Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Instruments and Amps Everything that makes music, Especially including instrument amps.

Suggestions for a Medium Gain Guitar Amp Topology...
Suggestions for a Medium Gain Guitar Amp Topology...
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
Thread Tools Search this Thread
Old 1st February 2018, 10:28 PM   #21
Gnobuddy is offline Gnobuddy  Canada
diyAudio Member
Join Date: Mar 2016
Originally Posted by LeftHandFool View Post
From memory, the Tetrode added a screen between grid and anode, while the Pentode went one further
As I understand it, the original tetrode design was largely unusable, because it had those huge "kinks" in the curves that you mentioned.

Phillips fixed this with the third grid, called the result a pentode, and patented the concept. Pentodes were hugely superior to triodes in many applications (RF and power output among them), and everybody wanted to manufacture them. But you had to pay Phillips a licensing fee to do that.

Then someone figured out how to modify a tetrode in a superficially very subtle way (line up the grid wires precisely) that actually produced spectacular results: the kinks in the characteristic curves went almost completely away, and now these new-fangled tetrodes were as good as a pentode, and best of all, you didn't have to pay Phillips a licensing fee to build them!

The good properties come from the ballistic properties of the searchlight-beams of electrons flying between the aligned grid wires inside the valve, so these valves are variously called "beam tetrodes" or "beam power tubes". The theory behind their operation is considerably more sophisticated than preceding generations of valves, but that seems to have been forgotten later, and nowadays most people seem to just call them pentodes. Which, quite literally, they are not!

You might also find references to "beam deflecting plates", but these actually have nothing to do with the actual beams that cause the "beam tetrode" characteristic. Those deflecting plates were only used in early generations of beam tetrodes, which had flattened cathodes which only emitted significant numbers of electrons in two directions. Later, and better, beam tetrodes used cylindrical cathodes, with electrons emitted radially in all directions. No beam deflecting plates in sight.

I got interested in valve guitar amps a few years ago, and at first I didn't know there was any difference between one pentode and another. But I was very curious why EL84s sounded so different from 6V6s (I like the latter a lot more).

So I dug deeper, and found that a lot of good information had been largely lost to history in the last several decades. It turns out pentodes and beam tetrodes do have some fundamental differences in characteristics. From our guitar-centric point of view, the one most likely to matter is the inherently different harmonic spectrum.

(There is a research paper by Schade on the theory and design of the original 6L6 beam power tube; somewhere in there is a rueful acknowledgement that beam power tubes have more 2nd harmonic distortion, but the authors go on to say that this is not as much of a problem as it seems, because even harmonics cancel out in a push-pull output stage.)

This, of course, was when tubes were used for Hi-Fi and RF work, and linearity was a good thing.

Things are different for us guitarists!

Originally Posted by LeftHandFool View Post
, but there were copyright issues which caused some valves to be 'mismarketed', shall we say .
I can imagine Phillips management grinding their teeth after their patents ran out, and pesky American manufacturers began to sell "pentodes" that didn't even have three grids!

Originally Posted by LeftHandFool View Post
From my side of the pond, the 6V6 is a pesky American 'tube'
They'd be overpowered for this application, would they not?
The 6V6 is a beam power tube / beam tetrode, and as you say, overpowered for a preamp application.

Which is why I mentioned that there are small-signal beam tetrodes out there, often cheap. But the datasheets almost always called them "pentodes", so it is hard to tell which types they are!

One way to tell for sure is by inspection with a magnifying glass (or, I suppose, cracking the envelope open )

Theory says that there might be detectable differences in the characteristic curves (a sharper corner at the top left of each curve for beam tetrodes), but I've looked at a lot of datasheets where the differences aren't evident to the eye.

My research so far has turned up two little "pentodes" that are actually beam tetrodes: one is the little 7-pin 6AG5, the other is the supposed "pentode" in the 9-pin 6JW8.

While manufacturers datasheets call the 6AG5 a "pentode", one website does agree with my magnifying lens: it's a beam tetrode! (6AG5 @ The Valve Museum )

I like the sound of a 6AG5, it seems to easily produce lovely shimmery clean tones. They tend to be dirt cheap, too, at least in the USA. To me, they are kinda-sorta small-signal 6V6 equivalents. And a lovely little secret to have, given the insane pricing of NOS 6V6 valves.

I have the feeling you prefer a lot more distortion than I do, so a true pentode (like the EF86) might be more to your taste than the 6AG5 - it's more likely to growl, like a good rock guitar amp should.

In the USA, there are also small-signal (true) pentodes that are unloved and dirt cheap, but which sound great. The little (7-pin) 6AK5 is one example. Nobody ever used one in a "classic" guitar amp, so nobody thinks they are worth anything...

  Reply With Quote
Old 1st February 2018, 10:39 PM   #22
Gnobuddy is offline Gnobuddy  Canada
diyAudio Member
Join Date: Mar 2016
Since this thread is about getting good distorted guitar sounds at neighbour-friendly SPL levels, I'd like to throw in a surprise contestant.

Take a listen to this guitar clip. (Not the somewhat erratic playing technique, just the guitar tone): http://www.diale.org/mp3/v1.mp3

What sort of awesome valve guitar amp do you think made those sounds?

Maybe this one? (try 9 minutes into the clip): YouTube

Well, no. The real answer is here (including the link to the first sound clip in this post): 2BJTE - 2 BJT triode emulator

(Disclaimer: This is NOT my work. And I have not built and tried this out yet. But it is definitely on my list.)

  Reply With Quote
Old 1st February 2018, 11:55 PM   #23
LeftHandFool is offline LeftHandFool  United Kingdom
diyAudio Member
Join Date: Oct 2017
Location: Carlisle, UK
Originally Posted by Gnobuddy View Post
I have the feeling you prefer a lot more distortion than I do...


This is the big question, isn't it. What you consider 'hi-gain' might be 'early break up', as far as I'm concerned.

So difficult to describe in words, but we may not be so far apart. I prefer to drive the 'clean' channel of an amp into power amp overdrive - think Vox Ac15 or Marshall 18W - for a crunchy 'classic' rock sound. Back off the volume for the cleaner passages, step on a treble booster for the leads.

I'd call thet a Classic Rock set up, and you're spot on about the EL84's. That was never a conscious choice though, just most British amps use 'em.
I had a Selmer Treble 'N' Bass 50, with EL 34's, which may have been my favourite amp of all, but was old and kept finding new ways to die.

Marshall JCM800/Soldano Super Lead style cascading gain stage amps are not my bag normally, but I'm coming around to the idea that not all pre-amp gain is equal. It's becoming increasingly apparant that to get a 'good' sound at bedroom levels, I'll need more pre-amp gain than I'm used to.

Aside from that, many thanks for another informative post. I've still got a lot of thinking to do on this project, but I feel like I'm getting close to a workable starting point.


  Reply With Quote
Old 2nd February 2018, 12:36 AM   #24
Printer2 is offline Printer2  Canada
diyAudio Member
Join Date: Apr 2010
Originally Posted by zenelectro View Post
A couple of things:

a/ That schematic is copyright information
b/ It's a great sounding amp

I would take the front end of an '800 (2203) with master vol and marry it to the 1W back end.


The schematic is drawn up by Matek and has about five different revisions, which one or part are copyright information?

Your welcome.
  Reply With Quote
Old 2nd February 2018, 02:09 AM   #25
Bigun is offline Bigun  Canada
diyAudio Member
Bigun's Avatar
Join Date: Jan 2009
Location: Waterloo, ON or Herefordshire UK
Suggestions for a Medium Gain Guitar Amp Topology...
what about just running the amp at full tilt, put the speaker inside a larger box designed to reduce the sound level that escapes into the listening area. It won't be at all easy to get it to sound 'right' but who knows.....
"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 February 2018, 05:38 AM   #26
PRR is offline PRR  United States
diyAudio Member
PRR's Avatar
Join Date: Jun 2003
Location: Maine USA
Originally Posted by Gnobuddy View Post
...I dug deeper, and found that a lot of good information had been largely lost to history in the last several decades....
Still lost. You might re-review the multi-grid family tree.

First: at least in practice, there are two applications. Small signal and power.

Small triode radio amplifiers oscillate from plate-grid feedback. If you put a grid between and ground it, no electrons flow to the plate. If it is an open grid and tied to a medium-high voltage, electrons flow and feedback is greatly reduced. The plate can't swing below G2, but in small-signal that is a minor problem compared to a radio that mostly transmits.

In Power, it is a problem, not having full plate swing. Actually for HIGH voltages (actually high impedance, which is easier in radio than audio), you can put G2 at 400V and Plate at 2,000V and get near-full swing, OK. But audio tends to like fairly high current with its voltage.

The problem is that the G2-P space fills with secondary electrons splashed off the plate. If plate is not most-positive, they go to G2.

A third grid at low potential can shield this electron storm. Now the plate can swing very low.

It does not have to be what we call a grid. Actually there can be a Critical Distance from G2 to plate where a negative region forms. Or two side-rods can induce a negative region. While it is all about the same to the electrons, it looks different on the patent drawings. RCA bought several of these ideas.

Then there is Aligned Grid. G2 tends to steal a significant fraction of the cathode current. Overall you come out ahead of a triode because of the deep plate swing. However as long as RCA was re-inventing (to dodge patent) they also bought the idea to hide G2 in the shadows of G1. You would think that just winding the same pitch would do that. But if you wind grid wires around grid rods of different spacings, the slant is different. RCA built a machine which cuts diagonal notches in the G2 rods so the grid wire does not just wrap, it jogs, and ends up same-slant as G1 wire. This is a fairly expensive thing to do in production, which is why the EL34-family remained popular even after these patents expired and competitors built grid-jog lathes for 6L6-family tubes.

All of these tricks can be bent one way or another. I don't think you can "tell" which construction from data. The pentode knee can be high or low (relative fraction of Vg2), sharp or broad, dippy or not.

There were a few TV Tuner tubes which were odd. Low noise (snow) suggests triode. Good stable gain with that leads to two triode cascode. But there were some "tetrodes" with internal side-fins which reduced grid-plate feedback low-enough, and without reducing current, and apparently without being kinky. These may only work good for the mV signals of a tuner; I have never played with one.
  Reply With Quote
Old 2nd February 2018, 10:49 AM   #27
tristanc is offline tristanc  United Kingdom
diyAudio Member
Join Date: Dec 2015
Location: Oxford
Originally Posted by LeftHandFool View Post
I was looking to go down something similar to your Firefly route.
However, given the advice I've had here, and the fact that you weren't too smitten with the Firefly I'm starting to reconsider. Biting the bullet and accepting a little more preamp gain seems the sensible route. After all, there's no substitute for genuine power and ear crushing volume...

What was it about the Firefly you didn't like?
I didn't get on with the _original_ Firefly topology as per Doug's schematic with some alterations (AX84 Firefly-style build – Tristan Collins ).

I think this was due to:
- no master volume
- the cascode at the front hitting the next stage too hard
- VVR of the whole amp was a truly terrible idea

So whilst it gave nice clean tones with the boost off, I didn't like the boost and the thing was way too loud for my needs.

Hence I went down the simpler 3 stage + CF tried and tested route. Lots of preamp gain / distortion for playing whilst watching TV etc. This thing sounds huge for its size. Lower preamp voltages help with that I believe.

Also, lowering the CF load resistor away from 100k helps as Merlin discusses in the book - no need to hit the output stage with a huge swing and this might make up for the lack of phase inverter distortion.

Merlin's Bone Ray tone stack is really useful as well. I end up running the tilt at 90% and about 20% on the middle for heavy stuff. And more around 50% for normal rock. I can set it for heavy sounds and clean up / control from the guitar easily - no need to move from the sofa or put pedals up front!

It was a fun build (my first point to point), and didn't take too many tweaks to get it 'right'.

