Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Length of wiring and firing switches: myth or truth?...

  1. Feb 18, 2016 #1
    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. :oldconfused:

    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... :oldbiggrin: ) 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.
  2. jcsd
  3. Feb 18, 2016 #2
    nuclear not chemical reactions.

    I don't know all the details, but the detonation must shock the plutonium or whatever from a sub critical to a critical density....that does take a very precise detonation.
  4. Feb 18, 2016 #3


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    The chemical reactions must occur in the explosive lenses precisely, otherwise there are no nuclear reactions.

    The plutonium core must be imploded using a fully symmetric shockwave created by the explosives, otherwise the chain reaction which creates the nuclear blast will not be set up, or only part of the plutonium will fission, creating what is known as a 'fizzle'.

    Early implosion weapons were physically quite large, as can be seen from this image


    The spherical assembly of nuclear material and explosives in this early Fat Man device must fit inside the bomb casing, which is approximately 60 inches in diameter. Literally dozens of detonators must be set off at precisely the right instant to create the shockwave which compresses the sphere of plutonium at the center into a critical mass, thus yielding the nuclear detonation. The lengths of the cables running from the trigger mechanism to the individual detonators had to be carefully designed in order for this particular device to work.

    Many experiments and much theoretical work on the hydrodynamics of explosions was carried out by the scientists and engineers at Los Alamos when it became clear that a critical mass of plutonium could not be assembled quickly enough using the gun-type of design being developed for the uranium bomb.


    Modern nuclear devices are much smaller than the old-style Fat Man implosion design. It's not clear that the same technology is used in these devices to create the critical mass of plutonium or uranium which results in the nuclear blast.
  5. Feb 18, 2016 #4
    Now I'm not sure what type weapon the OP questioned....

    Does Post #3 address an atomic bomb ....??
    Anyway here it is for an H bomb....

    Either way, Clancy likely has it correct.....he has great expert contacts as already noted.

    A thermonuclear weapon is a nuclear weapon that uses the energy from a primary nuclear fission reaction to compress and ignite a secondary nuclear fusion reaction. The result is greatly increased explosive power when compared to single-stage fission weapons. It is colloquially referred to as a hydrogen bomb or H-bomb because it employs fusion of isotopes of hydrogen.

  6. Feb 18, 2016 #5
    Okay, to everybody... I wish to underline that both my questions only cover the issues relative to firing the conventional (chemical) explosives: what happens later on in the "physics package" is beyond my interests. I am just talking of bridgewire detonators, wiring, switches, and conventional high explosives. The :radioactive: part is above me, I am not any nuclear physicist.
  7. Feb 18, 2016 #6


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    An H-bomb requires the detonation of a fission bomb primary to create the conditions necessary for fusion. Some fission devices are fitted with small amounts of tritium, the fusing of which can boost the yield of the overall explosion. Although small amounts of fusion occur, the fission devices so fitted are not traditionally considered to be H-bombs, where the majority of the yield comes from fusion rather than fission.

    I think where Clancy gets carried away in his description is that with modern small fission devices (the smallest can fit inside a 155 mm artillery shell), the bulky design of the Fat Man device imploding on a sphere of plutonium is now obsolete. This design was chosen during the war because it was the simplest to analyze, given the crude tools available (remember, no computers, only platoons of technicians armed with desk calculators) and the lack of time in which to experiment with other implosion configurations. The design had to work right the first time.

    Later information made public about H-bomb development suggests that these devices use one or more long, slender plutonium rods called "spark plugs" to provide additional energy to the blast when these rods are compressed into a critical mass by the fusing hydrogen. Compressing a rod of plutonium into a critical mass for a pure fission weapon would save a lot of space, but it might be more difficult to design the explosives required and how they are detonated.
  8. Feb 20, 2016 #7
    Ok, thank you, but... my two questions still remain both unanswered... :biggrin:
  9. Feb 20, 2016 #8


