A couple of weeks ago, I complained that the Los Angeles Times‘s Ralph Vartabedian wrote a misleading story on the W76-1 stockpile life extension program (W76 Problems Seem Overblown, June 3, 2009).
Bascially, Vartabedian used a relatively minor delay in the delivery of a “classified part” — the arming, firing and fuzing (AF&F) system — that had held up delivery of the first W76-1 in order to reprise now resolved concerns about the manufacture of a different “classified component,” aka FOGBANK. (For more on FOGBANK, see FOGBANK, March 7, 2008).
The two are problems are very different, since the AF&F system is not part of the physics package and can be replaced without testing. NNSA, to my mind, badly managed the FOGBANK production process and prematurely released a self-congratulatory press release, but neither sin justifies Vartabedian’s alarmist warning “about the Energy Department’s ability to maintain the nation’s strategic deterrent.”
After Vartabedian’s story, I was surprised that Peter Stockton from the Project on Government Oversight (POGO) doubled down on the issue, insisting that the problem was not with the AF&F, but rather the production of FOGBANK (Show me the FOGBANK, June 7, 2009). I was (and am) very skeptical of that claim:
Now, Peter Stockton at POGO tells Munger that the hold-up is still related to FOGBANK:
[snip]
So, which is it? What Stockton said certainly was true, at least until recently. GAO stated while awaiting FOGBANK, Pantex remained “in ‘stand-by’ mode, which includes maintaining the skills of the technicians who will assemble refurbished W76 weapons.” (I suspect “stand-by” mode refers to the process described by Stockton.)
But is it still true in Spring 2009? I have to say, I would be very surprised if the problem continued to relate to FOGBANK. My money is on the AF&F system.
Well, it seems my suspicions were correct — or at the very least, NNSA is sticking to its story.
Frank Munger, of the Knoxville News-Sentinel and the excellent Atomic City Underground, reports an “unspecified number of W76 warheads are fully assembled, certified and ready for delivery to the military” though delivery won’t take place for several more months. “[NNSA public affairs chief Damien] LaVera reaffirmed earlier reports,” Munger also writes, “that there is an issue with the warhead’s arming, fuzing and firing (AF&F) system.”
Here is the text of LaVera’s email, relating to the issues dating back to first production unit of the W76, that Munger posted on his blog:
After the first production unit was completed in September 2008 and before the Nuclear Weapons Council accepted the W76 into the stockpile, a specific vulnerability in one of the AF&F components was discovered. This was a minor design tolerance issue that only impacts the operation of some of one component. The component is a safety and surety feature that is designed to permanently lock the component and disable the warhead if an incorrect arming signal is sent. The component is one of several redundant safety features that is designed to permanently lock and disable the warhead if an incorrect arming signal is sent in an accident. As a result of this design tolerance issue, under a very improbable accident condition, a locked AF&F component could unlock, though the warhead would remain unarmed.
As the AF&F components are produced, they are screened to determine whether this design variance will impact performance. If it is determined that it will, the AF&F component is pulled aside and used for other purposes. If AF&F performance is not affected, it is installed on the warhead. None of the AF&F components with this problem were ever installed on any of the refurbished W76 warheads. This problem did not delay delivery of W76s to the Navy beyond the fall 2009 timeframe we discussed.
We are currently implementing a design modification to permanently address the issue. While the problem with this AF&F component has not delayed delivery of the refurbished W76 to the Navy, correcting the design tolerance issue will remove the need to screen the AF&F components.