I have noticed this and I keep it unplugged when not in use, but I wondered WHY it was designed this way? Is there a good reason?
I don't have an answer to that question, but at the very least it is a UI/UX issue: one would expect that a button in the "Off" position would turn off power entirely, while here it only turns off the motor's power.This said, I have success to report: I managed to get the magnet energized!I resolved to transplant the Q1 transistor from one SMD board to the other. This did not appear to change anything, and I figured that maybe neither of my boards were good.I tried adjusting the "rheostat", to no avail. Output voltage varied between about 2 V and 3 V (but see below).Then I decided to measure the resistance of the magnet, and I measured 6 Ohms. Finally it clicked: I had measured about 3 V on the output of Q2, which I had assumed was incorrect (IBM machines have solenoids which work at 24 V or even 48 V). Good old Ohm's law says that the current should be 0.5 A, which is exactly what the schematic says. So, with the swapped transistor, my driver was working, but still the magnet was not energized!It turned out that the issue was the way the magnet was adjusted: there are two screws which allow positioning the magnet assembly. I hadn't put them back perfectly, so the magnet was actually energized, but not well enough that it could pull the metal plate. One simple adjustment later, and it worked!After swapping back my transistors, the previously working SMD board stopped working. So my original Q1 transistor is definitely toast.All in all I had 3 issues so far:- dead electrolytic capacitor- dead Q1 transistor- misadjusted magnetBut I can now move on to the next set of issues ;)I'd still be interested in finding a replacement Q1 transistor, so I can have a second working SMD board for my other Model 33 (eventually).If anybody is interested, I have pictures of the work in progress here:Eventually I'll write a blog post or two about it as well.
-Erik