The theory behind vacuum tubes

[nexekho]

Been reading up on how a vacuum tube actually works and I *think* I have it.

There's an electron gun called the heater which fires electrons down the tube, through a mesh.  This mesh is connected to your input and can either pull the fired electrons out of the tube by applying a negative voltage and attacting them to the mesh or repel the electrons so they slip through the gaps in the mesh.  Down the other end is a collector ring with which the electrons collide, collecting them for output.

Is this right?  Anyway, I was browsing the pages on the wiki to do with tubes, and came across a tetrode:
http://en.wikipedia.org/wiki/Tetrode
If what I understand of tubes is right, could a tetrode be used to make a modulator/tremolo?  You'd attach the input signal to one mesh and the modulating signal to the other.  End result would ideally be when both signals push, electrons pass both meshes and reach the collector ring.  If either signal pulls, it'd absorb the electrons and deaden the signal.  Or do I have tubes completely wrong. 

[therecordingart]

Let's look at a vacuum tube diode to start. Just like a silicon or germanium diode you have a cathode and an anode (plate). In an ordinary diode, when the anode is more positive than the cathode you are forward biased and electrons can flow.The tube diode works in a similar way. The heater of a vacuum tube will boil electrons off of the cathode. The positive charge on the anode (plate) attracts the electrons. What happens when you want to control the number of electrons that make it to the plate? You add a control grid which is a mesh grid between the cathode and plate. This creates a triode. The more negative you make the control grid the less electrons make it to the plate (the negative charge of the control grid repels the electrons). The tetrode adds a screen grid and the pentode adds a suppressor grid.

I'm at work and will elaborate on this later if nobody else does.

[Salvatore]

Lots to be said about tubes, cause there are so many different kinds with totally different functions and specs.

I'll ad my knowledge about them...

The myths about them are outrageous, it's claimed they make everything sound good, cause they give cleaner results, and ad distortion, and magic, and can be used as a high voltage weapon to scare diy maniacs, make musicians horny and turn them into nerds, talking about there vintage equipment for that modern old school revolutionary new sound making everything warm (the sound and the temperature), and light (talking photons here), and fat, and smooth, and, and, and forces our brains to melt and want to put tubes in everything just for the heck of it, it often was witnessed that musicians experience euphoria just watching tubes starting to glow, warning, highly addictive and dangerous...

(With special tube designs it's also possible to make sound totally awful)

The bad rap,,,
The stuff tends to get a bit costly (tubes and transformers), and a tad heavy, and easily damaged, and draws more electricity, and it is said that one might experience a sting when licking the 300 volts transformer (never tried it).

[R.G.]

It's interesting to suppose: what if the governmental policies that exist in today's California had existed in the early 1960s?

It is likely that vacuum tubes would be *banned* because they waste too much electricity, wear out too fast, and incorporate too many politically-incorrect alloys. Not to mention the embodied energy that is lost when they wear out. 

When deadly non-switching wall warts are outlawed, only outlaws will have linear wall warts.
When light bulbs are outlawed, only outlaws will have light bulbs.
When vacuum tubes are outlawed, only outlaws will have vacuum tubes.

Hmmm... I think there's a song in there.

[nexekho]

It's interesting to suppose: what if the governmental policies that exist in today's California had existed in the early 1960s?

James May once said the same thing about cars.

"So you want millions of people to individually control tonnes of metal capable of topping a hundred miles per hour, running on a substance that burns well collected from dedicated stations?"

[R.G.]

TRA's comments are correct.

To understand tubes, you need to understand "electron ballistics".

Every conductor has a sea of electrons lurking under the surface. These stay in the material because one high-tech-ish bit of math called the work function at the surface keeps them inside. Imagine the electrons as a sea of pingpong balls bouncing around inside a walled courtyard.

Whether a pingpong ball (electron) ever gets over the wall or not depends on two things: the speed/energy with which they're bouncing, and how high the walls are. Temperature is literally a measure of how fast electrons are bouncing around. In insulators, they can bounce forever and not come loose. But in metals, they are fluid, and can flow from atom to atom and bounce into things. If you turn up the temperature, more and more of them start getting enough energy on "high bounces" to get over the wall. Air forces them back in. In a vacuum, they can wander around out in the vacuum for a little bit before they fall back in by the electrical attraction of the positive charge they left unbalanced in the metal when they left.

If you turn up the heat enough, more and more of them get out. To get even more, you can lower the walls.

By lowering the work function by using low-work-function materials on the surface, the wall gets lower. Tungsten has a low work function, and even better can stand very high heat without melting. All a little thorium and it gets lower. And indeed, lots of older high power radio tubes use thoriated-tungsten cathodes which are heated to a yellow-white heat to boil off enough electrons.

For lower power (and probably shorter life) you can use even shorter walls. Barium and strontium oxides have even lower work functions than thoriated tungsten. They provide useful electron emission even at only dim-orange temperatures. This led to the invention of the indirectly heated (by a non-emitting tungsten heater) oxide (of barium, strontium, and others) coated cathode. This is what is in all of the tubes in guitar amps.

Oxide coated cathodes boil off huge quantities of electrons per unit area at low temps. In all modern (i.e., current in the 60s) tubes, this emission is enough to keep a cloud of electrons floating over the cathode in the absence of any other voltages. This is the space-charge cloud you may have read about. All the electrons eventually fall back into the cathode from the cloud, but are replaced by others in a continuous cycle of boil-off emission into largely-ballistic trajectories in the surrounding vacuum, excepting only when they run nearly into each other or some wayward ion contaminant that floated in.

