Texas Instruments TI59 weird card problems

[voltaage]

For quite a time, I've been looking for a TI 59 to play, so when I saw one complete kit at a local flea market, I couldn't resist. After taking it home I noticed that was indeed a very nice model. The old batteries had already been removed and there was no corrosion anywhere. Apart from the normal amount of dust you'd expect for something like this, everything was in place. So I rebuilt the battery pack and gave it a clean with a moisted towellette. After making sure there was proper voltage at the battery terminals, I placed the pack and voila! It turned on right away. All the keys worked, all the operations gave the correct results, and even I could read some of the cards written by the previous owner. Ran the diagnostic card and there it was: diplay full of eights!

So I decided to take a deeper look into the other blank cards included, so I could retrieve any data stored on them. Turned out that whoever was the previous owner took great care of the device and/or it had very little use since there where only 4 custom written programs, and most of the pre-written cards looked intact. Wrote two test cards (a program counting from 1 to 9 and another one counting from 9 to 1) and read them with no problem. Tested the ROM module with success. That when I decided to place the key-code mat over the keys when, by magic, I sticked one of my previously written cards to the side but it refused to read. Not even spin the motor! I noticed that the display, when a card was inserted, a "C" on the left side as well as for a fast-blinking "8" on the right side showed up on the screen. If the card was removed (manually because there was no motor activity) then the "C" went away; the number "8" stayed in place but this time with a slower blinking pattern (similar to an error condition). You could press CLR to reset the operation weirder things happened: Let's say you were to do 25+3 you'd get 28, but if you wanted to do sin90 then you'll end up with a "0". Additionally, if you wanted to access the program mode by pressing LRN, at first you'll get garbage symbols; if you tried again, instead of "000 00" pattern, it showed "000000".

At first I thought it was all due to a misconnection between contacts inside the card reader mechanism. I downloaded the service manual and read the of that part. To my surprise when I got to open the case, everything seemed allright!? No corrosion, no acid, anything. Tested several times those contacts but everything worked accordingly. Tested the flex connectors in seek of shorts/misconnections but all ok. Lastly ran some voltage over the motor (2V) and it was ok!

After scratshing my head during 2 hours, I came to the conclusion that I don't know what is going on. My clues point where nothing makes sense. How it was working nice, but all of a sudden it collapsed?

If anyone has a similar experience I would really appreciate your help.

Sincerely, voltaage.

[Klaus Overhage]

A weak point of the TI-59 are the two metal strips that protrude from the circuit board as flexible contact tongues and establish the connection to the battery pack. They break very easily if you try to bend them a little higher, but they have to be high enough to safely reach the battery pack.

I once had the problem that newly purchased, fully charged batteries were apparently too weak to supply the TI-59 with enough power. I then placed the battery pack next to the TI-59 and connected it to the metal strips with two cables with alligator clips. Since the TI-59 then worked without problems, the cause was found: It was due to contact problems with these metal strips. I cleaned the tips of the metal strips and the contact areas on the battery pack with an alcohol cloth and bent the metal strips up very slightly.

[voltaage]

Hello Klaus!

In fact, that was what I initially did when the calculator arrived home. I built a battery package using only aluminium tape to make the contacts. Although it worked for the first 30 minutes, It started to get an inconsistent connection due to the high pressure printed by the contacts onto my poorly-crafted aluminium contacts (eventually they got chewed-up). The next day I did a better job with a thicker sheet of can aluminium; I just had to make sure that the negative battery contacts had enough pressue to keep them connected. Also I had to remove the varnish of the sheets and sand them a little bit. Unfortunately what I described at the top of the thread was just after that mod. Even I got to power my calculator directly to a power supply with a pair of cables, but the problem was still there.

Thanks for your reply but seems that something bigger is going on.

(10-27-2021 08:49 AM)Klaus Overhage Wrote:  It was due to contact problems with these metal strips. I cleaned the tips of the metal strips and the contact areas on the battery pack with an alcohol cloth and bent the metal strips up very slightly.

[Ren]

(11-01-2021 01:10 AM)voltaage Wrote:  Also I had to remove the varnish of the sheets and sand them a little bit

Many alumin[i]um cans have a plastic coating inside.
An acquaintance of mine, to illustrate that point, will put an unopened beer can into a bowl of Hydrochloric (?) Acid to about midway up the can. After a while he lifts the can out to reveal the alumin[i]um portion of the can that was in the acid has dissolved, but the beer is still contained in the plastic that remains.

[voltaage]

(11-03-2021 02:01 PM)Ren Wrote:  

(11-01-2021 01:10 AM)voltaage Wrote:  Also I had to remove the varnish of the sheets and sand them a little bit

will put an unopened beer can into a bowl of Hydrochloric (?) Acid to about midway up the can.

Whoah that is quite impressive. Certainly I don't have those chemicals at my reach. The can I used to make the contacts was an unbranded starbucks cookie straws container (like an OEM product: only sold for their franchises). Both sides (inner and outer) were covered with that thin plastic lacquer of some sort, but there was no label paint as you would normally encounter with soda cans. That made the job easier since I only had to remove the film with lacquer thinner, then sand them to make the surface conductive.

I don't think that is the problem beacause I also tested my calc with a power supply: same results. Also as I mentioned earlier that can is way more thick than alluminium tape, which I used to in the initial rebuild.