The first question goes to tuning the amplifier. I am using
a Vacuum Variable
cap, 3 - 30pF on the plate and the original air variable
plate cap has been
moved to the load position. With 60 watts of drive, I
achieve full amplifier
output of approximately 900 watts with a grid current of
65ma and a plate
current of 575ma. This happens with the plate cap at
minimum and the load cap
at ½ its range or set in the 'middle'. I can achieve the
same output power
results by 'screwing in' the plate cap and turning the load
cap counter
clockwise. This setting yields a grid current of 100ma and
a plate current of
450ma. The latter produces currents more typical of how it
operated as an HF
amplifier.>>
Without putting your amp on a network analyzer or taking
some time to cut and try, it's hard to tell what is going on
in the tank.
The first tuning method produces a more 'linear' gain
reading up to
800 watts output. In other words, if I graph the input
wattage against the
output wattage, the resulting gain stays very close up to
the 800 watt mark. >>
You measured transfer function. The fact transfer function
was linear indicates the amp is very linear up to 800 watts
and 575mA. That's exactly how the 3-500Z's I've tested
behave. They actually are more linear loaded that way. Of
course on CW or long carrier you should not run more than
400mA or so, you may get into heating problems if you do.
(Thanks by the way for verifying the amp can be linear at
575mA! That's an important fact for other people to note.)
The second tuning setting results in this gain measurement
fluctuating up and
down and drops of more at the 800 watt mark. Any comments
as to how this
amplifier should be tuned?>>
It's odd that you have a second setting. That probably
indicates the tank is going into two modes, or perhaps that
the amp has significant feedback that is changing as you
change tuning settings. You might even have a parasitic
going on. A parasitic is often hidden except for little
quirks like jumping power levels or spurious frequencies.
The second question is that the internal watt meters show
about a 100 watts
more output (forward and reflected power) as compared to an
external Bird
wattmeter. Can the internal watt meters be compensated for
50MHz to bring them
closer to a true value?>>
The directional coupler isn't designed for 50MHz. The
primary problems are ground loops in the groundplane. It
just gets by on ten meters, although you could recalibrate
only for six. That's what I did in my AL800H that I
converted to six (2kWout on six).
The third question, when using the ALC, the power cuts back
as expected in CW.
However, when transmitting SSB, the ALC set point has to be
set close to Zero
to have an effect on the exciter's drive power. While
transmitting on SSB, the
grid current will only deflect about 20 or 30ma so the ALC
never sees the 100ma
that I had it set to cut back at. Is this a typical issue
on SSB? >>
Shouldn't be an issue. The ALC should respond to peak grid
current, not average, unless something happened from the
early production stages.
Lastly, while transmitting in SSB, the 3-500Z does not show
any color. I can
transmit a long burst, 30 or 40 seconds, and never see a
color change. If I
key the amp in CW, I can see light orange well before this
time period. Is it
ok to operate in this manner and periodically operate CW to
produce enough heat
to 'condition' the tube?>>
3-500's are gettered on the anode coating. In order to
getter the tube, the anode must show some color. If you
never heat the anode red, the tube will likely fail
prematurely from gas. This is especially true if the pins
are contaminated by being in a moist environment and the
tube sits for extended periods.
Lack of operation that colors the anode for extended periods
of time will often cause an arcing tube, or a "big bang".
Sometimes the arc energy will break down the gas and heal
the tube, but it is better to not let one go that far.
73 Tom
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|