>
>> load, of course. Bird 43 with a 5kW slug was of no use,
>> it is way too slow and does not respond to the 50 wpm
>> dits.
>
>Even if the metering system does not indicate peaks, it is generally
>good for geting optimum tuning on long pulses. The only
>requirement is the rise and fall be short compared to the flat time at
>the top.
>
>There is no magic with speed or duty cycle used, except as it
>relates to average dissipation in the elements.
// Since a tube's optimal RL is basically E/I, an amplifier needs to be
tuned up close to the voltage the anode will see during actual operation.
Thus, one reason to use pulse tuning for SSB operation is that there
are exceedingly few 1500w output Ham radio amplifiers that have an anode
supply which does not have excessive voltage-sag when operated in locked
key / AØ mode. An example of an amplifier that has little sag in AØ is
the Henry Radio 3K-A, which has a 2500w anode supply. Unfortunately, the
3K-A has a tendency to intermittently oscillate c. 95MHz. [to measure
this frequency, couple a dipmeter to either side of the DC blocking cap.
- preferably with the amplifier's power plug pulled]
>
>> Anyway I found that once the tune and load controls were
>> at resonance, they did not seem to need to be tweaked
>> as the drive was increased beyond 100 watts or so.
>
>The load is not resonanted. It sets the anode load impedance. The
>plate control is used to set resonance.
>
>> max C). With 200 watts drive, factory data shows tune dial at 160,
>> and the load C is lower, at a dial of 460 or seemingly significantly
>> heavier loading of the output. Maybe I should do this all over again
>> and just use constant key down for the final tweaking.
>
>A pulse should be fine with almost any meter, as long as the rise
>and fall times are short compared to the on time. The meter does
>NOT have to indicate peaks to let you tune the PA properly on a
>string of pulses.
>
// Good points, Mr. Rauch. In a pulse-tuning comparison between an
oscilloscope and a garden variety, diode-detector-meter, there was no
observable difference in the tank settings. I found that 25 - 30 pulses
per second -at 30% duty-cycle- is fast enough to stabilize a meter needle
and not cause excessive HV supply sag, yet not fast enough to clip the
peaks.
>> Well, maybe all of this fiddling is of no point, as I lower
>> the drive well down before going to the antenna set up!
>
>Since most exciters have considerable overshoot, it is always a
>good idea to overload the tank somewhat.
// In my experiences, using an oscilloscope to observe the RF output
waveform, overshoot can be virtually eliminated by running minimal ALC.
>You can do that by
>opening the load control slightly past the point of maximum output
>at full carrier drive. The plate control should always be peaked for
>maximum output, unless the manufacturer tells you to do
>something else because of unstable feedback or something equally
>odd.
// Uh-oh.
>
>Grounded grid amps are really very forgiving, but it is best to load
>on the heavy side if you do make an error.
>73, Tom W8JI
>W8JI@contesting.com
- R. L. Measures, 805.386.3734, AG6K, www.vcnet.com/measures.
end
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|