View Full Version : PC100 printout with missing column

October 27th, 2011, 12:41 PM
Dear TI folks,

I'd like to share my solution to a nagging problem with my PC100B printing cradle for the TI59/58/57 and SR52/56 calculator line.

My PC100B consistently refused to print the 4th column. After a thorough cleaning of the printhead (which left no *visible* corrosion) still failed to fix the problem, I initially suspected the ribbon cable, which however checked out ok with an ohmmeter. I then concluded one of the (apparently unobtainable) SN96912N driver ICs had failed and the printer was a write-off.

Last week I went back to it and decided to test the print head contacts systematically for continuity. Lacking a service manual (anyone actually found one online?), I figured out the print head connector assignments, consisting of a matrix addressing the 20 columns and 5 dots per column on the print head:

Connector (0=right) Column/Dot (0=right)
0 = col 0
1 = col 1
2 = col 2
3 = col 3
4 = col 4
5 = col 5
6 = col 6
7 = col 7
8 = col 8
9 = col 9
10 = col 10
11 = col 11
12 = dot 0
13 = col 12
14 = dot 1
15 = dot 3
16 = col 13
17 = col 14
18 = dot 2
19 = dot 4
20 = col 15
21 = col 16
22 = col 17
23 = col 18
24 = col 19

Note the columns and connectors are numbered right to left facing the print heads along the top edge, corresponding to the usual left to right numbering on paper. Likewise, the 5 dots per column are also numbered right to left.

Checking for continuity between any one of the dot connectors and all columns revealed open circuit on the 4th column, despite an apparently clean contact. Even after working the contacts with a fine sandpaper I still got open circuit. Finally I resorted to scratching the contact (none too gently!) with a precision screwdriver , and THAT did the trick!

Note that the PC100B heads are actually diodes rather than simple resistors, and polarity DOES matter; the anodes are on the dot connectors (reverse probes to make sure)! The indicated resistance depends on the forward voltage during measurement, but should be finite and consistent for all columns.

Hope this helps anyone with similar nagging problems.


November 6th, 2011, 06:53 AM
This is a very useful tutorial for those of us with PC100s! Many thanks for sharing your experiences with it, as it will come in handy when checking out a problem with mine.