Sir Brian. . . . . .
Ab-so -rootley, on the possibility of C38 leaking and pulling down the Sub B+ level a bit.
What you might try is to monitor the Sub B+ at the origin of its tap off at voltage divider bridge of R14-R17.
Then you unplug 6K8 and 6K7 tubes to see what the sub B+ rises up to without loading . . . Then plug in one tube at a time and recheck the voltage level. One of those individual tubes will have loaded and pulled the voltage down a wee bit more than the other.
I am thinking that it will be ~ 12 ma consumption for the 6K8 and ~11 ma for the 6K7. Seems like C38 is over to the side of the chassis and down under a lip.
That C38 would have to be seriously leaky down in the k's of resistance to be effecting us on that degree of a voltage drop.
The potential other thing aparrent on that supply line, is leakage within the the C41---5 ufd filter , which we hope that you have already replaced with a good unit.
You did not make any reference to the 1k power res sub for a speaker field coil, so I guess that situation was on somebody elses set. And, you also still have the original speaker in your unit, and no one has subbed in a speaker with a lower value of field coil reisistance / inductance.
BTW have you ever heard stations picked on this radio ? That initially being even before you started changing parts, considering that you made an initial g r a d u a l power up, thru a Variac, to see if there were initially any power electrolytic filter issues in the unit.
2 questions included.
Standing by . . .
73's de Edd