[Musaeum]

I swapped half a dozen of caps. Electrolytics first, that improved things a bit. Then I swapped those paper caps that connect stages (2x 15nF) and 2 or 3 others that are in a low-impedance high-voltage path. They all tested fine, worst has been 5nF instead of 15nF between audio stages. The swapping did not result in any tangible difference in performance. In particular the audio stage still distorts a lot when its getting a bit louder.

All other paper caps measure fine, most display more capacity as they should, like 30nF instead of 10nF. And not a single resistor was out of spec! Even the 1M's all show 1,1M at most. I tested in-situ, cross-checking with the circuit diagram to see if the reading should be exact or could be influenced by other components. Unsoldering components from the lug for checking is a nightmare. I bought a solder pump, but that doesn't work. I hate cutting component-leads just for checking, but for some components there seems to be no other way.

Next I'll empty the bakelite thing in the mains and put new caps inside. Perhaps that will relieve the on-station-hum somewhat. According to a thread in "Readiomuseum.org" this is due to RF on top of the mains 60Hz coming into the set through the power cord. Some suggest clamping 10-100nF directly across the rectifiers.

cheers
Martin