|
Hi Jim, Alberto, Rik and others,
I agree that the frequency stability of the mains
network is a point to consider further. However, I think there are two
interference mechanisms at play in the LF band:
1. Genuine harmonics of the mains frequency,
from the likes of peak rectifiers in domestic equipment. These are a
mixture of 50 and 100 Hz components, but are "phase locked" within themselves,
so the net result is an apparent set of 50 Hz "lines"
2. Sidebands of "asynchronous" switch mode
power supplies, where the relatively unstable switching frequency is "hum
modulated" by the rectified mains (mostly 100 Hz sidebands for a full wave
rectifier, which most are). Worst case is for the "carrier" of the switch
mode supply to be in-band with the LF frequency of interest.
The sidebands of a switched mode power supplies
seem to wander over any part of LF, and look like "snakes" on an Argo
screen. The pic from Alberto (see below) gives a good hint that there is a
mains harmonic as well as a sideband from an asynch. switch mode supply.
When I first found the line very near 137.7000
kHz on 120 second dot Argo, during the evening, I phoned up my friend
Andrew ZL2BBJ to see if he was experimenting with his LF DDS, and he was.
He had it set to 137.7000 kHz as a marker for forthcoming UA9OC tests. He
had no antenna connected, but he did have scope probes and wires on the bench,
so I thought my receiving setup was performing very nicely as his QTH is about
10 km from mine. However, the weak line I got carried on after he assured
me that he had turned off power in his shack! I doubt very much if there
is a radio transmitter on 137.7000 with virtual world wide coverage, and that is
when I considered that high order mains harmonics may not be negligible when we
are digging well down on conventional noise floors.
My present opinion is that the New Zealand mains
network does have fairly tight frequency stability, but I'll look at doing
further tests. However, if the mains frequency has worse stability in
other countries then "avoiding" mains harmonics becomes less of an
attraction. 136 kHz is the 2720th harmonic of 50 Hz. For harmonics
to not sweep plus or minus 25 Hz means the stability would need to be better
than about 0.009%
73, Bob ZL2CA
----- Original Message -----
Sent: Tuesday, March 25, 2003 2:43
AM
Subject: Re: LF: Selection of spot
frequencies for DX
James Moritz wrote:
Not sure about ZL, but here in the UK the 50Hz nominal mains
frequency drifts around quite a bit - perhaps a few tenths of 1%. Seen on a
fairly coarse resolution spectrogram, the harmonics give a series of wavy
lines spaced by 50 or 100Hz, which drift enough to give overlapping coverage
during a period of a few minutes. But with very narrow resolution
spectrograms, this just adds to the "wide band" noise level, rather than
producing well-defined spectral lines, because the frequency does not remain
within one FFT box for long enough to produce a line. Just for
curiosity, I visualized the 53rd harmonic of the 50 Hz here in my QTH,
and
this is the result. An unidentified signal is also present.
73
Alberto I2PHD
![]()
|
