Hi guys, as in my recent post in General discussion, titled "New Old Radio, 1920's" I have identified the set and ordered new speaker grille cloths and a service manual from the "Radiola Guy"
I have in the mean time, been studying the circuit, having found a schematic on this following link:-

http://www.rsp-italy.it/Electronics/...RCA%20Radiola%2033.pdf

Having only worked on more modern radio's, I am wondering how these older valves actually work? As they only seem to have heater filaments, grid and anode??? If the heater filament doubles as the cathode, wouldn't the voltage destroy the filament? The rectifier appears to be supplied with the filament AC voltage but the rectified DC voltage would go through it too? how does that not cook the filament? I hope this isn't too much of silly question for somebody to explain this too me so that I may grasp the concept and thus understand the schematic!

Cheers.

Hmmm, where do we start?

Firstly, you are looking at what's called "directly heated" cathodes. The impedance of the cathode circuit is much, much lower than the anode circuit so current flowing from the anode would have only a very tiny effect on the cathode energising voltage and can be safely ignored.

You'll notice that the AC supply to the cathodes is centre tapped. The cathode wire inside the valve itself is in two strands like an inverted V, the intention being that the AC voltage superimposed on the cathode (or "filament" as it was called) would cancel out. It never did, not completely anyway, so you'll notice that the detector valve (which would be most affected because of the following audio gain) has an "indirectly heated cathode" where the filament (now called the "heater") is insulated from the cylindrical cathode.

How could they get away with using directly heated valves in the RF stages? Any 50 Hz signal injected by filament circuit imbalance would be well and truly out of band so it wouldn't get through the tuned circuits.

With the rectifier, centre tapping of the filament supply was abandoned early because the 5 volts had so little effect on the hundreds of volts the rectifier was handling. Again, the anode current would pass through the filament but because of the low impedance of the filament and its supply it will have negligible influence.

It's all a function of what currents are flowing, not the voltages. DC circuit theory 101. The rectifier filament, for example, needs about 1500mA to heat it. The rectified high voltage that also appears as a common-mode signal on the cathode of the rectifier (and so is sort of cancelled out anyway) is only about 50mA at this point. Because of the big difference in current it can be ignored.

Fast-forward about 20 years or so from this time to AC-DC battery valve portable radios that also used directly-heated valves. Because of the series filament strings used by these sets and the lower filament currents used by the newer more efficient valves, you started to need to take the influence of anode currents into account. Some designers did, adding a resistor here and there to compensate. Many designs ignored the effect, and they seemed to work just as well!

Then there's the issue of bias voltages and AGC offsets in these series filament sets. It gets interesting!

Indirecly-heated valves made life so much easier because each cathode could be isolated from the heater supply and the thermal inertia of the cathode removed any influence of the AC supply.

Hope this helps.

Ok yes, thank you, that makes sense! Interesting!