Joe Subich's note about which tube to use hit the nail on the head.
W3LPL's recent post looking for 3-1000Zs illustrated how Eimac's exit from
the glass tube market narrows our choices immensely.
One amplifier characteristic I have never seen discussed is a particular
tube's sensitivity to load impedance changes. I suppose this is not an
issue in commercial service, where the antenna is connected and the final
is tuned up on one frequency and it runs forever. (The recent note about
the 8877 in KH6-land running 1900W out on 106 MHz day in and day out is a
good example.)
However, we amateurs change frequency routinely in the course of our
operating, which means that the load the amp sees is changing too. If the
operating frequency is high, or the change in frequency is small, the
effect on the amp is minor. Most amps have some sort of trip circuitry to
prevent damage in the event of operator errors during bandswitching.
However, what about when switching antennas? My station is set up here for
contesting and we have more than one antenna on most bands. Unless you're
very meticulous about antenna tuning, and unless your feedlines are exactly
the same electrical length (or with additional half wave multiples) it can
be very hard to switch antennas without driving your amp nuts.
Two examples from the K8CC log:
1. Ten-Tec Titan on 10M. Three antennas: stack of widespaced 4Ls, a long
boom 6L, and a small 4L fixed south. Find a spot in the band where the
Bird 43 says the SWR is approximately the same (pretty close to 1:1). Tune
up on antenna #1 to 1500W out, grid current under control. Antenna #2 is
1000W out, no grid current. Antenna #3 pins grid current, shut off before
I bother reading the wattmeter.
2. Homebrew 8877 on 40M. Three antennas: full size 3L, N/S dipole, E/W
dipole. SWR's are not perfectly matched, but reasonably close (all under
1.5:1). Setting up to use on 40M in ARRL SS. Tune up on antenna #1, 1500W
out, grid current under control. After much trying I could not get the amp
into a state where all three antennas could be used safely (i.e., not set
off the grid trip).
In situation #2, I pulled the 8877 amp and stuck in a Viewstar PT-2500A.
No problem getting 1500W plus or minus 200W on all three antennas. The
3-500Zs were saying "Did you change the load? We didn't notice."
>From these experiences I've concluded that the modern, high-gain power
tubes are very sensitive to load impedance changes. My friend K3LR gets
around this at this contest station by loading his single-band 8877 amps
heavier than normal, which keeps the grid current down and the tubes can
take the extra power dissipation. My experience with homebrew single
4-1000A and 3-1000Z "wart hog" amps (refers to construction methods, not
power output) is that these tubes are very tolerant to load impedances.
This is what a contester needs in an amp. I don't want an amp that
protects itself, which will take itself off line when I switch antennas and
jump into a pileup. I want an amp that doesn't need to be protected!
Solutions with a 3CX3000, 4CX3000 or larger are OK, but the filament
demands are excessive as well as the key-up idling current, even when
running 1500W (yes, some of us do). It might be tolerable for one amp, but
I have six amps, one for each band. Its hard to manage 2KW of filament
heat in or near the ham shack.
Does anyone have any experience with grid-driven tetrodes with regards to
load impedance sensitivity?
73,
Dave/K8CC
--
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
|