The Best measurement I ever got was using a dummy load of know resistive
accuracy. I used a new "Cantenna" from heath kit. I said "new" because if
over heated once they can change resistance. The older one seemed to be
better.
I use a dummy load made of a dozen 600-ohm Glo-Bar resistors in parallel.
The composition material looks like it is tinned at each end, and plugs into
clips just like a cartridge fuse. Not exactly sure the power rating of
each, but they are hollow composition tubes about 1" in diameter and 18"
long. They were new in the box, dated 1945, apparently WW2 surplus intended
for use as rhombic antenna terminating resistors. I can load the
transmitter up to 1 kw output and run that into the load for 30 minutes or
more, and although the resistors get very warm, nothing looks like it is
anywhere near the self-destruct point. I have the resistors mounted
vertically to produce a chimney effect for convection cooling without a fan.
When cold, the parallelled resistors measure exactly 50 ohms with my Fluke
DVM. But if I run the load hot for a while, the DC resistance changes a few
ohms (don't remember if it increases or decreases), but the SWR meter still
reads exactly 1:1.
If I want to measure the power output of a transmitter, I load it into that
dummy load, measure the rf current with a thermocouple meter, and calculate
using ohms law.
On the air, none of my feedlines look anything near like 50-ohms
nonreactive.
On 160m, I use an outboard L-network to make the transmitter see a 50-ohm
load, since the el-cheapo Gates is designed to work into a very narrow range
of 50-70 ohms (much like a ricebox), whereas other BC transmitters of the
same era were rated to work into 30-600 ohms or so.
If you use an outboard L-network, beware of transmitting into it with the
feedline disconnected. I did that twice. Once when the flexible stranded
copper lead on my T/R relay failed, and once when I forgot to re-engage the
antenna switch following a thunderstorm. Each time, I blew up the rf
ammeter mounted in the transmitter, wired in series with the rf output line.
Apparently, just working into the L-network without a proper load on it
generates ENORMOUS circulating rf current and blows the thermocouple in the
meter.
I just strapped across the output rf ammeter in the transmitter. That meter
needs to go between the last element of the matching network and the
feedline itself. It would be nice to be able to read rf power output
directly using the rf ammeter, but thermcouple meters are too rare and
expensive to blow up every time the transmitter is accidentally keyed up
without a load.
The broadcast station where I once worked had a matching network between
transmitter and tower, and I never remember blowing the rf ammeter. They
had a disconnect switch to remove the meter from service when readings were
not being taken, to avoid lightning jolts wiping out the thermocouple.
Of course, since it was a broadcast station designed to run 24/7, there was
no disconnect switch to remove the tower from the output network, nor any
T/R switch in the line, so the L-network never worked without a proper load.
Don
k4kyv
_______________________________________________________________
This message was typed using the DVORAK keyboard layout. Try it - you'll
like it.
http://www.mwbrooks.com/dvorak/
http://gigliwood.com/abcd/
