Yesterday, I wrote:
To test your theory, I can probe the junction of CR1373 and CR1371 with a DC coupled scope with the beam finder. I would >expect to see the voltage clamped to +24 V during the negative swing of the cycle. During the positive swing, I would expect to >see it clamped to somewhere between +82V and +132V roughly, depending on the setting of the grid bias potentiometer.
I should be able to vary the grid bias pot and change the positive waveform peaks to anywhere between 82V and 132V, with no >effect on the lower peaks at +24V. Likewise, releasing the beam finder, I should be able to vary the least positive peaks (during >the unblanking period) by changing the intensity control without affecting the amplitude of the most positive peaks.
I ran the experiment this morning, and posted photos (475 Grid Bias). I found that the beam finder voltage does indeed get clamped to +24V during the negative swing of the cycle. However, during the positive swing, I was only able to control the clamping between +83V (OK) and +120V (too low). That second figure is about 12V lower than what I expected. It's interesting that 12V is also roughly how much higher I have to set the intensity to be visible (+35V) than what should be normal (+24V).
At first I thought this was because the output voltage from T1320 was too low, as I had mentioned in a previous post. Turns out this was a mental error, I was measuring the voltage on the other side of C1326. The voltage out of T1320 is a steady 300V p-p. However, just on the other side of C1326, it drops to something like 230V p-p, independent of the setting of the grid bias pot. That's a 70V drop across C1326, which at 50 kHz has a reactance of about 67K ohms, so about 1 mA though it. That doesn't make sense, however, because 1 mA through the 390K series resistor R1326 would be a voltage drop of almost 400V, and if that were happening, nothing would be working at all.
This leads me to suspect a bad C1326. Unfortunately, I don't have a replacement handy to try. So I'll have to get one on order.
I'm going to think about this some more and try to understand whether it explains the original observed problem with the beam finder. Any thoughts would be appreciated!