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

Input impedance experimentations, aiming to build an A2 stage
Input impedance experimentations, aiming to build an A2 stage
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 8th July 2018, 04:22 PM   #11
smoking-amp is offline smoking-amp  United States
diyAudio Member
 
smoking-amp's Avatar
 
Join Date: Dec 2001
Location: Hickory, NC
For class A2 P-P there is an interesting possibility. I discovered this surprising technique in a LV tube operation thread recently.

Notice from your graph of grid current versus grid voltage how close to linear the current increase is. Now with 2 tubes in class A2 P-P, the sum of the grid currents would be near constant. So use a center tapped interstage xfmr to drive the two grids with a CCS pull-up on the center tap ( of the average sum current ). The driver stage will now only have to supply the mis-tracking between the two "linear" current draw curves. So will be much higher input impedance.

Edcor WSM or XSM matching xfmrs could do this inexpensively.

This obviously makes for a novel way to operate tubes at low B+ voltages, using Class A2 operation to get decent plate currents, without the grid current penalty. An XSM xfmr should allow driving output power stages in P-P class A2 also.

Last edited by smoking-amp; 8th July 2018 at 04:33 PM.
  Reply With Quote
Old 10th July 2018, 05:48 PM   #12
Gnobuddy is online now Gnobuddy  Canada
diyAudio Member
 
Join Date: Mar 2016
Quote:
Originally Posted by smoking-amp View Post
...how close to linear the current increase is. Now with 2 tubes in class A2 P-P, the sum of the grid currents would be near constant.
Only if the quiescent operating point was so "hot" that both tubes were biased with the grid quite positive, i.e. deep into (control) grid current flow. For this grid current cancellation trick to work, you would actually have to stay in positive grid voltage territory even on negative signal peaks.

That is a bias so hot that the output tubes are limited to operating within quite a small region of their normal characteristic curves.

Doesn't that seem far more limiting than normal AB1 operation? Not to mention, with an ultra-hot bias like that, how are you going to keep the quiescent operating conditions below the maximum anode dissipation curve? You'd have to drop to an absurdly low B+, at which point, output power becomes drastically limited.

I can understand why this mode of operation might work in the weird world of ultra low B+ tube circuitry. Operating entirely in grid current territory is not unusual for these types of circuits.

But I can't see how the "bias it so hot it's always at positive control grid voltage" idea would work well for an audio amp running on anywhere near normal B+ voltage.

To me it makes a lot more sense to just use a MOSFET source-follower to drive the output tube grid. It has a low enough output impedance to comfortably drive the grid positive without difficulty.

-Gnobuddy
  Reply With Quote
Old 10th July 2018, 07:14 PM   #13
smoking-amp is offline smoking-amp  United States
diyAudio Member
 
smoking-amp's Avatar
 
Join Date: Dec 2001
Location: Hickory, NC
All very valid points you make. Certainly for a triode, one would have to operate at rather low plate voltages to keep dissipation (both plate and grid 1) in bounds.

And the usual Mosfet follower/driver for class aB2 makes the best sense for practical __2 operation.

However, the thread begins:
Quote:
Input impedance experimentations
so I just thought I would mention this off beat idea.

A 26A7 tube could work better for this "always positive" grid1 idea. Since the screen grid can be kept at low voltage to simulate a low plate voltage (as if a triode at LV). Or a TV Sweep Tube operated at below normal plate dissipation and low screen V could work.

One just needs to restore the normal operating current range by lowering Vg2 to compensate for positive Vg1. But be careful not to melt down grid 1 with positive current. A tube intended for positive grid 1 operation (4D32) or one with a big grid 1 cooler fin (40KG6, 6HJ5) should help. Something like 26V to 48V on grid 2.
  Reply With Quote
Old 10th July 2018, 10:00 PM   #14
PRR is offline PRR  United States
diyAudio Member
 
PRR's Avatar
 
Join Date: Jun 2003
Location: Maine USA
Quote:
Originally Posted by smoking-amp View Post
...for a triode, one would have to operate at rather low plate voltages to keep dissipation (both plate and grid 1) in bounds....
Or a very high Mu.

