Here is what I've been working on recently, the HP 5325A was pulled out of storage to help with the repair of the broken HP 5325B I acquired locally last year, however I noticed something wasn't right with the 5325A either, the overflow lamp was on all the time.
This was traced to transistor Q1 that is supposed to ground the supply from R2 going to the neon & extinguish the lamp when not required. The transistor still measured as two diode junctions with the DMM but wasn't working in circuit, it was part no. 1854-0232 which is apparently a 2N3440, one of those was fitted & the overflow lamp problem was fixed.
Onto the 5325B, I had previously cleaned the switches (as recommended on here), but it had made no difference as I mentioned before and it still refused to count anything, not even the self check.
The first problem was traced to the A4 function selector board, it has a few minor differences to the earlier counter (for the 20MHz spec rather than 12.5MHz, plus some resistors in series with the output BNC's), I verified it was the cause using the older counter. And using the 548A logic clip I traced the fault to IC6 1820-0380 (marked SN74H53 on the underside) which had no output, this was
temporarily permanently replaced with a standard 7453 as I haven't got any of the 'H' version.
That fixed the lack of any count appearing on the display, but also revealed a second problem, it would only count using the first five display tubes and one of those would only display a few numbers, mainly 4,5,6 & 7 then blanked again before repeating.
This took quite a bit longer to diagnose the failed IC's with the 548A & 546A, I removed the board & powered the 5V rail from the 6237B while fault finding, due to the amount of wires that would have needed disconnecting/reconnecting in the counter.
Two of the decade counter IC's & one of the buffer/storage IC's were faulty, with either missing outputs or were overloading the outputs of the preceding IC. The buffer/storage IC had been attacked by a previous owner, a pin cut off & random diode attached & also cut through (it still didn't work when reconnected).
All the IC's for the counting & display section are custom parts, thankfully I had a parts board that donated some replacements.
Here are the two counters after repair measuring the 1MHz output from the Racal 9839 and the faulty parts. I still have one problem with the 5325B, that is the crystal oven not getting warm.
David