Tubes and frequency response

[aro]

I have another dilemma regarding the functioning of the tube...

We have a guitar with certain tonal characteristics, for example due to treble-y pickups. The signal from there hits the grid, which manipulates the plate current and eventually we have amplified voltage at the plate output due to the increased flow of electrons from cathode to plate. So far, so clear...

I assume that the Miller effect is responsible for making sure that the tonal characteristics of the pickups are not lost, but exit the tube at the plate output.

What I don't understand though is what happens in the tube, how does the frequency "make it" from input to output? I understand that it is the flow of electrons that makes possible voltage amplification, but what's the "vehicle" for the transfer of the pickup's tonal characteristics from input to output?

Sorry if it's lame question, but it's obsessing me.

[HotBluePlates]

I assume that the Miller effect is responsible for making sure that the tonal characteristics of the pickups are not lost, but exit the tube at the plate output.

Nah, Miller Effect is something entirely different.

Ya gotta first read up on basic electricity and electronics. I suggest googling "NEETS" which is a series of books to teach basic electricity and electronics to Navy folks, and some of the early volumes will serve you well. I wouldn't suggest bopthering with the stuff about radio propagation, waveguides and all that mess.

Two conductors seperated by an insulator create a capacitor. How much capacitance exists depends on a numbers of things, including how large the conductors are (bigger = more capacitance) and how closely they are spaced (closer = more capacitance).

Vacuum tubes are bits of metal with a vacuum space between them. By itself, the vacuum is a fairly good insulator. But the bits of metal are comparatively far apart, and fairly small when compared to the metal that forms the plates of a common cap, so the capacitance between the elements is fairly small, on its own.

The capacitance from the grid to the plate in a triode, when the tube is actually being used, is many times bigger than it is when measured with the tube out of the circuit and not energized. If memory serves, it is the amplification of the tube plus 1 times bigger than the static measured value. This capacitance reduces the effective gain of the tube at high frequencies, so it rolls of highs.

The total input capacitance of a typical 12AX7 stage is usually something around 90-120pF, depending on the gain of the stage in the actual circuit and on stray wiring capacitance. This capacitance interacts with resistance in the surrounding circuit to roll off highs, however the resistance commonly used in guitar amps is low enough that the roll-off takes place above the audio range (or at least above the guitar range).

[HotBluePlates]

What I don't understand though is what happens in the tube, how does the frequency "make it" from input to output?

You do understand that in properly-functioning guitar amp circuits, nothing is intended to actually ever flow into the grid, right?

Many folks outside the U.S. call a tube a "valve". You crank the handle of a faucet outside your house back and forth, and the water from the garden hose flows out in greater and lesser amounts. You're controlling the water with a valve handle rather than physically sticking your hand in the pipe to stop the water from flowing.

An energized vacuum tube wants to have a steady stream of current flowing from the cathode to the plate. You can't stick your hand inside the tube to alter that current, so the tube comes with its own "valve handle" in the form of a control grid. If a voltage is applied to the grid that is more-negative than the cathode, it reduces the flow of current, until the voltage is negative-enough to completely stop all current flow.

If we assume a perfect, ideal tube (we're perhaps not ready to consider real-world imperfections yet), then whatever voltage signal is applied to the control grid results in a current through the tube which tracks every variation and nuance of the input signal. So whatever is present on the input is also present in this flow of current from cathode to plate.

There is no useful output at the plate (useful in preamp situation is most often a voltage output) until we stick a largish resistor in between the plate and the power supply. A resistor with a little current through it results in a little voltage dropped across the resistor; if a big current flows through it, the a big voltage is dropped across the resistor.

Our ears do not hear all the way down to 0 Hz. So for us to hear something, there has to be variation in it. The plate output of a common preamp stage is the result of a varying current flowing through the tube (varied by applied the control grid voltage) flowing through a plate load resistor, which then creates a varying voltage drop across the resistor. The varying voltage now present at the plate is passed through a coupling cap (which passes a varying voltage, but blocks a d.c. voltage) to the control grid of the next tube.

If the tube is perfect (and we said ours would be just that), the output signal is exactly like the input, but bigger. With our collection of perfect tubes, no alteration of the frequency or characteristocs of our guitar signal are changed, except by the effect of external components like bypass and coupling caps.

[aro]

I read the Navy manuals and I read about the tubes capacitance in Merlin's book. I understand the concepts, but I was under the false illusion that tone, or frequency, travels through the tube through an "intermediary", just like current is an intermediary between input and output voltage.

You do understand that in properly-functioning guitar amp circuits, nothing is intended to actually ever flow into the grid, right?

