Len, I can't believe you don't know this.
You fix the set with a piece of wire. Cut the traces going to the filament pins and wrap the wire through the flyback. Two turns does it alot. You need either a true RMS meter to read the 6.3 volts or on a scope on an NTSC set you are looking for 23 volts peak to peak.
Make sure you isolate those pins completely, that there are no compenents felt from the original filament supply. One might be grounded on quite a large foil, in this case some type of small grinder is useful, because you need to grind all the way around the solder pad in some cases.
You never know what you are going to get, you might not want to completely on the hot side because you might want to use the resistor. If emission is strong, you do not want to feed it any more than about
26 peak to peak, that is like already having a brightwener on it and will kill these new CRTs.
And when you wire this contraption up, don't be too neat. Don't strap the wires all up nice and neat, that causes capacitance and that is what will smear that color.
Keep them loose, or go the other route, tighten them up and apply equalization to that particular output stage.
To do that you must locate some resistor, most likely in the emitter circuit of the video output for the affected color. Then you need to find a small capacitor to put across that resistor. The resistor must be high enough in value to allow for the changed frequency response desired. What's more, put a 10K resistor between the cathode and the heater, that simulates the short, and will keep the short from shorting if it were intermittent.
Last time I did this it was on Panasonics. I had to because it seems their video outputs were more of a current drive than a voltage drive. They did niot brute force it like alot of manufacturers, they actually matched the impedance and equalization to the CRT. As such, when you start transmitting the red video from the wire wrapped around the flyback, it causes a capacitive load.
If you do not want to deal with putting in some pre eq on the video output, you can just reconverge the set, with the 10K resistor in place.
You DC supply loaded it down alot more because of mass as well as it's HF coupling to the power source. It will not be as bad with a piece of wire and even just reconverging it, you have a usable set.
I actually did not invent this process, but I damn well reinvented it. If you want to know more just RSVP.
JURB