Receiving VLF Radio Signals
By Andrew Westcott: MØWAN
Back To Radio Intro > VLF Signals
Contact And Location Info
80m Antenna On A Postage Stamp
40 & 80m combined inverted V
• Receiving VLF Signals •
CB Radio
PLT Interference
How I Got Interested In VLF Phenomena
I have many odd interests, but possibly none rival this particular subject for sheer 'oddness'. Few people are aware of the presence of naturally-occuring Very Low Frequency (VLF) radio signals and earth currents, and probably even fewer could really care about it. However, there are those who have an interest in such things, and actively pursue the monitoring and recording of the effects, whilst theorising on how they might be caused, so at least I can rest reasonably reassured that I'm not completely on my own.
I'm no expert in natural VLF phenomena, but I enjoy the opportunity to set up my home made detection equipment somewhere away from built up areas and to just sit and listen to the variety of sounds, waiting for that really unusual one to come along; I find it strangely relaxing, a similar effect to listening to the sound of the sea or a stream.
I first became interested in these natural VLF signals when I accidentally stumbled across them whilst conducting experiments in earth current signalling, after reading about it in an old copy of Practical Wireless. The article mentioned how bases were set up during the First World War for communication between the trenches, and I found the technique used very interesting to the point of attempting it myself.
Without going into too much detail, a signal is fed into the ground via two earth stakes set some 25 metres or so apart, and the signal can be received some distance away by using another pair of electrodes and some amplification.
During a period of manic enthusiasm I actually built a self-contained earth current transceiver. The receiver side consisted of the necessary amplification to receive earth signals, and the transmitter consisted of a CB microphone driving an output amplifier feeding an old valve radio single-ended speaker transformer connected in reverse: the amplifier fed the low impedance side and the high impedance winding connected to the earth probes.
The whole assembly was switched via a relay by using the PTT button on the mic, just like a real transceiver! It worked quite well actually; the resistance between the probes measured around 1KΩ and by connecting a 1KΩ resistor in place of the earth probes, a significant amount of power was transferred to it, heating it nicely showing that the impedance match was, by chance, reasonably good.
I'd like to have built another transceiver and to have persuaded a friend to chat with me through a hill, but ended up building a dodgily-designed inverter which injected a tone of sorts into the ground, which I then attempted to receive.
As it happened, I managed to get a reasonable distance during the tests, perhaps a couple of hundred yards, but range was ultimately limited by the mains hum which masked the quieter audio signals, and the system may well have done rather better in a more remote location. Although I can consider my earth current communication experiments to be a moderate success, the project also introduced me to the sound of atmospherics and my attention began to turn to investigating these strange sounds in favour of further earth current communication experiments.
Back to top
About VLF Signals
A diagram of the Earth's magnetosphere
Credit: NASAA diagram of the Earth's magnetosphere. Reproduced with the permission of NASA
Our planet, Earth, generates a magnetic field around it known as the magnetosphere, and this is known to extend for a minimum of around a quarter of a million miles out into space, but as the diagram shows, the solar wind distorts the shape of this field in such a way that it is dragged out behind the Earth in a direction away from the Sun, and in this direction extends for well over two million miles. This 'force field' surrounding the earth represents one of the ingredients for some interesting effects.
It has been estimated that as many as one thousand thunderstorms are active in the atmosphere at any one time, and each of the lightning strikes releases a huge amount of electromagnetic energy which ranges in frequency from almost zero right up to light and beyond, and are responsible for the constant crackling heard in a VLF receiver.
If the conditions are just right, these lightning-initiated radio bursts may leave the planet and propagate for an enormous distance around the magnetosphere, and by various means return some seconds later, exhibiting a degree of frequency separation in much the same way as light passing through a prism is split into an ordered spread of frequencies, or colours. This frequency splitting can result in some interesting phenomena.
