Fixing TRS-80 Model III keyboard problems

[d_a_parker]

Hi All,

 

Looking for a little help from folks who are really good with circuits...

 

I've had a TRS-80 Model III for a few decades. A bunch of keys on the keyboard stopped working, so I assumed they were dirty and went through the process of desoldering/cleaning/resoldering all of them, only to discover that the same keys still don't work. I then found the Model III service manual online, which has a diagram of the keyboard interface on the CPU board. See the attached PDF for the diagram.

 

From the diagram, I discovered that lines 9, 10, and 15 account for all of the keys which do not work. So the problem is most likely on the CPU board and not the keyboard itself. The problem is, I'm no electrical engineer and I might be a bit out of my depth here. Looking at the diagram, it looks like each of those lines is connected to a resistor (RP5) and maybe a diode (D0A, D1A, etc.) but I don't know if I'm even reading that correctly. I can't really tell if those are all discrete components or if some of them are the same component shared by multiple lines, though. So I'm wondering which parts are likely the culprit here, and how I might locate them if I pull the CPU board from the computer.

 

I tested the 20-pin ribbon cable end-to-end and it seems fine, so the issue isn't in the keyboard or it's connection. It definitely seems to be something on the CPU board based on this diagram.

 

Thank you!

trs80-model-3-keyboard-diagram.pdf

[d_a_parker]

Update:

 

The components on the CPU board are not the problem. Looking again at the diagram, it occurred to me that the grid of horizontal lines (1- and vertical lines (9-16) indicates which two pins of the ribbon cable are connected when the specified key is pressed. For example, when you press the 2 key in the number row, the resistance drops between pins 5 and 11.

 

I examined the board carefully and found that the lines which do not work (9, 10, 15) are all routed through switches in the numeric keypad. These keyboard switches seem to have constant continuity between leads which are across from each other, and each lead has continuity with the one diagonal from it when the switch is pressed. When I manually short the switch lead connected to pin 9 to its neighbor, the "broken" keys on line 9 suddenly work again. So, there seems to be a continuity problem across the leads of some switches. That's progress, at least.

[d_a_parker]

Removing and cleaning the numeric keypad switches solved the problem. In hindsight, I think that the loss of continuity across a few of those switches was the whole problem in the first place. Anyway, all of the keys are working and the computer seems to work fine. Success!