Strange keyboard behaviour

[Paul]

My daughter has been using a usb keyboard on her laptop and now, suddenly,
some of the keys have started typing 2 items at once, here are some:
press m and you get m,
press , and you get ,m
press j and you get jk
press k and you get jk
y]
ui
ui
y]
6=
78
78
6=
I have tried the keyboard on my desktop and it still does the same. I asked
if she had spilt anything on it and she said no, not that that is true.
Is there anything I can try before I bin the keyboard.

Many thanks Paul

 

[Brian A.]

My daughter has been using a usb keyboard on her laptop and now, suddenly,
some of the keys have started typing 2 items at once, here are some:
press m and you get m,
press , and you get ,m
press j and you get jk
press k and you get jk
y]
ui
ui
y]
6=
78
78
6=
I have tried the keyboard on my desktop and it still does the same. I
asked if she had spilt anything on it and she said no, not that that is
true.
Is there anything I can try before I bin the keyboard.

Many thanks Paul

Check for any keys that may be stuck and give the board a good cleaning.

--

Brian A. Sesko
Conflicts start where information lacks.
http://basconotw.mvps.org/

Suggested posting do's/don'ts: http://members.shaw.ca/dts-l/goodpost.htm
How to ask a question: http://support.microsoft.com/kb/555375

 

[Paul]

My daughter has been using a usb keyboard on her laptop and now, suddenly,
some of the keys have started typing 2 items at once, here are some:
press m and you get m,
press , and you get ,m
press j and you get jk
press k and you get jk
y]
ui
ui
y]
6=
78
78
6=
I have tried the keyboard on my desktop and it still does the same. I asked
if she had spilt anything on it and she said no, not that that is true.
Is there anything I can try before I bin the keyboard.

Many thanks Paul

Keyboards use matrix scanning. That saves pins on the controller IC.
If you needed to sense over 100 individual switches, with 100 pins on
an IC, that would cost a fortune. If you instead, made a 10x10 matrix,
just as an example, that would take 20 pins. The chips used, don't
always use a square matrix, and you might find an 8x18 array or something
similarly weird. And the wiring pattern on the keyboard, seldom makes
sense - the keys sharing a wire may not all be adjacent.

(An example here, of the concept. This isn't an actual keyboard chip.)

http://www.atmel.com/dyn/resources/prod_documents/doc2532.pdf

As a result of the matrix scanning, if there is a malfunction, there may be
evidence of the multiple keys, that share a wire. That may be what
you're seeing. It could be that two wires from the keyboard chip are
shorted together, such that one key press causes two of them to be
detected.

*******

On the keyboard I'm typing on, the construction of the keyboard is very nice.
I remove a handful of screws, and the whole mechanism is accessible. There are
three plastic membranes, the middle one being a "mask" that only allows
the outside ones to touch at carefully defined points. The keys press
the membranes together, and that is detected as a key press. So the
"switches" in the matrix, are conductors printed on plastic.

If I remove the plastic membranes, they are "welded" at a few points.
So the membranes can't be completely separated from one another.
That is a good thing. That is to guarantee that the orientation of
the plastic sheets is maintained, after a guy like me has finished
fooling with them All the conductors and holes, continue to line up.

I've washed spilled coffee off those membranes. I carefully propped the
membranes apart, while a table fan was located next to the membranes to
dry them. If it is a "spilled fluid" problem, you may be able to
do something like that. If it is a "dead controller chip", then well,
no repair is possible for reasonable money. There is usually only one
big chip inside the keyboard.

So, buy a replacement keyboard. A USB one might cost $20. Make sure that
new keyboard works first, then disassemble the duff one and work on it.

I have an older keyboard that cost $150, and it is assembled internally with
steel rivets. It was impossible to work on, in any reasonable way.
(I've tried drilling steel rivets in the past, and they just spin
around on you.) It is possible a cheap keyboard, may be easier to
work on, than an expensive one. You'd be surprised, even on a
well maintained keyboard, how much junk gets inside there.

Paul

 

[SC Tom]

*** reply in line.

My daughter has been using a usb keyboard on her laptop and now,
suddenly, some of the keys have started typing 2 items at once, here
are some: press m and you get m,
press , and you get ,m
press j and you get jk
press k and you get jk
y]
ui
ui
y]
6=
78
78
6=
I have tried the keyboard on my desktop and it still does the same. I
asked if she had spilt anything on it and she said no, not that
that is true. Is there anything I can try before I bin the keyboard.

Many thanks Paul

Keyboards use matrix scanning. That saves pins on the controller IC.
If you needed to sense over 100 individual switches, with 100 pins on
an IC, that would cost a fortune. If you instead, made a 10x10 matrix,
just as an example, that would take 20 pins. The chips used, don't
always use a square matrix, and you might find an 8x18 array or
something similarly weird. And the wiring pattern on the keyboard,
seldom makes sense - the keys sharing a wire may not all be adjacent.

(An example here, of the concept. This isn't an actual keyboard chip.)

http://www.atmel.com/dyn/resources/prod_documents/doc2532.pdf

As a result of the matrix scanning, if there is a malfunction, there
may be evidence of the multiple keys, that share a wire. That may be
what you're seeing. It could be that two wires from the keyboard chip are
shorted together, such that one key press causes two of them to be
detected.

*******

On the keyboard I'm typing on, the construction of the keyboard is
very nice. I remove a handful of screws, and the whole mechanism is
accessible. There are three plastic membranes, the middle one being a
"mask" that only allows the outside ones to touch at carefully defined
points. The keys press
the membranes together, and that is detected as a key press. So the
"switches" in the matrix, are conductors printed on plastic.

If I remove the plastic membranes, they are "welded" at a few points.
So the membranes can't be completely separated from one another.
That is a good thing. That is to guarantee that the orientation of
the plastic sheets is maintained, after a guy like me has finished
fooling with them All the conductors and holes, continue to line
up.
I've washed spilled coffee off those membranes. I carefully propped
the membranes apart, while a table fan was located next to the
membranes to dry them. If it is a "spilled fluid" problem, you may be able
to
do something like that. If it is a "dead controller chip", then well,
no repair is possible for reasonable money. There is usually only one
big chip inside the keyboard.

So, buy a replacement keyboard. A USB one might cost $20. Make sure
that new keyboard works first, then disassemble the duff one and work on
it.
I have an older keyboard that cost $150, and it is assembled
internally with steel rivets. It was impossible to work on, in any
reasonable way. (I've tried drilling steel rivets in the past, and
they just spin around on you.)

*** If you're very careful (and quite steady), you can use a Dremel to grind
the blossomed part on the rivet until it's thin enough to reach under the
edges with a knife point, bend the pieces up, then snip them off with
jeweller's dikes. I've used this method on a couple of different things, one
being an old VCR where the internal circuit boards were riveted to the
chassis. I put double-sided tape arond the area I was working, then used an
aerosol can (like for blowing out keyboards) to clean it all off after
removing the tape that caught most of the dust.

 

[Paul]

*** If you're very careful (and quite steady), you can use a Dremel to
grind the blossomed part on the rivet until it's thin enough to reach
under the edges with a knife point, bend the pieces up, then snip them
off with jeweller's dikes. I've used this method on a couple of
different things, one being an old VCR where the internal circuit boards
were riveted to the chassis. I put double-sided tape arond the area I
was working, then used an aerosol can (like for blowing out keyboards)
to clean it all off after removing the tape that caught most of the dust.

Thanks for the tip. I guess I just give up too easy
It's from all that experience busting stuff as a kid.

Paul