This was occasionally suggested back in the day. 6A6 data shows a zero-bias condition. The grid impedance has already dropped at zero bias. Under swing, the one side impedance falls while the other side rises, making a non-constant but less-kinky driver load. But note that the grid input is nominally 0.451 Watts, and must face a varying load, so we would want a driver capable of a good Watt easy. The grid impedance gets near 2K at peak swing which will guide our driver transformer (10K CT?).

Type '46 is well documented for zero-bias operation.

3CX1500D7 is rated for zero bias, 0Vgk and at a zippy 3,000V on the plate. (The 500W dissipation is only 1/3rd what this coffee-mug tube is rated for.)

There was a battery-radio tube designed for zero bias and low idle current with reasonably constant Zin in such condition. Dang if I remember which one. Maybe 1/3rd of a Watt?

Last edited by PRR; 10th July 2018 at 10:02 PM.
  Reply With Quote
Old 10th July 2018, 11:39 PM   #15
smoking-amp is offline smoking-amp  United States
diyAudio Member
 
smoking-amp's Avatar
 
Join Date: Dec 2001
Location: Hickory, NC
Type 52 tube got some mention here:
Question about type 52 tube

If one has the internal Mu factor between grid 2 and grid 1 of some pentode, then the input V range should be translatable by shifting Vg2 accordingly. And watching grid 1 dissipation for safety max.

Last edited by smoking-amp; 11th July 2018 at 12:00 AM.
  Reply With Quote
Old 11th July 2018, 03:56 AM   #16
Trajan is offline Trajan  United States
diyAudio Member
 
Join Date: Apr 2014
I want to see more A2 01A's. Having seen curves from an EE text circa early 1920's they are quite linear. Never published in datasheets apparently, though. Maybe someone can trace some.
  Reply With Quote
Old 11th July 2018, 06:34 AM   #17
Wavebourn is offline Wavebourn  United States
Designer & Technologist
diyAudio Member
 
Wavebourn's Avatar
 
Join Date: Sep 2006
Location: Pleasant Hill, CA
Input impedance experimentations, aiming to build an A2 stage
Here you go, GU-50 when all 3 grids are strapped together, behaves as a "Right-Handed" triode. High Mu, pentode-like curves.
Attached Images
File Type: png Gubernator_Mk_6b_3.png (18.2 KB, 149 views)
__________________
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 15th July 2018, 07:43 PM   #18
Elerion is offline Elerion  Spain
diyAudio Member
 
Join Date: Sep 2017
Following Gnobuddy's advice, I considered a MOSFET driver stage. As I like to do, I started from the begining, with a minimalist design.
I simulated a simple 6V6 amplifier, and it seems to works great (as expected) bumping up from 1W (single ended triode, A1) to 4.4W(< 0.4% THD, A2).

The problem is that, a strong enough signal makes de output tube's grid more and more positive, with no limit (until MOSFET supply voltage is almost reached)., and the max grid dissipation will surely be surpassed. The sensitivity is around 0.6 V pk, but for 1V pk input signal (not that high), the output tube drive signal is overkill. Just look at the simulation attached: 300 mA grid current !!

I read some interesting information about this kind of circuits, like MOSFET Follies, a book, some threads around,... and a high voltage bipolar supply is usualy used for the driver stage (be it MOSFET or tube), and no limiting mechanism.

How can I avoid a strong signal to basically destroy the output tube's grid (and even the MOSFET, if not properly heatsinked)?

The only workaround I found is to use a +25V/-70V supply for the driver. This way, the grid signal clips at a safe voltage, much the same as when it clips at 0V when using a common A1 driver stage, and I get plenty of clean power.

Suggestion?
Attached Images
File Type: png mini_a2_nmos_overload.png (52.0 KB, 120 views)
  Reply With Quote
Old 15th July 2018, 09:51 PM   #19
Gnobuddy is online now Gnobuddy  Canada
diyAudio Member
 
Join Date: Mar 2016
Quote:
Originally Posted by Elerion View Post
Following Gnobuddy's advice...
Credit where due: George (Tubelab_com) was doing this long before I had the same idea. I only got interested in valves around 2010 or so, shortly after I got back into playing guitar. Once I'd encountered the problem of blocking distortion, I thought about using a MOSFET direct-coupled to the output valve grids, Googled it, and found links to George's website. George had already been doing it for years in multiple designs of his own.

