deskthority

Hello,
I recently received my solenoid and key caps for my F77 from Elipse. I ran into some rather peculiar issues with it though, and I'm curious as to whether anyone has any idea what might be up.

Originally, I'd installed the solenoid per the guide. However, I found that the screw protruded too far out the bottom with the rubber isolators. This meant that the keyboard was sitting on the screw rather than on the cork feet. Thinking this was for mechanical isolation, I opted to forego it. However, I'm now thinking this also served as electrical isolation instead.

In this configuration, everything tests out alright, until I put the keyboard assembly in the case and screw it together. After this, the 56tyghb[space] row always stops working, with the 78uijkn row also sometimes not working. Otherwise it seems to work alright. I figured this was a bad solder joint or a ribbon cable failure for the key matrix, since I'd moved it to install the solenoid driver. (Can we acknowledge that the ribbon cable for the solenoid driver doesn't actually fit properly? It's too big!)

Dreading having to take the keyboard assembly itself apart to replace the ribbon cable or solder joints on that side (I restored an XT Model F before, I never want to do that again! Elipse actually sold me a part for that restoration.) I decided to look further. What I found was... weird.

Firstly, I saw a small spark between the chassis and the key assembly when I put them together, which immediately told me there was some very real issue. I then found that unplugging the solenoid from the solenoid driver fixed the issue. If I touch the assembly to the solenoid, but not the case, the issue appears, and if I reverse the polarity on the solenoid and touch the keyboard assembly to it, the solenoid fires. So clearly there's some weird electrical path happening that shouldn't be. The fact that it's only effecting one or two columns is bizarre, but just coincidental I guess.

I tested continuity and resistance between both solenoid wires and the solenoid chassis in case one was shorting to the chassis. The coil is perfectly isolated though from the solenoid chassis! I next checked the pins on the solenoid driver, in case I don't have sufficient isolation with the VHB tape I'm using to attach it to the case. That is also not making contact.

I next isolated the solenoid from the keyboard case and connected the chassis of the solenoid to an oscilloscope referenced to the solenoid driver ground, which has continuity to the case ground via the solenoid driver ribbon cable. I figured maybe the solenoid had some waveform inducing a voltage onto the chassis. There is no waveform though, only a constant 9V DC voltage, which I confirmed with my multimeter as well. Given this is a 9V solenoid, that number sticks out like a sore thumb.

So the conclusion I can make is that the solenoid is dumping 9V onto the chassis. Since I have the solenoid chassis bolted to the keyboard ground, it's causing some fault in the controller. Fine, that's all well and good. But if there is absolutely no continuity (completely open circuit) between the coil and the chassis on the solenoid, and the ground for the solenoid driver is also referenced to the case via the controller board, how is the solenoid chassis getting 9V across it relative to everything else?

In the meantime, I'm going to go hit up Home Depot. I have a couple ideas to fix this. I've used these rubber feet on another project (A replica SCELBI-8B mini computer) and they'd both give the Model F an angle like most keyboards and allow me to use existing screw. (It'd do double duty holding the solenoid and holding the taller foot on) Or I might just find a flatter screw that won't protrude so much.

But yeah, this is really confusing me, and even if I have a solution, I'd be very interested if anyone knows what is going on!

I believe I have solved the issue! So I wasn't being totally accurate when I said I had separated the solenoid from the case. In reality, I had separated the top and bottom cases, with the solenoid being the only thing on the bottom case. This becomes important.

As it turns out, the supplies screw, without the mechanical isolation, is too long. It ends up going through the mounting bracket, wearing through the protective plastic over the coil, and shorting on the coil. This ends up making the chassis of the solenoid sit at 9V DC.

I've found that a washer or two (depending on thickness) combined with the additional offset that the rubber feet I'd linked to in the original post effectively shorten the screw to the right length to mount the solenoid securely to the case without shorting to the coil. I also found that the case is barely big enough to fit the solenoid as it is, so any isolation underneath just makes it so you can't close the case properly. Therefore, I recommend the washer(s) go on the outside.

I've typed this entire comment on the F77 with the solenoid going with zero issues! (: