As the esteemed Mr. Ehrhorn said, when you hit the "off" switch, the
sequence is usually that the bias supply goes to zero before the anode
supply. The tube biases on, and the anode current bleeds down the HV
supply. I've seen this in medical and broadcast designs as well. In
fact, in microprocessor-controlled medical amps, I've seen this done
intentionally. The microprocessor, operating from stored charge in its
power supply capacitors, interrupts AC to the HV supply, then drives the
grid bias to zero, forcing a bleed of the HV capacitors through the path
most capable of bleeding that charge - the tube.
This is all great for a triode. And with a little care, it can work
well for a tetrode. But you must be very careful that as the anode
supply bleeds down, you make sure the screen supply goes with it. If
the anode supply bleeds low enough, and the screen supply is still "up",
then you'll draw substantial screen current. On some tubes, it may be
damaging screen current. Sophisticated designs will interlock the
screen and anode voltage, or sometimes the screen and grid bias, so that
screen voltage will remain zero unless there's cutoff bias to the grid.
For a simple amateur design, deriving the screen voltage from the anode
voltage is one way to avoid this problem.
There is also a startup risk, whether anode or tetrode. If you shut
down the amplifier, then turn it back on before the tube cathode has
really cooled off (probably not a big concern with thoriated tungsten
filaments), it is possible for the anode voltage to appear before you
achieve blocking grid bias. In this event, the tube draws current from
the anode supply as it's coming up. I have seen a prototype commercial
amplifier where this happened. In that case, the monitoring circuits of
the amp detected the current being consumed from the anode supply before
it was expected and faulted the amplifier off. The redesign ensured
that the grid bias was at cutoff before allowing the anode supply to be
powered. In an amateur design, this may not be a major risk. With
typical amateur tubes (as opposed to 8171s and the like), even zero bias
produces cathode current that can be tolerated by the tube for the time
it takes for the HV supply to come up.
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|