Poke around diyAudio, there are tons of posts by George involving MOSFET drivers and class AB2 operation, including schematics of most of his designs, measurement data, 'scope shots, etc, etc.
Quote:
Originally Posted by Elerion View Post
... a high voltage bipolar supply is usualy used for the driver stage (be it MOSFET or tube), and no limiting mechanism.
This is okay for Hi-Fi, where the output is never intentionally driven into clipping.

But for guitar, I too think that something else is required.
Quote:
Originally Posted by Elerion View Post
The only workaround I found is to use a +25V/-70V supply for the driver.
This seems like a perfectly reasonable solution to me.

For my own project (a guitar amp in the very early conceptual stage), I had two solutions in mind. The first is exactly what you described, limiting MOSFET drain supply voltage (to about 15 - 18 volts in my case.)

The second solution I was considering is simply inserting a resistor between MOSFET drain and the MOSFET positive supply rail. The resistor value will be chosen so the MOSFET saturates due to limited Vds before the output valve grids become overdriven to the point of overdissipation. There should be minimal effect on distortion, as the MOSFET is relatively insensitive to changes in drain voltage.

There is the question of MOSFET reverse transfer capacitance (Cgd or Crss), which can vary with drain-to-source voltage, especially at relatively low values of Vds. Given the nonlinearity of the valves in the amp, I have my doubts that tiny variations in Crss will affect the amplifiers distortion performance to any significant extent, but it may be something to keep in mind, at least until you know it's not a problem.

As a third alternative, in a post a couple of months ago, George mentioned using 1k grid stoppers for the output pentodes in a test-bench prototype, driven by his MOSFET driver board. Since these resistors sit between the driver MOSFET source and the output valve control grid, nonlinear grid current through them does affect THD significantly once into the region of positive control grid voltage.

George found that it took about 15 mA of current flow into the control grids of those particular output valves to achieve full power at the output, and those 15 mA of non-linear current flow through the 1k resistor added a fair amount of distortion to the amplifier, progressively increasing with drive signal level. Which may be okay, or even good, in a guitar amp - many guitarists want progressively increasing distortion to create touch-sensitive amp response.

Obviously, this is not a good idea for a Hi-Fi amplifier.

-Gnobuddy
  Reply With Quote
Old 16th July 2018, 01:54 AM   #20
smoking-amp is offline smoking-amp  United States
diyAudio Member
 
smoking-amp's Avatar
 
Join Date: Dec 2001
Location: Hickory, NC
The resistor in the Mosfet Drain plus a Zener from cathode to grid should stop any overdrive. Suggest using the smallest Mosfet (low Crss) that will do the job.

You would need to know the grid current curve reliably with just the Drain resistor (may change with tube ageing), so the Zener is an insurance policy.
Gets used for Crazy/Twin Drive protection.

If you are experimenting with positive grid drive, don't forget there is also the possibility of using a current drive of the positive grid. Crazy/Twin Drive uses this for grid 1. (Screen grid V likely will need adjusting down for positive grid only type drives) One can make vacuum tube current mirrors this way.

Last edited by smoking-amp; 16th July 2018 at 02:00 AM.
  Reply With Quote

Reply


Input impedance experimentations, aiming to build an A2 stageHide 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
Input impedance of grounded grid stage? 50AE Tubes / Valves 8 16th March 2018 12:10 PM
Experimentations on Regulators peufeu Solid State 56 20th September 2015 11:46 AM
Input impedance on cathode follower stage? Vingborg Tubes / Valves 19 11th February 2015 10:38 AM
Input impedance of inverting stage of TPA6120? Mateo Jun Chip Amps 5 4th February 2012 01:44 PM
How to design input stage with specific input impedance/pot value mahleriana Tubes / Valves 0 16th November 2009 09:37 PM


New To Site? Need Help?

All times are GMT. The time now is 05:49 AM.


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