I do, hence my initial confusion. The initial signal doesn't enter the grid, so how does it "pass" its tonal characteristics to the output signal?

I also understand your description of all the processes and components involved in getting a proper output signal. The Rider book was of invaluable help.

If we assume a perfect, ideal tube (we're perhaps not ready to consider real-world imperfections yet), then whatever voltage signal is applied to the control grid results in a current through the tube which tracks every variation and nuance of the input signal.

This is exactly what goes to the heart of the matter. How is the current able to "mimic" all the variations and nuances of the input signal? What enables the current to do that? I realize this knowledge is not needed in order to understand a gain stage, but it's intriguing.

[jjasilli]

The sounds that you hear come from waveforms in the air.  A sine wave is a "perfect" wave form.  A tuning fork produces a sine wave.  It is pleasing, but a bit satiating.  A square wave has sharp edges.  It produces a harsh, grating sound, but it gets your attention -- like a modern ambulance siren.  The same musical note sung by a little girl sounds pleasing, by an old man more gruff.  This is timbre.  It all winds up in the unique shapes of various waveforms in the air, as they deviate from perfect sine waves and take on shapes which correspond to their particular tonal characteristics. 

Guitar strings are like vocal chords in a human being.  The tone they generate is affected by the size, shape and type of wood and/or other materials to which they are attached.  The vibration of metal strings induces an AC signal in the pickups.  This electrical signal has a waveform with unique shapes, just like waveforms in the air.  The amplifier takes weak waveforms and makes them taller (more powerful), by increasing their amplitude while (more or less) preserving the particular shapes of each waveform.

[aro]

Thanks, guys. Your replies have kept pushing me in my search, and it looks like I found the missing link (missing in my head).

I believe the answer is modulation, in this case, amplitude modulation. NEETS Module 12. Now I have something fun to read this rainy Sunday :read: :toothy2:

[HotBluePlates]

I believe the answer is modulation, in this case, amplitude modulation. NEETS Module 12.

You could look at it that way, but the overall concept of modulation as presented in that book will confuse you somewhat. That's because they're talking about radio signals.

But you will get a parallel to what you're asking, in that the idle current in the tube is modulated by the signal voltage applied to the grid, which you could think of as similar to a carrier wave being modulated by a signal in a radio wave.

How does the grid voltage really do it? Electrostatic force.

The electrons are (according to us) negative. When the cathode gets hot, electrons are released. If nothing else happened (no other voltages applied to the tube elements), they would just sit dormant near the cathode, or float aimlessly in the vacuum. But in actual use, the electrons are pulled to the plate by a large positive voltage.

The grid sits in the path between the cathode and plate, so the electrons must flow in the spaces between the grid wires to get to the plate. Like charges repel, so if the grid is made a little negative, the current gets a little smaller. If the grid is made very negative, the current stops.

The basic error might be thinking that "tone" is some discrete thing, rather than the shape of the voltage wave produced by your guitar's pickup.

[jjasilli]

I believe the answer is modulation, in this case, amplitude modulation. Per, Hotblue, you seem to have the right idea but "modulation" is the wrong label.  Modulators modulate.  Demodulators demodulate.  Amplifiers produce gain, not modulation.  Gain = multiplication (not mere addition).  Therefore gain is a geometric, and not a mere arithmetic, operation.  Building on what Hotblue stated, an amplifier multiplies the voltage of its input signal.  Multiplication is not modulation.  If you look at a picture of a waveform, it is a curve. If you amplify it, it gets taller.  If the shape chages in the process, that's distortion.   As the curve gets taller, the area under the curve increases.  The top of a curve is its peak voltage.  The area under the curve is its power.  

[aro]

The basic error might be thinking that "tone" is some discrete thing, rather than the shape of the voltage wave produced by your guitar's pickup.

OK, now I really do see the error in my assumptions. The tonal characteristics of a pickup are inherent in the voltage wave. Duh... That's what you guys have been trying to say. Why do I have to make things more complicated than they are...

Anyway, that one phrase explained everything. Thanks, HotBluePlates and jjasilli.

[PRR]

> The tonal characteristics of a pickup are inherent in the voltage wave.

Thanks for getting that. I saw your confusion and went to bed instead.

The "tonal characteristics" are hard to understand. However whatEVER they are, if we can make a duplicate signal which is EXACTLY the same only BIGGER, we get nearly the same tonal characteristics only LOUDER.

So the amplifier only has to be "exact".

And it only has to be better than the final use of the large wave. Our guitars don't make output below 82Hz. Our ear-bones don't work much past 20,000 shakes per second; for other reasons, guitar need not go this high. The ear is not a precise judge of amplitude. An amplifier that has reasonably consistent gain from 80Hz to 10KHz, and does not add significant amount of other tones (hiss, hum, buzz, distortion) is "perfect enough".

