I've just invented The Oscillating Oscilloscope! Most analog oscilloscopes have a nice sawtooth oscillator for the horisontal sweep and it's an interesting and surprising thing to play with. The trigger circuit will make the sweep generator lock to the signal to the Y-input, but it will not be the same frequency. It can be chaotic and sound terrible, but depending on levels, settings and ratio between sweep and input frequency, there may also be wounderful subharmonics with more or less distinct separation. It may sound quite musical, with a nice chaotit touch of dirt.
Then I realised that the vertical amplifier must be possible to turn into an oscillator! The good old wien-bridge circuit seems to work best, and it's easy to control over a wide frequency range with a stereo potentiometer or any other pair of tracking resistors.
A vacuum tube oscilloscope works with hundreds of volts just about everywhere, and thousands of volts closer to the cathode ray tube. Even a transistor-based oscilloscope have the same high voltages around the CRT, so knowing what you're doing is a good idea to avoid potential lethal electric shock! The voltage to the vertical deflection (the Y-plates of the CRT) is in the order of 20-100 volts. The final amplifier stage is a balanced differential amplifier but it doesn't seem to be bothered about driving the single-end load of the feed-back over a wien-bridge. Even if the input of a wien-bridge is capacitive, it may be a good idea to incorporate a larger DC-separating high voltage capacitor close to the vertical amplifier's output tube. The DC level at this point may be anything between 50 and 200 volts, so be careful! A transistorised oscilloscope may have a lower DC level, but since the CRT's are basically the same in all analog oscilloscopes, they still need quite a lot of AC voltage on their deflection plates. The oscilloscope in this video have up to 80vpp on each of the output tube's.
To make the experiment even crazier, I added two LDR's in parallel with the potentiometer in the wien-bridge, to see if the oscillator frequency can be controlled by a simple homemade double vactrol. -And yes, that works great! The LDR's are so sensitive that the CRT trace itself is bright enough to change the frequency, and that in turn changes the frequency and phase of the sweep generator, resulting in something of a "trace-o-phobic" or autophobic oscilloscope that prefer to avoid it's own trace. :-)
This far I have only done experiments to the vintage oscilloscope seen in the video. Most of the oscilloscopes in my collection are based on vacuum tubes and are quite simple single-channel oscilloscopes. With a dual trace oscilloscope it may be really interesting to see if two different wien-bridge circuits can be driven from the vertical amplifier. I guess there's a fair chance that the oscilloscope will play two alternating tones while set to alternating mode with a long timebase on the sweep generator.
Then we have off course the possibility to set a dual channel oscilloscope in XY-mode for two completely different wien-bridge oscillators, and seing lovely lissajous figures on the screen. For musical purposes this may be a good help to find proper musical intervals even on tones in the low frequency region. It's most likely also possible to cascade two or more oscilloscopes, adding ONE extra tone for each extra oscilloscope, while feeding the X-input with the audio from a previous oscilloscope.
I don't expect it to be easy to turn an old oscilloscope into a perfectly CV-tracking synthesizer, but it's worth a try! Oscilloscopes may run quite warm, particulary all-tube oscilloscopes, and heat causes oscillators to drift around. Since the wien-bridge can operate on fairly low signal levels (depending on the gain and modes in the vertical amplifier input) it will at least be safe to experiment with, once you've put that DC-separating capacitor rated to more than say 150 volts between the CRT deflection amplifier's output and the experimental circuits.