The large range of signals which may be heard in an ELF/VLF receiver can initially be divided broadly into two classes:
Signals Of Man-made Origin
Technology leaves its imprint on the VLF spectrum, and these signals form part of what can be received.
Mains Hum
This is caused by fields radiated from the mains electricity supply lines, known in the U.K. as the National Grid. These alternating fields have a fundamental frequency of 50Hz here in the U.K., and unless steps are taken to site the receiving equipment in a remote location, the interference they generate represents the bulk of what is detectable and indeed can present a major obstacle to detecting the fainter, natural signals. It is relatively easy to filter out the fundamental, but the higher harmonics can still present problems, extending as they do well into the audible range.
Received mains hum waveform showing a high fifth harmonic content
Above is shown five complete cycles of mains hum as detected by my receiver, in this case showing a very high fifth harmonic content (250Hz), along with a lot of other noise; note that the mains proper isn't quite this bad! This kind of power pollution is generally blamed on modern electronics and can cause additional heating of transformers and motors, and can possibly cause other problems by exciting resonances which the fundamental would be too low to do; quite apart from this, it makes the hobby of VLF listening rather more difficult.
VLF Radio Transmissions
Radio transmissions in the VLF part of the electromagnetic spectrum manifest themselves as tones of varying frequencies, these generally being above the hearing range of humans, and are encoded using a small frequency shift to carry data. These signals are mainly for communication to submarines such as those transmitted by the Russian RSDN-20 station, as such signals can penetrate for several metres below the sea surface meaning the submarine doesn't need to surface to receive them. Other signals could be for ionospheric research, and some years ago navigation aids such as the now silent Omega could also be detected.
A sonogram of VLF data transmissions, probably intended for submarines
The sonogram above shows 13 data streams being transmitted concurrently; the transmissions are represented by the horizontal wavy lines and presumably originate from different sites. The frequency scale is vertically on the Y axis, covering between 18KHz and 44KHz and the time scale is along the X axis, spanning half a second; there were no transmissions detectable below this. The vertical components are atmospherics of natural origin.
Reducing the intensity of these signals, assuming they are of no interest, may be in order as at times they may be so powerful as to overload parts of the equipment causing audible intermodulation products which sound rather like a telephone modem, so a good measure of filtering is recommended; I find second-order filtering above 10KHz to be the bare minimum requirement at my location.
Signals Of Natural Origin
Signals resulting from natural phenomena are many and varied, with various theories devised to explain them; many conveniently fall into the audio range, making it relatively simple to detect them.
Most of the natually occurring signals can be classified according to how they sound:
Atmospherics
Often abbreviated to 'sferics', these signals form the bulk of the soundscape and sound rather like the crackling of a bonfire. Caused by multiple lightning strikes around the world, you should be able to hear this all the time, and this is a good indication that the receiving equipment is working. The recording below is typical of what you should be able to hear, and was made using the equipment detailed further down the page.
Tweeks
These are the sounds of lightning strikes which have undergone a form of frequency separation due to the considerable distance they have travelled around the planet, and rather than a simple click, the sound has been split so that the lower frequencies arrive a fraction of a second later, exhibiting a characteristic 'pew pew' sound.
Whistlers
These eerie noises are again believed to be the result of a lightning discharge, but this time the signal consists of a breathy sound falling in frequency; although whistlers can be heard regularly if the conditions are right, good examples are relatively rare.
Unlike the tweeks, in this case the signal has travelled an enormous distance around the magnetosphere, and the signal has become split to a far greater extent than with tweeks, the higher frequencies reaching us first, with the lowest ones lagging behind by a few seconds.
No doubt lightning strikes at various locations around the globe account for most of the noises and studies have shown that whistlers are probably initiated by such strikes.
