While we're on the parasitic subject (sorry to pump the ol' oscillators out
there):
I once had a disagreement with a competent RF designer of many years
experience, who made big RF dielectric heaters using TPTG configuration,
about parasites. He insisted that the oscillator could not have a parasite
simultaneously and still work on the designed frequency. I said it could,
that it could be simultaneously be making "non-harmonic" energy that could
break things, or created uneven heating and hot spots. Also arcs around the
tube and output electrodes. Does anyone have evidence to refute such
advice? That an oscillator can, by nature, only work on one frequency at a
time?And I don't mean XTAL oscillators but LC circuits and tubes.
I had this experience with a 5 KW cavity amplifier on VHF. The second
harmonic filters kept burning up, yet they were being powered within the
manufacturers rating for the fundamental. The F2 component did not look bad
on a spectrum analyzer, within what one should expect for a Class C
amplifier. But when magic occured, poof when the filter. No arcs in the
amplifier cavity, since it did not use a 'meatslicer" air variable
capacitor. It was a coaxial line circuit with generous spacing between the
ID and OD. Finally was able to catch it, a parasite was happening, at
certain settings of the loading and tuning, and it would only happen at
some power levels. It was very close to F2, which explains the melted
filters. Without a filter, the XMTR was stable into a wideband Bird 50 Ohm
load. With a filter, the PA would get tempermental, and a parasitic would
come alive, the magic. Took us a while to chase it down, and nail it down.
Solution was not something that HF amplifiers would normally use, a small
low Q trap circuit to dampen/kill it, capacitively coupled at the voltage
peak on the line for that frequency. A choke and R circuit would not have
worked here, as it was not the traditional parasitic of the plate tuning
capacitor, the ciruit and tube inductance. At least it couldn't be thought
of that way, being a distributed circuit without actual capacitors and
inductors. Nothing on the input side of the tetrode would stop it. The
tuned ckt on the output side did, but had to be careful not to couple into
fundamental power.
John
K5PRO
--
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
|