|
Dear LF, Jean Pierre, Laurence,
the attached image is a false colour assembly of
grabber captures above 19 kHz from the last four days. The timescale
is 10 minutes per pixel from April 1, 06:00 until April 4, 14:00 UT,
with midnight UT highlighted by arrowheads.Frequency resolution was
1.46 Hz. The strong red signal is GBZ Anthorn, with maxima around midday
and midnight, and a very sharp sunrise minimum at 4:30. NWC is a much
weaker 200 Hz band around 19.8 kHz, green to olive, with the ripple effect
visible during the first half of each day. It comes out best in the
GBZ sideband nulls around 19.73 and 19.83 kHz. GBZ was off for a few hours on
April 1st, and the receiver was disconnected during my VLF transmit test
yesterday afternoon.
I am now quite sure that the effect is real, and
that the proposed explanation is correct, by the following reasons:
- the ripple pattern repeated every day in a
deterministic way. It's not local interference.
- the time delay causing between the
components (1 / 21 Hz) matches the geometrical path difference (14.4 Mm /
c).
- the ripple is strongest between 4 and 12 UT.
During this time, a large part of the long path across America is in darkness,
whereas the short path across Asia is attenuated by daylight.
- the diurnal
pattern shift shows a cyclic variation of phase difference. When the
difference is larger (ie. long path is more delayed) the pattern shifts
downwards in frequency. The largest difference occurs around 9
UT, with lower ionospheric height at noon over Asia advancing the
short path, and midnight delaying the long path.
- the short path is mainly overland, while the
longpath is favoured by going across two large oceans. This helps
to emphasize the pattern by making the levels more comparable. On
the web there are some hints about magnetically induced nonreciprocity,
favouring west-to-east propagation. But I have no idea whether that
could contribute significantly at this frequency.
All my observations were on the usual passive
E-field antenna. I would predict that a loop would probably swap maxima and
minima in the ripple, by introducing a phase inversion between the two
components arriving from east and west.
I think this unusual and interesting effect is not
difficult to see for anyone. All you will need is a spectrogram with narrow
enough frequency resolution, and preferable some incoherent averaging to flatten
the speckle from the modulation content. Of course being a bit further away from
GBZ helps.
Best regards,
Markus, DF6NM
----- Original Message -----
Sent: Thursday, April 01, 2010 4:12
PM
Subject: VLF: VK to Eu - both ways!
Dear LF,
the Australian VLF transmitter North West Cape (NWC, 19.8 kHz) can be
received regularly in Europe. This morning I had a detailed look on the power
spectrum of the MSK signal, and found interesting 21 Hz periodic ripples. I
believe this is an interference effect, due to the same signal being
simultaneously received on both short path and long path. The geometrical path
difference (12.8 vs 27.2 Mm) does indeed match the observed ripple
period.
I have switched my VLF grabber http://www.mydarc.de/df6nm/vlf/vlfgrabber.htm
to zoom in on the signal, and have slowed it down to be able to see diurnal
changes. During the last hours between 7 and 13 UT, the fading pattern has not
disappeared completely. The frequency of the minima has shifted upwards by
one-and-a-half ripple periods, most likely due to a variable differential
phase shift between the two routes.
I'd be interested to see if others observe this effect as well. If the
explanation is correct, the null positions should depend on east-west receiver
location on a half-wavelength scale.
Fascinating stuff!
Kind regards
Markus (DF6NM)
PS - This is not an April
joke.
|
