His failure is hardly due to running the supply right at its maximum continuous rating. NEXT?<---I would submit that metal migration within the diodes was a factor in their failure. (I first heard about this when I worked at TRW in the 1970's and RF power transistors were still kinda "new")http://www.ridgetop-group.com/docs/prognostic_techniques.pdf
I'd guess it isn't the diodes, but is one or more Pass Transistors shorted. Just a guess.... Kevin
