A
krytron is, essentially, a
triggered switch. It was first built by a U.S. firm (
EG&G). There are a few facts about krytrons that make them more interesting than, say, a
vacuum tube or a
diode. First of all, let's look at how the krytron does what it does.
A krytron consists of a sealed gas tube with a cathode and anode inside. The full name for the krytron device type is a 'cold cathode trigger device.' The class of devices it is part of is called 'Pulse Power Switching Devices.' Here's how it works. The krytron is a switch, which means that it is designed to deliver an electrical charge when a 'triggering' current is fed into it. Inside the sealed glass tube is an ionizing gas, much like in a neon tube or fluorescent light tube. When a sufficient charge differential is built up across the gap between the cathode and the anode, the gas will form an ionization path between the two, and current will jump the gap. This is almost exactly how a fluorescent light tube works; a 'starter' provides an extremely high-voltage current when the switch is pressed in order to form an ionization path from one end of the tube to the other; once this path is in existence, a lower-voltage sustaining and fluorescing current can be applied.
So, to return to krytrons, a krytron can be used to separate a circuit until the appropriate trigger voltage has been applied to one side; at that time, the switch opens, and current flows. So why are these special? In this case, krytrons are special because a) they can thus handle (in-line) large power flows, and b) they are extremely fast at switching. One of the reasons for this is that the gas in the tube is kept permanently slightly ionized already by the presence of a small (less than 5 microCurie) radiation source emplaced within. This means that the formation of the ionization path once sufficient current has been applied is very very fast; some krytrons can trigger in around 5 ns! Also, krytrons tend to have extremely consistent trigger time from one use (called a shot) to the next. Variance in times between uses (called jitter) is often in the 5 ns range, which is very small for a switch. They may only last for a few hundred 'shots' before the gas ionization begins to fail; however, for those uses, they will provide high-current, fast-response switching.
Okay, so why are they important? In the 1980s, the krytron was the subject of much secretive concern, policy, factual events and fiction. If you've seen the Harrison Ford thriller Frantic you'll remember that the movie's McGuffin was, in fact, a krytron switch. This is because the primary use of the krytron is in triggering very fast, precisely timed events; most notably, the detonation sequence of an atomic weapon. In order to set off the warhead of an ICBM at precisely the correct altitude for a maximal-damage airburst, when the weapon is travelling downward at approximately Mach 4-6 (near sea level), you need a very fast switch. Enter the krytron, which not only was fast enough but could handle enough current flow to pass the full current detonation signals of the warhead and, in smaller form, the signals to detonate the various conventional high explosive lenses that surround the core of an implosion-type weapon.
In June of 1985, the magazine Newsweek reported that a Grand Jury had been impanelled in Los Angeles to look into allegations that several persons had conspired to bypass U.S. Export control laws by shipping krytrons to Israel. The devices have become less of a problem, apparently, because they are almost freely available, although likely not in the capacity required to trigger weapons. They can be (and are) used to trigger other forms of fast, high-energy events such as the triggering of thyristors, and flashlamps for use in lasers and other strobe applications such as modern Xerox machines.
There is a variant of krytron specifically intended for use in environments which are subject to high levels of ionizing radiation. Since such radiation might cause the gas in a krytron to ionize prematurely, and thus form a current path (closing the switch) without the requisite triggering voltage being applied, the sprytron switch contains a spark gap inside a sealed vacuum environment. The sprytron does away with the 'keepalive' radiation source but also requires a much more powerful triggering pulse, since the current needs to arc directly from anode to cathode rather than traversing an ionized path. The sprytron is able to handle larger inline loads and is slightly faster in some phases of operation than the krytron, but its requisite higher triggering pulse means that the krytron is still preferred for secondary switching operations.