Thermionic Valves - Electron Tubes - Vacuum Tubes - Lest we forget

Instantaneous long distance wireless communication was made possible by Thermionics. The Electron Tube or Thermionic Valve reigned supreme for about fifty years before semiconductor technology began to erode its position of dominance. This hub looks back over thermionics, lest we forget.

The 50 great years of thermionics were from 1920 to 1970. During this period, thermionic valves were at the heart of radio, and later, television production, transmission and reception, audio amplification and public address, radar and radio telescopy. Valves also made possible early tape recording and mainframe analogue and even digital computers.

Towards the end of this era, semiconductor technology (transistors) began to replace thermionic valves in most 'routine' applications, though in specialised areas valves hung on for many years, most notably in image gathering (TV cameras) and displaying (TV picture tubes). Today, they have all but disappeared, except in Geiger-Muller tubes (geiger counters) and a few hi-fi audio applications where their continued use owes more to fashion than function.

What's inside a valve?

The typical valve is a glass envelope enclosing a vacuum, or as close to a vacuum as can be achieved. Inside are a number of electrodes, accessed (from outside) by wires that pass through the glass and terminate in connection pins in the valve's base.The electrodes are:

1. the cathode - this is heated by an electric filament (indirect heating) or may actually be the filament (direct heating). When heated, the cathode emits electrons into the surrounding vacuum.
2. the anode - this is maintained at a positive potential (with respect to the cathode) and attracts and collects the emitted electrons, causing a current to flow.
3. the grid (or grids) - this is a wire mesh or helix through which the electrons pass on their way to the anode. The flow of electrons is controlled by the grid potential.

The drawing on the right is the symbol for an indirectly heated triode valve, as used in circuit diagrams. A triode is a valve with three electrodes, cathode, anode and grid. The heater is not classed as an electrode. The symbol for a diode valve (two electrodes) is the same, but without the grid. The symbol could give the impression that the electron flow is from the bottom to the top of the valve, but in fact, the internal construction is cylindrical, with the cathode in the middle. The filament (heater), not shown, is inside the cylindrical cathode. Electron flow is radial, outwards from the cathode, past the grid, to the anode. Apart from early experimental devices, the only valves that work axially instead of radially are television camera and display tubes.

How do valves work? - the basics

1. The Diode Valve

The cathode is heated to a dull red heat. This causes negatively charged electrons to leave the surface. These liberated electrons are then accelerated by the radial electrostatic field, crossing the vacuum to land on the (positive) anode. Electron flow is cathode to anode. If the polarity is now reversed, i.e. the anode is made negative with respect to the cathode, no current flows. Why not? Because electrons leaving the cathode are driven back to the cathode surface by the reversed electrostatic field direction. Also the anode, being cold, does not emit electrons, so there is no source of electrons for reverse current flow. The diode is therefore a device that passes current in only one direction. It has two main applications in radio receivers:

• Rectifying Diodes are used in power supplies to convert mains alternating current (AC) to direct current (DC) to feed the rest of the receiver circuitry.
• Detector Diodes are used to extract the audio signal from the received amplitude modulated (AM) radio frequency (RF) signal.

2. The Triode Valve

In a triode valve, electron flow is still from cathode to anode. However, they must first pass the grid, which is held at a negative potential with respect to the cathode and therefore tends to repel electrons back towards the cathode. This negative grid voltage is called grid bias. So, how do electrons get past the grid? The fact is, electrons don't just 'fall off' the cathode; they come flying off with a wide range of velocities. The slower ones will be turned back by the grid, but the faster ones will manage to pass through the grid into the pull of the anode. In the language of Physics, their initial kinetic energy must be sufficient to overcome the potential barrier of the grid. The grid acts like a tap (faucet) to control the cathode to anode flow. A small change of grid voltage can result in a large change in anode current. This is the principle of amplification. Triode valves have three main applications in radio:

• Triodes can amplify low level audio signals to drive headphones or loudspeakers.
• Triodes can amplify low level RF signals from an aerial, prior to diode detection (as described above)
• Triodes can function as 'local oscillators', to produce the continuous RF wave signal required in superheterodyne receivers. (To be explained!)

The Triode as an Amplifer

The next circuit diagram shows how a triode valve can be configured for use as an amplifier of small a.c. signals, e.g. the output of a microphone.

The DC conditions - Electron flow is from ground through R_C to the cathode, then from cathode to anode, and through R_A to the positive supply rail (+V). The voltage drop across R_C causes the cathode to be slightly positive with respect to the grid which is held at ground potential (0 volts) by R_G. (No current flows in R_G). The anode resistor R_A is chosen so that the voltage drop across it is about half of the supply voltage. For example, if the supply voltage is 200V and the anode current is 1 mA, we could have R_A = 100KΩ making V_A 100V, and R_C = 10KΩ, making V_C 10V. With the grid at 0V, the grid bias is effectively -10V.

