The Black Bellows

An alert TW reader “points out”: that, as anticipated, there was a “centrifuge-related announcement”: on Iran’s 3rd National Nuclear Technology Day:

Iran’s President Mahmoud Ahmadinejad announced on Thursday on the occasion of the 4th [sic] national day of nuclear technology that the country has tested two types of new high-capacity centrifuges.

The Head of Atomic Energy Organization of Iran (AEOI), Gholam Reza Aghazadeh said that the centrifuges were 5 to 6 times faster than the older ones. He gave no further details.

Some of us have been waiting for this announcement more or less since _last year’s_ NNTD, or before. You might recall a certain “open-source intel bonanza”: right around then. There was an odd, unresolved detail in Jeff’s analysis:



The picture also contains a hand holding what looks, to me at least, like it might, might be a carbon fiber bellows — although I don’t have any reference images to compare.

A bellows, if you don’t know already, is a short cylinder meant to connect two lengths of rotor in a centrifuge. A notch (or “crimp”) allows the entire assemblage to withstand otherwise destructive vibrations when accelerating to (or decelerating from) operational speeds.

The IR-2 and IR-3 machines — prototype devices first mentioned in IAEA reports in “Feb. 2008”: and “May 2008”:, respectively — are said to have one rotor each. But there were “too many pictures”: from last NNTD seeming to depict “multi-rotor”: machines — either disassembled or partly assembled — not to wonder.

One possible answer: because newer, multi-rotor centrifuge models were still to be fully assembled and tested. Seizing on key details in the Feb. 2008 IAEA report, plus the same mystery photo at the top of this post, ISIS flagged this possibility back in “May 2008”:

According to the February 2007 [sic] IAEA safeguards report, inspectors visiting Kalaye Electric were given information on four different centrifuge designs, including two subcritical rotor designs, one or more supercritical rotor designs with bellows, and a more advanced centrifuge, which is undefined in the report. The IR-2 and IR-3 are the two subcritical centrifuges. The IR-2 is an experimental model that contains a single composite rotor made from carbon fibers. The other parts of the rotor assembly are modified P-2 components (see figure 1). The IR-3 is an experimental model that seeks to increase the enrichment output by increasing the centrifuge’s length somewhat and by varying the cooling of the centrifuge rotor (see figure 2).


Although not mentioned in the [May 2008 IAEA] report, there appears to be a third advanced centrifuge at the pilot plant. It appears to have the same diameter as the IR-2 and IR-3 but to have double or triple the length of the IR-2. Thus, it would hold two or three rotor tubes, connected by bellows (see figure 5). Prior to Iran’s suspension of the Additional Protocol in 2006, Iranian officials told the IAEA they could not make P-2 bellows. Iran has apparently overcome this obstacle (see figure 6).

(Thanks to Paul for pointing this out.)

Information has been leaking out around the edges. The AP’s George Jahn “mentioned the IR-4”: last month in a story about carbon-fiber importation. (Does this mean that the mysterious fourth new centrifuge doesn’t use carbon fiber? I suspect we’ll learn soon.)

And now the NCRI has “told the WSJ”: that the new machines are “being assembled” in the basement of a “reception center” at Natanz. Too bad they weren’t exhibited to the press this year.

Update: Back in February, the AEOI’s Agazadeh said that new-generation centrifuges would soon be installed at Natanz, as “Paul has pointed out”:

Leave a Reply

Your email address will not be published. Required fields are marked *