Sorry I don't have any pictures, I did these repairs at work and didn't have a camera. No capacitors involved!
Most inexpensive keyboards today use an internal construction like the one shown in the first picture here:
http://en.wikipedia.org/wiki/Keyboard_technology
The 3rd layer down is the internal wiring from the keys to the logic board. The logic board is often (but not always) located in the upper right corner of the keyboard, above the keypad. (Generally where the numeric lock, caps lock, and other LEDs are seen.)
We had a batch of Keytronic KT800USBBUS-C keyboards, which were failing one by one, in every case losing the space bar and certain other keys. So I opened one up to see what was going on. It turned out that the connection from the internal wiring layer to the logic board consisted of two sets of exposed contacts on plastic (mylar?) sheets. These were held together solely by the pressure of a foam rubber "spring", which pushed on one side on the back of the case, then on the internal wiring layer, the extension from the logic board, and finally the front side of the case. After seeing this, I found that all of these keyboards would work correctly if one pressed down on the top of the keyboard above this connection.
The first attempt at a fix was to cut a thin piece of cardboard (the backing from a yellow lined paper pad) and fasten it to the back of the keyboard above the spring. This was done by roliing up some scotch tape into tubes, sticky side out, and taping the cardboard in place. This repair worked for a while (months) but eventually failed.
What else to try? Well, there were hollow cylindrical standoffs coming up from the front of the keyboard on both sides of the spring. These appeared to be screw mounts, but were unused. It looked like some earlier incarnation of the design had used a physical clamp over the spring. This had been eliminated, leaving the two closest screws to the clamp a couple of inches away in either direction. Over time the case plastic would loosen up, the spring wouldn't have enough oomph to keep the connectors together, and some keys would stop working. Of course, it worked long enough to test good at the factory, and in most instances, long enough to outlast the warranty.
To repair this problem on a test keyboard two screws were scavenged from another failed keyboard. A lot of careful measurements were made on the positions of the two unused posts with respect to the used screw posts, and then this was mapped onto the back of the case, in this case using the corresponding screw holes, to determine where the two new holes were to be drilled. Then the new holes were drilled (with a drill press for the first one, later I just used a hand drill). Reassembled the keyboard, screwed in the two borrowed screws, and voila, the keyboard worked again.
A small cardboard template was made from the first repaired keyboard. It indicated the new screw hole positions with respect to other features in the molding. For the subsequent keyboards the keyboard was disassembled, the template used with a center punch to mark the two new holes, the holes drilled, the cardboard spacer added (probably not necessary with the clamps, but it does up the spring pressure slightly), and the keyboard reassembled with screws in the two new holes. In every instance this let the keyboards function again.
The extra screws were 4-24 pan head hi-lo screws from McMaster-Carr
http://www.mcmaster.com/#97975a110
These were very similar to the other screws in the keyboard, except with a different finish and driver (Philips instead of T8 Torx). I only needed 18 of them (2 * 8 keyboards plus 2 missing screws) but the smallest number that could be ordered was 100. So I'm set to repair a large number of crappy keyboards should they ever come my way.
The materials cost was about $10 for the screws, which was a lot less than 8 new keyboards would have run. I think this fix should last a reasonably long time because the front and back parts of the case are now firmly held together right at the edges of the rubber spring. Most likely several years from now the rubber in the spring will get old and not push hard enough, but I will cross that bridge when I come to it.
Most inexpensive keyboards today use an internal construction like the one shown in the first picture here:
http://en.wikipedia.org/wiki/Keyboard_technology
The 3rd layer down is the internal wiring from the keys to the logic board. The logic board is often (but not always) located in the upper right corner of the keyboard, above the keypad. (Generally where the numeric lock, caps lock, and other LEDs are seen.)
We had a batch of Keytronic KT800USBBUS-C keyboards, which were failing one by one, in every case losing the space bar and certain other keys. So I opened one up to see what was going on. It turned out that the connection from the internal wiring layer to the logic board consisted of two sets of exposed contacts on plastic (mylar?) sheets. These were held together solely by the pressure of a foam rubber "spring", which pushed on one side on the back of the case, then on the internal wiring layer, the extension from the logic board, and finally the front side of the case. After seeing this, I found that all of these keyboards would work correctly if one pressed down on the top of the keyboard above this connection.
The first attempt at a fix was to cut a thin piece of cardboard (the backing from a yellow lined paper pad) and fasten it to the back of the keyboard above the spring. This was done by roliing up some scotch tape into tubes, sticky side out, and taping the cardboard in place. This repair worked for a while (months) but eventually failed.
What else to try? Well, there were hollow cylindrical standoffs coming up from the front of the keyboard on both sides of the spring. These appeared to be screw mounts, but were unused. It looked like some earlier incarnation of the design had used a physical clamp over the spring. This had been eliminated, leaving the two closest screws to the clamp a couple of inches away in either direction. Over time the case plastic would loosen up, the spring wouldn't have enough oomph to keep the connectors together, and some keys would stop working. Of course, it worked long enough to test good at the factory, and in most instances, long enough to outlast the warranty.
To repair this problem on a test keyboard two screws were scavenged from another failed keyboard. A lot of careful measurements were made on the positions of the two unused posts with respect to the used screw posts, and then this was mapped onto the back of the case, in this case using the corresponding screw holes, to determine where the two new holes were to be drilled. Then the new holes were drilled (with a drill press for the first one, later I just used a hand drill). Reassembled the keyboard, screwed in the two borrowed screws, and voila, the keyboard worked again.
A small cardboard template was made from the first repaired keyboard. It indicated the new screw hole positions with respect to other features in the molding. For the subsequent keyboards the keyboard was disassembled, the template used with a center punch to mark the two new holes, the holes drilled, the cardboard spacer added (probably not necessary with the clamps, but it does up the spring pressure slightly), and the keyboard reassembled with screws in the two new holes. In every instance this let the keyboards function again.
The extra screws were 4-24 pan head hi-lo screws from McMaster-Carr
http://www.mcmaster.com/#97975a110
These were very similar to the other screws in the keyboard, except with a different finish and driver (Philips instead of T8 Torx). I only needed 18 of them (2 * 8 keyboards plus 2 missing screws) but the smallest number that could be ordered was 100. So I'm set to repair a large number of crappy keyboards should they ever come my way.
The materials cost was about $10 for the screws, which was a lot less than 8 new keyboards would have run. I think this fix should last a reasonably long time because the front and back parts of the case are now firmly held together right at the edges of the rubber spring. Most likely several years from now the rubber in the spring will get old and not push hard enough, but I will cross that bridge when I come to it.
Comment