December 1966 QST
Wax nostalgic about and learn from the history of early electronics. See articles
QST, published December 1915 - present. All copyrights hereby acknowledged.
This the the forth and final installment on a series written by
Paul Rockwell discussing station design for long distance (DX) communications.
Covered was antenna selection and siting
I), economics and construction(Part
II), station configuration and receiver topics
III), and propagation quirks and this one - operating
Station Design for DX - Part IV
Part IV - Station Configuration and Receiver Topics
By Paul D. Rockwell, W3AFM
The Geochron Map-Clock. Hams at first stand transfixed
before this - then steal occasional looks. It shows the
sun and daylight zones imperceptibly moving across the Earth,
with diurnal and seasonal corrections built in.
Discussion of propagation will be mainly on (a) use of the long
path, (b) the twilight zone, (c) use of meteor bursts for quick
identification of weak local signals, and (d) use of WWV advices.
It is well known among DXers that signals frequently come in
better the long way around the earth. This applies mainly to paths
exceeding about 4000 miles the short way. Under some conditions,
the optimum path flips from s.p. (short path) to l.p. (long path)
in a few minutes, and it is difficult to choose optimum propagation.
The neatest station-design to ascertain the better path is that
in use at W6AM. He brings each end of each rhombic, through transmission
line, into his shack. By appropriate relays, the path may be tested
or operated on in less than a second simply by flipping a switch.
Similar technique can be (but practically never is) applied to driven
arrays and Yagis.
The question of when, in a longer term, it is desirable to search
for long-path openings, is not easily answered. Recent experience
gives the operator his best competence. However, some general guidance
is to look along the twilight zones.
The twilight zone, globally, has an important relation to h.f.
propagation. For example, on the long hauls, about 6,000 to 20,000
miles (s.p. or l.p.), phenomenally good transmission can be realized
for small portions of the day, on paths nearly parallel with this
zone. According to the relation between the maximum useable frequencies
(m.u.f.) and operating frequencies, propagation may be better on
the day or night side, or directly along the twilight zone. Seasonally,
the zone runs due N-S at the equinoxes; mornings NW-SE in summer,
NE-SW in winter; evenings NE-SW in summer, NW-SE in winter. The
NE and SW directions are for northern latitudes. Long propagation
paths perpendicular to a single intermediate twilight zone, on the
other hand, tend to be poor, especially when the zone is near mid-path.
This is because MUFs are usually much different in night and day
zones - sometimes called the"contrast" problem. An appreciation
of these phenomena is useful in estimating diurnal and seasonal
openings to various parts of the world.23
A chart, from information prepared by Frank Smith, W5VA, is presented
as Table III. Frank has maintained daily schedules for several years
over difficult DX paths - notably with VU2JA, plus 4S7NE as well.
Sometimes signals arrive from unexpected directions - neither
s.p. or l.p. Particularly, arrivals from north and south have been
reported for DX signals as much as 60° displaced in geometric azimuth.
Signals reflected from the aurora zone are usually characterized
by a gravelly sound. To a lesser extent, this can be observed on
signals propagated through the aurora. When Europeans are heard
working the Far East, it's a good sign that Far East propagation
will be good, later in the day, from the United States.
Some propagation phenomena are useful for preliminary identification
of signals in DX work. One of these phenomena is signal enhancement
by reflection from meteor trails. The enhancements may be 20-to-40
db. in amplitude, and are typically about one second in duration.
Unfortunately they are not always present. When they are, they serve
to distinguish weak nearby (non-DX in the 20-meter skip-zone) signals
from bona fide weak DX. An appreciation of this saves the time of
waiting for the weak station to identify himself by sending his
call letters. Another phenomenon, useful in recognition, is the
well-known rounding of keying envelopes by multipath transmission.
That is, keying which has passed through several propagation hops
and has arrived by a combination of several paths, is likely to
sound softer (in the sense of less clicky) than ground-wave or single-hop
It is conservative of one's time in general listening, to spend
more time in good conditions than bad. WWV sends propagation advices
every five minutes. While these are for the North Atlantic only,
their extremes are indicative of conditions in general. A few minutes
at "N7" payoff better than an hour at "W4." At W3AFM, a receiver
is kept on WWV. A digital clock permits turning this receiver on
a second or two before the announcement. Thus the check-up takes
only about ten seconds. An automatic timer could be installed. City
power runs typically ± 3 seconds of accuracy.
To be guided wholly by NBS-CRPL dicta, however, can (a) subtract
the fascination of the completely unexpected, and (b) cause loss
of country contacts. On rare occasions, a choice DX find will turn
up on an otherwise dead band. Sometimes these signals last only
a few minutes, and are heard by only one or two W stations, within
a hundred-mile distance of each other.
