> I was wondering what a meaningful test of the relevant issues
> would look like? I happen to have two 220's that I am willing
> to use as test beds.
>
> What do you think? Would this be helpful to the discussion,
> or not?
I think everyone is worn out on this, and I know I certainly am. I wouldn't
object to one more objective test session assuming it was scientific. I
enjoy learning, and sharing knowledge, but I really hate wasting time.
A meaningful static test would require some somewhat uncommon test
equipment, or careful thought and use of less sophisticated gear. There
would have to be a way to measure the actual anode system and impedances at
resonance. Even though I have the latest Agilent Technologies vector network
and spectral burst measurement gear, anything I or anyone who agrees with me
might say would be declared "false", so it would do no good.
This is how I would do a test:
*The claim VHF oscillations cause capacitors and switches to arc is easy. We
can find several examples of amplifiers alleged to have failed this way in
published data. All that is needed is to inject the anode system with a
sweep source in an amplifier that has that alleged problem.
By injecting the anode with a HF through VHF sweep signal (it could even be
an antenna analyzer with a level output as a source) and measuring voltage
distribution in the tank, the results would easily prove whether extreme
voltages are likely in the tank at any frequency but the operating
frequency.
My experience is any PA having a problem with voltages not being heavily
attenuated (bypassed by low impedance capacitors) is a TVI and harmonic
nightmare. They never even come close to passing FCC testing.
*The claim instability damages a filament is very easy. All we need to do is
look at the peak emission of the tube and do some simple math. I've done
that on 3-500Zs, and I'm amazed no one else seems to have done the same.
Once you eliminate peak emission current causing necessary stress, you
eliminate the need for any further testing. If the tube can't even come
close to supplying the necessary peak current, it can't bend the filament
for any reason related to emission and that would clearly include
oscillations.
*Tube arcs are another matter. They have been the subject of great
postulation for many years, because once the event happens it is gone and
may not ever return. The very action of the plasma in an arc has a gettering
effect on gases, and tubes are even intentionally arced in manufacturing to
assist in removal of residual gasses. Impedance and voltage measurements
would help, because any large anode swing would require high anode Q or high
impedances...not just in the tank alone but in the tank with the tube's
shunting capacitance in the system.
A comparison of possible HF voltages to possible VHF voltages, knowing the
available peak emission currents and other system characteristics would
certainly indicate the likelihood of a VHF problem causing an arc, or
disprove it as a likely cause.
None of this is very difficult, and with some planning and proper use even
simple test gear could give meaningful results. I routinely do exactly these
sorts of tests using sweep equipment, because it speeds design time and
insure a clean (harmonic-wise) product. It actually is pretty easy to see
what type of potential problems a system might have in a cold test, and the
time savings is why rapid-test equipment (even though expensive) is worth
the investment.
None of this is really that esoteric or difficult, and I suspect many people
have made similar measurements.
As for experiments, you'd have to A-B test the system. Put the suppressors
in, take them out. Put them in, take them out. I don't think many people
would want to do that, if parts damage was involved. I do know manufacturers
who have, in desperation when there were tube problems, tried nichrome
suppressors with no improvement in tube life. But this doesn't prove that
some other cases might be corrected by nichrome.
I know that Clipperton L's behave better with nichrome, or any lossy tank
system on ten meters. That is a upper HF/low VHF stability issue caused by
the tubes not being neutralized at the operating frequency. So there are
cases, where the system is unstable at lower frequencies, where de-Qing the
anode and tank system will at low VHF and upper HF will make the amp more
stable. Of course that isn't the only cure possible, but it is one cure. It
is a cheap and simple patch for the Clipperton, and while it still can be
made to oscillate with light loading on ten and 15 meters it at least is
considerably more stable with nichrome.....and the fix is easier than adding
neutralization.
It would be nice if some other members of the amps reflector would do actual
measurements and post results, or even calculations (such as the filament).
I think this could be an interesting and really provide useful insight into
how PA systems work. Collectively most people would understand the results,
and that would go a long way towards improving the quality of knowledge in
this group.
73 Tom
|