The incandescent bulb is a gas filled monode. It has a tungsten filament, run in a gas filled envelope that is tinkered to force evaporated tungsten to recondense on the filament. If you coat the filament with the right oxides, and suck a vacuum on the bulb, the incandescent bulb becomes a thermionic monode.

If you put another plate of metal inside the monode, you get - surprise!! - a di-ode, literally, "two odes". Wait. I mean "two terminals", "ode" being Greek or Latin or Geekish for "terminal". Nothing happens until you hook up the extra plate to a voltage.

If you put a voltage on the plate that is negative with respect to the voltage on the electron-boiling cathode, the electric field tries to force the electrons back into the cathode. For sure, none of them reach the plate. However, if you make the plate positive, the electrons start... falling... falling... into the plate, lured by the intense and entrancing beauty of a positive charge to cancel with. The higher the voltage, the more of the electrons fall into the plate. In fact, you can measure and curve fit, and you come up with equations that describe the current flow in geometric terms based on the electron charge and voltage. The electrons speed up as they fall toward the plate, driven by the accelerating force of the electric field

This is a vacuum diode.

If you put a screen between the cathode and plate, with vanishing small wires, essentially none of the electrons making the fall into the plate hit this grid of wires, and so they fall right past it. If you make the grid positive, some of the electrons whipping past come close enough to a wire to get sucked into the wire by its positive charge. This is plate current. If the grid voltage is the same as the cathode, the electrons largely ignore it. The grid makes it a tri-ode. We're going to go through a lot of 'odes.

If you make the grid negative, the electrical field on the grid wires acts with the negative electrical charge on the electrons like trying to push two same-pole magnets together. Effectively, the field on the grid wires forces the electrons away from the wires. But there's another wire not far away, and it's pushing electrons away too. The voltage on the grid pushes electrons away and has the effect of making the holes between wires smaller. At some voltage, the field between the wires is more than the attraction of the (relatively distant) plate, and no electrons flow. This is cutoff.

If the grid is positive, it sucks on electrons. But since it doesn't have much surface area, most of the electrons sucked upon by the grid fall through it, look around and see an even-more positive plate over there, and head off for the plate. But a fairly large share hit the grid, causing high grid current.

The electrons between the space charge cloud and the plate follow ballistic trajectories, much like cannon balls, except where they're bent or forced back by the grid charge.

That was section 1.

[R.G.]

James May once said the same thing about cars.
"So you want millions of people to individually control tonnes of metal capable of topping a hundred miles per hour, running on a substance that burns well collected from dedicated stations?"

Oh, I'm pretty sure that if California didn't already have cars, they wouldn't let them get started.

[Earthscum]

OT a little, but...

If you put another plate of metal inside the monode, you get - surprise!! - a di-ode, literally, "two odes". Wait. I mean "two terminals", "ode" being Greek or Latin or Geekish for "terminal".

Doesn't "node" comes from? N-ode. In like 5th grade I remember something like that, N being something like "zero".

[R.G.]

Doesn't "node" comes from? N-ode. In like 5th grade I remember something like that, N being something like "zero".

The Online Etymology Dictionary says it's from the Greek "hodos" = way, in the sense of path. Anode is from anodos, "the path or way up", cathode from kathodos "the path or way down", and diode was coined in 1886 from "di" (twice) plus "hodos" (way) for bise... er, two ways. 

Node seems to be from the Latin "nodus" or knot. But I'm sure that Geeks could conflate and empirate them into N-odes, literally "unstatably many poems".

[PRR]

What R.G. said.... but you must have read other essays almost as good. So let me poke at your misunderstandings.

> There's an electron gun called the heater which fires electrons down the tube, through a mesh.  This mesh is connected to your input and can either pull the fired electrons out of the tube by applying a negative voltage and attacting them to the mesh or repel the electrons so they slip through the gaps in the mesh.  Down the other end is a collector ring with which the electrons collide, collecting them for output.

CRTs have "electron gun". Ordinary amplifier tubes do not.

They have a cathode, which can be a hot naked lamp filament, but is often two parts: a heater and an oxide-coated "cathode" surface.

Electrons come off the hot cathode at ZERO (average) speed/energy.

They won't do anything interesting unless you have a positive electrode, what you  accurately call "collector", also called anode or plate. Now electrons flow from cathode to plate if plate is positive (and not if plate is negative). Not just a collector, but an attractor.

Just like that you have a uni-directional diode, rectifier, useful for some things.

It gets more interesting if you have this current flow and add a fence. If fence ("grid") is zero volts relative to cathode, it has almost no effect: current flows. If fence is negative relative to cathode, electrons won't pass to plate, they just hang around the cathode.

And the interesting thing is that a 100 volt 10mA plate circuit can be controlled with just 10V and zero mA on the grid. Weak interesting signals modulate strong boring battery power to make big interesting signal.

> could a tetrode be used to make a modulator/tremolo?

Yes. That is an astute observation.

However.

You can modulate just fine by adding your modulating signal to your audio signal. Plain triodes are often used in tube guitar tremolos.

True tetrodes are rare because they have problems solved by pentodes. (Power tetrodes exist but are awkwardly huge for simple modulation.) If you actually wanted a tetrode today, you'd buy a pentode and "mis"-wire it.

There are dual-control pentodes where one of the extra grids is fully specified as an additional control electrode.

There are tubes designed for modulation. Penta-grid can combine oscillator, amplifier, modulator all in one.