What I did in my tetrode amp was to put an 0.25 ohm 50 watt resistor in
the power supply box between the negative output of the power supply and
ground. This was clamped with 6A10s both in the power supply and at the
point where the control cable plugs into the amplifier. Maximum plate
current is 1.5A, so that corresponds to a voltage of 0.375V.
I was concerned that RF might be picked up on the metering line and be
rectified by the diodes, so I heavily bypassed it at both ends, and the
control cable is shielded. I didn't notice any problem, but I can see
how this could be a concern, especially at 160 meters. Multiple parallel
bypass capacitors are required -- note that the reactance of an 0.1 uf
capacitor at 1.8 MHz is about 0.9 ohms! Luckily the voltage is low at
this point so the capacitors can be physically small.
I learned my lesson about meter protection the hard way when I
accidentally grounded the screen in my previous amplifier: back-to-back
diodes directly across a meter do not necessarily protect it, and it's
necessary to arrange the circuit so that the diodes will start to
conduct when the current through the meter is no more than about 1.5
times full scale.
On 3/13/2014 5:06 AM, Jim Garland wrote:
Hi All,
Handling the B- return is straightforward with triode amplifiers, such as
Paul's converted Alpha 77D, but it can be a bit tricky with tetrode
amplifiers, which typically have cathodes directly tied to chassis ground.
In those cases, the B- wire should be routed from the power supply into the
RF deck and grounded at the same point where the cathodes are tied to the
chassis, thus providing an unambiguous B- return path for the cathode
current. The subtlety comes about because one typically measures the cathode
current by placing an ammeter in the B- return or, alternately, measuring
the voltage across a low-value resistor in series with the B- lead.
In my latest tetrode amplifier, I used the latter method. I connected a 1
ohm/10W resistor in series with the B- return, with one side of the resistor
grounded at the same point where the cathodes were grounded. I measured the
voltage across the 1 ohm resistor in order to determine the cathode current.
Thus in normal operation, the cathode current flowed to ground, up through
the 1 ohm resistor, and then through the B- wire to the power supply. The
problem came about when one of the tetrodes suffered an internal arc, thus
momentarily shorting the B+ (2500V) to ground and placing a 2500V potential
across the 1 ohm resistor. This immediately destroyed the resistor, thus
disconnecting the B- return from the RF deck -- a highly unsafe condition
which places the entire RF deck at the B+ potential, with no way for the
current to thread its way back to the power supply. Even if the operator
avoids danger, the flashover is nearly guaranteed to wreak havoc with other
amplifier components.
The fault with my design was in choosing a 1 ohm resistor to measure the
cathode current. Since the normal cahtode current ranged from 0-2A, the
voltage across the resistor ranged up to 2 volts. This voltage was too high
for me to clamp it with a power diode (though I could have used seveal
diodes in series). My fix was to replace the 1 ohm resistor with a 0,1 ohm
resistor and change the divider network in the metering circuit to
compensate for the lower voltage. Now the voltage drop across the resistor
is a maximum of 0.2V, which meant I can clamp the voltage with a single
power diode, in my case a 6A10. The anode of the 6A10 is grounded at the
same point as the tube cathodes, which is also the same point as one end of
the 0.1 ohm resistor. Now, in the event of an internal tube arc, the diode
will handle the flashover current and safely route it back into the power
supply. The 6A10 has a high pulse current rating which, in compbination with
a 25 ohm safety resistor in the B+ lead and an HV fuse in the B+ lead at the
tubes, means a flashover will not create a safety hazard.
A related issue with tetrode amplfiers is safely protecting the screen
voltage supply and bias regulator circuitry in this situation, not a trivial
matter because an internal tube arc momentarily raises both the screen
voltage and bias voltage to the full B+ level. In my own amplifier, I spent
considerable time designing regulator circuits that could handle this
momentary voltage pulse without damage. One also has to be careful to
protect metering circuits and to make sure low volage bypass capacitors in
the amplfier aren't subject to damaging voltage spikes. I'm finding that
tetrode amplifiers present a number of unique design challenges. Tetrode
amps work great, but definitely require more thought and care in their
design than triode amps.
73,
Jim W8ZR.
--
73,
Vic, K2VCO
Fresno CA
http://www.qsl.net/k2vco/
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|