Semi-dead black keyboard repair?

[retro486]

I recently picked a pretty bad shape silver/black TI-99/4a (like out in the open of a damp garage bad) and the upper right portion of they keyboard doesn't seem to work, from the enter key to below the number keys.

 

I've disassembled it and cleaned it up, removed the keyboard (which is in surprisingly good shape for how dirty the thing was) and checked the continuity on a few of the dead keys and the switches are working perfect, and no shorts from what I could see, so I suspect something else is up but I wasn't positive about further disassembling the keyboard since I don't have any spares handy.

 

I suppose I couldn't make it worse but do these symptoms give a clue what might be wrong? Do the controllers go bad? From what I've seen, most people seem to have broken stems or missing contacts, neither of which are the issue on this one.

 

Thanks!

 

 

[Sinphaltimus]

Just throwing it out there - in case you want to continue taking it apart to try and fix it.

http://atariage.com/forums/topic/242816-usb-keyboard-adapter-warduino-style-components/

 

EDIT:FIXED URL

Edited June 21, 2017 by Sinphaltimus

[SignGuy81]

This may help you in finding the issue.

 

Edited June 21, 2017 by SignGuy81

[retro486]

Thanks for the info on the USB conversion but I'd rather try to repair the original board if possible; I do like it and it's switch style (slider/spring?).

 

SignGuy, I think that map will be incredibly helpful, thanks! Looks like I've got some poking to do.

 

Just looking at that I see most the dead keys are leading to the same pin: 1, which means that trace is now highly suspicious.

[+arcadeshopper]

This may help you in finding the issue.

 

 

 

thanks I was searching for that schematic

 

If the keyboard has solder points on the backboard, you are in luck! It can be repaired if it has dirty switches etc..

 

If the keyboard has just a pcboard without any solder points except for the caps lock key then you have a mitsumi and you may as well toss it..

 

You can test the keys by testing from the connector for continuity and eliminate the keyboard.. the keyboard connects to the 9901 so that may be bad if the keys test out right.

 

 

Greg

[+arcadeshopper]

 

GOOD (or like it, there's like 3 models with switches)

 

 

BAD

[+arcadeshopper]

Just throwing it out there - in case you want to continue taking it apart to try and fix it.

 

http://atariage.com/forums/topic/242816-usb-keyboard-adapter-warduino-style-components/

 

EDIT:FIXED URL

 

http://ti994a.cwfk.net/TiUsbKeys.html a better URL

[SignGuy81]

Look really close at the solder joints. Especially near pin 1 on connector and the letter "P"

Edited June 22, 2017 by SignGuy81

[SignGuy81]

From your description I believe you have one that isn't the mitsumi which is good, but just want to let you know that those may be repairable as well. I've repaired that type(traces on the sheets of plastic) with conductive pens in the past.

[retro486]

Alright, I've attached a photo of my keyboard PCB, it's the solder-point style.

 

I tested the affected keys for continuity and there aren't any shorts. I also checked the connector cable and all wires are solid and no shorts between the 15 pins either.

 

Is there a way to test the 9901?

 

Edit: The solder on the keyboard connector on the motherboard looks good, as do the connections to the 9901 on the board but I'll so some basic continuity tests.

Edited June 22, 2017 by retro486

[+arcadeshopper]

nice, well then did you try the keyboard again after cleaning the pins on the motherboard it connects too?

 

9901 is soldered in I believe, so you'd have to replace it ..

 

Greg

[retro486]

Yup retested and even checked continuity to the 9901 (I tried to follow the traces) from the keys. Alright ordered a replacement and some sockets (always add a socket, right?). I'm curious if a bad 9901 would ignore keys on a single pin (pin 1 in my case). I'll also try resoldering/resetting the existing 9901. Thanks for the tips!

Edited June 22, 2017 by retro486

[Sinphaltimus]

Thanks for the info on the USB conversion but I'd rather try to repair the original board if possible; I do like it and it's switch style (slider/spring?).

 

 

No switch, stays on and it doesn't mean giving up your original keyboard. It's just an optional add-on that only takes away "completely original unmodded" from your console description.

[ti99iuc]

please, check very well the wires soldered on the board from the connector... often it is happened that some wires can be disconnected from the board and you have to resolder them.

