Plate-Grid vs. Plate-Cathode local NFB - Is it the same? - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

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 18th June 2017, 02:44 PM   #1
rongon is offline rongon  United States
diyAudio Member
 
rongon's Avatar
 
Join Date: Jul 2009
Location: Across the river from Rip's big old tree...
Default Plate-Grid vs. Plate-Cathode local NFB - Is it the same?

I've been reading the many (many, many, many) posts about local negative feedback around an output stage. I picked up somewhere that local feedback performed by taking signal from the plate of an output tube can go equally well to either that output tube's grid (plate-grid local feedback, aka "Schade") or to the cathode of the driver tube (RCA Tube Manual style). Are these basically the same thing, done differently?

I was also reading about the various ways of putting NFB into a circuit. Series-derived, series-applied, parallel-derived, parallel-applied, in all combinations. I'm sure I'm just being dense, but I'm not following the meanings of those definitions, although I get that applying NFB reduces gain in all cases (as opposed to positive feedback, which boosts gain once applied).

Anyhow, here are two circuits I modeled in spice, to illustrate the question. R10 and C4 feed back the audio signal from the plate of U2. To keep things simple, I AC-coupled the feedback loop from the output tube's plate, so that the high voltage there wouldn't influence DC conditions elsewhere in the circuit. The top schematic is plate-cathode, the bottom one is plate-grid.

--
So, the question is, of these two ways of applying NFB around the output stage of a simple SE pentode amp, is one better than the other? Is one better than the other in certain situations, but not in others? Are there downsides to using one or the other when driving the complex load presented by the OPT with a speaker connected to its secondary?
--

PS - Yes, DC-coupling the feedback to the drain of the input DN2540 is possible, by taking R10 directly from the 6BQ5 plate and going to the DN2540 drain, with no DC-blocking C4 (and removing the decoupling network R9/C5). I wanted to keep this comparison apples to apples, with identical DC conditions for both examples.
Attached Images
File Type: png LocalNFB_compared.png (34.7 KB, 171 views)

Last edited by rongon; 18th June 2017 at 03:00 PM.
  Reply With Quote
Old 18th June 2017, 03:12 PM   #2
diyAudio Member
 
Wavebourn's Avatar
 
Join Date: Sep 2006
Location: Pleasant Hill, CA
Send a message via Skype™ to Wavebourn
Series feedback around 2 stages VS parallel feedback around 1 stage?
How can they be the same?
__________________
The Devil is not so terrible as his math model is!
  Reply With Quote
Old 18th June 2017, 03:13 PM   #3
rayma is offline rayma  United States
diyAudio Member
 
Join Date: Apr 2011
Location: Wisconsin
Quote:
Originally Posted by rongon View Post
local feedback performed by taking signal from the plate of an output tube can go equally well
to either that output tube's grid (plate-grid local feedback, aka "Schade") or to the cathode
of the driver tube (RCA Tube Manual style). Are these basically the same thing, done differently?
Not the same, since Schade is feedback around a single stage.
http://www.electronics-tutorials.ws/systems/sys29.gif

Using the driver cathode is feedback around two stages.
More distortion, more phase shift, error signal derived through a device junction.
https://upload.wikimedia.org/wikiped...erting.svg.png

Last edited by rayma; 18th June 2017 at 03:19 PM.
  Reply With Quote
Old 18th June 2017, 03:48 PM   #4
diyAudio Member
 
SpreadSpectrum's Avatar
 
Join Date: Nov 2007
Further complicating things is the fact that Schade used series applied feedback in his paper but everybody here refers to parallel applied feedback as "Schade feedback." Just be aware of that in your quest for knowledge and understanding.
__________________
My Blog: http://tubeswithatwist.blogspot.com/
  Reply With Quote
Old 18th June 2017, 03:59 PM   #5
rongon is offline rongon  United States
diyAudio Member
 
rongon's Avatar
 
Join Date: Jul 2009
Location: Across the river from Rip's big old tree...
Thank you kind sirs!

