Saba Villingen 9

Greetings:
I have Villingen 9 that has a problem with low B+ voltage and loss of sound output under a particular condition. B+ at turn on is about 167Vdc on the plate of ECL 82 (pin 6). It should be about 200Vdc. I have replaced the main filter capacitors but left the Power supply rectifier in place. There is just a few volts drop across it so it doesn't appear to be a problem. When I turn on the radio it works and if I don't turn the Bass potentiometer (P3) it will continue to work on MW and UK. The bass pot (P3) is different from what I have seen. It appears to be divided into two sections of 3 Megohm on the schematic. I measure about 10 Megohm on each section. When I turn P3 to get more bass the volume decreases to nothing after a few seconds when it passes from one section of P3 to the next. I also notice that B+ on the plate of (pin 6) of ECL82 goes down to about 100vdc. I have checked resistors around the audio section and they seem fine. I replaced all the paper capacitors in the radio. Have the P3 type poteniometers been an issue on these radios? Here is the schematic Any advice will be appreciated.

One thing I can add. I removed the ECL 82 tube and measured the voltage at the plate pin (6). The voltage was around 210VDC which is about right. I wonder if the ECL 82 is bad? Perhaps a thermal problem .

Just a follow up. I found the problem. One of the capacitors I replaced (C33) is the type with 4 leads... two for the signal and two for grounding for stray capacitance. I usually have no problem replacing these capacitors with a regular axial capacitor. The schematic doesn't show it but this cap also provides the ground connection for R37 which is part of the bias for the grid of ECL82. The bias for the grid was way off. I'm posting this so that others who see these type of caps should check to see if they are also providing a Ground path for other components.