HP Forums

So I saw this Texas TI-59 for sale, complete with charger, hard case, french manual, Master Library Module ROM 1 complete with the full 25 program cards, a bunch of magnetic cards with some of them have blank clean labels and probably never used, and at last but not least, some cleaning cards, one of them still sealed.

The bad news is that the calculator was another victim of battery leaking and someone tried to repair it, having done an incomplete job on removing the electrolyte and have replaced the original NICd battery with one modern mobile NiMH cells soldered to the PCB.
And to add insult to the injury, the new battery is dead having two of the three cells in short circuit.
So when I opened this machine, I saw a total disaster in the keyboard domes, as the battery leaking had time to attack under the domes, and the calculator is doomed as it is.
I needed to find a keyboard donor, and a TI-58 is under his way to rescue.

Pictures of the card reader.

Pictures of the card reader (cont.)

Power adapter / Charger


I finally found someone more curious than myself.
This fellow did open the charger to see what's inside, spending great effort to separate the two plastic parts that forms the sealed case.
Nothing was touched inside, all solder joints are original.
Then sticky tape was used to joint the two parts together.

I guess this fellow didn't believe in the outside label:
Model AC 9900/H
Output: 6.2V~ / 200mA
So this is the typical Texas charger supplying an AC output to the calculator.

This is a European charger operating with an input of 220V~ / 50Hz / 3VA, made in Germany.
And it is the original charger according to the owner.

(11-11-2014 10:06 PM)jebem Wrote: [ -> ]I needed to find a keyboard donor, and a TI-58 is under his way to rescue.

Before you do the keyboard surgery, test the keyboard in the 58! Either a keyboard is bouncy or it isn't. If you've got a bouncy one, just wait for another.

(11-12-2014 08:01 PM)Marcus von Cube Wrote: [ -> ]Before you do the keyboard surgery, test the keyboard in the 58! Either a keyboard is bouncy or it isn't. If you've got a bouncy one, just wait for another.

Thanks for the advice, Marcus!
I was reading plenty of complaints about this bouncy kb issue - Fortunately the Texas prices are low enough to keep buying additional items until one has enough spares to build a good one

The Battery issue


The original battery using NiCd technology was designed to survive a few charging cycles (500 to 1000) using a basic charger circuit.

By replacing it with different NiMH cell technology battery soldered directly to the calculator battery terminals, meaning that it will be charged by the original calculator built in basic charger circuit (a resistor that is out of circuit when the calculator is powered on, meaning that the full charger current is applied to the battery), one can only expect the worst.




The wrong NiMH cells used for a while in the calculator. Battery leads soldered to the PCB terminals.




After cleaning. But it was to late for the keyboard domes.

Solid State Software Library and Magnetic Cards


I believe this is a first image published for the "presaturated" cleaning cards - a full original set, having one of them sealed in the bag.

Documentation

Seller receipt. The original owner bought it at a very popular electronics gadgets shop at the time in Lisbon.
This is the French version of the calculator sold in Portugal.
The cost was 21980$00 (equivalent to 110 Euros in 1983) - a very high price for a student or even a middle class working person to pay at the time in Portugal.



French owners guide. Warranty stamp from the seller shop with the correct calculator serial number.



Remaining documentation details.

Thanks for the pictures! This brings back fond memories. The 59 was my third calculator. It followed an SR-51A and an SR-56 (both with perfect keyboards). The 59 inherited the cheap keyboard design of the early TI-30. This has started a lot of trouble.

I still own my original 59 but the card reader can no longer read cards. :-(. My collection now consists of several 58Cs and 59s. I haven't used them for a while but I'm still hoping that I at least one of them is still working. If not, there are the recent "true" emulators.

Marcus, i followed a similar path.

SR-16-II
SR-51A
SR-52 (cheap hand me down)
A TI-58C

but then the HP 41c came out and I never looked back. :-)

That said, next year's HHC conference will have at least two presentations from the "dark side".

Marcus and Gene,
Thank you for sharing your past experiences and knowledge.

I started with a Casio FX-29 (or a 31, I don't remember the exact model) but moved to HP RPN machines as soon as I get money for them, having owned a HP-25C, a HP-55, and a HP-67.
Texas was the "other side" and only recently I have acquired a few of them, as part of my new calculator collection that includes only four brands so far (HP, Casio, Texas and Elektronika).