Operation and Subjective Elements
Time and keen operating practices can be traded off, to some
extent, for station technical-effectiveness. Especially in this
era of well-equipped DXpeditions, it is more important to be active
on the right frequency at the right time, than to have the ultimate
in DX e.r.p. Thus, a station at home, a home within 20 minutes drive
of the place of work, and a job that doesn't require being out of
town on trips, can add more to the countries total (if that's your
criterion of performance) than a 200-foot tower and 50-foot boom.
A mountain-top summer-cottage, well sited and well equipped, may
be the answer (a) for contest or weekend DXing, (b) if there are
motivations to get the hobby out of the home, or (c) if the home
location is DX-wise impossible. Remote control of the country sites
is not out of the question - but is seldom put to practice, because
of the costs and difficulties. W6YY u.h.f.-remotes transmitters
etc. at Mt. Wilson, elevation 5710 ft., from his home at La Canada,
Neighborhood relations and station esthetic appearance are often
problems. Even if it would fit on the home lot, a hundred-foot lattice
tower and fifty-foot boom may be out of the question. In cities,
the type of antenna beyond which troubles are likely to develop
with neighbors is often something like a 45-foot telescoped-pipe
self-supporting mast, with a 25-foot boom lightweight Yagi. Such
a practical compromise, however, will not compete in contests, no
matter how good the operator, against a fully equipped, well-sited
station. The latter, of course, usually has a good operator along
Flag-poles, if of interest for good-looking mast construction,
are made by John Lingo and Son, Inc., P. O. Box 1237, Camden, New
Jersey. Pneumatically telescoping masts are marketed by Andrew Corp,
P. O. Box 807, Chicago, Ill. 60642. Costs of the latter range from
$2,000 to $8,000 for heights of 30 to 100 feet.
Table III W5VA's LP/5P DX
1. East Coast subtract 1 hour:
West Coast add 1 hour to times above.
2. Months may be read
February 15, etc. for W5 land; February 1, etc. for northern U.
3. 4th Quarter l.p. sigs under South Pole reach
W5 1 hour earlier than W8/1/2/3; sigs over North Pole 1 hour later.
4. November-February 01-02Z; India and Gus/Asia l.p.; FB8XX
589SP - Band sounds dead, actually is not.
Much has been written on operating practices for DX effectiveness.
Listening is far more productive than calling CQ. In this connection,
the tradeoffs represented by Figure 6 are recommended over the more
conventional ones of Figure 7. (Part II, October QST).
It is better to listen several times a day for short periods,
than to listen for the same total time in one session. In tuning,
as from 14,000 to 14,100 kc. repeatedly, it is slightly preferable
to "snap back" (as is done by oscilloscope horizontal sweep circuits)
than to tune uniformly back and forth. In contests, it is preferable
to tune from high to low, as "the pack" predominantly moves the
other way. In pile-ups, a short call precisely timed and on the
right frequency, can be more effective than a longer call at higher
Logging, filing, and QSL procedures cannot be neglected. DX intelligence
("G-2") is very important. DX-alert nets are good. But when they
are not available, it, is sometimes possible to exchange alerting
services over telephone landlines. Particularly useful are DXers
who have retired from full-time employment and spend several hours
a day scanning the bands. Having tools, test gear and spares close
at hand can be a practical advantage in the event of breakdown at
a critical time - some wrong Sunday afternoon.
It is desirable to be capable of zeroing-in with an accuracy
better than 100 c.p.s. on top of the station communicating with
the desired DX station. This can be accomplished most quickly by
the use of a very sharp receiving filter (say, 200 c.p.s.). It is
very helpful if a second receiver can be employed to permit the
zeroing operation to be performed without detuning the desired DX
station, which may be very weak and fading into and out of the noise.
It is especially convenient if the second-receiver v.f.o. can be
cross-coupled, transceiver style, to the exciter, so that the transmit
frequency automatically follows that of the zeroing receiver.
A problem that plagues some exciters is that the spotting zero
is different from the key-down zero. Such a situation can seriously
impair DX effectiveness.
Receiver frequency calibration should have accuracy of ±
1 kc. or better over the 14,000 - 14,100-kc. range. This is to permit
prompt action on DX news-bulletin or other DX tips.
Full break-in is a very desirable feature. Use of even the best
available t.r. electronic switches may degrade station DX performance
materially. This is because the noise figure, at best, is inferior
to that of a good receiver. However, these devices are sufficiently
good to be useful during transmissions. A way of avoiding degradation
during the most critical listening periods is to bypass the t.r.
switch automatically when the transmitter is switched off. Figure
8 (blocks 3, 4 and 5) in Part III, November QST, illustrates the
interconnections and components. The B & W Model 381 has proved
satisfactory at W3AFM; it permits full c.w. break-in by signals
S6 or better. With it, one can tell while sending the approximate
level of clutter on the calling frequency.