Is "Schade" feedback any kind of NFB around a single stage, whether series or parallel applied?
  Reply With Quote
Old 18th June 2017, 04:02 PM   #6
diyAudio Member
 
SpreadSpectrum's Avatar
 
Join Date: Nov 2007
"Schade" feedback is an imprecise term and will mean different things to different people. If you are asking questions about specific different types of local feedbacks it is best to avoid that term, IMHO.
__________________
My Blog: http://tubeswithatwist.blogspot.com/
  Reply With Quote
Old 18th June 2017, 04:04 PM   #7
rongon is offline rongon  United States
diyAudio Member
 
rongon's Avatar
 
Join Date: Jul 2009
Location: Across the river from Rip's big old tree...
One thing I'm seeing in spice sims is that if you use a triode as the driver stage, plate-grid (parallel?) NFB doesn't work well. Replacing the triode with a depletion mode MOSFET (orders of magnitude higher gm) makes the plate-grid feedback work very well. The MOSFET is also super-easy to implement. Using a pentode as the driver works pretty well for plate-grid NFB, but never as well (or as easily) as the depletion mode MOSFET.

However, if you must use a triode as the driver, then plate-cathode (2-stage, series) NFB works very well, at least in the sims.

Does that make any sense?
  Reply With Quote
Old 18th June 2017, 04:14 PM   #8
diyAudio Moderator
 
jazbo8's Avatar
 
Join Date: Jan 2011
Location: In Transient
Yes, it makes sense, as many of us have been saying for years, despite Kitic and the RH followers who insist otherwise... Use a pentode or MOSFET for the driver if you want to use plate-to-grid feedback.
  Reply With Quote
Old 18th June 2017, 04:22 PM   #9
diyAudio Member
 
SpreadSpectrum's Avatar
 
Join Date: Nov 2007
You want to use something that is a very linear voltage to current converter working into a vertical load line. If you look at triode characteristics, you can see that a triode wants a horizontal load line, but a vertical one will give awful results as far as linearity goes. Pentode is much more linear into a vertical load. Cathode degeneration will further improve that linearity but will reduce gain.
__________________
My Blog: http://tubeswithatwist.blogspot.com/
  Reply With Quote
Old 18th June 2017, 04:57 PM   #10
rongon is offline rongon  United States
diyAudio Member
 
rongon's Avatar
 
Join Date: Jul 2009
Location: Across the river from Rip's big old tree...
Great stuff, thank you everyone.

OK, I'm finding it very easy to get a good looking simulation going with a DN2540 driving an EL84, with plate-grid (parallel) feedback. That's the schematic in the bottom half of the attachment (post 1). The thing is, I'm finding it very difficult to get similarly good looking results with a pentode as the input stage. I figured 6AU6A would work, but it seems to need high voltages at plate and screen. With a B+ of only 270V, there's not much for it to work with. If I use 6J9P (E180F, 6688) then I get reasonable results, but the gain is very, very high (much higher than with the DN2540 in the first stage).

How do I set up the pentode so that it's working into a vertical load line? Do I want to use a relatively small value plate resistor and adjust plate+screen current by adjusting values of the screen grid dropping resistor and cathode resistor?

I've read that plate-grid NFB works from the plate into the combined load of the output pentode's grid leak resistor in parallel with the rp of the driver pentode. If I reduce the value of the input stage pentode's plate resistor (say to 10k), how will that influence the feedback voltage divider? Will it load it down?
--
  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
Suppressor Grid tied to plate and not ground rsumperl Tubes / Valves 22 20th September 2016 08:57 PM
Dual grid and plate pins MrCurwen Tubes / Valves 7 28th October 2012 07:20 AM
Cathode/plate resistors ?? Minion Tubes / Valves 20 17th April 2011 11:33 PM
cathode for plate ccs ? jj2 Tubes / Valves 3 23rd February 2007 04:32 PM
Free Plate Amplifier for local pickup in Seattle motherone Swap Meet 6 31st January 2006 01:46 AM


New To Site? Need Help?

All times are GMT. The time now is 06:13 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki