# Electron Flow in a Vacuum Tube Guitar Amplifier

#### Gerry Rzeppa

Joined Jun 17, 2015
170
I'm trying to understand the actual flow of electrons in a guitar amplifier circuit. Consider the partial circuit below.
Everyone seems to agree that in the wire marked with pink patch (A) the electrons actually flow in both directions (alternating current).

And everyone also seems to agree that in the wire marked with pink patch (B) the electron flow is in a single direction, as shown by the arrow, though the voltage varies (pulsating direct current).

The question is, Which way do the electrons flow in the wires marked with pink patches (C) and (D)?

#### #12

Joined Nov 30, 2010
18,210
Both ways.

The capacitor makes the DC component irrelevant so only the AC component travels in that line...except for the moments of start-up and shut-down.

#### Gerry Rzeppa

Joined Jun 17, 2015
170
Both ways.
That's what I thought. Now can we say a similar thing about the output stage of the amp?

Specifically, that the primary side of the output transformer sees electron flow in just a single direction, but that this pulsating DC induces actual bi-directional electron flow (AC) on the secondary side of the transformer?

#### #12

Joined Nov 30, 2010
18,210
Sure. Just remind yourself that it is impossible for a vacuum tube to run (current) backwards. As soon as you remove the DC component, by capacitor or transformer, it's all back to the back and forth of the intended signal.

#### Hypatia's Protege

Joined Mar 1, 2015
3,226
I'm trying to understand the actual flow of electrons in a guitar amplifier circuit. Consider the partial circuit below.View attachment 87272
Everyone seems to agree that in the wire marked with pink patch (A) the electrons actually flow in both directions (alternating current).

And everyone also seems to agree that in the wire marked with pink patch (B) the electron flow is in a single direction, as shown by the arrow, though the voltage varies (pulsating direct current).

The question is, Which way do the electrons flow in the wires marked with pink patches (C) and (D)?

As an aside: It's more meaningful to discuss charge flow -- The actual behavior (and, in fact, very nature) of electrons is rather complex...

Best regards
HP

Last edited:

#### Gerry Rzeppa

Joined Jun 17, 2015
170
Sure. Just remind yourself that it is impossible for a vacuum tube to run (current) backwards. As soon as you remove the DC component, by capacitor or transformer, it's all back to the back and forth of the intended signal.
Great. You've been very helpful. Let me try to sum up. This is the description I hope to give to my ten-year-old to help him understand how a tube guitar amp works:

"You strum the guitar and the movement of the metal strings through the magnetic field around the pickups makes the electrons run back-and-forth through the wires. The faster they change direction, the higher the note (pitch, frequency).

"These wires are connected to the gate (grid) in the middle of the tube, and this back-and-forth movement of electrons opens and closes that gate. When the gate is open, a whole bunch of electrons run upward from the pool at the bottom of the tube, through the gate, and out the top.

"The (anode) resistor prevents the electrons from escaping too quickly back into the electron pool (ie, ground via the power supply), so the electrons alternately bunch up (and un-bunch up) against the capacitor's "rubbery wall" (as the gate in the tube opens and closes). This causes the electrons that are sitting on the other side of the capacitor's rubbery wall to move back and forth in sync with the ones from the guitar. Only there's more of them. And they move in the opposite way (180 degrees out of phase).

"Ditto for the power amp stage, only we leave out the resistor and let the electrons rush back to the pool as fast as they can because (a) the wire in the transformer is itself a kind of resistor, and (b) the varying concentration in that wire is enough, all by itself, to generate the back-and-forth electron movement we need on the secondary side of the transformer to make the speaker -- which in a lot of ways is the opposite of the guitar pickup -- wiggle in sync with the strings.

Sound reasonable?

#### AnalogKid

Joined Aug 1, 2013
8,536
That's what I thought. Now can we say a similar thing about the output stage of the amp?

View attachment 87273
Specifically, that the primary side of the output transformer sees electron flow in just a single direction, but that this pulsating DC induces actual bi-directional electron flow (AC) on the secondary side of the transformer?
Yes, except that for audio the plate current is not pulsing, but changing as a function of the input voltage, so it has both an AC and DC component. The transformer primary sees *varying* electron flow in just a single direction. Algebraically, the current is varying from one positive (depending on how the meter is connected) value to another, never decreasing to zero or reversing (going negative). Thanks to Faraday, the transformer secondary does not see the DC component. The transformer core does, and that static flux must be accounted for in its design.

ak

Joined Jul 31, 2013
218
Dare I say that an explanation of a tube circuit based on electron flow for a ten year old is a complicated approach. In fact after more than 60 years as hobbyist I still only think of "conventional current flow" - positive to negative. Surely consideration of electron flow goes hand in hand with proton flow and is mainly of concern to the scientist.
I would describe that circuit to the youngster in terms of voltages, having first described the difference between AC and DC.
In so doing it would have become clear that in this particular case the input stage, as drawn by the OP, is not correct. The bias to a tube is not applied in parallel with the source, but in series with it. The "bias" would normally be in the cathode circuit to ground.
This means that the grid "sees" a DC voltage with a varying voltage from the pickup superimposed. At least in that circuit as drawn. Insert a capacitor between the source and the grid and the explanation is more accurate.
The signal on the anode of the capacitor coupled output is then the same, but amplified, as the grid signal. ie A varying DC voltage where the variation is the wanted signal. Or am I being pedantic ?
as an edit: Back in the valve/tube days we really only thought about voltages. The tube is a very high impedance device and there is no current flow at the grid in linear (class A) operation. So the assumption in the first line that current A flows "in both directions" is a false premise. There is no current flow.

Last edited:

#### #12

Joined Nov 30, 2010
18,210
I disagree.
Tube circuits based on electron flow = complicated? Is it easier to understand that a voltage applied here causes a larger voltage to happen there, while ignoring what in the world makes one voltage connect to the other? Maybe I'm a late bloomer, but I had the curiosity and intellect to ask about and understand that one thing has to control the other when I was 14 years old. There is no magic in a box called, "tube" or, "transistor". I am looking at a CRT right now, and I know from decades of measurement that current flows through it. Ignore that at your peril.

As for being pedantic? I say you answered a question that was not asked, based on your assumption that the box labeled, "bias supply" could not contain the "C" battery used in early radios.

Sorry if I sound harsh, but you are posting theories that contradict my personal experience. Apparently I went through this process very differently than you.

Joined Jul 31, 2013
218
I'm sure we both understand well enough how a tube circuit works. Within the tube a varying current flows - controlled by a voltage. It would take forever to find where we do disagree so I won't go there.
But I made no assumption about the "bias box". If it contained a cell, dc current would also flow through the pickup coil....unless of course it was a piezo....A high impedance voltage generator capable of producing virtually no current and having infinite resistance.

#### #12

Joined Nov 30, 2010
18,210
I can agree with that. A guitar pickup has several K of ohms, but that is a huge waste of current compared to what is needed to control the grid. If only a capacitor was in the first drawing, this part of the conversation would be moot.

#### Gerry Rzeppa

Joined Jun 17, 2015
170
My goal is to give the kid some kind of tangible picture, void of abstract and technical terms as much as possible, and full of words he already knows. A picture that he can use to guess what will happen if we change the circuit in this way or that.

Case in point. Having explained the circuit as I have, I ask the kid what he thinks will happen if we increase the value of the (anode) resistor in the first circuit. Using nothing but the electron flow picture that I've given him (and his knowledge of everyday objects), the kid should be able to say, "Well, it will be harder for the electrons to go that way and easier for them to head off to the rubber wall in the capacitor. Which should make the amp louder. But it will also tend to make the electrons bunch up by the rubber wall (since it will be harder for them to get back and out) and that will probably goof up the sound." Then I ask him about the effect of decreasing the value of the resistor, and he should be able to say, "Well, I don't think that will goof up the sound, but it will probably make it softer and if the resistance is too low, all the electrons will go up and away and none of them will bounce off the rubber wall in the capacitor and we won't hear anything."

To trained ears, a crude picture no doubt. But a picture, in familiar terms, nevertheless. Compare that with this description that I just pulled off another forum in answer to a similar question: "a larger plate resistor will cause the tube to draw less current. But, because there is a larger voltage drop across a larger value resistor, you will obtain a larger signal (more voltage swing) off the plate." THAT, I'm sure, would mean nothing to my kid.

The good part is now that I have a picture in the kid's head, I can start defining terms, not in terms of other terms, but in relation to his picture. I can, for example, tell him that a "goofed up" sound is a distorted sound. Or that voltage is (in our picture) simply the difference in the number of electrons in one spot versus another. Though I'll have to warn him about all the meters being wired backwards, of course!

#### Gerry Rzeppa

Joined Jun 17, 2015
170
I apologize for the confusion regarding the "bias supply". The graphic was from a different discussion. That box should been just a grid resistor. But while we're on the subject, let's discuss the electron flow in that part of the circuit. Here's a better diagram:

Exactly how do the electrons flow in the left part of this circuit? Seems to me, when the guitar is strummed, the electrons first run one way (applying pressure to the rubber wall in Cg), and then they go the other way (releasing that pressure and simply piling up in the ground pool at the bottom -- I'm ignoring the Rg leg of the circuit for the moment).

Meanwhile, on the right of Cg, the electrons are going back and forth, in and out of the tube as the "rubber wall" in the capacitor flexes. So what's the function of Rg? Seems to me Rg "regulates" the amount of fluctuation on the right side of Cg by providing a path to the ground pool when there are too many electrons, and by providing additional electrons from the ground pool when there are too few. If Rg is small, the tube's grid will tend to have extra electrons (since they'll be easy to suck from the pool); this will make it harder to open the gate. If Rg is big, the tube's grid will tend to have fewer electrons (since it will be hard to suck them from the pool); this will make it easier to open the gate.

Yes?

Joined Jul 31, 2013
218
This goes to the heart of your present understanding of an amplifier tube circuit. In this correctly drawn version the input impedance of the tube IS Rg. The purpose of Rg is simply to connect the grid to gnd to make it negative to the cathode by the voltage developed across Rk. Thus "biasing" the tube so that a constant specific current flows through it. (in the absence of a signal)
It's bad to use the word gate when talking about a tube. The fixed current flowing through the tube is controlled up and down by the varying input voltage from the pickup superimposed on the bias voltage. As long as the input voltage never goes more positive than the -ve bias voltage NO current will flow into the grid. The grid controls by voltage only.
The input impedance of the circuit is set by Rg, typically 1megohm. True, a tiny AC current will flow from the pickup (via the cap) through Rg but that's not intrinsic in the operation of tube amp.
If for example a -ve bias cell was inserted between the lower end of the pickup and ground, thus presenting the grid with a varying DC voltage, no current at all would flow in the input circuit. (assuming Rg, Cg, Rk and Ck removed)
That was my point in querying an explanation with current flow. Now, if that had been a basic transistor amp stage things would be different.

Last edited:

#### #12

Joined Nov 30, 2010
18,210
As long as the input voltage never goes more positive than the -ve bias voltage NO current will flow into the grid.
I just want to sharpen that point. If the grid ever becomes more positive than the cathode, the grid will act like the plate of a rectifier tube and start loading electrons on the grid side of Cg. If you keep this up faster than Rg can release those electrons, the average voltage of the grid becomes more negative, thus suppressing average current flow through the tube. That reduces gain and might eventually arrive at goofy sound. It doesn't hurt anything. It's a sort of self protection in this circuit. That's why you can add a stomp box and not melt the input tube.

#### Gerry Rzeppa

Joined Jun 17, 2015
170
This thread is about the flow of electrons. So statements like, "...If the grid ever becomes more positive than the cathode, the grid will act like the plate of a rectifier tube and start loading electrons on the grid side of Cg. If you keep this up faster than Rg can release those electrons, the average voltage of the grid becomes more negative, thus suppressing average current flow through the tube..." are helpful because I can easily convert them to electron flow statements: "If the grid ever has fewer electrons than the cathode, electrons will flow from the cathode through the grid to the right side of Cg. If this happens faster than the electrons can escape through Rg, electrons will start to pile up on the grid and the gate will start to close."

On the other hand, phrases like, "the input impedance of the tube IS Rg" and "the varying input voltage from the pickup superimposed on the bias voltage" and "a -ve bias cell" are less helpful (in this immediate context) since it is not obvious how they translate into electron flow terminology. If you could provide translations that would be great.

Remember, I'm trying to describe the whole thing in terms of, and only in terms of, the flow of electrons. To a ten-year-old. In the spirit of Ken Amdahl's delightful book, There Are No Electrons -- Electronics for Earthlings ( http://www.amazon.com/dp/0962781592 )

#### #12

Joined Nov 30, 2010
18,210
Thank you for your kind words. I have made a hobby of describing electronics on the level appropriate for a 10 year old. Sadly, it still doesn't make good dinner conversation with 99% of the people in the real world.
Congratulations on having a kid in the top 1%.

#### nsaspook

Joined Aug 27, 2009
7,470
Remember, I'm trying to describe the whole thing in terms of, and only in terms of, the flow of electrons. To a ten-year-old. In the spirit of Ken Amdahl's delightful book, There Are No Electrons -- Electronics for Earthlings ( http://www.amazon.com/dp/0962781592 )
The best thing you can do is to NOT fixate on electrons but to explain 'Electricity' in terms of charge and potentials. The physical movement of particles is important to explain the physics of device operations but it's A part of the system and usually is not the most important aspect of the system when analyzing a circuit.

http://amasci.com/miscon/eleca.html

Last edited:

#### Gerry Rzeppa

Joined Jun 17, 2015
170
Thank you for your kind words. I have made a hobby of describing electronics on the level appropriate for a 10 year old. Sadly, it still doesn't make good dinner conversation with 99% of the people in the real world. Congratulations on having a kid in the top 1%.
You're quite welcome. And thank you. Chuckles is the child of our old age: mixed up in a dish, frozen, thawed, and implanted in my 56-year-old wife's well-past childbearing womb. He's now almost ten (this is a photo from a couple of years ago):

We taught him to read with our www.rhymingreader.com system. But back to electronics.

These "electron flow" descriptions were originally intended to be part of a no-solder tube guitar amp kit. I put the idea out on KickStarter some time ago, but failed to get funded. So it's just the little guy and me for now. The project (I think you'll like the song) is here:

https://www.kickstarter.com/projects/1335354839/banana-jack-amps-no-solder-all-tube-guitar-amp-kit

#### Gerry Rzeppa

Joined Jun 17, 2015
170
The best thing you can do is to NOT fixate on electrons but to explain 'Electricity' in terms of charge and potentials.
The problem is the kid has no way to picture "charge" and "potential" and so those words, to him, are words without corresponding pictures -- in other words, just words. We need a picture, a model, that is accurate enough so it can be used (a) to explain the behavior of the system, and (b) to make predictions about how changes in the system will affect that behavior -- but at a level of abstraction that can be easily grasped by a ten-year-old. Opening and closing gates, water flowing through a pipe, line-dancing-electrons moving left and right, rubbery sheets stretching and springing back -- these are the kind of things the kid has first-hand, tangible experience with. "Charge" and "potential" and (from another post) "voltage superimposed on a bias voltage" are things he has no experience with and thus cannot picture in any meaningful way.
http://amasci.com/miscon/eleca.html

I studied that site rather extensively when I first started my Banana Jack Amp project, and I agree it's chock-full of good stuff. But as Bill himself admits ( http://amasci.com/miscon/elteach.html ), "This is the one thing I lack, good replacements for the incorrect material in K-6 textbooks." I'm searching for, and working on, that elusive "good replacement" -- at least in the context of a simple guitar amp.

The best alternative I've seen at a more general level is the book I mentioned in an earlier post ( http://www.amazon.com/dp/0962781592 ). My wife -- who has no interest in electronics -- was unable to stop reading it when I asked her what she thought about it! And she was able to ask sensible questions about circuits when she was done reading it. So there must be something to this kind of approach.