Hm interesting. My E4402 (that also doesn't match the Artek schems) has totally different part numbering; there is nothing like? "U54", all is numbered by group spaced by hundreds, so U8xx, U7xx around the pll charge pump, etc. If you have a good photo of your RF board that points to the location of U54 maybe we can narrow things down somewhat.
?
Something I didn't get around to trying when I had unlock problems on mine, was to somehow force the YIG drive signal to different levels, and look at the prescaler output (map frequency + power) to see how it behaves. I am not equipped to measure directly the YIG output so it would have been of limited use maybe.
I agree that an IC running high temperatures like that is suspicious, but not necessarily wrong. If by chance it is a prescaler IC, for some reason these seem to be the bane of HPAK gear; they run hot and fail often.
fenugrec