Reply #3: The controller was the component effected by the virus - impurities in UF6 gas were the result

[leveymg]

These impurities would have been detectable after the first batch was altered by irregular motor speeds. Once they ruled out a problem with the motors and feed gas, the controllers would be the next thing to check. The solution to a bad controller is to swap out the programmable logic card. If the virus effected only the controller -- which is what was widely reported -- then that would take care of the problem.

The P-1 and P-2 design centrifuges used by Iran are not very sophisticated. They're based in 1940s and 1950s designs stolen by Pakistan and sold to Iran and several other countries by AQ Khan in the late 1980s. These are also not very reliable units, so the Iranians have plenty of experience detecting and solving problems with the Uranium Hexafloride (UF6) gas cycle, which they've been toying on and off with for more than two decades. The programmable industrial controllers that were reportedly infected are off-the-shelf units produced by Siemens or are knockoffs, probably Russian copies. They also are not particularly complex, and are designed to be easily programmed and tested. If there is a problem, the circuit boards quickly slide in and out for replacement.

I read up on the Iranian nuclear program a couple years ago, and if there is a specific assertion you need a citation for, I will try to find it for you.