The tube is stupid. Imagine we custom-build a stupid demon. His left arm has a very delicate finger which can rest on your string and follow its vibration without affecting the string. The right arm is powerful. We train the demon to shake his right arm "exactly" the same as his left finger, only stronger. We tie that big-shake arm to an air-pusher. Small string vibrations are turned into large air vibrations.

(In concept, a person could do this. In practice, a human's fastest arm-shake is slower than a slow string vibration. Our frequency response is poor. Our finger-arm coordination is sloppy too. My imaginary demon could be imagined to be as fast and as precise as needed. Tubes are good for low gain to 1,000,000Hz, high gain to 10,000Hz, so audio vibrations are not hard for them.)

Note that the demon is not doing anything "smart", does not have to understand "tonal characteristics". Just has to output what comes in only stronger.

All this strong work makes the demon hungry. Amplifiers have to be fed. Our demons (tubes and transistors) eat a steady electric current such as from a battery (or similar derived from wall-power). I assume a demon has a liver to store food energy between meals. Tubes don't have livers. On battery power, the battery can supply energy as needed. On wall-power, food only comes out of the wall in pulses every 8/1000th seconds, we need a power capacitor to "liver" the energy between food-pulses.

[aro]

I saw your confusion and went to bed instead.

LOL Now I can see why you would do that

Fortunately for me, HotBluePlates and jjasilli had the patience to put up with me a little longer. If courses on this were taught somewhere, there would be less stupid questions on online forums. I read anything I can get my hands on, and there's lots of information online, but nothing can replace a real person when you invariably have lame questions in the beginning.

[HotBluePlates]

Believe me, I took a lot longer to catch on to some ideas that (I now know) are very basic to tube circuits. Everyone deals with the learning curve on this stuff, especially when you don't have someone to answer your specific question.

Per, Hotblue, you seem to have the right idea but "modulation" is the wrong label.  Modulators modulate.   

Well, if you already have a good understanding of modulation, then it is a good description. If tubes are hard to grasp, comparing an RF circuit to the internal action of a tube could lead to needless confusion.

I'll go on the record now as saying it is very valid to view a vacuum tube as passing a plate current which is modulated by a voltage input (regardless of which tube element you choose to use as your input).

If you already know tubes, that definition makes sense; if you don't, the water valve analogy is much easier to grasp quickly.

[HotBluePlates]

We have a guitar with certain tonal characteristics...

I probably shouldn't even mention it at this point, but have you ever heard your guitar's unadulterated sound?

I used to have a dual-cassette deck that my dad bought a long while back (maybe mid-80's). It was a nice Marantz deck with a pair of recording inputs with level controls. Sometimes, when I wanted to play my guitar softly (not through the big amp I had at the time), I'd plug the guitar cord into one of the inputs and play through a channel of the stereo system.

Well, if you like funk music, or you like a very clean, trebly and thoroughly uninteresting sound, you'd like the sound of the guitar pickups being reproduced with no added coloration or help.

What real amp's do is take a fairly boring sound and chop it up (maybe reduce/boost bass, highs, mids) and mangle it up (add distortion). The resulting sound becomes much more interesting and quite a bit different from the original.

But just like a recording or live-sound engineer can't take a lousy-sounding or out-of-tune guitar and EQ it to sound great, you have to start with an instrument with the right raw material to have a good resulting tone from the speaker. It's also why the colorations inherent in some amp designs will sound better to some ears when fed with a Strat, Tele or Les Paul (and often only one of these guitars, rather than all of them).

[Shrapnel]

I probably shouldn't even mention it at this point, but have you ever heard your guitar's unadulterated sound?

[ ... ]

But just like a recording or live-sound engineer can't take a lousy-sounding or out-of-tune guitar and EQ it to sound great, you have to start with an instrument with the right raw material to have a good resulting tone from the speaker. It's also why the colorations inherent in some amp designs will sound better to some ears when fed with a Strat, Tele or Les Paul (and often only one of these guitars, rather than all of them).

Funny you should mention this. Because, in addition to your running it through the tape-deck to a Hi-Fi Amp, just running your guitar straight into the mixing board gives that same sterile, unadulterated sound.

[aro]

We have a guitar with certain tonal characteristics...

I probably shouldn't even mention it at this point, but have you ever heard your guitar's unadulterated sound?

[...]

What real amp's do is take a fairly boring sound and chop it up (maybe reduce/boost bass, highs, mids) and mangle it up (add distortion). The resulting sound becomes much more interesting and quite a bit different from the original.

Point taken. So true...