If you're interested, I have some PCBs for general preamp stages / tone stack / CF left over - Merlin and I designed similar things at the same time. His no doubt better... You could whip up something really quickly with them, and come up with a few interchangeable options... AX84.com - The Cooperative Tube Guitar Amp Project
  Reply With Quote
Old 2nd February 2018, 02:31 PM   #28
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
Tubelab_com's Avatar
Join Date: Jul 2005
Location: West Virginia panhandle
You might re-review the multi-grid family tree.
What PRR said about pentodes and patents is 100% correct. Much of this is explained in the old RCA tube manuals. Pages 5 through 10 in the RC30 edition are worth reading. I have a few additions.

Most EL34's are true pentodes with three spiral wound grids. They typically eat more screen current than "beam power tubes" when operated under similar conditions. I have seen some EL34's that do have aligned grids, and their screen current is far lower. I have also seen a few "EL34's" that were beam tubes, but they don't exist right?

The EL34 was a pure pentode of UK and European origin. When they became popular in the USA, Sylvania simply rebranded Mullard made tubes.....for a while. Then they introduced the 6CA7, which was a beam power tube, and according to Sylvania, it was compatible yet "better." For use in Marshall amps that ate tubes, the "fat bottle Sylvania 6CA7" is prefered by many users for longevity. The cross breeding of the lineage occurred when vendors started making tubes with both numbers on them. These could be mongrels with anything inside them. Some of these mongrels are just labeled EL34.

The true pentode, big or small, has three spiral wound grids. A "beam tube" has two spiral wound grids and a beam forming plate, which is a closed metal box, connected to the cathode, ground, or in special cases a small positive voltage. This box has a window opening in it through which all plate current must flow. As stated this was primarily "invented" to dodge a patent, but has been tweaked for good purpose over the years. I call these "beam tubes" because the terms "beam tetrode," "beam pentode", or just "pentode" have been used to describe the these tubes over the years since the pentode patent expired.

A true pentode with a round cathode and round plate will have electron flow in nearly a circle, yes the flow is greatly reduced in the shadow of the grid rods. This is desirable in many small signal pentodes for noise control, especially at RF frequencies. For a power output tube it can be a problem because the electron flow will not be evenly distributed across the entire plate structure unless everything inside the tube is perfectly concentric. No tube is perfectly aligned, so hot spots will develop in areas where the electron flow is the highest.

The beam forming electrode can be used to form the electron flow into a beam which is confined to a smaller area of the plate. On high powered tubes this is usually the seam area where the two plate halves are joined. The seams are often folded outward and extended to create heat radiators. This is the reason for the plate dissipation increase from the 6L6GB to the 6L6GC. Some vendors have added additional heat radiating fins in this area to further improve heat dissipation.

Note that the outer shell that you see when looking through the glass on many small signal tubes, particularly those originally intended for use as RF or IF amplifiers is NOT the plate. It is an outer shield usually internally connected to the cathode or G3. The plate in these tubes is usually two small rectangles inside this shield to reduce it's capacitance. These tubes are sometimes actually beam tubes regardless of what the book says.

Here are pictures of two beam power tubes that I have dissected. As you can see they were quite dead before I smashed them.

The Sylvania 6V6GTA was gassy so I just turned up the power supply and watched it fry. I have peeled back the plate (black) to show the bent cathode, what's left of the control and screen grids, and the beam forming "box" with its window (silver). Note that the cathode in this tube is nearly round. Beam control is done entirely by the beam forming electrode.

The 6BQ6GA died from a screen drive experiment gone wrong. This little guy was half of a push pull pair busy cranking out 125 watts before a tube arc ended the fun and blew up a bunch of other parts in the process. The other tube survived. Here I tore the plate completely away and opened up the beam former to find out where the arc took place. I would guess that it was right where the screen grid rod melted in half (upper right).

