I'm trying to repair weak color on a Motorola TS-918 chassis with weak chroma. The chroma signals to the separate grids of the 6LE8 demodulator are good and can be varied beyond the level given in the schematic, so the color IF section is most likely OK and not the source of the problem. The 3.58MHz at the cathode and G1 is a different story, however. The schematic shows roughly 10V P-P of mostly continuous 3.58MHz signal at both electrodes, but I have what looks like a short burst followed by a few volts P-P of 3.58MHz. The output signals are very low and consist mostly of 3.58MHz "grass" on the scope. The DC voltages are all within reason. The 6LE8 was the only tube in the color section I did not change out; I'll have to wait until next week before I can order new ones. I had a book years ago describing how these circuits work, but no more and I don't remember anything. I don't see how only a cathode and grid make an oscillator here. Perhaps this set is a ringing crystal setup similar to what GE used on a lot of their designs and I just have a dud crystal, and just enough 3.58MHz is being created by the tuned circuits to create the color signals. The crystal seems to be in series between the burst amp and demodulator.

Not to state the obvious, but have you checked the tube? I would rule this out before touching anything else - new crystal will probably mean redoing the alignment, which is absolutely no fun.

Unfortunately that tube type is a bit too new for my tester, so the only way to test is by substitution and I don't have another tube right now. Looking at the schematic again I am convinced this is a ringing crystal type circuit that isn't ringing like it's supposed to. I will check components in that area until I can get another tube for a test. Realignment if needed will not be a problem since I have the service manual and the needed test equipment.

I've seen a weak burst oscillator kill color in other TVs, I think subbing the tube before doing anything else makes a whole lot of sense.

I checked the crystal, it's OK. Everything points to the tube itself, so I will order one as soon as I can. I blew it on my previous tube order and only got the color IF tubes, not thinking this one could be the problem. My question still is, does this tube really act as the oscillator? I'm just trying to understand how the circuit works. I know how it does the demodulaton, just not how it oscillates with only two tube elements used as an oscillator.

http://www.videokarma.org/showthread.php?t=93056&highlight=sodpil

Although the tube can be used as a ringing circuit, in this case it is an injection locked electron-coupled oscillator.

Motorola called it SODPIL - self-oscillating phase injection locked.
Edit: left out a word: self oscillating detecting phase injection lock

Problem could be either crystal or tube, I guess. I would wait to try the new tube.

http://www.vias.org/basicradio/basic_radio_30_01.html

I think I understand it now. The 6LE8 is an oscillator that is synced by the color burst every line. I will replace the tube and go from there. Should have ordered this one before.

Unfortunately in that old thread the scans are too low of a resolution to even be legible.

It's also a high level demodulator, saving parts count by not needing the demod stages.

I think I understand it now. The 6LE8 is an oscillator that is synced by the color burst every line. I will replace the tube and go from there. Should have ordered this one before.

Unfortunately in that old thread the scans are too low of a resolution to even be legible.

With the normal injection level, it doesn't sync exactly every line, but pulls in over several lines much like a phase-locked loop, even though it isn't one.

When I click on the scans, I get a photobucket version with a magnifier that makes it readable - not sure what you're seeing.

I had to view the scans in desktop mode instead of mobile. I will try the adjustment procedure later today to see if it can be made to work, or if it just can't be brought into operation due to tube problems. In any case I'll still replace the tube since I already replaced the others in the color section anyway.

Going through the alignment procedure for the demodulator/oscillator brought the color back. I'm still going to replace the tube to get the best performance, which I understand isn't very good compared to other sets using straight ahead demodulator circuits. The color rendition I see now is what I remember from the last Motorola set I had with this circuit (and a weak 23EGP22), deep colors sometimes appear washed out. But now I can oversaturate the color with the intensity control and back it off to normal.