|
I brought your comment from the YouTube page over here, since John Staples has a reply which I can also post, which might possibly generate some further discussion. Probably easier to do that here than on YouTube comments.
Your comment regarding the Westinghouse color TV use of a non-crystal controlled oscillator to track the chroma bursts:
"Regarding the Westinghouse crystal-less color oscillator: The main reason for using a color crystal VCO and not an LC is that the crystal can be guaranteed not to drift so far (actually much less than one-half the horizontal rate), so that side lock to the burst is impossible, even in a home set without a crystal oven to control the temperature. An LC oscillator could be tuned off frequency by a multiple of horizontal, in which case the hue would go through a stable rainbow pattern (or multiple rainbows) from left to right. So, the Westinghouse circuit required an unusually stable LC oscillator that would drift very little even when in the hot chassis environment."
John's response (via email to me earlier today):
"Thanks for the first comment which add some more information about how the different bandwidths of the I and Q signals are handled.
"The second comment is very interesting as it may shed some light on what the APC circuit is. The point is that the Hartley oscillator may lock on to a frequency that differs by the horizontal frequency or something else and would give a rainbow.
"Here is the schematic of that circuit which I should have included in my talk, although it is almost unreadable.
"Note the usual upper 6AL5 phase detector in the PLL, but also note the lower phase detector excited at 90 degrees which feeds the color killer, which I spoke about, but also an additional low-frequency coupled loop through the 12AX7 back to the reactance tube. This may be a way to detect that the Hartley oscillator is locked by some multiple of the horizontal frequency off, and a beat note may kick the oscillator back on frequency. Just a speculation.
"What do you think?"
Last edited by Tom Albrecht; 05-31-2024 at 11:22 PM.
|