My only guess is probably the original problem was just wear spots in the carbon touching the same spot all the time, and both domes got rotated to a new random position, making 2 new contact points on each carbon.
It could possibly be something deeper like one key has a high resistance trace making it hard to get a good reading from it, and switching the carbons made it just a little better so now it's just barely working good enough to register every time, and the other key had a good trace where it can afford to have a weak carbon and still register. bkw On Mon, Jul 7, 2025, 1:56 AM Thomas Morehouse <[email protected]> wrote: > Brian - thanks for the video. My 102 has a different type of key, but I'd > think the swapping of the rubber/stud would be the same. > > Any idea why swapping yours got both keys to work, and didn't just fix one > key but then "break" the other? I'm surprised that one key didn't work > with its original rubber/stud, but swapping two of those didn't cause the > other key to fail. > > Tom M. > > > On Sun, Jul 6, 2025 at 11:58 PM Brian White <[email protected]> wrote: > >> Forgot to say in that video the point was I fixed a partly dead T key by >> swapping the dead key with another key that sees much less use, and which >> you can tolerate being weak because you need it less. In the end not only >> the T key got better, but the doner key didn't get worse. Both keys work >> every time now. >> >> And it did not require taking the machine apart. You can pull the keycap >> off with a key puller and then use a 90 degree pick to get underneath the >> clips and release the top of the switch body, all from the outside without >> taking anything apart. >> >> bkw >> >> On Sun, Jul 6, 2025, 11:43 PM Brian White <[email protected]> wrote: >> >>> There are silver plated metal contacts with indeed a little stud, and a >>> carbon pad on a silicone dome. >>> >>> https://youtu.be/n_oyDYRDYzs >>> >>> bkw >>> >>> On Sun, Jul 6, 2025, 6:27 PM Andrew Ayers <[email protected]> wrote: >>> >>>> I've never taken a 102's keyboard apart, nor have I seen the "insides" >>>> like you describe them, but that little thing in the rubber cap you >>>> mention as a "metal stud"...likely isn't metal. >>>> >>>> Based on keypads and other similar kinds of "button devices" I've taken >>>> apart (remote controls, calculators, etc) - it's actually some kind of >>>> plastic or rubber compound impregnated with carbon. It does actually >>>> complete the circuit, as you stated. >>>> >>>> But I've also found when I've tried to bridge such a circuit with just >>>> a >>>> piece of metal...getting it to work can be "finicky" - assuming in your >>>> case it works at all, and the break isn't somewhere else. >>>> >>>> Cleaning of the contacts as suggested is a good step, though I don't >>>> know if you need to necessarily use an eraser; they tend to be >>>> abrasive, >>>> which in the case of cleaning contacts you'd think would be okay, but >>>> realize that the metal layer of those contacts is very thin. One such >>>> "cleaning" won't harm it, but do it enough times...hopefully that won't >>>> be needed or necessary, of course. >>>> >>>> You might try just using the alcohol (and a swab), first (I have a >>>> "thing" when cleaning things, especially if solvents and plastics are >>>> to >>>> be involved: start with the least harmful first, then move up the chain >>>> - that's usually either soap and water, or rubbing alcohol; end of the >>>> chain, of course, is a right angle grinder and/or "blue wrench" and/or >>>> BFH). >>>> >>>> Regardless, try also some rubbing alcohol on that little stud, as it >>>> can >>>> sometimes get dirt and/or other deposits that can prevent the circuit >>>> from being made. >>>> >>>> Good luck! >>>> >>>> Andrew L. Ayers >>>> Glendale, Arizona >>>> phoenixgarage.org >>>> github.com/andrew-ayers >>>> >>>
