[R-390] Maintenance: mains voltage

[2002tii]

Charles wrote:

>He pointed out "thorium stripping" of the 26Z5 heaters and of the other
>tube cathodes as a reason why this should not be done..

Thorium stripping happens only on filamentary tubes with thoriated 
tungsten filaments (transmitter tubes, really old pre-octal triodes, 
and some larger rectifiers, mostly).  It is brought on by prolonged 
high current operation, when the filament is run without the 
protection of its "secondary cathode" (the cloud of electrons around 
the physical cathode).  Tubes with indirectly heated cathodes are not 
susceptible to this (they do not use Thorium to increase emission, at 
least not in receiving tube types), although they, too can be damaged 
if they are run without the normal secondary cathode.  However, for 
this to happen in a working cuircuit just because the filament 
voltage falls is extraordinarily unlikely.  Well-nigh impossible, 
actually.  Yes, the cathode emission goes down as the cathode cools 
off, so if the tube continued to draw the same current, eventually 
you would reach the point where the static current draw would keep 
the secondary cathode continuously depleted and you would risk 
degradation of the cathode.  In most cases this would not happen 
until the heater voltage was WAY low -- much, much less than half 
rated voltage.  However, even this won't happen in nearly all actual 
circuits, because the cathode current will also fall as the cathode 
cools even if the B+ is regulated.  (The exception might be where the 
tube operates into a well-regulated current-source load, as some 
tweak hi-fi circuits do, or a filamentary rectifier like the 5U4 is 
run into a capacitor-input filter with an insane amount of 
capacitance, like upwards of 20,000 uF, and turned on and off repeatedly.)

Indirectly heated cathodes can also suffer "cathode poisoning" if 
they spend a lot of time cut off (drawing zero current).  This was a 
problem for some early vacuum tube digital computers, and there are 
"computer versions" of some of the popular miniature twin triodes 
that have cathodes designed to minimize this problem.

Running tube receivers on lowish voltage is not going to cause any of 
these problems.  At some point, your power supply will drop out of 
regulation (if it is regulated to begin with), and you may find that 
oscillators are ever so slightly more stable at their design voltage 
than at 90 volts, but you are not going to damage the tubes.

Best regards,

Don