Back to top
The Equipment I Use
The signals of interest to me fall into the audio range, and therefore a receiver doesn't need any frequency conversion to make them audible. There are two relatively easy ways to detect these VLF signals: either by a pair of earth probes to detect the current flowing in the ground which replicates the passing radio wave, or directly from the air using an antenna, which detects the electrostatic field. My first experiments used the probe method but I discovered a high level of mains interference which seemed impossible to get away from so I turned my attention to trying to detect the electrostatic field of the radio waves using an antenna which offered a lower level of mains hum in most cases.
I've seen plenty of weird and wonderful designs for VLF receivers on the internet in recent times, and in most cases they are way more complicated than they need to be; a few appear to be theoretical only, with little chance of actually working in the field so are evidently untested and quite possibly AI-generated rubbish.
I've gone into some detail with the circuitry I ended up using as some may wish to re-create the circuits used, or modify them for their own use. This equipment used whatever components I happened to have to hand, so the perfectionists out there may like to improve or tidy up what I've done, but what I use is the result of many years of trial, error and modifiation to get the results I wanted. The receiver was designed using discrete components and built in a modular manner; a pre-amp module, a filter module and the output module; this allows easy modification should it prove necessary.
The 'Front End' or Pre-amplifier
All that is required for receiving signals via a pair of earth probes is an audio amplifier, but the situation is a little more complicated when designing for reception via an antenna, simply because of the need for a far higher input impedance. JFETS are a good choice here due to their inherently high input impedance, low noise and simple associated circuitry; the device of choice for me was the 2N3819, cheap and readily available as of 2025.
Circuit diagram of the front end of the receiver
The diagram above shows the circuit of my receiver's 'front end' module. The requirements for the first stage of this circuit were a high input impedance and moderate gain, exactly what a JFET offers; FET1 does double duty here by presenting the high input impedance, and also provides moderate voltage gain, which is fed to the next stage to amplify the signal further.
The total voltage gain of this module in practice was measured on the bench at a surprising 100 times, with the unit outputting 100mV with an input of 1mV, (confirmed by an output of 1 volt for 10mV input) and capable of a clean 4 volts peak to peak maximum output before distortion became apparent. However, too much gain within this module risks feedback and instability problems, so pretty much rules out a conventional bipolar transistor in the second stage which could result in excessive gain.
This section is really the heart of the receiver, with the following sections being optional to a degree. A small circuit board for this stage was made to fit inside an air rifle pellet tin which provided excellent screening from later stages, and the rear of the antenna input terminal was passed through the tin to make connection directly with the board, and also acted as the board's mounting. The filter switching was a later addition and located just outside of the tin. There's no reason why a buffer stage couldn't be added to this, which could allow sensitive headphones to be used without the need for the other modules, therefore keeping size down.
PREAMLIFIER CIRCUIT DESCRIPTION:
The signal coming in from the antenna first encounters the neon lamp, which limits any voltages to around 60 - 90 volts, protecting the DC blocking capacitor which we'll come to next.
The 0·1µF 400 volt capacitor is there primarily to block the standing voltage present when using ground probes (which I've measured at over a volt) from altering the bias point of the FET, but also serves to block any static charge below neon strike voltage which may build up on the antenna.
The 47KΩ variable resistor and the four switch-selectable small capacitors form a first-order low-pass filter for eliminating radio station break-through, with a position available on the switch to have no filter capacitors in circuit, for comparison. By selecting different capacitors and different settings of the variable resistor, a range of filtering combinations can be selected allowing for optimum filtering for any particular situation; too much capacitance reduces the sensitivity, so it's best to use the minimum necessary to remove any radio stations which may be heard.
The 10MΩ resistor is the usual gate resistor, required for basing; leakage current through the FET's gate is miniscule, but can start to become significant when the gate resistor is made much higher than around 10MΩ, which can potentially render the bias point unpredictable so considering 10MΩ as a maximum is a sensible precaution. The two diodes are there to protect the FETs gate from high voltage transients, particularly those created by the striking of the neon lamp and any voltage spikes below the lamp's strike voltage, by limiting them to around 0·7 volts; these could otherwise damage the transistor. Ideally ultra-low leakage diodes would be selected for this position in respect of the high impedance environment they are operating in. For many years I ran the receiver without the limiting neon or the protection diodes without incident, but their inclusion is probably a worthwhile safeguard considering they can probably be found in the junk box.
