The recent discussion about under sized tank inductors struck close to
home. Hans' idea about using an infrared viewer would have saved me several
hours of troubleshooting time.
The 4 x 813 amp that is shown on the AMPS PIX page is an experiment
designed to see how far I could push components. The photos show the amp in
a monoband configuration. Having solved a tank circuit problem, I
re-configured the tank for operation on 160, 80, and 40. The tank inductor
had been would with #12 copper wire on a ceramic form. The form had
"threads" on it which provided a turns spacing of about one-half wire
diameter.
Now, I knew that the #12 wire was marginal for the power levels involved,
at least on 40. Since the form was ceramic and the coil itself was in the
air stream ala Heathkit, I reasoned that loss in the inductor would not
matter other than affecting efficiency.
While using the amp in a 160 monoband mode, indeed, there was no problem.
Until, that is, I started stress testing. Having just constructed a larger
power supply with a Variac on it for the YC156 project, I proceeded to see
just how far I could push the 813's. With only a modest increase of anode
voltage over what I had normally run, I found that power would suddenly
drop off after about 30 seconds of transmission (key down). Letting things
cool down for 15 or 20 seconds gave full power again, for a few seconds. I
assumed that it was a problem with one of the 813's and resolved to figure
it out later when I had more time.
With the amp now all set up for band-switching, I discovered that the
problem was much worse on 40 meters. On 40, the power drop off was preceded
by a bit of an arc and occurred at a much lower power level. This was
obviously a symptom of RF heating, but where.
I noticed some discoloration on the coil caused by heat. Operating the amp
with the cover off and following the sound of the tiny arc soon provided an
answer. At the 40 meter tap point, the coil had heated to the point where
the shiny copper turned black for 1.5 turns each side of the tap. The
heating caused the copper to expand and loosen on the form to the point
where one turn came up out of the "thread" and moved over to short against
it's neighbor. The same thing was happening on 160, but without the severe
localized heating at the tap point.
Since this coils dimensions were dictated by the size of the enclosure and
the placement of other parts, it was quite a job to build a different
inductor and fit it in the space available. Live and learn!
The problem was solved by construction of two coils with #10 wire, a
smaller one for 40 with a larger one for 80/160. Increasing the air flow in
the tank compartment probably didn't hurt either. Now, I can run "full tilt
boogie", and really make those 813's blush. As soon as I get caught up, I
will post photos of the "finished" amp on the PIX page.
The curious part of this whole business was the location of the heating.
With turns on BOTH sides of the tap involved, I have to wonder what,
exactly, was going on there. In the past, when I've melted down tank coils,
the heating (and deformation) was confined to the center of the active area
of the coil. Has anyone else seen this effect?
73,
Larry - W7IUV
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|