Card Reader testing - Part I
While I wait to get a good keyboard replacement for my HP-59, I had a closer look to the card reader and did some testing.

DC Motor:
DC resistance: 2 ohm

Current consumption without load (Tol. +/- 10mA):
41 mA @3.0 VDC
52 mA @4.5 VDC

Current consumption under load (Tol. +/- 10mA):
113 mA @3.0 VDC
130 mA @4.5 VDC

4-Track Magnetic Tape Read/Write Head:
Read/Write heads resistance (Tol. +/- 1 ohm):
L1: 270 ohm, L2: 272 ohm, L3: 265 ohm, L4: 267 ohm

Read/Write heads inductance (Tol. +/- 5 mH):
L1: 60 mH, L2: 60 mH, L3: 60 mH, L4: 60 mH

Signal output using several prerecorded cards of unknown state:
Signal read directly at the read heads with a oscilloscope (Tolerance is high here, like 10%).
Amplitude: 9 mVpp
Wave shape: Clean, it looks good to me.The encoding looks like a basic pair of frequency signals using FSK or maybe PSK (Phase Enconding), but I couldn't find information about it.
Higher period: 1.25mS (800Hz)
Lower period: 2.50 mS (400Hz)

According to the service guide, the magnetic head output signal is about 3 to 4 mV peak, so my readings (9mV peak to peak) are probably within the required values for a good card reading.



EDIT: Video showing the reading operation.

Keyboard replacement

Just received a TI-58, made in 1978 Holland, to be used as spares for this TI-59.
The keyboard needs to be removed on both machines, using a good soldering gun and some caution to not destroy the PCB.
It turns out that the PCB quality is first rate on both machines, resisting to a lot of abuse without apparent damage.

TI-59 PCB with the keyboard removed. 14 empty holes waiting for the replacement keyboard.



The original TI-59 keyboard (from 1982) taken apart and cleaned.
I believe I can repair this keyboard as well. Well, I will have a try one of these days.
IMHO the bouncing issue with these keyboards are related to the fact that some of the dome's crimping connections are getting lose with the usage.



The keyboard replacement. Despite being an older unit (1978), it works flawlessly.

Initial testing.

In order to test the machine without having to do a full assembly, three jumpers are needed in order to a successful power On.
On the 8-pin female connector, from left to right in the picture:
- Connect together the 2nd, 3rd and 4th pins (power on switch);
- Connect the 5th and the 8th pins (card reader sensor);
- Connect the 6th and the 7th pins (card reader sensor).




Repairing the Battery connector:

Another spare part taken from the TI-58: one battery connector.



Assembling:

A new keyboard foam was cut to the right size from a large sheet.
This foam is important to eliminate the extra gap between the plastic keys and the domes.
The right pressure on the keyboard is required to fit it into the plastic slots on the top cover, as the foam is pushing it out because it is a little on the thick side.
The back cover will put all in the right place, though.



The battery set was donated by the TI-58 as well.

Final testing:


The Diagnostics from the solid state software Master Library gave me a "1" in the display after 10 or 20 seconds later, by typing:
2nd Pgm 0 1 SBR =



Now, this is really amazing.
After a small adjustment in the card reader speed (I had to slow down it a bit by rotating anticlockwise the single potentiometer in the PCB), I could successfully read the magnetic cards written by this machine 30 years ago!!!
In the example below, I loaded a so called gamma program.



Trying 5! ...



And the live demo...

(11-22-2014 06:19 PM)jebem Wrote: [ -> ]IMHO the bouncing issue with these keyboards are related to the fact that some of the dome's crimping connections are getting lose with the usage.

I'd be glad to see a reliable fix for the issue.

(11-22-2014 07:29 PM)Marcus von Cube Wrote: [ -> ]

(11-22-2014 06:19 PM)jebem Wrote: [ -> ]IMHO the bouncing issue with these keyboards are related to the fact that some of the dome's crimping connections are getting lose with the usage.

I'd be glad to see a reliable fix for the issue.

When I have the time, I will try to restore my Texas TI-58 that is missing his original keyboard donated to the 59.
Now I have a dismantled 59 keyboard to play with.

Looking to the keyboard white plastic frame, we can see a number of small holes in front of the crimping/welding points. I think Texas made a less than perfect job when welding these joints, resulting in a lot of failing keyboards.