FETs can under certain conditions break into oscillation, and the usual remedy for this is to install a "gate stopper" resistor of a few hundred ohms as close to the FET's lead as possible to damp this tendency; ideally the resistor would form part of the FET's gate lead itself if room permits, as has been done in this case with both FETs here. The gate stopper of Q2 in combination with the 500pF capacitor and Q1's (unknown) output impedance form a high-cut filter which starts to take effect from around 30KHz to reduce RF which has made it this far, before further amplification is applied.
NOTES:
I initially tried both a variable capacitor and ceramic capacitors in the filter, but both types were highly microphonic in this high impedance part of the circuit, producing audio/mechanical feedback from the receiver's speaker even at moderate sound levels; the solution was to use fixed value film capacitors in the filter.
The 2N3819 FETs tend to have a fairly wide specification spread so the source resistor of any particular FET used should be chosen by trial and error, and I aim for around 0·5 to 1mA of current flow, which seems to give good performance; this can be easily calculated from the voltage drop across resistors in the circuit. In practice, with the samples I have here, 2·7KΩ resistors worked well in most cases, but varied between 2·2KΩ and 3·3KΩ, so I chose the value which gave the closest to, or just under, 1mA.
The Filter Section
Circuit diagram of the high pass & low pass filters
The above circuit shows detail of the filter sections and as usual, this circuit doesn't represent perfection in design, but it's good enough to get the job done; this section can be considered optional, but I do find it useful to have the filtering available.
FILTER BOARD CIRCUIT DESCRIPTION:
The medium impedance signal from the preamp passes into the 'input' terminal, where it is buffered by Q1, providing a high impedance input to the incoming signal and a low impedance output to drive the following first-order low-cut filter stage, formed by the 0·022µF capacitor and the 27KΩ resistor. From the first filter section, the signal feeds into buffer Q2, which drives a second first-order low-cut filter feeding the high-impedance input of Q3. Both filter sections are isolated for best performance but together form a second-order low-cut (high pass) filter with a cutoff frequency of around 268Hz for reducing the level of mains hum present and much of the unwanted low frequency content. A DPDT switch can be provided which switches in 0·1µF capacitors across the existing 0·022µF filter capacitors which drops the cutoff frequency to around 50Hz, effectively disabling the filtering should it not be required for any reason.
The low impedance output of Q3 drives the first-order high-cut filter, formed by the 10KΩ resistor and the 1600pF capacitor which feeds into the high impedance input of buffer Q4, this providing a low impedance signal to drive a second first-order high-cut filter section, which feeds into the high impedance input of Q5. As with the previous filter sections, these are isolated for best performance and in this case together form a second-order high-cut (low pass) filter with a cutoff frequency of just under 10KHz for reducing the intensity of any supersonic or radio frequency signals getting fed to the output stage. A DPDT switch can be provided which switches in a 2·2KΩ resistor across the existing 10KΩ filter resistors which increases the cutoff frequency to around 55KHz, effectively disabling this filtering should there be an interest in VLF transmissions.
Besides providing a high impedance load for the final filter section, Q5 also provides a voltage gain of around 5 times in my circuit which more than compensates for losses in the previous filter and buffer stages (100mV in gave 500mV out). The output of Q5 feeds buffer Q6 which outputs a low impedance signal ready to drive a pair of 10KΩ potentiometers, one for volume and one for recording level.
Design notes:
If I were designing this board now, I'd make the high-cut section a third-order or even a fourth-order filter to suppress the submarine transmissions further; the second-order filter as it is only just about does it, and could do with being more aggressive.
