Hi Gary – you have a number of subjects here so I will try and address them one at a time.
First, you ask where does the 35 volts for the heater of the 35W4 come from? There is a good explanation on the angelfire.com website here:
http://www.angelfire.com/electronic/funwithtubes/AA5-1.html
but in short, the heater of the 35W4 is in series with the heaters of all the other tubes and the string is across the AC line. The voltages of the tube heaters in series adds up to approximately the line voltage (50+35+12+12+12=121 volts). It’s like the old-fashioned Christmas tree bulbs in series; if each bulb is 6 volts, 20 bulbs x 6 volts each = 120 volts.
Next, how is the dial light connected? Again, the best explanation is on the Angelfire.com website, which says:
‘The tube manufacturer had provided a tap on the heater of the 35Z5/35W4to operate a 6 volt light for illumination of the dial. But the lamp filament also needed .15 amps so it just couldn’t be connected to the six volt tap. The voltage would drop way down and part of the heater would be under heated while the rest would be over heated. The currents were equalized by having the B+ current flowing through the parallel combination of the lamp and portion of the heater. So the plate of the 35Z5/35W4 connects to the tap as shown in the figure. All of the tubes had been designed so the B+ current would be right for the lamp.’
As for working under the chassis, you will get better with practice about probing around the bottom of tube sockets when the wiring is crowded! Just take your time and maybe invest in a few different shapes of probes and wire clips.
And finally, how to stop burning out the 35W4. John, above, has started you with the right measurement and you say that the resistance between pin 7 of the 35W4 and the B- rail is only 1.2 ohms. If you remove the 35W4 tube (by now it could have an internal heater-cathode short) and measure resistance from pin 7 on the tube socket to the B- rail, do you still get 1.2 ohms after 5-10 seconds? If so, either the 30uF filter capacitor is shorted, or the output transformer primary winding is shorted to ground (B-). You can check each of these by disconnecting them and measuring them separately. If your resistance measurement on pin 7 of the 35W4 socket increased to 1500 ohms and stopped, the 50uF filter cap is shorted. If the resistance slowly increased into the megaohm range, all is well with the new filter caps. At this point, we start looking at excessive current draw on the B+ line, starting with the 50C5. But first, tell us the results of your measurements from pin 7 on the 35W4 socket.
Other forum participants, do you have any other suggestions to help Gary? Please join in here.
Gary A