HP 65 Repair Effort

[ClausB]

I recently got a broken 65 off eBay. On power up it displays all 1s for a second and then goes blank. It has been repaired before because the logic board ROMs have later dates (7437) than the serial label (1333A 08230) and the main IC has a sticker dated '78. Also the card reader is clean and its wheel has new o-rings. Would anyone here happen to know the history of this unit?

I printed the logic board layout from the May '74 HP Journal article and checked the power supply output with a DC voltmeter. I get these readings:

Nominal _ Actual

3.75 _____ 3.8 (batt)

6 ________ 6.2

7.5_______ 7.7

-12 ______ -12.2

These seem OK, no?

I also measured 2.9 DC volts at phi1 and 3.4 at phi2. Sound reasonable?

I guess the next step is to scope some signals. Any other ideas?

Edit: Can that serial number be right? I thought the 65 came out in '74.

Edited: 2 Apr 2010, 5:59 p.m.

[Eric Smith]

DC voltages OK. To really tell whether the clock is OK, you'd have to look at it on an oscilliscope. The main signals to look at are:

          test    hybrid    ROM

signal   point     pin      pin

------   ------   ------  ------

 Phi1       6       38      12

 Phi2       5       39      10

 SYNC       1       32       7   sync, pulse active 10 of 56 cycles

 IA         3        8      15   instruction address, CPU -> ROM

 IS         2       33       6   instruction, ROM -> CPU

 RCD       10       34      n/a  reset cathode driver, every 15 words

 WS         9       30      14   word select, which digits are affected by instruction

 DATA       8       35      n/a  56-bit data word

Pin information is from my own personal notes, but I think it matches Tony's schematics.

[ClausB]

That's very helpful. Thanks, I'll check those.