PTCO Sol-20‎ > ‎

Sol-20 page 3 – Keytronics keyboard

None of the keys worked on the Sol-20. This is a common failure. 

The key mechanisms use a capacitive sensor. Each plunger pushes a metalized plastic
disk into the PCB board. The disk is held to the plunger by about 0.2" light, open-cell foam. The foam disk sits beneath the sides of the plunger assembly, well out of contact of the PCB. When a key is pressed, the foam pad is displaced beyond the bottom of the assembly and (should) contact the capacitive pads on the PCB. Unfortunately, the foam degrades over time and loses its compliance. The pads are simple, if time consuming, to replace. Some Sol owners choose to purchase an old Sun keyboard, which uses the same foam inserts. Making my own replacements felt much less wasteful rather than scrapping another keyboard (although the Sun keyboards would mostly likely be consigned to an e-waste pile, anyway).

First up was to remove the keyboard PCB. The screws holding the key assembly to the PCB were Torx T8. The PCB was labeled "KTC 65-01558-" and "PCB-002A" along with "Mechanism by Key tronic Corp." These labels are in the plated metal. Additional black silk screen labels indicate part numbers for each IC and component, "Copyright 1976 by Processor Technology Corp", and "Sol-KBD". 

Two of the foam inserts degraded enough that the metallized disk fell out immediately (the "S" and "U"). Others appeared to be fragile, but holding. In this version of the keyboard, the foam pad is held in place by a sticky membrane, similar to, but thinner than, double stick tape, which adheres to the bottom of the plunger. It is not mechanically attached by a bottom plastic disk like others describe. 

These are the replacement materials that I sourced to make new foam pads:
  • Frost King 0.25 inch thick, "maximum compression" weather stripping with adhesive backing. This is the lightest, most compliant foam that I could find. (I didn't search that hard. My dad found it in his basement gopher hole.)
  • Mylar safety blanket, making sure to use the non-conductive side facing the circuit board
  • Double stick tape
  • 7/16" punch (General Tools 1280L Hollow Steel Punch)
Mylar is a thin PET film, specially processed to produce extra toughness, with a layer of metal deposited on one surface. Others report using mylar balloons and wrapping paper. I taped the mylar to a board, then the foam strip, tape, and backing of the foam and punched through this using the punch and a rubber mallet. This produces consistent pieces that can be stored until later use. The adhesive layer of the weather stripping sticks to the mylar. The double stick tape anchors each pad into the key plunger mechanism. The process was a therapeutic evening of pounding out the 85 replacement pads, plus a few spares. Below is a photo of the half-way point and the finished products in the keyboard. After reassembling, the keys all worked. They are nice and snappy and with a great feel. The only change that might be worth making in the future is to use a slightly thinner foam, which should give the pads enough pressure to make "contact" but reduce the overall stiffness of the keypresses. (It isn't quite as buttery smooth as the Apple IIe keyboard, one of my favorites.) 

After a few days of using the keyboard, the "U" and "return" keys got a bit unresponsive. I disassembled the keyboard again. The foam pads looked fine and intact. Earlier, I had used an eraser on the capacitive pads and I noticed some eraser material still sticking to the pad traces. This could have interfered with the contacts  so I rubbed the traces down with a cloth. After reassembling, the keys functioned normally, and have so-far worked well.

One of the things I like about the Proc Tech components that I've worked on, especially the keyboard, is their small but thoughtful design choices. The keyboard cable, which attaches via the J1 socket on the Sol-20 PC board, is one example. The power and ground connectors are on both rows and are symmetric, so that accidentally reversing the cable won't cause damage to the components. The PC board power supply connector shares this feature as well.

 

More on the Sol-20 keyboard 
Over the past year, the Sol keyboard has had intermittent problems with unresponsive keys. I've had it apart a few times, but disassembling it makes me nervous, in particular the screws holding the keyboard PCB to the frame. These feel like they could strip out, and one or two have become quite loose. 
One thing I've noticed when keyboard issues crop up, though, is that removing the keyboard and putting it on a flat surface immediately fixes the unresponsive keys. I'm beginning to think that there is a subtle issue with cold solder joints on the PCB. Since the keyboard is held only at the ends, some sagging, especially as it is used, may affect PCB through these points or other issues with the traces. I haven't been able to confirm this, though. I guess I could summarize my suspicions as the origin being mechanical more than electronic or with the capacitive elements.