The whole thing is quite complicated. The vertical deflection in the original design is driven by the voltage appearing on one side of the mains transformer’s HT winding. The HT is bridge rectified so, as you can imagine, that deflection waveform is at 50Hz and very far from sinusoidal (and to get a true linear ‘oscilloscope’ deflection it would need to be something like sawtooth of course, corrected for the nonlinear properties of the electrostatic deflection assembly). So much for the vertical deflection.
The two horizontal deflection plates are connected essentially to the signal feed to the speakers - one plate to the left speaker feed and one to the right. So in stereo operation the horizontal deflection represents the difference between the left and right signals. It is possible to mute either the left or right channel in which case the trace is limited to just the left or right half of the display and then shows the full music signal on that channel, albeit still swept up and down in an unhelpfully non-sawtooth way.
All that said, the gain control for the horizontal deflection (it has its own little amplifier) is calibrated in watts and there’s a procedure described in the manual for measuring audio power by adjusting the deflection gain knob until the trace extends sideways a fixed distance over the tube (they talk about marking it up with wax crayons !). You’d have to be pretty determined to get anything useful out of this though, and you’d also have to have nothing better to do whatsoever !
EDIT: I should say I filtered the vertical deflection signal somewhat to remove the worst of its distortion away from 50Hz-sine, but the trace is still pretty crowded when it’s at the bottom of the screen.
