Jan Jozef Oksiuta
Brisbane QG62lk
Australian Amateur Radio Station

DC to light, homebrewing, minimalist antennas and projects, QRSS, QRPp and less
HomeELF-ULF - Signal Gallery 01

Signal Gallery 01 - Sources and locations

Below is a random collection of various patterns recorded around a suburban home, using either an induction coil or earth probes (right-click on the image to examine it in high resolution).

In order to get closer to locating the sources I started simultaneous recordings from several locations. I used a set of earh probes near the house as a reference point (two copper earthing stakes driven 1.2m into the ground approx. 20m apart). The signal after amplification and filtering (50Hz notch and 40Hz low-pass) was supplied to the left channel of a professional USB sound card (E-MU0204). The right channel was simultaneously recording from an induction coil (a rather large solenoid 1.2m long, with 9kg worth of grain-oriented steel core and 7kg of copper wire) also with some amplification and filtering. The signals were displayed in real time on a rolling spectrogram through the Spectrum lab software.

Recording in locations further away from home were performed with either a portable data logger and collected files analysed retrospectively, or through an HF FM radio link in real time at shorter distances.

Initially, another set of recording probes was placed ina nearby park about 50m away from the reference probes.

A number of simultaneous recordings was then compared, with one random sample below.

I could observe that most signals were present at both recording sites. Some, however, appeared stronger at one site and others (even with similar morphology) appeared stronger at alternative site at different times.

One of the portable recorders was then taken on multiple overnight missions further into the field up to 4km away from home. A number of signals were recorded resembling those observed closer to home - however parallel home recordings failed to convincingly produce any signals that would be observable at both sites simultaneously, once the separation between recording sites exceeded several hundred meters.

These observations suggest that the majority of observable signals are localised in small areas. I conducted then another series of recordings, this time all confined to my suburban block of land. An 8-channel data logger with 10-bit resolution and eight sets of earth probes were used in addition to my reference recordings as illustrated below. Each set of probes had a relatively close spacing (around 2 metres) to reduce the capture area and provide more location-specific readings.

Tent pegs were employed as earth probes with satisfactory results. The very first recording session already provided interesting results (right-click on image for better resolution).

Signals visible on the reference recording were also present with varying intensities on all eight channels. The signals were most pronounced in channels 5 and 6 - indeed they appear to have a better S/N ratio than the 16-bit sound card recording. Of course these two channels are located in direct vicinity of my main earth which leads to the inevitable conclusion - the majority of these mysterious patterns are caused by signals "leaking" to ground from the earthing conductor and subsequently from the neutral conductor of the mains distribution network - whatever the primary origin of these signals might be.

More examples are available by right-clicking the thumbnail images below.


It was no surprise then that channels 5 and 6 went "quiet" as soon as my additional earhing stakes were disconnected from household earth wiring. Interestingly, signals on the reference probes also dropped in amplitude. Of course there was still enough "supply" of signals from earthing installations in nearby houses to maintain good "reception" on the widely-spaced probes.

Just to confirm my guesses I connected the data logger iinput (at high gain) to two points along the main earthing conductor. Even across a very negligible resistance of a thick conductor I was able to measure voltage signals reflecting the currents flowing between the household earth wiring and the ground - and most of the observed signals reflected those collected through earh probes. An example below:

- and two more.

The data logger, with no lower bandwidth limit (essentially a DC recorder with maximum sampling rate of 180 samples/second/channel) appears to also show quite a multitude of signals tightly packed between 0.5 and 1.5Hz that are not visible on sound card recordings. This comes as no surprise, with sound cards generally not intended for operation below 20Hz... On-board sound interfaces, particularly those found in laptops, are unlikely to give any useful results for direct ELF recording. However, a number of high quality external USB audio devices seem to work quite well down to several Hz as exemplified on these pages.

Concluding remarks

The great majority of ELF signals observed through earth probes in a suburban environment appear to propagate along the man-made power supply infrastructure, and the reception appears most pronounced in the vicinity of safety earth conductors.

Signals observable with earth probes are also present in recordings made with induction coils. However, relatively large amplitudes of individual signals together with background noise generated by a multitude of weaker signals seem to preclude reception of very weak natural signals such as Schumann resonances, even with a high-sensitivity induction coil whilst in a suburban environment.

Observed signals appear to have similar characteristics when recorded in various locations. However, an attempt to record the same signal simultaneously  at locations  separated by distances greater than hundreds of meters proved elusive to date. Signals entering the ground from what can be considered a point source(individual safety earth installation) appear to undergo severe attenuation with distance.

Signals in the TLF (tremendously low frequency - below 3Hz) range are difficult to directly register using a PC audio interface (sound card). A DC data logger is preferred.

editing in progress - visit again soon. Last edit 20140622

Signals Gallery 02 - Signals and the mains supply 1 - amplitude-modulated 50Hz mains (coming soon)
Signals Gallery 03 -
Signals and the mains supply 2 - other curious sideband patterns (coming soon)
Signals Gallery 04 - Signals and the mains supply 3 - identified signatures of some mains-powered appliances (coming soon)
Signals Gallery 05 -
TLF - Below 3Hz (coming soon)

Preamplifiers and filters project - home recording with a sound card (coming soon)
8-channel ELF data logger
HF radio link project - for use in remote ELF recording (coming soon)

back to top
back to ELF-ULF
back to Home

Created with CompoZer. Last updated 2014-06-21