Vacuum Tubes
A Micro History of the Vacuum Tube
Starts with the electric light. A practical electric light required, amongst other things, a vacuum in which to operate. The filament required two wires to exit the vacuum and connect to a source of electricity.
The Edison Effect was noted when a third wire was placed within the vacuum. Electric current was observed flowing even though no physical connection took place between the wires. The electrons were flowing from the heated metal, through the vacuum, and onto the open wire. This phenomenon remained a mere curiousity for a long time.
When it was realized that electrons were flowing through the vacuum, the diode was born. A diode represents a one way path for electrons to flow. This is a very useful mechanism.
When yet another wire was added, another phenomenon was observed. An element placed between the heated wire and the collecting wire could control the flow of electrons. Now the world had an extremely useful mechanism: an electronic valve. A three element tube is called a triode.
Four and five element tubes, dubbed tetrodes and pentodes respectively, followed suit.
Basic Tube Operation
Diodes
When a tube is just a diode, current flows if:
-Filaments are hot and heating the cathode
-The plate voltage is much more positive than the cathode
Triodes
A diode can be shut off, or controlled, by inserted a new electrode between cathode and plate.
The new electrode is the "control grid."
In order for the grid to control the flow between cathode and plate, the grid must have a negative voltage with respect to the cathode.
How much more negative the grid voltage is with respect to the cathode will determine to what degree the tube is "on" or "off."
The schematic above shows the use of 3 independent batteries, labelled "A," "B," and "C." These terms are still in use to denote the filament supply, plate voltage supply, and bias supply, respectively. Most tube amp schematics or analysis will refer to the "B+," which is the high voltage sent to the plates, analogous to the VCC power rail in transistor circuits.
The A Battery supplies the filaments. The above tube is an indirectly heated cathode triode. When the cathode is indirectly heated, the filament is generally not shown in the main schematic section. Typically, all the filaments will be shown together in the power supply section of the schematic, usually given their own dedicated secondary winding from the main power transformer.
Note the polarity of the C battery. A tube circuit's bias voltage will sometimes still be referred to as the "C-" voltage, but this is a relatively obscure term these days.
The vast majority of tube preamps, and many tube power amps, simplify biasing by using the "self bias" arrangement shown above. We still must keep the control grid voltage negative with respect to the cathode, but we're going to flip our thinking around. We are going to keep the cathode positive with respect to the control grid.
It is assumed that the reader understands voltage divider circuits. If not, this is a good example to whet your appetite for them. The R Cathode, "R Tube", and R Plate resistors are all in series with the B battery. This means that the voltages on each side of "R Tube" can be intuitively understood like this:
B+ > Plate voltage > Cathode voltage > B- (0V)
And R Grid, which is connected to B- and effectively open circuited at the control grid is assumed to be at the B-, or 0V, potential.
The arrangement conveniently sidesteps the issue of deriving a dedicated C- supply for the grid.
We also get negative feedback (NFB) from the voltage that develops across the cathode resistor. The NFB reduces the voltage gain of the amplifier. We can bypass the cathode resistor by adding a capacitor in parallel to it. This is know as a bypass capacitor. It is made large enough so that its reactance appears as a short circuit to all frequencies of interest. If it is made purposefully small enough to only bypass some frequencies and not others, it creates a shelving EQ effect where high frequencies receive more gain than low frequencies. This is a useful effect, as bass doesn't continually roll off as it would with a RC high pass filter.
Speaking of RC high pass filters, the input and output caps create exactly that, so they must be large enough to not present a large impedance to bass frequencies, unless that would have a desirable effect (such as when creating a filter network).
As a third choice in biasing arrangements, you will see something like the above schematic. In these arrangements, R Grid must be very large, typically 2.2M to 10M. The grid will develop a slight negative voltage with respect to the cathode. Why this works is more of a physics question, and I simply take this effect for granted. I don't see this arrangement too often, except for the input stage of very old amps. Since there is no cathode resistor, gain is maximized, and we've saved two components over the previous design. On the other hand, the negative voltage developed at the grid is very small (maybe 0.5V), so the input signal must be kept appropriately small to keep the signal from upsetting the biasing. Hence why it is mostly seen at the input stage.
Tetrodes and Pentodes
These tubes introduce the screen grid and the suppressor grid. The screen grid should be at or near the plate voltage. This helps move electrons along towards the plate. If the electrons hit the plate too hard, they can bounce right off. The suppressor grid is kept negative, usually the same potential as the cathode, to prevent electrons from bouncing away.
The suppressor grid was patented by Philips. Supposedly as a way of getting around their patent, the concept of "beam forming plates" developed that function similarly to pentodes without using the suppressor grid. Tubes with "beam forming plates" are known a "beam power tubes" (6L6) or "beam tetrodes" or "kinkless tetrodes" (as in KT66/KT88).
Tetrodes and pentodes are usually seen as power amp tubes, but sometimes they are seen as preamp tubes as well.
Voltage Gains of Tubes
Tubes are built by hand with the aid of machines. The parts are assembled on a human scale (as opposed to microscopic) and the electronic parameters can be finely tuned by mechanical means. Due to the macro scale construction, tubes have very reliable gain figures. A simple, single tube amplification stage can be made from 3 resistors and 2 capacitors with highly predictable results.
Click on the image for an animation of the input and output signals.
The input and output of a grounded cathode voltage amplifier will be 180º out-of-phase. A positive going pulse at the input causes a negative going pulse at the output and vice versa.
Gain is realized due to the small voltage swing at the grid required to create a large voltage swing at the plate. Although over simplistic, a basic understanding of the effect can be achieved by replacing the tube with a "voltage controlled resistor" in your mind. The imagined resistor replaces the plate and cathode connections, while the grid voltage controls the ohm value of the imaginary resistor. A positive going grid voltage makes the imaginary resistor small, and thus reduces the voltage observed at the plate. A negative going grid voltage makes the resistor large, and thus increases the voltage observed at the plate. If this is not intuitively understood, then you should review voltage divider circuits.