Hi all,
A few weeks ago, I mentioned that I had an AL80-A that I had modified for a
higher B+ voltage, mainly for the educational value. After putting it back
into service after a 7 yr hiatus, I had some fireworks and asked for
advice. I thank everyone who responded. That aspect has now been tamed. I
did some other mods, such as added a DC "brushless" fan in place of the AC
fan, which moves more air and is much quieter, but must be shielded from
the RF deck, lest it stop when I transmit.
I hadn't had much chance to use the amp immediately after my mod, because I
relocated to Oklahoma from Colorado. So, I was fuzzy on its operating
characteristics. About all I recall is that, after the mod, I wasn't very
happy with it on the low ends of 160 m and 80 m, but that 75 m and all the
other bands seemed fine. The move, new job, and fresh biological harmonics
left little time for radio, and it, along with all my other
equipmentcollected dust .
When I started this, about 7 years ago, that I fully realized that 1)
modifying perfectly fine, working commercial equipment is often a fool's
gambit, 2) raising the B+ from 2700 V to 3700 V (key-down) is pushing the
components to their limits, and 3) an extra 1000 V nets me only a nominal
1.4 dB increase in power output, which is of no practical
significance. Yet, in the face of all that, I had the itch. I wanted to
*do* this, and see, hands-on, how all of this stuff works. I don't have the
tools, and only barely the metalworking skills, to fabricate the hardware
needed for a complete homebrew amplifier, and I don't have the
facilities. So, I was left with fixing stuff that ain't (yet) broke.
After the mod, I knew that the output impedance of the 3-500Z would be
higher, but decided to "just see" how things went. Sure enough, all the
knob positions were different and I couldn't tune the amp well on the low
end of 80 and 160 m. It was as if I needed more tune and especially load
capacitance. Fiddling a bit with the equations and a simple program showed
me that the change in output impedance could not be easily ignored on the
low bands, especially.
The output network (a Pi-L or L-L, take your choice) now had too high a
loaded Q (probably around 16-18 or so) and needed far more overall
capacitance because the inductance of the Pi part remained constant. I
guesstimated how much to move the taps to increase the inductance and bring
the loaded Q (and so capacitance requirements) back to a reasonable range
for the low bands (40 - 160) and everything worked great, except
160. There are no unused turns on the Pi inductor, so I had to add some.
Along the way, I found that the resistor in my parasitic suppressor had
been baked to a value of ~10 ohms, and so I replaced that.
There's not much room to work with, and I couldn't make the low-band
inductor any longer, so I added a coil perpendicular to the others. It
worked great on 160 m, and all other bands remained OK but it was clearly
too close to the 10-15-20 m coil, and made 10 m unusable. So, I the
inductance of the coil I'd added that worked so well on 160 m, pulled out
a T200-2 powdered iron toroid core and wound the equivalent inductance on
it. Since toroids are self-shielding, their placement is not critical. It
fits nicely under the low-band output coil, epoxied onto a 0.125" piece of
lexan. The toriod stays cool, the amp is now well-behaved, stable, and it
makes a bit more than 1 dB more output (yeehaw!) than when I started.
I wouldn't recommend that anyone repeat what I've done if they want more
power output. It might be considered for the educational value. It was an
expensive 1 dB (over the years, with false starts and blind alleys, about a
dollar per watt of input), but I'm glad I did it. The practical knowledge
I've gained was worth the price of admission.
Thanks to all who provided ideas and input.
Kim Elmore, N5OP
Kim Elmore, Ph.D.
"All of weather is divided into three parts: Yes, No, and Maybe. The
greatest of these is Maybe" The original Latin appears to be garbled.
|