I set up the scope, and it looks like the red gun is indeed failing. It has
the highest drive signal, and the lowest current. When I increase G2 a tiny
bit, the picture stabilizes like you suggest. When I watch it on the scope,
I see that increasing G2 decreases the current draw of B and G while leaving
R unchanged. So you are exactly right on.

It seems to me that there are two possible ways to cheat the system on this
one. One is that I could add a high value resistor (Maybe 1 to 5 megohms)
between the 200V supply and the output of the red video drive chip
(Conveniently the pins in question, 6 and 8, are about the distance of a
surface mount resistor apart). This would flow some extra current into the
output and might make it happy. The second way to cheat would be lower the
value of the series resistor between the IK output pin on the red video
driver and the IKB buffer. If either of these techniques worked, I might be
able to adjust the color back in with normal controls, and not get the
"washout" of using the G2 adjust.

Worth a try??

Thanks

Dave Miller
wrote in message
oups.com...
Not exactly. Sony uses an AKB (auto-kine-bias) which during the
vertical blanking interval sends out a red line, green line, and a blue
line. The AKB circuit senses the current and as long as all three are
close and within specifications for the circuit, the video is unblanked
when the tube is warmed up. By blanking the video during the warm up
time it prevents the 'stripping' of the cathodes that happens in vacuum
tubes when full cathode voltage is applied before the tubes heat up all
the way. This circuit also fine adjusts the tube bias for each color
to keep the color temperature close to correct over the life (useful
life) of the tube.

The problem you have is either one of two:
1. One of the guns is significantly weaker than the other two, making
the AKB circuit not see the tube within specifications.
2. All three guns are weak causing the blanking.

If you have access to a good o'scope, you can sync the scope off the
vertical blanking pulse and monitor the AKB line with the other
channel. With the tv totally cold, turn it on and watch the three akb
return pulses come up. You will likely see one of the guns come up
much slower than the other two.

BTW the reason the video blinks is that with no video on screen, the
cathode current makes it just inside the window to unblank the video.
As soon as the video comes on, the sudden drop in cathode current
during the sense scan line drops below specification and the video
again blanks out. This is why a slight increase of G2 voltage will
compensate for this for a time as it tends to average out the current
draw between blanked video and un-blanked video, but it also washes out
the picture some.