Back to main page
I've had a lot of visitors but few comments! Let me hear from you! Email me at: Gene!
This article will deal with 1) memory problems, 2) print head cleaning, 3) keyboard problems, 4) electrical contact problems, 5) improper paper advance, and 6) card reader problems.
Some TI-58/59's display bizarre malfunctions. Palmer Hanson reported finding one at Honeywell that worked fine but started developing an anomaly in the Sine function after being on for about thirty minutes. Problems like this will almost impossible to find. Most problems aren't that hard to find, although a machine will sometimes work fine and other times not.
Memory Problems Page 43 of the Programmable TI-58/59 Service Manual describes a program that stores 1/9 (all 1's) into each of the 100 memories and then sums -1/9 into each memory, checks the contents of each memory for zero and prints out the contents and register number for each failure. The test takes about three minutes.
However, machines have been found that will pass this test and yet not properly work for other numerical values. Palmer Hanson modified the TI program to perform the test with the series of values from 1/9, 2/9, up to 8/9. Palmer stated that he had not found a memory problem that went undetected by this program. A fast mode variation of this program was later published that ran in only 8 minutes. Fast Mode TI-59 Memory Test
To test program memory from steps 000 to 159 (the part of memory not tested by the 1/9 program above), load each of those steps with a 41 keycode (SST). This can be done, albeit rather tediously, by keying in STO 41 into all of program memory, going back and deleting each 42 keycode and leaving all 41 keycodes. You will need to do this a few times to get program memory completely filled with 41 keycodes. Record this to a card for future use. To perform this diagnostic, press CLS, RST and SST. The program should go all the way through memory and will stop if any step is not a 41. (Well, actually, it will not stop if a step is somehow changed into a 31 keycode, but that is not TOO likely).
Print Head Cleaning The PC-100 manuals described a print head cleaning technique for cleaning foreign particles which may have collected on the print heads. The program listed below will do a pretty good job. To use the program, press A and see I's printed. Press R/S and then B to see II's printed. Press R/S and then C and see square O's printed. Press R/S and then D and see 8's printed.
Print Head Cleaning Program Listing:
Step Instruction Keycode 000 OP 69 001 05 05 002 RST 81 003 LBL 76 004 D 14 005 1 01 006 1 01 007 LBL 76 008 E 15 009 * 65 010 1 01 011 0 00 012 1 01 013 0 00 014 1 01 015 0 00 016 1 01 017 0 00 018 1 01 019 = 95 020 OP 69 021 01 01 022 OP 69 023 02 02 024 OP 69 025 03 03 026 OP 69 027 04 04 028 RST 81 029 LBL 76 030 A 11 031 2 02 032 4 04 033 E 15 034 LBL 76 035 B 12 036 7 07 037 4 04 038 E 15 039 LBL 76 040 C 13 041 3 03 042 2 02 043 E 15
Keyboard Problems: One suggestion for helping keys make contact properly and to reduce "bounce" is to take the calculator apart and reverse the foam under the keys. This has apparently had some success.
Contact Problems: Sometimes, where electrical contacts are present, such as where the plug in roms make contact on the back side of the rom and within the printer, scrubbing the contacts with a pencil eraser has been known to provide success.
Improper Paper Advance: Some PC-100's develop an inability to properly advance the printer paper. You can often provide proper paper advance by lifting the top cover and pressing on the metal support which holds the print head against the paper, but that is tiresome. You can make a temporary repair by cutting a wedge from an eraser (the soft white ones seem to work best), and pressing the wedge between the metal support and the printer case until proper paper advance occurs. Of course, you will have to do that all over again each time you install a new roll of paper. This temporary repair can become semi-permanent, however.
Card Reading Problems: From the very beginning, TI-59 owners seem to have encountered read/write problems with the card reader.
Some problems seem to be with contaminated read/write mechanisms. Other problems seem to be with contaminated cards. TI provided two card reader cleaning cards with each TI-59. The head cleaning card contains an abrasive and can be used to remove buildup of oxide or foreign material on the read/write head. TI cautioned that this card should be used sparingly. Palmer found very little use for this card; only in one instance did use of the card definately improve read/write characteristics.
The drive roller cleaning card seems to be no more than a piece of cardboard. Palmer's experience was that the card did not seem to help much; furthermore, the card deteriorates rapidly to a condition where it is simply unusable.
Got a card that just will not read? Try spitting on it! Palmer discovered that saliva, believe it or not, works many times when you have a card that just won't read. Wet the card with saliva and then gingerly wipe the card off. You might find that it works. Others believe that a form of isopropyl alcohol works well.
There are some machines that will not read cards recorded on other TI-59's due to the variations that exist in reading speeds as the card is passed through the calculator. Palmer's experience was that MOST TI-59's would read cards written by other machines. Make sure that the card will read on ANY TI-59. It is always possible that the card is damaged. Just because the card won't read, doesn't mean that anything is wrong with the calculator.
That's it. Hope this is helpful to someone!