In my case, these mechanical/electrical joints were broken in several positions, and I believe this was causing both bouncing and missing keys as well.
Pressing a key can either cause multiple contacts spaced in time in a way that the debouncing software routine can't handle, or a total failure to register.

After reading so much reports from others here and there, I have noted that some people can fix some of these keyboard issues by just manipulating the original foam that sits between the plastic keys and the domes.
This makes sense to me, as the foam makes some pressure to the domes forcing them to make a better contact with the keyboard frame wires.
A thicker foam should give better results as well. I did it on one TI-57 some time ago with some success.

To test my hypothesis, I was thinking on soldering all the dome's connection points where originally it were crimped/welded at factory.
To do this, the plan is to lay down and solder one row of domes at a time.

Then, apply some sort of self adhesive plastic film, similar to the original one. I have yet to find the appropriate one for this application, though.

Hi Jebem ! Thank you very much ! I really enjoyed reading your article. Very good job.

Marcus about TI30 : long time ago, I owned a TI-30. Every time I was pushing a key on the keyboard I was afraid that the calculator breaks. So many years after, I still have in my ears the sinister noise of my TI30 keyboard (Krkkrrrrkkraccc)

(11-30-2014 09:53 AM)Gilles Wrote: [ -> ]Hi Jebem ! Thank you very much ! I really enjoyed reading your article. Very good job.

Marcus about TI30 : long time ago, I owned a TI-30. Every time I was pushing a key on the keyboard I was afraid that the calculator breaks. So many years after, I still have in my ears the sinister noise of my TI30 keyboard (Krkkrrrrkkraccc)

You are welcome, Gilles!

Yeah, the classic LED Texas calculator series sports keyboards and casings that are on the cheap side... however I find the electronics to be very reliable, and the PCB quality is first rate, second to none IMHO.

Hi Jose,

where did you connect the scope to measure that wave ?
Any specific point at the head terminals or at the LM324 or at the 594 ?
thanks for help !!!

(11-15-2014 09:41 PM)jebem Wrote: [ -> ]Marcus and Gene,
Thank you for sharing your past experiences and knowledge.

I started with a Casio FX-29 (or a 31, I don't remember the exact model) but moved to HP RPN machines as soon as I get money for them, having owned a HP-25C, a HP-55, and a HP-67.
Texas was the "other side" and only recently I have acquired a few of them, as part of my new calculator collection that includes only four brands so far (HP, Casio, Texas and Elektronika).


Card Reader testing - Part I
While I wait to get a good keyboard replacement for my HP-59, I had a closer look to the card reader and did some testing.

DC Motor:
DC resistance: 2 ohm

Current consumption without load (Tol. +/- 10mA):
41 mA @3.0 VDC
52 mA @4.5 VDC

Current consumption under load (Tol. +/- 10mA):
113 mA @3.0 VDC
130 mA @4.5 VDC

4-Track Magnetic Tape Read/Write Head:
Read/Write heads resistance (Tol. +/- 1 ohm):
L1: 270 ohm, L2: 272 ohm, L3: 265 ohm, L4: 267 ohm

Read/Write heads inductance (Tol. +/- 5 mH):
L1: 60 mH, L2: 60 mH, L3: 60 mH, L4: 60 mH

Signal output using several prerecorded cards of unknown state:
Signal read directly at the read heads with a oscilloscope (Tolerance is high here, like 10%).
Amplitude: 9 mVpp
Wave shape: Clean, it looks good to me.The encoding looks like a basic pair of frequency signals using FSK or maybe PSK (Phase Enconding), but I couldn't find information about it.
Higher period: 1.25mS (800Hz)
Lower period: 2.50 mS (400Hz)

According to the service guide, the magnetic head output signal is about 3 to 4 mV peak, so my readings (9mV peak to peak) are probably within the required values for a good card reading.



EDIT: Video showing the reading operation.

Hi Alberto,
I can't remember the exact procedure after all these years, but after reading my posts, I am quite confident that those readings were taken in open circuit, no load at all on each coil, when I was testing the motor and checking the coils impedance values, when the card reader assembly was disconnected from the main PCB. So I must have connected the gnd terminal of the scope probe to one coil wire, and the input scope probe to the other coil wire, checking each one of the four coils for a reading.