[3830] TBDC K1LT Single Op HP

[webform at b4h.net]

                    Stew Perry Topband Challenge - Warmup

Call: K1LT
Operator(s): K1LT
Station: K1LT

Class: Single Op HP
QTH: 
Operating Time (hrs): 11:22

Summary:
Total:  QSOs = 347  Total Score = 1,247

Club: 

Comments:

Since 160 is my favorite band and the Stew Perry Topband Distance
Challenge Warm-Up is the first contest of the season, I have a lot to
report for those who track these updates.  Hopefully this stuff is
more informative than boastful.

Last year I built a 2 element broadside array of 2 element end-fire
arrays composed of short verticals.  The 2x2 BS-EF array seemed to
work about as well as my 1.5 wavelength Beverage in the same
direction.  The 2x2 BS-EF array can be steered, but the pattern falls
apart fairly quickly.

Since the 2x2 worked better than expected, and the 4x2 works so well
as to be unbeatable, I thought a 3x2 might be an acceptable
compromise, since the lay of my land makes a 4x2 pointed toward Japan
a bit difficult.

In August I started carving some 50-foot diameter radial fields out of
the weeds and small trees, aligned with the European Beverage.  I like
trees so cutting them was a bit painful, but I like antennas, too.  I
laid out the array with the same 290 foot spacing as the existing
array, because that's what would fit.  Wider spacing up to about 420
feet gives a better pattern.  With 290 foot spacing, I can reasonably
fit 3 pairs of verticals, with an option for a fourth pair if I get
really ambition about carving holes out of the forest.

Through September I built 6 more short verticals out of PVC conduit.
Each is 23 feet tall with 4 23 foot top hat wires that also function
as guys.  Each vertical has 16 23-foot radials which seem to provide a
very stable ground plane in that the feedpoint impedance stays the same
regardless of weather.  Each pair of verticals requires a relay and
some transformers in a PVC box to form the end-fire array feed and
provide for direction reversing.

I already had a pair of 4-channel Softrock-style receivers in a box
left over from the 8-circle array experiments and the old 330 foot
feedlines for that array still work and amazing, are just able to
reach the most remote element, with a 4-foot extension that doesn't
contribute too much phase error.

The entire system was complete by the Wednesday before the contest,
and I began changes to my software to integrate the two arrays into a
single receiving system.  Friday afternoon I discovered that for some
reason the 8-channel ADC driver (ICE1712 driver for the Envy24 chip
set as used in various Midiman (now M-Audio) Delta Pro_Audio sound
cards) was dropping bits or somehow corrupting samples.  The effect of
this corruption is somewhat similar to using a noise blanker in the
presence of very strong signals.  If the band is quiet, the receiver
functions normally, but if the band is filled with strong signals,
considerable broadband noise appears.

I spent Friday evening and part of Saturday morning trying every trick
in the book to improve performance: built a Linux kernel with the
real-time patch, diddled the process priorities, put the CPU in
"performance" mode instead of "power saving" mode (3.6 GHz
clock
instead of 1.6 GHz), but nothing helped.  This problem must have
appeared when I upgraded the motherboard back in March, but I didn't
notice any issues when I operated the summer Stew.  Maybe I couldn't
hear the "corruption noise" through all the static.

Since I couldn't fix the problem, I decided I could live with it.
Most of the time when signals are the weakest, the band is usually
fairly quiet.  But I did a quick walk of the Beverages to make sure
they all worked.  The European Beverage needed to have the SW ground
reconnected and a couple of feedline connections needed to be cleaned,
but the full set of 12 Beverages became available.

I started the Pre-Stew using the phased array system.  For the first
couple of hours, there was no noise of any kind: no power-line noise,
no static, and no "data corruption" noise.  After twilight signals
got
stronger and the "data corruption" noise got bad.  So I switched to
using the Beverages.

Conditions seemed to be mediocre, and I worked only 2 Europeans early:
CN2R (I know, Africa, not Europe) and DL4UNY.  A little later I worked
G4AMT.  Around midnight there was a brief "opening" of sorts to
eastern Europe and I worked US8ICM and UY0ZG for my best DX (18 and 17
points, respectively) and several others.  I also worked RU3UR but I
couldn't get his grid, which would have been nice.  The rest of Europe
sunrise brought a few more contacts.

While working the European DX, I was watching the waterfall display
from the phased arrays.  But none of the Europeans showed any trace on
the waterfall.  Perhaps the "data corruption" erodes sensitivity.

During the Eu sunrise period, I also worked K9FD/KH6, KH6LC, and KH6AT
along with many 6's and 7's.  When the 7's seemed numerous, I pointed
the phased array receiver that direction and observed that the 7's
were equally copyable on either the vertical array or the Beverages.
Then when Europeans were calling, I put the array in that direction,
but didn't see any waterfall action.  KL7RA popped in quickly enough
that I didn't have time to throw the new phased array at his signal to
see if it worked.

Around 0700 the rate got very slow so I took my 3 hour nap.  At 1030Z,
there were many new signals on the band, although they were mostly
locals.  I did work KL7KY on the Beverages and I found KL7RA again.
SO I studied his signal for a while.  There was rapid QSB, but the
Beverage seemed to produce the same signal quality as the phased
array.

It was strange to work K7CA without a CE1/ prefix.  No VK, ZL, JA or
other Oceania here-say or heard.

The various power lines noises the often plague me have stayed away so
far this fall.  I have equipment ready to go hunting when there is
something to hunt for.

I got an email from an operator in Florida complaining that I was too
close to C82DX's frequency.  Sorry!  But I heard no trace of him at
all (and I need Mozambique on 160) nor did anyone complain on the air.
The C82DX signal must not be getting over land at all.

Two weeks ago the new Alpha 8410 blew out one of the three paralleled
plate blocking capacitors while I was proving its "key down"
worthiness for the Juno flyby.  New caps are on order from Mouser but
they have not yet arrived.  Fortunately, I still have K8ND's ETO 91B
so I pressed that amp into service for this test.  Its also has a
much better acoustic noise spectrum (constant amplitude and skewed
towards lower pitches).

So the plan is to analyze the data corruption coming from the Delta
audio cards and also to see if they work in a 64-bit OS.  Hopefully I
don't have to resurrect the old motherboard.  After the data
corruption problem is fixed, I intend to improve the audio routing so
I can use both vertical arrays simultaneously.  When that works, then
I can try the automated signal finder again.  (The automated signal
finder does the equivalent of pressing all of the antenna direction
buttons until a signal pops out of the noise.)

DX worked: CN, DL (2), E7, FM, G (6), GM, GW, HI, I, KH6 (3), KL7 (2),
KV4, LY (2), OE, SP, UA, and UR (2).

Equipment: K3, P3, 8 element and 6 element SDR receiving system (still
need a name for this concept), ETO 91B (thanks K8ND), 6 2-wire
Beverages for coverage every 30 degrees, 65 foot Tee antenna with 75
125 foot radials under the grass.


Posted using 3830 Score Submittal Forms at: http://www.3830scores.com/