[Lowfer] W1VD modified Amrad whip tested (matching jfets not needed?)

[Steve Ratzlaff]

Probably this is the last test. I picked two jfets more or less at random
from my bag of J310s (probably around 60) and put them in the circuit and 
measured the IMD at 24.0 and 13.5 volts. (I picked two fets of different 
manufacturers to try to make it more random.) (I did measure their Idss just 
for the info.)
>From my previous tests I picked 60 mA for the 24.0 volt setting and 38 mA
for the 13.5 volt setting. For each voltage I simply adjusted the bias for
that current, and checked the IMD without trying to optimize the second 
order IMD as before.

24.0 volts, 60 mA
1 MHz +58 OIP2
2 MHz +33.2 OIP3
5 MHz +33 OIP3
7 MHz +58.1 OIP2

13.5 volts, 38 mA
1 MHz +71.4 OIP2
2 MHz +33.5 OIP3
5 MHz +33.7 OIP3
7 MHz +69.4 OIP2

Since the thread is getting long I've deleted previous measurements and
comments, you can compare them from your previous emails. Simply adjusting 
for
specific current appears to give reasonable results. If you have an IMD test
setup you can get (much) better 2IMD results for the specific jfets you
have.
Note that this antenna (as well as the original Amrad circuit) will more
than likely overload from strong nearby AMBC stations with a 36-39 inch
whip. That can be stopped by shortening the whip, generally 1 inch at a time
until the overloading stops. I would class this type of single-ended circuit
as "medium performance" for IMD/overload resistance. Much better
overload/IMD performance can be obtained with other (balanced output)
circuits.

(If anyone would like this board I've been testing, email me privately and
we'll discuss a (minimal) price for it. It will have a 90 volt gas tube on
the input, and you can choose which voltage you want it to operate at and
I'll optimize the bias with the IMD test set for lowest IMD. The board is a
small scrap of PCB, built ugly-style, about 1.75" x 2.5". Of course you'll
need to put it in an enclosure, add a whip mount and coax connector, make a 
receiver coupler and a power supply.) (And I'd be happy to optimize the 
bias/IMD on anyone
elses board for the cost of return postage.)

And for the record, my IMD test set system intercepts are about +60
OIP3/+120 OIP2 for the 3/4 MHz test tones and IMD frequencies used for these
tests. There are very few active whips that begin to approach such
performance, though the active whip by Jack Smith at Clifton Laboratories
(as well as the active whip that I also build) probably come the closest to
testing the limits of my IMD test set. If you're a Lowdown/LWCA subscriber,
John Reed has the first of several articles in the latest issue where he's
tested some active antennas. His IMD test setup uses off the air signals and
isn't the same as a bench IMD test setup but his results have shown some
interesting comparisons among the antennas tested. For broadband devices, 
higher second order intercepts are generally more important than higher 
third order intercepts, though in every circuit I've ever tested the two are 
directly related. If you can
improve third order intercept then usually second order intercept goes up as
well. Larger intercept numbers are better, i.e. +40 is better than +30, when
looking at intercept results. A general rule of thumb is second order
intercept should be two times or better what third order intercept is, for a
well-behaved circuit.
73,
Steve AA7U
NE Oregon