Shaping of Keying Characteristic
It is well-known that the corners of the r.f. envelopes of keyed
characters should be rounded to prevent clicks. For intelligibility,
the leading corner should be less rounded than the final corner.
What is less well-known is that a considerable improvement (as much
as 3 db. in effective output-power) may be built in the forward
part of each character without loss of the rounded-corner. This
is done by use of a very high value of filter capacitor in the high-voltage
power supply. At W3AFM, 120 μf. are used. It should be noted
that, in the case of linear amplifiers, output is not a direct function
of plate voltage as in the case of Class C grid-driven amplifiers.
For grounded-grid amplifiers the gain is nearly always constant
(below saturation) at 10-13 db. (output power 10-20 times input
power). Thus, to use this technique effectively, the driving stage
and conceivably its driving stage should also be provided with high-capacitance
filters. The technique should also be useful for handling modulation
peaks. It has no value unless, as is usually the case, the h.v.
power-supply regulation is imperfect.
The station clock should always run on GMT, and logs should be
kept that way. Digital 24-hour types, such as the Tymeter Numechron,
are preferable to round-face clocks. The map-clock, made by Geochron,
is an interesting and useful guide to propagation and is a hobby
in itself. It shows automatically the sun's position, daytime and
nighttime zones, and corrects for seasonal changes.
Size is about 3 feet wide by 2 feet high by 4 1/2 inches thick.
A Mercator transparency moves imperceptibly, according to time of
day, across a red dot at the middle of the map, representing the
sun's zenith position. The red dot moves even more slowly, ±23 1/2°
in latitude and a few degrees in longitude, forming annually a thin
figure of eight (the analemma), Most impressive, the precise daylight
and nighttime zones are continually displayed. The clocks are made
by Geochron, 2515 Palms Place, San Mateo, California, 94401
In preparing this series, lots of helpful correspondence developed
in connection with sending out preliminary draft texts and accumulating
information. My special thanks to W1WHS, K2HLB, W2GhK, W2JT, W2PCJ,
W2VCZ, K3OKX, K3TVU, W3BMX, W3GRF, W4AO, W4BPD, W4FFV, W4KFC, W4YHD,
W5VA, W6AM, W6SAI, W8BRA, W9HHA and KH6DVD.
Not all that was in my head got onto paper. Not all of what did
get on paper is clear. Also, a couple of errors are worth correcting.
Transmitters should be at least be mentioned in a tract like
this. Essentially all top DX stations in W/K land use kw. finals.
About half these are home-built, It may help those who plan home
construction of this kind, to relate a point-of-view developed from
experience. This is, that it pays to buy components of first quality
right from the start. For example: Westinghouse Oz-Paks, Ebert mercury
power relays, B & W Type 800 chokes and Linemaster 632-S foot-switches
are fine products now in use at W3AFM but each was preceded by a
cheaper one. The predecessors now are in a junk box ... a total
loss. Cheaper does not mean more economical.
Referring to Part I, Sep., on antenna siting. There are three
zones under consideration, namely: 1. Near-zone (I2R
losses) under the antenna, 2. The reflection zone, 3. The far-zone
(horizon clearance). Take a site such as W3CRA's on Fig. 2. Frank
Lucas has perhaps the strongest signals coming out of W/K land.
His near-zone I2R loss is negligible; he uses a balanced
horizontal radiator and a reasonable antenna height of λ/2.
His take-off lobe is formed within the first fraction of a mile,
on a nearly ideal sloping forezone. This low-angle lobe is able
to clear the horizon because of his high altitude in reference to
surrounding terrain. If the antenna were situated back over the
ledge of the hill on a level plateau, so that the antenna could
not see the sloping foreground, then his take-off angle would be
only that determined by the height of the antenna over the plateau.
Some questions have been raised about the curved coordinates
in Fig. 2. This is 4/3 earth-radius paper. On it, radio rays passing
through the refraction of standard atmosphere are straight lines.
To construct such paper, draw a level straight line. From the center
of this line mark off distances in miles. Then drop down for various
heights according to the formula:
dmi = 1.4 √(dft)
The opening remarks of Part III were mostly intended for Table
I rather than Table II. Table I appears in Part I, September.
Contesters are more heavily equipped than DXCCers.
In Part III, 75A4 mod (4) refers to the r.f. stage.
23 - Persons interested in h.f. propagation
should acquire and study an excellent value in this field: Ionospheric
Radio Propagation (NBS CRPL), 1965, From Supt. of Documents, USGPO,
Washington, D. C. 20402, $2.75. See also "Simplified CRPL DX Predictions,"
QST, July 1957, p. 28.
Posted May 27, 2014