The plate is also shown. Note that there are actually 3 pieces of metal in the plate sandwich. The center piece extends slightly into the internal plate cavity. This is not for heat radiation, but to cure an oscillation mode particular to TV sweep circuits. Some larger sweep tubes may have two or three internal fins. Often these fins do extend outside the plate area for heat radiation purposes. It is between the screen grid rods, and both of these internal fins that the arc happened.
Attached Images
File Type: jpg 6V6.jpg (162.8 KB, 89 views)
File Type: jpg 6BQ6GA_no_plate.jpg (163.8 KB, 88 views)
File Type: jpg 6BQ6GA_plate.jpg (153.1 KB, 89 views)
Tubelab, it's 5 year mission. To explore strange new tubes, to seek out new circuits and topologies, to boldly go where no tube has gone before......
  Reply With Quote
Old 2nd February 2018, 05:48 PM   #29
Gnobuddy is offline Gnobuddy  Canada
diyAudio Member
Join Date: Mar 2016
Thanks to PRR and George (Tubelab). Lots of fascinating stuff there.

And we haven't even mentioned some of the weirder vacuum devices like heptodes. There was another weird one I stumbled across a few years ago that used control electrodes to steer electrons (from a single shared cathode) between either one of two anodes. I can't remember what this was called, or what its intended application was (RF mixer?).

My understanding is that the "beam" in "beam power tube" is actually the sheets of electrons squeezing between the aligned grid wires. The beam forming plates used in early beam power tubes constrained the electron flow in the perpendicular direction to simplify manufacturing in earlier devices, but were evolved away in later generation beam power tubes, which had full cylindrical symmetry, and no beam forming plates.

I found a well-written 'Web page that described this evolutionary history of beam power tubes a few years ago, but have since lost the bookmark, and the newly ruined Google can no longer find it.

  Reply With Quote
Old 2nd February 2018, 09:04 PM   #30
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
Tubelab_com's Avatar
Join Date: Jul 2005
Location: West Virginia panhandle
Here is the picture from the 1939 article by RCA introducing the new 6L6 tube. It has been reprinted in almost every edition of their tube manual since then. The picture shows the electron streams seperating as they pass through the grids, then recombining as they head toward the plate.

mentioned some of the weirder vacuum devices like heptodes.
Heptode......5 grids (7 elements) most often found in the form of a pentagrid converter, a tube used in AM radios as the mixer / oscillator. The common flavors are 6SA7 old octal radios, and 6BE6 in mewer 7 pin miniature radios. The 6BE6 makes a nice triode is you wire all the grids together.

Hexode......4 grids (6 elements). To make matters more confusing, there exists at least one "beam hexode", the 6GU5. They make nice preamp tubes, I plugged them into a 6CB6 pentode socket to get more gain.

I stumbled across a few years ago that used control electrodes to steer electrons
There are a few gated beam, or beam deflection tubes around for various uses. The most common was for demodulating a phase modulated color signal in TV sets. The best undocumented use? An audio compressor. There are types (I can't recall the numbers right now) that stack the steering plates on top of a pentode. Use the pentode to amplify the audio, then steer that audio stream to one of two plates, or spread it between them. Use one plate for the output. That way you can vary the output level without the distortion associated with variable Mu devices.
Attached Images
File Type: jpg BeamTube.jpg (344.4 KB, 70 views)
Tubelab, it's 5 year mission. To explore strange new tubes, to seek out new circuits and topologies, to boldly go where no tube has gone before......
  Reply With Quote


Suggestions for a Medium Gain Guitar Amp Topology...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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Variable Gain Control Topology JFace Tubes / Valves 16 2nd March 2015 08:33 AM
Advice: medium power bass guitar amp/speakers dizmayed Instruments and Amps 7 30th June 2011 06:48 AM
Medium power Class D battery bass guitar amp SEclassA Class D 5 14th December 2010 11:01 AM
high gain 2nd order high pass active filter topology sreten Solid State 21 23rd March 2006 01:26 AM
DIY preamp design - gain topology, IC vs. discrete, OPA627 vs. Zen... ? dorkus Pass Labs 18 10th July 2001 05:26 AM

New To Site? Need Help?

All times are GMT. The time now is 01:11 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio