Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki 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

A new perspective on current-sharing hybrid amplifiers
A new perspective on current-sharing hybrid amplifiers
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 1st January 2018, 04:59 PM   #1
Michael Koster is offline Michael Koster  United States
diyAudio Member
 
Michael Koster's Avatar
 
Join Date: Jan 2008
Location: Sunnyvale, CA
Default A new perspective on current-sharing hybrid amplifiers

Happy New Year!

Coming back to the design of tube-like hybrid amplifiers, one goal was to create an amplifier that has some desirable characteristics of all-tube amplifiers but optimizes away some undesirable aspects, frequently low "specific output" wrt. operating voltage, heat, weight, cost...

The approach outlined in this old thread was to "boost" a tube using solid state technology such that the tube characteristic dominates, ultimately arriving at the idea of current sharing between a tube and a feedback loop:

Another kind of hybrid

But what if we construct an alternate view, centered on the output transformer?

The circuit below contains the familiar chip-amp boosted tube stage, and goes another step to model it like a tube output transformer. I have separated the components to illustrate this.

The gyrator with extra B+ models the Primary Inductance (Lpri), allowing the tube to operate at a fixed DC point with high AC impedance.

The chip amp circuit models the step-down voltage ratio and current multiplication of the OPT, but multiplies the current by a very large factor. The tube only "sees" a high impedance resistive load.

The purpose of the Reflected Load is to set the current multiplication ratio of the modeled OPT to some desired value, putting some of the output load back on the tube. This can be set independently from the voltage ratio, allowing us to scale the tube down in current and voltage swing, while providing a scale model of the actual loudspeaker load for it to work into.

Some fraction of the AC current load at the output is reflected back to the plate circuit of the tube, to create a dynamic load impedance that looks more or less like what the tube would "see" through a transformer.

Some components are omitted to simplify the circuit...
Attached Images
File Type: png CurrentSharingAmplifier.png (78.8 KB, 319 views)

Last edited by Michael Koster; 1st January 2018 at 05:01 PM. Reason: Happy New Year
  Reply With Quote
Old 1st January 2018, 05:23 PM   #2
stocktrader200 is offline stocktrader200  Canada
diyAudio Member
 
Join Date: Jan 2008
Location: Calgary
does it remain stable ?
  Reply With Quote
Old 1st January 2018, 11:08 PM   #3
smoking-amp is offline smoking-amp  United States
diyAudio Member
 
smoking-amp's Avatar
 
Join Date: Dec 2001
Location: Hickory, NC
Happy New Year Michael!

The three inversions around the N Feedback loop (IC1, IC2, Q1) should be nominally stable at low freq. Some gain roll-off likely needed at HF. Especially with inductance in the speaker.

The tube current and speaker current are in phase. So seems natural tube operation.

-----------------

A possible simplification below using a Ferrite auto-xfmr (low turns, only Mosfet gate drive V signal handled within xfmr, teeter totter operation):

The current ratio should be close to the (inverse) turns ratio on the auto-former. Xfmr should be reasonably DC balanced, just AC gate drive voltage on it. (assuming the winding resistances are in similar proportion to the turns ratio.) (Option: could put a token resistance between the tube plate and its xfmr end, so the 1st order gate drive signal is automatically generated in the resistor without any teeter totter "V motion" required in the auto-xfmr. Auto-Xfmr only has to handle the "distortion" in the gate drive signal then, for really low turns )

Now to come up with a totem-pole P-P version...
Attached Images
File Type: jpg Hybrid_OT_simple1.jpg (18.6 KB, 280 views)

Last edited by smoking-amp; 1st January 2018 at 11:36 PM.
  Reply With Quote
Old 1st January 2018, 11:48 PM   #4
smoking-amp is offline smoking-amp  United States
diyAudio Member
 
smoking-amp's Avatar
 
Join Date: Dec 2001
Location: Hickory, NC
Fixed the simple version for ultra reduced xfmr turns:
N is turns ratio.
Rd provides 1st order linear gate drive signal for Q1, from V1 current across it.
Rd = N / gm_Q1

------------

And a P-P version. Woops, a coupling cap or Zener needed between V1 plate and the top CCS to allow for the steady state gate drive DC offsets.
Some more work yet to eliminate C1
Attached Images
File Type: jpg Hybrid_OT_simple2.jpg (19.3 KB, 260 views)
File Type: jpg Hybrid_P-P_OT_simple.jpg (28.4 KB, 259 views)

Last edited by smoking-amp; 2nd January 2018 at 12:13 AM.
  Reply With Quote
Old 2nd January 2018, 09:52 AM   #5
Bigun is offline Bigun  Canada
diyAudio Member
 
Bigun's Avatar
 
Join Date: Jan 2009
Location: Waterloo, ON or Herefordshire UK
A new perspective on current-sharing hybrid amplifiers
Michael - I like very much where you are going with this one.
__________________
"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 2nd January 2018, 12:30 PM   #6
indra1 is offline indra1  Indonesia
diyAudio Member
 