 

[mizapf]

Yes, I noticed that one too. The cables of the keyboard and their solder joints are very brittle; bend them once, and at least one cable breaks off.

[retro486]

I'll reseat the keyboard connector while I'm reseating things to be sure. A continuity test showed when I pressed one of the pin 1 keys, a signal was making it to at least the resistor bank on the mainboard but when powered on they were still unresponsive.

[retro486]

Nope. Resoldered the keyboard cable connector both on the keyboard and on the motherboard, also resoldered each point in the failing chain on the keyboard with no improvement. I removed and reseated all chips that could be reseating also with no improvement.

 

I've ordered another TI-99/4a and this one will be my parts machine. It'll also let me test if the keyboard is any good. Thanks for the help everyone!

[retro486]

Ok. I socketed the 9901 but I haven't replaced it yet (still the original 9901), resoldered the keyboard cable and the keyboard pins on the motherboard.

 

With no power, continuity checks out. Once I apply power, pins 1 and 13 short. These are the two that my dead keys use. All other keys work perfectly.

 

That should narrow it down but I have no clue what that means... a short in a gate that's only closed when powered on?

[SignGuy81]

It may be possible that the "I" switch is stuck closed, which is causing the problems because the input for that key is always on. desolder that key and then check to see if the rest work then.

 

Just to make sure I don't confuse it is the capital "i" not an L.

 

If you were doing a continuity test while checking the letters on that section you might not have caught if it stayed on all the time. Before desoldering that key, first unplug the connector and check to see if you have continuity between 1 and 13 even with the key not pressed. If so then desolder the key.

 

 

EDIT:

Last edit. See on this computer I have now I can type away leaving keys pressed down. But on my TI-99 computer I can't do that with some letters so I think that probably explains it. So if one key is stuck closed it can probably cause that issue.

 

EDIT:

 

I got pins mixed up I think 13 would have been "O" not "I"

Edited June 25, 2017 by SignGuy81

[retro486]

With the keyboard removed, there's no issue. Continuity test beeps only when I press the keys.

 

Now on the keyboard header on the motherboard, with power off all the pins appear disconnected (no continuity). Once I apply power, pins 1 and 12 (sorry I miscounted) short. A few others short with 12 as well: 6, 7, 8, 9.

 

I do have a spare ti99/4a coming late next week that I can test with but I'm impatient.

Edited June 24, 2017 by retro486

[SignGuy81]

 

Now on the keyboard header on the motherboard, with power off all the pins appear disconnected (no continuity). Once I apply power, pins 1 and 12 (sorry I miscounted) short. A few others short with 12 as well: 6, 7, 8, 9

How are you checking for short with power on? Just want to make sure you don't have a multimeter set on ohms. Instead you would want to check for voltage between pin 1 and the following pins: 7(P), 13(O), 14(I), 15(U), 9(Y), 12(ent)

 

Voltage presence between pin 1 and the others means no short and open. Pressing a key(if connector is plugged in) would short across and cause 0V. That is how it should look on voltage test.

[retro486]

Ok I was just checking continuity, I wasn't checking voltage. Thanks for the instructions, I did a voltage test and all keys show some voltage then drop to 0 once I press them. So I guess there's no short?

[SignGuy81]

Ok I was just checking continuity, I wasn't checking voltage. Thanks for the instructions, I did a voltage test and all keys show some voltage then drop to 0 once I press them. So I guess there's no short?

 

 

Each one to pin 1 should show around the same voltage, is any of them significantly more different?

[retro486]

Enter (12) is almost 3.5v. The rest are 0.12-0.15v...

[SignGuy81]

Enter (12) is almost 3.5v. The rest are 0.12-0.15v...

 

Yeah that doesn't sound right.

 

Just to clarify because I don't have a schematic and I want to know which way it is set up. When checking do you have black lead on pin one and red lead checking the others? I want to know if pin 1 is positive or negative. For example if pin 1 is connected to black lead of meter and checking 12 with red lead does it read -3.5V or 3.5V? I just want to know that out of curiosity it may help down the road, but yeah all of them should have that voltage on them without the keys pressed.

 

So also too is the "ent" key working by the way?