No, the radio is not accurate enough to give more than a ballpark idea of frequency. Plus, when I had the breadboard circuit hanging off the chassis and two oscillators (or parts of them) running in near proximity, the radio was more confusing than useful.
Today I found a cheap frequency counter on craigslist, and it showed that you were right once again. The oscillator was running at the wrong frequency, somewhat high. This is as low as I could adjust the frequency using the reactance coil:
The ideal frequency is 3.579545, not 3.579629. I tried clipping various caps in parallel with the 270-pf cap (C213). Adding a 100-pf cap worked the best:
The system locked on the lower frequency and I got color bars again:
I'm back where I was a while ago, in other words. But at least I got out of the rolling-bars rut! This evening I'll have time to tack the new cap in place and do some more checking.
Thanks,
Phil Nelson