Tom Clancy's first novels were known for being fairly accurate from a technical/scientific point of view. That doesn't mean that they were 100% accurate, of course, but the author reportedly had some really good insider sources and knew what to ask them. I remember to have read the same thing I'm going to explain (about cutting all the detonator cables the same lenght) somewhere else, not only in Tom Clancy's novel, so perhaps it's not all fiction. In "The sum of all fears", a thermonuclear device is being built from materials recovered from a salvaged Israeli fission warhead. The author explains some interesting aspect about the timing accuracy needed to achieve a precise symmetry of the implosion shockwaves: That left me quite puzzled, because... what the heck, we are comparing the speed of light vs the speed of a chemical reaction. Orders and orders of magnitude of difference. And in fact, in another (earlier) chapter of the book, we read: It seems that some figure is indeed out of place. For example, let's take Octol, a typical high explosive for nuclear applications, whose detonation speed is around 8,500 m/s. That means that in one "shake" (10 ns), a block of Octol would burn only about 85 µm (eighty five microns). The obvious question is: why should the designer worry so much about the lenght of the cables?... The figures are so different, they really doesn't seem to justify such a concern. The other point that I really can't understand is the problem behind firing all the detonators at the same instant. From a purely electrical point of view, the load seen at the output of the high energy pulse generator ("X unit") is a ohmic-inductive load. Each detonator shows quite a low resistance (the bridgewire), so if you put a bunch of them in parallel, you will obtain a R-L load, with a very low R (so many bridgewires in parallel) and a L that would depend on the lenght of the wirings (and how many of them, of course). So the problem is "just" to discharge a sufficiently high amount of energy into the load very quickly so all the bridgewires would vaporize at the same instant. Once you have a capacitor bank capable of storing sufficient energy at the appropriate voltage, and a suitable switch (of course), I don't see much of a problem with it. The capacitor banks used in those "X units" are calculated with ample margins about both the amount of energy stored and the voltage: it's a by-the-book example of "overkill". I would connect one side of a single high-rating switch to the output of the capacitor bank, and the other side of the switch to the n detonators, just connected all in parallel. Instead, I read of a number of krytrons (not just one) and other strange things... Okay, a krytron is essentially a controlled spark gap, with a very short switching time. In the past, I used to work with non-controlled (two-terminal) spark gaps for the high energy ignition units used in turbine engines. Those tubes can switch several joules of energy for at least one million discharges. What makes a non-controlled spark gap so expensive, in jet ignition applications, is its resistance to the electrical "wear" (erosion) of electrode surfaces through its lifetime. In a nuclear weapon, the switching element is by definition a one-shot device. Krytrons aren't the only high-speed high-energy switching tubes. So, once my switch can close the circuit with an internal resistance low enough in order to avoid an excessive voltage drop across its terminals, and quick enough to avoid a slow rising slope of the pulse (which might produce significant difference in detonation times for each detonator, but that should be really slow-rising), I guess that almost any switch could do the job, even a mechanical switch (I suppose the hardest task would be to find someone to throw it... ) And just one switch, not an array of switches. It is my feeling that if I introduce any element, of any kind, in series with each detonator (or group of detonators) instead of having all of them connected in parallel, I would introduce a potential source of delay difference, which is exactly what should be avoided in such applications. So why the firing circuit arrangement seems to be so critical and complex? What am I missing?... I am little interested in the so called physics package, I am much more interested in the electrical arrangement and, above anything else, in the safety features and concepts of these weapons: I think that the "philosophy" behind them can be really inspiring for many applications in other fields.