I could probably only see that as an old volume controls switch getting ready to fail . . . OR . . .if you had cleaned it, possibly a graphite residue could have ended up draining from the control portion into the adjunct switch contacts area and across them.
Cindy . . . oh Cindy . . . . . .
It was to then initially break down with the AC voltage and then after an initial arc path had developed, some serious conduction after the molded bakelite also then got involved and fully carbonized.
You then , as I am reading it, have got a different new control and switch in there, and the set is still remaining "garbled" .
Any chance that this "garble" is time related, like it is not being apparent when the set is stone cold and only after a warm up period.
If the latter, I would place the tuning dial off station and measure the DC plate voltages of the 2 AF tubes and log them down with the set stone cold.
Then tune in the set to a station and await for the trouble to onset, then place off station and check on how those voltages have changed when the sound is bad.
Leaning towards a leaky AF coupling cap. . . but probably not since you have changed them already . . . or possibly a def AF output tube , failing after hot.
73's de Edd