From Dave G3YMC

Dick PA0SE mentioned a professional report on CW copying which should be
interesting reading. I look forward to seeing extracts of this in Technical
Topics, I am sure Pat Hawker will be interested.

Although not having seen the report, I expect it will give an insight into
the many aspects of the brain's ability to copy morse which simply cannot be
explained by conventional theory.  The following comments from me may well
contradict the report's but may be useful.  Note that these are possibly
slightly biassed as they come from a 'seasoned' CW operator for whom the
code is very much a second language (that means that German has to be my
third language!).  They are based on copying morse directly by the ear and
brain, as I never ever write anything down from a QSO other than that which
goes into the log book for record purposes.  Unless you were a press
reporter, would you normally write things down when holding a normal
conversation?

1.  I have noticed that a weak station close or below noise is often much
easier to copy if one walks away from the receiver.  This may be related to
stereophonic reception characteristics of the ear, but whatever the reason
there are frequently times when I have gone out of the shack, even to the
other end of the house, and heard perfect copy of a station who seems
inaudible when I return to sit in front of the receiver.  A classic example
is when I was at our VHF Field Day station some years ago and happened to
walk past the 23cm station tent in the early hours to hear a PA0 calling.
When it was apparant the station operator had not heard him I put my head in
the tent and (with a little help) he then achieved a QSO.  Apparantly the
PA0 had been calling him all night!  Others have commented to me of this
effect.

2.  It has for a long time been the custom to use relatively slow speeds for
conventional CW on 136, in the belief that this is better when signals are
weak.  I would question the wisdom in that.  A couple of weeks ago when we
had the terrible thunder static on the band, Mal G3KEV was calling a series
of CQ calls, at I guess around 15wpm.  His s9 signals were reasonable copy
but significantly broken up by the QRN and sufficient to make a QSO hard
work (no, I didn't try...).  Suddenly Mal started sending his call at 30-35
wpm - copy was very noticeably better at this speed.  When the brain is
copying CW, as least how mine does it, it doesn't look for individual dots
and dashes, or even single letters, but integrates the whole lot recognising
patterns of words and even (for rubber stamp QSOs) whole phrases, much as we
understand speech.  If this overall pattern of words is interrupted by
static it makes the process harder.  Send the whole lot between the static
crashes and it works wonders.

3.  (related)  I usually have the receiver monitoring on 136 while I am
doing other things in the shack or around the house.  On Saturday I was
monitoring Jim M0BMU/P.  Jim was sending fairly slowly.  I realised after a
while that I had not been paying too much attention to the drift of the
QSOs.  Normally I can follow QSOs in their entirity while doing other
things.  However when the sending is slow the brain has to concentrate
fairly hard to keep track of the dots and dashes and form them into words -
not helped if there is the occasional keying error as well!.  At a faster
rate the brain is only conscious of the words and it is far easier to absorb
these without really thinking. I have in the past copied QRSS in my head,
but it requires a lot of effort to keep track of where one is in a callsign
(and the brain normally loses patience before it has received a complete
call!).

Perhaps we should review the custom of slowing right down when signals
become weak.  Under the right circumstances faster morse may well get
through better and the optimum speed may well depend on local noise
situations at the receiving site.  Of course the copying ability of the
operator must also play a part, and I appreciate some are less happy with
QRQ than others. We must accomodate the abilities of all, but slowing down
may not always be best.

3.  On the subject of receive notes.  I also noted that Harry optimises his
filter for 1kHz.  Many operators including myself prefer a lower note,
600-800Hz or even lower.  I suspect it is the optimisation of the frequency
relative to background noise, and with a narrow audio filter may be less
noticeable.  My rig has its transmit crystals adjusted for transceive with a
800Hz note.  I never monitor with a SSB filter since with USB CW this brings
DCF39 into the passband.  With a 300Hz CW filter copy of weak signals is not
a problem, and the main requirement for a narrow audio filter would be to
reduce adjacent channel QRM.

73s Dave
[email protected]
http://ourworld.compuserve.com/homepages/sergeantd