Jeffrey LewisIran and the Shahab 3

A couple of weeks ago, Dafna Linzer had a long, detailed story (and online chat) on intelligence found by German intelligence on a laptop computer stolen by an Iranian citizen in 2004. I am writing a four-part series looking at the key pieces of information: a schematic of a shaft that might be for a nuclear test, plans for an underground facility to produce uranium tetrafluoride (UF4), modifications proposed for the Shahab 3 ballistic missile and the acqisition of relatively advanced P2 centrifuges from Pakistan.

A couple of weeks ago, I started a four part series on the claims advanced by intelligence sources in Dafna Linzer’s Washington Post article. One of the questions I had concerned the modifications made to Iran’s Shahab 3, specifically “What was the diameter of the notional nuclear payload?” If the diameter is .6 m (as David Albright estimated based on publicly available photographs), the Iranians probably couldn’t build a bomb that small.

I love Paul Kerr’s work at Arms Control Today and not just because he blogs on this site. He managed to run the story down, and the answer is: No, Iran’s nuclear weapons would probably be too big for the Shahab 3.

But whether and to what extent such a re-entry vehicle design would improve Iran’s ability to deliver a nuclear weapon is unclear. The former State Department official said that a re-entry vehicle built according to the design that Libya obtained from the Khan network would be too small to hold a nuclear weapon. (See ACT, March 2004.) That acquisition has sparked concern that Tehran also may have obtained similar designs, but no evidence has emerged that Iran has actually done so.

Nevertheless, the official cautioned that “[n]ot enough is known about the Iranian bomb-making capabilities” to determine whether Iran is capable of building a warhead suitable for the re-entry vehicle described in the laptop documents.

This was what I estimated based on publicly available information, concluding that Iran’s Shahab 3 nose cone was just too small for a first generation Iranian device. (More on the method)

I think a second aspect of Linzer’s story was notable, namely confirmation that the design team working on the Shahab 3 nose cone was not terribly competent. “Experts at Sandia National Laboratories,” Linzer wrote, “determined … the modification plans, if executed, would not work.”

more later


  1. Pavel Podvig (History)

    You may find this useful: I’ve found in old Soviet archival documents a breakdown by weight for a couple of nuclear warheads that were developed for a SS-13 missile (ca. late 1960s).

    One was 385 kg, of which the nuclear charge was 175 kg + 40 kg of command electronics. RV body was 170 kg. On top of the 385 kg they had an RV stabilization and orientation system, which added another 25 kg to the weight of payload.

    A heavier warhead was 625 kg (of which 325+40 kg were nuclear charge and electronics and 260 kg – RV body) plus 35 kg for the stabilization system.

    I’m not saying, of course, that whatever Iran may be developing would be the same, but it gives some idea of what may be inside a warhead.

  2. Yale Simkin (History)

    Jeffry wrote:
    “One of the questions I had concerned the modifications made to Iran’s Shahab 3, specifically “What was the diameter of the notional nuclear payload?” If the diameter is .6 m …, the Iranians probably couldn’t build a bomb that small.”

    The question is based on the assumption that the upgraded Shahab-3 is a REPLACEMENT for the earlier large nose-cone Shahab-3. Maybe not. It is potentially a VARIANT of the missile.

    The older variant (with upgraded boost-phase active guidance and a spin-stabilized RV during ballistic flight) is a nuclear-armed COUNTERVALUE weapon, with room to carry an implosion assembly.

    The newer variant, with active guidance for the entire flight, is a conventionally-armed, COUNTERFORCE weapon.

    Altho the Wright-Kadyshev essay on the NK Nodong does not completely apply to the Shahab-3, let’s assume its figures. They gave a CEP of 2-3km.

    In 1945, Fat Man missed its aimpoint by MORE than 3km (and most of the city) and still caused at least 40K dead and 10s of thousands of injuries.

    Tel Aviv/Jaffa has 380K people and covers 50km2. More significantly, the metropolitan area stretches for 14km and has 2.2 million people.

    As Wright-Kadyshev point out:

    “Its ability to carry a one tonne payload could make it a more significant threat in the future if North Korea or Iran develop a deliverable nuclear weapon. However, even if armed with a nuclear warhead its accuracy is so low that it still could not be used against military point targets.”

    “military point targets” is what the upgraded Shahab is for.

    Also, if Iran earmarks 3 times the U235 it would use in an implosion warhead to a single device, then a double-gun uranium warhead would easily fit the space and weight requirements of the smaller RV.

    Here are some interesting images of some conjectured variants:


  3. Gerry (History)

    Assumptions, assumptions and assumptions….No one knows really what capabilities the Iranians have or have not. Whether AQ Kahn did more damage than we suspect or not.

    Assumptions… Where the hell is the agencies non-proliferation unit these days?

  4. John Field (History)

    I think the experts should footnote their congressional testimony with calculations. At the core of it, nuclear proliferation policy is deeply intertwined with freedom of thought. And that is something which we are bound by initial resolution to defend.

    It seems to me that Little Boy demonstrated that it is possible to assemble a uranium bomb at a center of mass closing speed less than about 500 m/s.

    [Why? Hydrocarbon propellants end up being more or less ideal gases with maximum temperatures around 3000 K(molecular dissociation of CO2) corresponding to a speed of sound in the range 1000 m/s. – e.g. 2x CoM speed (Please note, I think Wikipedia’s Little Boy numbers seem to be somewhat too slow to me)]

    Explosives do not suffer from the velocity limit above because much of the delivered impulse comes before expansion has entered a non-interacting gas phase. The speed of sound can be much higher.

    Nominally, the speed at which a flat plate will come flying off a thick backing of explosive is known as the Gurney constant. It is in the range of 2500 m/s. This number is scaled appropriately as the ratio of flyer plate mass to charge becomes non-zero.

    In an imploding geometry, additional corrections need to be made to consider the geometry as well as the thermodynamic state in a converging detonation front.

    Even at 10% charge to mass ratio, however, the flyer plate speed is still 30% or so of the gurney constant – around 750 m/s.

    So, here’s what the iranians can do :

    Quasi-linear implosion. Use 100 kg(!) of HEU surrounded by another 100 kg of HE with an egg-like shape and a single firing point at the tapered end of the egg tailored to converge on a chosen center closer to the far end. No reflector, no tamper(or maybe just a thin one). I’m not sure you would call this an implosion so much as a ‘big smash’. Even though the shock will be oblique, you will get much of the gurney velocity. Closing speed will far exceed little boy.

    To do it, you need a good flash radiography system, a laptop computer and time to experiment.

    It’s dangerous. It’s crude. It’s terribly wasteful of HEU. Yield will be low. But, it seems that it would work.

    Things are just so much easier with Uranium.

  5. Hass (History)

    See Iran’s response to to El-Baradie’s report, that raises certain interesting points, ie: the “individual” related to the Lavisan site who supposedly had not been made avaibable to IAEA inspectors had already been interviwed by IAEA inspectors.

  6. Yale Simkin (History)

    This is a link to images of the Sept 22, 2005 Martyr’s Week Parade in Iran.

    Three interesting Shahab missile vatriants are displayed.

    The top image is the Baby bottle with clipped tailfins.

    The middle image is a “classic” large cone with full-size tailfins.

    The bottom image is the most interesting. It is short-body with baby bottle nose and big tailfins.

    Because the images are from a fixed camera position, relative measurements are easy.