[R-390] 60 Hz HUM Why AC filaments?

[2002tii]

>I must confess that I don't have a lot of experience
>with tubes; But pumping AC into a tube seems to me
>to be a great way to inject hum. Why did they do it?

The heater inside an oxide-coated cathode is insulated from the 
cathode, so there is no galvanic connection between them (well, 
precious little -- no insulator is perfect).  The heaters are 
carefully wound in various helical and reverse-helical arrangements 
so as to be balanced with respect to radiated magnetic fields (the 
same principle that makes twisted pairs of wires resistant to the 
ingress of stray fields).  Finally, cathodes are generally 
low-impedance points, inherently resistant to injected signals.

I do use regulated DC heater supplies in some very low-noise 
circuits, but it is surprising how little difference it makes for the 
vast majority of tube circuits.  For example, I have successfully 
designed phono preamps for low-output moving-coil cartridges (90+ dB 
of gain at 60 Hz) with not a trace of hum audible in the 
output.  [For the record, no tube circuit is ideal for this 
application -- a properly-designed SS circuit will have a noise floor 
at least 20 dB lower than the best tube circuit.]

Finally, there are some tube types (or tube brand-and-types) that 
tend to have leaky heater insulation.  More often than not, these are 
tubes with unusually high heater voltages (26 volts and up), which 
must use longer than normal heater wires.

Finally, a point regarding terminology: strictly speaking, 
"filaments" are directly-heated cathodes (generally, made of 
thoriated tungsten).  Indirectly-heated, oxide-coated cathodes use 
internal "heaters" to make them glow.

Best regards,

Don