In preparing for an upcoming CONTEST expedition for WPX, I recently set out
to test my new Icom 706 with my Alpha 87A. I know this is a bizarre
combination, but I'm looking at the 706 as a backup rig. (Incidentally, my
apologies to all who were in the Sunday morning 40/20m pileup on KH6CC during
ARRL-CW and couldn't understand why I didn't answer them and just kept on
CQing. It turns out the receive oscillator on my main rig had suddenly taken
a sideways step. We finally found the pile-up by using a second rig and I
finished out the contest with a whole lot of RIT dialed in!)
Anyway, when I try to drive the 87A with the 706 I see some very unexpected
behavior. No matter how low I set the drive power on the 706, the very first
dit I send causes the power output lights on the Alpha to flash full scale.
Succeeding dits which are closely spaced will give the expected power
reading based on the amount of drive supplied. In SSB, a similar situation
holds. I can pull down PTT so that all the T/R switching is complete but
then the first bit of audio supplied causes a momentary full-scale reading on
the 87A's power meter. After that momentary flash, the reading is expectable
in the same way as it is in CW.
In the course of making these tests, I noticed one other interesting piece of
data. If I wait a half second or so after the first dit before sending a
second dit, then the second dit will cause a power reading which is still
excessive but less than full scale and reduced by an amount proportional to
the time elapsed since the first dit. To give an example of the pattern, a 1
second delay might show 1KW out, a 2 second delay 800w, and so forth.
I first contacted Icom Customer Service about this and they had no
explanation at all. Their representative mentioned K1KP's article in the Nov
'95 issue of QST on using a optocoupler for T/R switching, but even he
acknowledged that the problem I was seeing didn't sound like a timing problem
with T/R switching. This morning I contacted ETO Customer Service (Ray
Heaton), and he gave me what I think is a totally plausable explanation.
This is that the 706 is actually putting out a momentary spike when it first
starts transmitting that is full-scale (or worse) output. Some time delay is
required for the ALC circuitry in the 706 to kick in and cause the output
power to be cut back to the proper amount. This time delay is in
milliseconds, of course, so a peak-reading wattmeter is not fast enough to be
able to record it. The LEDs on the Alpha do show it, though, and Ray
suggested that I was risking damage to the PIN diodes in the Alpha if I
operated it with this initial spike coming down the line from the 706. He
suggested that what I really need to do is use a scope to measure the size
and duration of that leading spike and then go back to Icom with the
information.
This sounds fine, but I don't have access to a scope these days. So, for
those of you with the perseverance to read this far, I wonder if there's
anybody out there with a 706 and a scope who would be willing to take a look
at this for me? Alternatively, I'd eagerly be willing to go anywhere in the
SFO bay area to set up a test with my 706 if you have a scope we can use.
Bruce Sawyer, AA6KX
|