Signal (AC) amplification - The input signal V_in is applied to the grid via a decoupling capacitor (to ensure the DC conditions are not disturbed). A small increase in grid potential causes a large increase in anode current. As this current flows through R_A, the voltage drop across R_A increases causing the anode potential to fall. The output signal voltage V_out thus follows the input V_in, but the small fluctuations are amplified, and the sense is inverted (see sketch). Notice the bypass capacitor across R_C. This couples the cathode to ground for AC (signal frequency) and prevents the gain reduction that would otherwise occur if the cathode voltage were allowed to follow the input signal.

Radio frequency amplification

The same basic circuit can be used to amplify radio frequencies, but then it is normal to replace the anode resistor R_A with an inductor, called an RF choke. This is a coil of wire usually wound on a piece of ferrite. At audio frequencies and below, its reactance is low so no signal voltage develops across it, but at radio frequencies its high reactance gives rise to a high signal output. Taking this a stage further, using a parallel tuned circuit as the anode load produces an amplifier that is highly selective of one particular frequency. This principle is used for tuning a radio receiver to a particular station.

The triode is the fundamental 'active device'. The diode cannot amplify or produce signals; the triode can do both. It was the triode that made possible radio transmission and reception and audio amplification. But triodes have their limitations, and these led to the development of multi-electrode valves.

The Tetrode

One problem with the triode valve is the capacitance between the anode and the control grid. As the anode signal is inverted, the effect of this capacitive coupling is to reduce gain, especially at high frequencies. In some cases, when using reactive anode loads, this capacitive coupling can result in instability and oscillation. The solution is to place a second grid, called a screen grid between the control grid and the anode. This grid is held at high potential but is decoupled to ground by a capacitor. The effect is to screen the control grid from the anode and therefore eliminate the undesirable effects described above. Tetrodes therefore provide higher gain at high frequencies.

The Pentode

Unfortunately, the tetrode introduced a new problem of its own. When electrons strike an anode at high velocity, sometimes they 'dislodge' other electrons. This is called secondary emission. In a triode, these secondary electrons are recollected by the anode, but in a tetrode, some of them are collected instead by the (positive) screen grid. This causes a kink in the graph of anode current vs anode voltage, in some cases even a downturn of the curve, giving a region of negative dynamic resistance. Negative dynamic resistance results in instability or oscillation (the dynatron effect). To counter this, a third grid is added, closer to the anode, and held at a low potential. This has the effect of reflecting secondary electrons back to the anode, preventing their take-up by the screen grid. This extra grid is called the suppressor grid. In terms of performance and stability, the pentode is generally considered the 'standard' valve.

Special purpose valves

Triodes, tetrodes and pentodes can all be considered general purpose valves. Additionally, several valves were designed for very specific purposes, especially for radio receivers. The aim was often to reduce the number of physical devices by mounting two or even three electrode sets in a single envelope. This was particularly prevalent in the UK, where radio manufacturers paid a certain amount of duty per valve to the exchequor. Most of the compound valves combined some of the functions of oscillating, mixing and detection as required in superheterodyne receivers. Some of the commoner specials were the double-diode-triode, the triode-hexode, and the pentagrid.

In no particular order:

1. Cathodes are doped with thorium to increase electron emission at relatively lower temperatures. Before this discovery, the 'bright cathode' valves ran much hotter.
2. To improve (harden) the vacuum during manufacture, after the bulb is sealed, a substance called a 'getter' is vapourised inside (by electromagnetic induction). The getter captures residual ions and deposits itself (and them) on the inside of the glass envelope.
3. Cathodes wear out eventually becoming unable to source enough electrons for the required anode current. This is called cathode stripping.
4. Some valves have a 'cap' terminal on top as well as the pins in the base. This is the control grid. It is led out through the top to reduce capacitance with other electrodes.
5. Some valves are entirely encased in a metalised conducting 'paint' which is connected to chassis ground. This is to prevent RF radiation to adjacent circuitry, not to screen the valve from outside interference.
6. Valves can become microphonic - airborne vibrations (sound waves) if passed on to the electrodes, especially the grid, can result in an output signal. This is why if you tap a radio valve a sound may come from the speaker.
7. Valves are not better than transistors in guitar amps. It's simply that a lot of classic tracks were cut on valve amps and people are nostalgic for that particular type of distortion!
8. Valves are immune to EMP (electromagnetic pulse) as produced by certain nuclear weapons. This led to valves' extended deployment in some emergency broadcast facilities in nuclear bunkers.

Enough is enough

I could carry on forever about thermionics. What has kept my interest alive is my small but select collection of 1930s wireless sets which I keep in good working order and listen to regularly. This hub has barely scratched the surface of a fast fading but fascinating technological field. The Golden Age of Wireless might just have been our finest achievement as a creative species.

Thank you for reading!