Nice idea with the screw, Cubdriver.
I ended up not using the metal cans, but rather just the cardboard tubes. I used hot snot in the end. It worked fairly well. There was one tube in the corner I was not totally pleased with, but it looks pretty darned good. Once you put the mesh case on you would never see the 1/8 inch of glue sticking out from that one. It sounds a little over the top, but I strive to either do things in the best way possible, or the cheapest/fastest way possible. I hate it when things are not perfect if I have opted for the former. I expect thing will not be perfect with the latter.
I calibrated this unit up, really well, actually. At first I had a DC offset issue with the triangle wave output. It was much higher than the other waves, way up above the centerline. I started suspecting a resistor attached to the triangle wave DC offset pot. So I de-soldered one leg. It measured fine so I soldered it back on. Then it worked! I was in the right place, but looking at the wrong thing. I guess it was a bad solder joint.
So I fixed that.
I guess duty cycle meters and frequency counters weren't too popular in 1962. The calibration procedure uses the scope graticules (remember those scopes had no cursors) for timings.
The calibration procedure does have a very interesting method for setting the duty cycle of the square wave. It really is worth taking a look at (and trying with a circuit or function generator with an adjustable duty cycle). It has you toggle the trigger slope back and forth while making the adjustment. It was spot on according to the duty cycle meter after doing it with the scope method!
It then goes on to have you set the triangle duty cycle by giving you a voltage level to which to set the base of the triangle wave. This was very imprecise. After calibrating once I ended up going back and redoing most of the cal in order to set the duty cycle of the triangle wave correctly using a duty cycle meter.
So in the end, I ended up with a pretty darned accurate, low frequency function gen. It can go down to a very low frequency (one cycle every 1000 seconds I believe). It can only go up to 10kHz, but that is sometimes all you need. It can generate and output SIMULTANEOUS square, triangle, and sine waves, as well as the main output, which does those, but also has an attenuator and a DC offset control.
The thing can also be used to heat the room if it gets too cold in the winter! Those tubes are quite warm. I will post some pics soon.