+- HP Forums (https://www.hpmuseum.org/forum)
+-- Forum: HP Calculators (and very old HP Computers) (/forum-3.html)
+--- Forum: General Forum (/forum-4.html)
+--- Thread: Help needed: HP 67 repair (/thread-15496.html)
Help needed: HP 67 repair - Marcovecchio - 08-24-2020 03:58 AM
I'm trying to fix an HP 67. The condition is not good: no charger, battery leakage and someone already tampered with it. Before describing what I have done so far, here goes my main question: is there an easy way to make sure ROM0 and the 4 small ROM/RAM ICs are good?
Calculator status: I'm powering it with a bench supply, 3.8V. It's pulling 190mA, and display looks like this, almost all is lit:
[attachment=8707]
On the keyboard board, both chips seem the be working, because I have the display almost completely lit. Only segment c looks strange, it's dim on some digits, and lit on the exponent signal position, but I think there's a partial short somewhere, I'll try to find it. Battery leakage killed 3 tracks, but I repaired them.
On the processor board, I confirmed Vss and Vgg are ok (+6.2V and -12V). ACT has clock on pins 16 and 17, and I see activity on ISA and Data pins on the scope. Both look noisy, but noise amplitude is less than 1V, and signal has good 6V swing, so I assume it's good.
ROM0 seems to be good: I have strobe signal, and 7 segments activity. The only segment permanently off is the dot (segment H).
I replaced the smaller capacitors, the bigger tantalum ones I don't have at hand to replace, but I removed and measured them, they're good.
Well, I think I covered everything. I'm confident something is working, the calculator displays a minus on first position, and it clearly shows the space between mantissa and exponent, so it doesn't look like random garbage on the display.
The parts I have no clue on how to test, are the ACT, and ROMs/RAMs. I'm assuming the ACT is good, because on clock and ISA activity, but I'm not 100% sure. The 4 small ROM ICs are a mistery to me, and ROM0 seems to be partially working, but I have no idea on how to confirm it's good for sure.
Please, could you help me on how to proceed from here? I have a simple logic analyser, but I have no idea on how to decode the pulses to confirm I have good data on ISA and Data lines.
Another question: how does the Data line work? All 5 ROM ICs share it. Do they keep Data in high impedance state, with only one Rom feeding it at a time?
Any help is greatly appreciated, thank you!
Eduardo
RE: Help needed: HP 67 repair - teenix - 08-24-2020 06:16 AM
Does it power up with 0.00 displayed? If not there could be issues with the ACT and one or more ROMs. If numbers can be entered but displayed wrong, it may be PCB, Cathode/Anode driver or ROM 0 causing the issue.
Only segment G should light on both the sign digits. They have 4 transistors decoding and driving them from segments E and G. Because segment C is lit on the exponent sign, it may suggest something is "crossed" on these segments somewhere.
If the cathode driver is playing up, it should be easy to spot with a scope.
If you want to download this, I have some info on instruction decoding for Classics towards the end of the document which may help.
http://www.teenix.org/ClassicNotes.pdf
cheers
Tony
RE: Help needed: HP 67 repair - Marcovecchio - 08-25-2020 02:27 AM
Thank you very much, Tony! This document has a huge lot of information, I'll study it carefully and try to find out what's happening to my calculator.
Eduardo
RE: Help needed: HP 67 repair - Marcovecchio - 08-26-2020 02:29 AM
I still didn't study the signals, but the C segment problem seems to be related to the transistors near the display: when I removed them, segments G turned off, and segments C started to behave correctly. I'll try cleaning the board somehow, specially under the display LEDs, probably I have some kind of short between C and G lines.
And I think I have damage on the ROMs: the calculator behaves the same exact way with or without any of the 8 pin ROMs. But I'll try to investigate more and find out where exactly the problems are.
Thanks!
Eduardo