Join Date: Nov 2010
Location: Bogor
Michael, I prefer simpler power gyrator topology like below. Loading of tube can be tailored by varying R2. A buffer with low 4H and above distortion to avoid buffer stage dominant spectra. I have not yet found any solution other than class A bias. Comments?
Attached Images
File Type: gif 6BQ5 Power Gyrator R_MD11.gif (16.9 KB, 208 views)
File Type: gif 6BQ5 Power Gyratof FFT.gif (16.5 KB, 93 views)
File Type: gif EL83 CFP Follower.gif (18.4 KB, 77 views)
  Reply With Quote
Old 2nd January 2018, 06:51 PM   #7
Wavebourn is online now Wavebourn  United States
Designer & Technologist
diyAudio Member
 
Wavebourn's Avatar
 
Join Date: Sep 2006
Location: Pleasant Hill, CA
A new perspective on current-sharing hybrid amplifiers
Happy new year Michael;

glad to see you back!

Back in 2017 I designed a triode push-pul amp using 6S19P tubes, and QQE03/12 drivers. Positive feedback by current I took from cathode resistors of output tubes, and applied as a bootstrap to driver tubes. Such a way I killed 2 ducks by a single shot: increased damping factor, and added dynamic load to driver tubes without using any CCS or gyrators. ;-)
__________________
Nothing in the universe is perfect. The ideal things are the ones that are most optimal. Optimization criteria, what matters. When I hear "No Compromise Design", I want to take a sledgehammer and test how impact-proof it is.
  Reply With Quote
Old 13th January 2018, 04:05 PM   #8
Michael Koster is offline Michael Koster  United States
diyAudio Member
 
Michael Koster's Avatar
 
Join Date: Jan 2008
Location: Sunnyvale, CA
Quote:
Originally Posted by indra1 View Post
Michael, I prefer simpler power gyrator topology like below. Loading of tube can be tailored by varying R2. A buffer with low 4H and above distortion to avoid buffer stage dominant spectra. I have not yet found any solution other than class A bias. Comments?
Yes, there is definitely a huge circuit optimization to be had by modulating the gyrator element to produce the current feedback.

I like the ratiometric current mirror approach where a modulated current source drives a small current feedback signal to the gyrator, maybe around 1mA, which scales it up to the tube load current (10-20mA or more).

The big advantage using a modulated gyrator vs. the separate parallel anti-load is saving about 1/2 idle current in the output stage for the same tube operating conditions.

For push-pull topology and low quiescent current I am thinking class Ab (Big A, little b) with pentodes where each tube doesn't cut off but gets into the way low gm operating region. I also have a circuit design for a "Differential Gyrator" which works like the (idealized) coupled inductance of a tapped primary.

I will look into a modulated differential gyrator which may be able to present the scaled current feedback correctly to a push-pull pentode pair operating in asymmetric class A.

Note that the design using a parallel load can substitute an actual (iron and copper) choke for the gyrator.

PS since I am building mostly guitar amps now, and I have an immediate personal use for this, and discussion of instrument amps is not allowed on this list, I will probably take up this discussion over on the instruments and amps board.

Last edited by Michael Koster; 13th January 2018 at 04:13 PM.
  Reply With Quote
Old 13th January 2018, 05:13 PM   #9
Michael Koster is offline Michael Koster  United States
diyAudio Member
 
Michael Koster's Avatar
 
Join Date: Jan 2008
Location: Sunnyvale, CA
Default Scaled bigA, little b, asymmetric pentode load line

Here is a 6L6GC scaled down to 100V G2 to illustrate the idea of a push-pull output stage operating point, using this basic idea. We don't need to be a aggressive in biasing toward cutoff because we're operating at 12mA due to the scaled down Vg2 and scaled up load impedance.

I guess for push-pull, the current feedback can be inverse polarity and on the opposite side..
Attached Images
File Type: png 6l6-scaled.png (392.5 KB, 51 views)

Last edited by Michael Koster; 13th January 2018 at 05:17 PM.
  Reply With Quote

Reply


A new perspective on current-sharing hybrid amplifiersHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Bryston 3BIII SPICE Simulation - Current Sharing and Output Stage PB2 Solid State 39 4th May 2017 12:04 AM
Sharing current through diodes rtarbell Parts 1 3rd April 2007 03:49 PM
Current-Sharing on the DOZ Amp EchoWars Solid State 9 3rd November 2004 01:20 AM
FET/BJT hybrid amplifiers angel Pass Labs 6 19th June 2004 02:37 AM
current sharing in class AB outputs hitsware Solid State 5 19th October 2003 05:02 PM


New To Site? Need Help?

All times are GMT. The time now is 11:39 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki