state current being too high for the CL-60s NP used in the stock F5 (Balanced F5 question see post 735).

Instead of getting an answer, I was directed to search the internet (which I'd already done) or buy a book. The suggestion was there were numerous threads that already answered the question. If there were, I either didn't find them, or didn't think they answered the question for the Balanced F5.

NP ultimately answered the issue in one sentence. In my opinion, that's the sign of a true expert, one who is willing to spend the time to answer questions, even if they are simple compared to one's knowledge.

Quote:
Originally Posted by ashaw View Post
I happen to be a member of that "private club" and I can say that it is not as exciting as you think it is. All of the schematics are either public relatively straight forward or could be dangerous in the current state. We are just not very fast and are working on our own. There is just not that much to see.

Alexis Shaw
  Reply With Quote
Old 22nd January 2011, 07:14 PM   #24
NYCOne is offline NYCOne  United States
diyAudio Member
 
Join Date: Jul 2010
Blog Entries: 1
Which would give the better outcome, caps around each cascode grounded, or a cap between the bases linking them together. It's a trivial matter to re-etch a board.

Quote:
Originally Posted by Nelson Pass View Post
That would be the "fewer parts" edition. The cap from
base to base achieves much the same thing as the base
currents cancel each other.

This is a bandwidth issue, the object being to prevent
the cascodes from degrading the bandwidth due to high
impedances on their bases.

  Reply With Quote
Old 22nd January 2011, 07:33 PM   #25
Zen Mod is offline Zen Mod  Serbia
diyAudio Member
 
Zen Mod's Avatar
 
Join Date: Jan 2003
Location: ancient Batsch , behind Iron Curtain
you'll need golden ear to hear difference
__________________
my Papa is smarter than your Nelson !
clean thread; Cook Book;PSM LS Cook Book;Baby DiyA ;Mighty ZM's Bloggg;Papatreasure;Papa... by Mighty ZM
  Reply With Quote
Old 22nd January 2011, 07:36 PM   #26
flg is offline flg  United States
diyAudio Member
 
flg's Avatar
 
Join Date: Jul 2005
Location: North East
My opinion would be an individual cap for each side. I would also suggest bypassing the "nice big caps to ground" with a high a quality .01u-.1u.
__________________
"It was the perfect high end audio product: Exotic, inefficient, expensive, unavailable, and toxic." N.P.
  Reply With Quote
Old 22nd January 2011, 07:41 PM   #27
NYCOne is offline NYCOne  United States
diyAudio Member
 
Join Date: Jul 2010
Blog Entries: 1
Maybe, but if NP is willing to opine, I will re-etch, the best is the best...


Quote:
Originally Posted by Zen Mod View Post
you'll need golden ear to hear difference
  Reply With Quote
Old 22nd January 2011, 07:47 PM   #28
Zen Mod is offline Zen Mod  Serbia
diyAudio Member
 
Zen Mod's Avatar
 
Join Date: Jan 2003
Location: ancient Batsch , behind Iron Curtain
feel free to re-etch , than opine by your self

only that is valid
__________________
my Papa is smarter than your Nelson !
clean thread; Cook Book;PSM LS Cook Book;Baby DiyA ;Mighty ZM's Bloggg;Papatreasure;Papa... by Mighty ZM
  Reply With Quote
Old 22nd January 2011, 07:55 PM   #29
NYCOne is offline NYCOne  United States
diyAudio Member
 
Join Date: Jul 2010
Blog Entries: 1
Yes, but I'd hope someone has already run the experiment. It would save a lot of time if there is an answer.

You suggest the answer may be too close to call, if that's the case, then no re-etching needed.


Quote:
Originally Posted by Zen Mod View Post
feel free to re-etch , than opine by your self

only that is valid
  Reply With Quote
Old 22nd January 2011, 08:02 PM   #30
NYCOne is offline NYCOne  United States
diyAudio Member
 
Join Date: Jul 2010
Blog Entries: 1
This was a great explanation of the cascode. Personally, I appeciate the effort.

Quote:
Originally Posted by flg View Post
O.K. I have some time. I don't know exactly where people are having trouble but, lets see if I can shed some light on the F5 case of input cascode modification. Basically we can read about the benifits elsewere, as in the previous sited papers authored by N.P. or Borbely etc. I will just try to squeeze in a cascode device as designed into the CViller boards I have (rev2 I guess). The schematic can be found here diyAudio - cviller go down to F5. I'm sure if you only want to know what time it is, you should skip to the next post.
We will be maintaining something very close to the original qurrent through input JFETs Q1,2 but, we will place the emitter of the selected Bipolars Q101,102 to the drain of the JFETs through a second hole provided on the PCB and marked "casc" I believe. No change in the JFETs 10 ohm source Rs is necessary.
Pictured in most of these cascoded F5 schematics is a voltage divider from each rail to Gnd or maybe from rail to rail(R101,102,103,104 in CVillers board). I think in almost all the schematics I've seen, the voltage divider is setting up 1/2 of the rail voltage as a refrence for the cascode. That might be O.K. but, I seem to remember some of these "experts" saying these JFETs work the best at around 8-10Vds so thats where I would put my ref voltage. The drain of the input JFETs will see about .6V less than the refrence voltage you set at the base of Q101 and 102. It is a personal choice based not only on where you want the JFETs to operate but also what the rail voltage is. In the case of a 2X25VRMS xfrmer giving about 32-33V rails I would want my voltage divider to provide about 25% of the rail voltage to the base of the Cascode xistor. So where CViller and N.P. and others have 2 equal value restors in the R101,102,103,104 position you might get 16.5V at the base of Q101,102. Me though, I think I would use more like a 12k for R101,103 and a 4k for R102,104 giving me about 8V at the base of Q101,102. A reasonable limit for the JFETs Vds is probably around 20V due to the power dissapation it will inccur with 5mA through it. I think any more than about 100mW is starting to push it a little.
I think what I have described would be just fine for the 32-33V rail situation without any other changes than adjusting P1,P2 as normal. Hopefully I'm getting this discussion rolling a little. I have undoubtedly overlooked things, dare I say due to my intelegence but, I'm open for further questions, ideas, etc...
  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
Balanced F5 question NYCOne Pass Labs 1153 23rd February 2011 07:42 PM
F5 Williams Audio Pass Labs 9 11th June 2010 10:13 PM
I need Help with my F5.... rtate Pass Labs 15 2nd December 2009 12:07 AM
Anybody help with F5? yammy1688 Pass Labs 2 28th October 2008 04:50 PM
Yet another variation on cascoded balanced line stage jwb Pass Labs 10 27th October 2002 06:22 PM


New To Site? Need Help?

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