Apple //e keyboard issue [message #381504] |
Fri, 01 March 2019 15:03 |
|
Originally posted by: Bobbi
I have a very nice condition Apple //e which has a rather strange keyboard fault. I am seeking advice before I start tearing this mint machine to pieces. If it wasn't in such nice condition I would tear into it without further ado.
The numeric keys 1, 2, 3, 4 in the top row do not work very well. It used to be that if I 'exercised' these keys a few dozen times, they would start working. Now it seems even this expedient is not helping. If I press down on the affected key and rock it forwards and backwards then sometimes it will work.
This seems like a mechanical keyswitch problem, but why would four adjacent keys be affected when the rest of the keyboard works (and looks) like new? Is there a solder trace common to these four keys that I should be looking at? Perhaps there is a piece of debris lodged somewhere that could affect adjacent keys.
This is the original //e keyboard, without the numpad. I think it's made by ALPS.
Any ideas folks?
|
|
|
Re: Apple //e keyboard issue [message #381505 is a reply to message #381504] |
Fri, 01 March 2019 15:30 |
Michael J. Mahon
Messages: 1767 Registered: October 2012
Karma: 0
|
Senior Member |
|
|
Bobbi <bobbi.manners@gmail.com> wrote:
> I have a very nice condition Apple //e which has a rather strange
> keyboard fault. I am seeking advice before I start tearing this mint
> machine to pieces. If it wasn't in such nice condition I would tear into
> it without further ado.
>
> The numeric keys 1, 2, 3, 4 in the top row do not work very well. It
> used to be that if I 'exercised' these keys a few dozen times, they would
> start working. Now it seems even this expedient is not helping. If I
> press down on the affected key and rock it forwards and backwards then
> sometimes it will work.
>
> This seems like a mechanical keyswitch problem, but why would four
> adjacent keys be affected when the rest of the keyboard works (and looks)
> like new? Is there a solder trace common to these four keys that I
> should be looking at? Perhaps there is a piece of debris lodged
> somewhere that could affect adjacent keys.
>
> This is the original //e keyboard, without the numpad. I think it's made by ALPS.
>
> Any ideas folks?
>
I like your use of the failure of adjacent keys to propose a trace fault.
But it seems like only an intermittently broken trace could account for
just four keys being affected.
Another thing that’s spatially correlated is coke. ;-). It wouldn’t be
unusual for a beverage to spill int four adjacent keys.
Something that would almost positively confirm an intermittent break in a
trace would be that the correct functioning of the keys comes and goes for
all four simultaneously, rather than being able to restore function to one
while another continues not to respond.
If it *is* keyswitches failing separately, a good contact cleaner and lots
of exercise (dozens to hundreds of key presses) can restore most dirty
keyswitches to proper operation.
--
-michael - NadaNet 3.1 and AppleCrate II: http://michaeljmahon.com
|
|
|
Re: Apple //e keyboard issue [message #381509 is a reply to message #381504] |
Fri, 01 March 2019 17:09 |
|
Originally posted by: frank_o_rama
cold solder joint? try shorting out pins behind each switch to see if/how it registers. sometimes groups will go out because of impact damage.
bad keyswitch: vacuum/clean before using alcohol and deoxit.
slightly less likely: loose/bad keyboard ROM.
|
|
|
Re: Apple //e keyboard issue [message #381544 is a reply to message #381504] |
Sun, 03 March 2019 19:09 |
|
Originally posted by: groink_hi
Lemme see if I have this correct: IIe keyboard, and the 1, 2, 3 and 4 top row keys are faulty.
Detecting key presses are all about the keyboard encoder monitoring signals.. There are two rails for each key, which I normally call X and Y. When the encoder detects an X rail and a Y rail suddenly connected, then a key press occurred, and the encoder looks up the switch with both rails in common, and sends the ASCII code for that along.
Keys 1 through 9 are on the same X rail, while each of keys 1 through 9 are on their own Y rails. From the ribbon connector, the closest switch on this X rail is the 9 key, followed by 8 and so on, and ends at 1. If your problem is indeed a trace corrosion issue, the corrosion would have to be located between keys 4 and 5 on the X rail, because keys 1 through 4 are furthest away from the ribbon cable.
Luckily the PCB for the keyboard is only a 1-layer board, so locating the trace between 4 and 5 is easy. First thing I would do is place some flux onto all the solder joints for keys 4 and 5 and re-solder. The flux will help re-flow the solder. This would fix any cracked solder joints. If that does not work, then try soldering a wire between keys 4 and 5 on the X rail, and that would bypass the potential broken trace between these two keys.
|
|
|