    User Avatar

    Staff: Mentor

    Yes, the point was missed, but you are comparing the wrong things. The rate of travel of the signals/shock waves isn't what matters. What matters is that they arrive at their destination at the same time. Giving some explosives a head start by detonating them first means their shock waves will arrive at the target first.
  10. Feb 20, 2016 #9
    @russ_watters The point you described is perfectly clear. What I cannot understand is: given the enormous difference in the speed of an electrical signal (hundreds of thousands of km/sec) and the speed of detonation (thousands of meters per second), where is the point in cutting all the detonator cables the same lenght? The difference in arrival times of the signals would fall by orders of magnitude within the normal tolerance in the responsiveness of the detonators, and in any case that difference would translate in microns of difference in the combustion amount of the explosive lenses. So, is that lenght really important or is it just a myth?...
  11. Feb 20, 2016 #10

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    my guess is it's a way more simple practical matter
    each detonator wire must receive enough current to vaporize it
    so it's either a lot of reasonably sized switches or one huge one.

    If i wanted to assure simultaneous vaporization i'd first try one capacitor and one switch per detonator. You wouldn't want them shiftng current between themselves.

    But i'm not a weapons guy. I have no idea what they use.
  12. Feb 20, 2016 #11

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Any idea what's the actual variance? Those "lenses" were the secret that Fuchs sent to Stalin.
  13. Feb 22, 2016 #12
    In the story, they are simply controlling the variables they could.
    By making all 70 wires the same length (and from the same batch),
    they could be assured the signals would all arrive at, as close to the same time as possible.
  14. Feb 22, 2016 #13


    User Avatar
    Gold Member

    To answer that question I think you'd have to determine how much difference from the ideal spherical shape of implosion can be tolerated in the PU core and still achieve critical mass. From the Trinity test, some conclusions might be drawn about the degreen of concern over variation in chemical burn rate, given the finished explosives were inspected to find small internal bubbles, then hand drilled and filled with liquid explosive.

    I dont think you'll get anywhere close to an answer from speculation. The Manhattan Project people spent a lot of time on implosion, many weeks or months, with a lot failures despite the best physicists and engineers the US had to offer and unlimited resources. Thus it's a extraordinarily tricky problem and likely has a complex answer, which won't be found laying around (thankfully). Note that none of the professional nuclear engineers or physicists that frequent this forum have commented here.

  15. Feb 22, 2016 #14
    Yes, and I don't even think that anybody seeking information for other than pure curiosity would ever write in a forum (or use the Internet at all), rather (s)he would use offline records or books :oldwink: By the way, I don't even know if they cut all the cables the same lenght in the "Gadget" device, or not. I was just referring to the contents of "The sum of all fears" because the figures seemed way too different (and with variables - due to chemistry - that apparently overcame them) to raise any concern about a couple nanoseconds more or less in a chemical cap.
  16. Feb 22, 2016 #15

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    it's science of explosives not nucleonics.

    Only explosives textbook i ever saw had a disclaimer in the introduction to effect 'no weapons information in this book.'
  17. Feb 23, 2016 #16
    To put it even clearer: if I were a wannabe-nuke-maker, and I had a doubt that a difference of some feet in wiring lenght was important, I would not take a chance just for saving (from a huge budget) a few bucks worth of wire: cut them all the same lenght and that's all. But, if that wannabe-nuke-maker really had doubts about such an issue, he'd better go back to the library (rather than to a Tom Clancy's novel) first, and then the drawing board, because it's 100% sure that his design had bugs just everywhere.
    Last edited: Feb 23, 2016
  18. Feb 23, 2016 #17


    User Avatar
    Gold Member

    It is possible Clancy got misled/confused.
    It makes practical sense for all the wires to be the same length, if only for operational ease.
    So an author such as Clancy may have thought it was important or perhaps he just highlighted it that way because it helps the story sound informed.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Threads - Length wiring firing Date
March 1945 Hanford Substation Fire Sep 4, 2017
Worst Case Windscale Fire? Aug 10, 2017
MCNP5 error: Source distribution in 2 volumetric cells (help please) Jun 28, 2017
NPP fire codes Jun 16, 2011
Melt wire experiments Oct 31, 2007