Connect with us

Keypad buttons not responding

Discussion in 'Troubleshooting and Repair' started by johnc83, Jan 23, 2015.

Scroll to continue with content
  1. johnc83

    johnc83

    5
    1
    Jan 23, 2015
    Hi everyone, absolutely newbie here with electronics but keen to learn. Ive attached a picture of a keypad which controls an old Windows PC.

    The A1 button stopped working so we had a company solder a new one in. When it came back it still didn't work - we always have the option of sending it back but I would like to tinker with it myself so I could maybe repair things like this in future.

    During my investigation, I've found that all the buttons with red writing do not work (as in they don't beep when pressed like all the others do). I've tested the buttons (with a multimeter to see if, when pressed, they allow signal through) and they seem to work. I'm not sure if they are daisy chained so if the top one doesn't work then the ones below won't.

    Would someone be able to give me a nudge in the right direction please?

    Thanks
    John

    p.s my mistake for missing an 'A6' button
     

    Attached Files:

  2. Gryd3

    Gryd3

    4,098
    875
    Jun 25, 2014
    They are typically wired in rows and columns.
    Each row is activated one at a time, and the signal is sent back to the appropriate key's column. (or vise-versa)
    If the pin or controller responsible to reading a column, or energizing a column fails... then you end up with a row of keys that no longer work.
    The picture you linked... is that the only part you sent in?
    Is there more to the circuit?
     
  3. johnc83

    johnc83

    5
    1
    Jan 23, 2015
    Hi Gryd3, many thanks for your reply. Your explanation has helped me locate the problem. This circuit board connects to another board (with LCD panel) via a 15 pin ribbon connector. I identified which pins routed to which buttons and helped me work out that pin 1 (farthest left) was supposed to control column 1 of buttons. If you look at the attached picture and the yellow line, this is a copper wire that I ran from pin 1 over to that button. Once this link was put in place then all of the buttons in column 1 worked perfectly!

    So, my plan now is to solder a wire exactly like that yellow line. Do you think that would be the best next step?

    Thanks
    John
     

    Attached Files:

  4. Gryd3

    Gryd3

    4,098
    875
    Jun 25, 2014
    Glad you got that tracked down!
    Soldering a wire in place to bypass the 'open' track on the circuit board would be a good repair. Make sure you can fasten the wire to the board as to avoid it from getting dislodged or caught in the housing when you put it back together.
    Additionally, you could try to resolder the leg on the A2 button that connects to the trace on the board. (This is less likely to work if the actual trace has been damaged, you may be able to see a small hairline fracture.)
     
  5. johnc83

    johnc83

    5
    1
    Jan 23, 2015
    Thanks, I'll make sure to fasten the wire. I've made a note of what you said about the trace being fractured and re-soldering the leg on A2 and I'll look into that. Is there any was of being able to test if a trace is working (there doesn't look to be any hairline frature but I wondered if there was a way to narrow down where along the line it could be, if there was one?

    Thanks
    John
     
  6. Gryd3

    Gryd3

    4,098
    875
    Jun 25, 2014
    Grab a multi-meter and set it to measure continuity (Ω).
    Leave one probe on the incoming wire from the ribbon cable, then test the leg of each button that is daisy chained together. The meter should show 0 (or close to) for every proper connection, then will show a much higher (infinate or overload) once you cross the broken portion.

    Not that those buttons on the board (you may already know) have 4 legs, they are used in pairs, when the button is pressed, the first and second pair will connect to each other.
    Because of this, each button can be used as a small 'jumper' to hop over traces on the board. Make sure that when you do the above mentioned test, that the test the leg on each side of the button. The problem could be with a trace, or with a button.

    (Note that the wire you soldered is a replacement for only a trace... It's more difficult to find 'where' along the trace it is broken. If you are determined to find the break... you can very carefully scratch a little bit of the green mask off to expose the copper trace, then test the exposed portion compared to the ribbon cable. If you get a low reading (or 0) then the break is further away, otherwise the break is close to the ribbon cable. If scratching the mask off does not appeal to you, you can also use a small pin or needle... something to pierce the green mask applied tot he board to make contact with the copper)
     
  7. KJ6EAD

    KJ6EAD

    1,114
    157
    Aug 13, 2011
    I'll offer a guess, and some of this may be what Gryd3 was saying.

    There's a cracked solder joint at one of the switch legs and two of the legs on each switch are being used as a jumper to provide continuity down the row to the next switch on the matrix. This construction method provides lots of potential failure points but allows the column traces to run under the switches between the row legs of the switch and makes it possible to build the keyboard cheaply from a single layer board.
     
    Last edited: Jan 28, 2015
    Gryd3 likes this.
  8. johnc83

    johnc83

    5
    1
    Jan 23, 2015
    thanks everyone for their input. I've now soldered the wire in and its all working great.

    I'm sure I'll come back to this to read the notes on other projects.

    John
     
    Gryd3 likes this.
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-