After a short warm-up, the set still continues to perform as described. It is now often enough so that I was able to look at the waveform of the signal which drives the HOT. During the cycle to the OFF period, the waveform callapses and the HOT during this period of time is not being cut OFF. The waveform at the input of the driving transformer also collapses, but this may not be surprising since one of the B+ volltages is derived from the horizontal output transformer. In fact, any waveform with horizontal periodicity at many of the test points collapses. There is a rather high B+ voltage of over 160 volts which doesn't seem to be effected at all. As I said earlier, I do not have the schematic diagram so I don't know what circuit is supplied with what B+ source.
Perhaps as someone suggested it may be the horizontal output transformer, but if that is the problem, I cannot understand why spraying the HOT with a component coolent would seem to get rid of the problem at least temporarily.
I've gone over the circuit board with a jeweler's loupe in order to see if there are any suspicious solder joints. Yes, I know that is not 100% effective. Anything that looked even the sightest bit suspicious, I resoldered it.
As Sam Goldwasser states a problem with blowing HOTs may require substitution of components in order to solve it. I'm not sure how much more I am willing to invest in view of the age ot the set.
Thanks to all. Dan