Wayne,
I took a shot with the scope triggered on horizontal rate, channel A is the h-drive wave form, channel B is video from the CRT. As you can see the horizontal blanking bar is right is the middle of the drive wave form, which is exactly what I'm seeing on the face of the tube. What I would have expected to see is the blanking bar triggering the end of sweep like it's supposed to, so the blanking bar should roughly line up with the negative going (retrace) portion of the drive wave form.
Since that's not happening, I can only conclude that for some reason the horizontal time base isn't being properly triggered by the horizontal timing pulse. The way things sit at the moment, retrace is being triggered right in the middle of active video and it shouldn't be. No matter what I do with the discriminator transformer it refuses to behave, I can adjust the phase slug to where the bar is nearly off screen but it's very unstable there and the picture gets wavy.
I really don't know what is triggering horizontal prematurely, I can't see anything else that would cause it to happen. Sync pulses coming from the sync section are well defined and have more than enough amplitude, so it's not a sync problem. I don't see any noise with no video present, so that's not it. Next thing I might try is injecting pure horizontal sync pulses to see what will happen, if it behaves like that it might give me a few ideas.
Sweep angle between the 2 CRT's aside, the only issue I was expecting was an overscanned raster horizontally. Vertical is fine, the height pot has enough range to make things work out.