The components for the four filter sections where carefully tested, and only those pairs of capacitors and resistors which tested as virtually identical were used. This was to ensure that when the two sections of each filter were cascaded, their response curves would be as near identical as made no difference; not really necessary in this basic application, but I'm happy in the knowledge! The seemingly arbritary value of the 1600pF filter capacitors were ones I happened to have to hand, which together with the associated 10KΩ resistors provided the filter slope I wanted. Any other values which provide the necessary response could be used, such as a 15KΩ resistor coupled with a 1000pF capacitor for example, which is probably more widely available.
It is a good idea to install gate-stopper resistors in FET circuits like this, but this board seemed very stable without them with no hint of parasitics on the 'scope; however, due to the long trace connected to the gate of FET1, I decided to cut the trace and fit one at that location just in case.
Filter board top |
Filter board bottom |
The above pair of photos show the circuit board I made for this module. It is a fairly simple circuit for which I used a piece of 100mm by 70mm laminate, giving plenty of room to lay out the board without cramping things up too much; as I draw my tracks manually this means I don't have to draw very thin tracks. (I enjoy doing this manually, as if I need to give a reason.)
The Output Stage
The output from the filter board passes into a conventional TDA2003 chip amp which provides around 40dB of voltage gain (X 100), a fair chunk of the gain this receiver provides, and enough power to drive a loudspeaker. I believe this chip is now obsolete, but may still be obtained as new old stock. For an output of just under 1 watt, you could use an LM386 instead and if I was starting from scratch, this is what I would now use.
Output stage circuit diagram
The above diagram shows the circuit for the output stage, based around the TDA2003. There are plenty of diagrams for this available on the internet, most based heavily on the one published in the data sheet for this device, as is this one.
OUTPUT STAGE CIRCUIT DESCRIPTION:
This is the classic circuit suggested in the data sheet for the TDA2003 audio output chip, with a few minor modifications. The input capacitor is now a non-polarised 1µF ceramic capacitor rather than the (unnecessarily large) 10µF capacitor suggested. This low value does reduce low frequency response somewhat, but this is of no consequence in this application.
The feedback capacitor is shown as a 10µF film capacitor rather than the 470µF electrolytic capacitor suggested. This much smaller value of capacitance causes the low frequencies to begin rolling off from around 100Hz, but this is fine for this application. In addition to this, there is less than a volt of bias across this capacitor so a conventional electrolytic may eventually become a bit leaky in this position, and leaks aren't good in a feedback path.
Finally, the high frequency roll-off is controlled by the 0·047µF capacitor and the 39Ω resistor. The capacitor is slightly larger than the 0·039µF shown on the data sheet as I happened to have that value available. With this slightly larger value, the high frequencies now begin gently rolling off above around 13KHz, which again is fine for this application.
Output board top |
Output board bottom |
The above pair of photos show the output board I made. As can be seen, it is a very simple circuit with the minimum of components which I spread out over a piece of 100m by 70mm laminate, as I have several pieces of that size and I didn't want to be bothered cutting any. The heatsink may seem marginal, but as this amplifier draws less than 50mA when idle, the heatsink remains completely cold under those conditions. Even under test when supplying 2 watts of power into an 8Ω load the heatsink only became slightly warm, so is expected to run cold under normal use.
A quirk of the TDA2003 is that the output seems to need to pass around 30mA of current continuously via the 220Ω resistor to ground to keep it stable, which also forces the output stage to operate in class A mode for low level signals, presumably to keep crossover distortion at an acceptable level. This current wastage isn't a problem for in-car use as was intended by the designers, but may become an annoyance when running off a battery. Attempts at increasing the values of the 220Ω and the 2·2Ω resistors while keeping the ratios the same simply resulted in severe oscillation, so further tests along those lines were abandoned. I see this as a basic flaw in the design of the chip but one I can live with, especially as there don't seem to be any other true single-rail amplifier chips out there with this power output.
So in summary, we now have an output stage adding around 40dB of gain to the total, with a frequency response which starts to roll off above around 13KHz, pretty much ideal for this application; heck, I can barely hear 10KHz these days anyway!
That reminds me:
fairly recently both me and a mate became slightly concerned about the lack of grasshoppers. We used to hear these things chirping away on summer evenings, but these days, nothing at all. The concensus of opinion was that insecticides must have killed most of them off, until one evening when out on a walk by the coast with my wife, she happened to mention the vast numbers of chirping grasshoppers she could hear and yet all I could hear was the gentle sound of a breeze blowing through some marram grass. Slowly the penny dropped, and I realised that at my age I could no longer hear frequencies that high; the revelation was a little upsetting at first, but apparently this is normal for males of my age so I have no choice but to accept it.
The Completed Unit
The various modules were mounted on the back of an aluminium panel in a very amateurish way, which fits onto the front of a reclaimed steel enclosure. Some additional bits of circuitry were tacked on over the years, these being made in the classic bird's nest' style for simplicity and to allow easy changing of component values should it become necessary, the whole unit being the result of many years of gradual evolution.
Front panel controls |
Rear of front panel |
I believe the message, if any, to take away from this is to not worry what the end result may look like, just do it!
Back to top
The Practical Side Of Setting Up A Receiver
Parked up in my favourite spot, just 5 minutes from home. Note the fork stuck in the ground.
Theory and circuit diagrams are all very well, but some pointers on how to set up the VLF receiving equipment may be in order.
YOU'll NEED A RECEIVER...
Of course you will, and at this point I assume you've equipped yourself with something suitable, and a suitable power source; you may not be able to use the car's battery - see further down.
FIND A GOOD LOCATION
The most important thing to sort out is where you are going to set up. 11KV power lines radiate an intense electrostatic field for some distance around them, so you need to find somewhere as far away from them as possible as this will just swamp your radio with 50Hz hum, obliterating the far weaker natural signals. Given that much of the UK has houses scattered around the place, for best results you'd need to find somewhere remote; here in the South West I can go to Dartmoor, where some regions are sufficiently far away from those power lines to allow decent reception.
I've also recently found a good spot up on the Blackdown Hills, where the power lines are at least 3 large fields away in every direction, meaning the level of mains hum is sort-of acceptable considering the location is 5 minutes from my house. Forget trying to listen in any built-up area, it'll be a waste of your time; I doubt even dedictated filtering software would be able to clean up that mess.
INTEFERENCE FROM YOUR CAR
During a recent resurgence in my interest in VLF, I stuck a CB antenna onto a magnetic mount on the car roof, as I had done before. I was somewhat irritated to find that modern cars (even something as old as my Corsa) radiate interference themselves when parked up even with the key out, making things a bit difficult. Locating the antenna away from the car reduced the intensity of the interference a small amount, but the only sensible remedy was to disconnect the damn car battery which I now do, and have got the procedure down to 30 seconds or so.
THE ANTENNA
You'll need an antenna of some kind; it doesn't have to be anything fancy and could even be just a couple of metres of wire suspended from a branch or similar. This isn't a resonant tuned antenna; it's more like an electrostatic probe up in the air, and the longer the antenna, the stronger the signals will be. You should also poke an earth stake into the ground and connect that to the ground terminal; I do this as a matter of course as it makes the whole unit more electrically stable.
My antenna mounting idea
I found (borrowed from my wife - permanently) a garden fork, and mounted a CB antenna bracket onto the handle to allow an 8-foot antenna to be fitted. The fork can be pushed into the ground as you do with forks, and the aerial connected to the receiver. An earth wire is also fitted, with one end clipped to the metal part of the fork, and the other end earthing the receiver. Siting the antenna near trees will reduce the signal strength somewhat, but if your receiver is sensitive enough, it may not be an issue; just something to bear in mind when choosing a location.
THE FEEDLINE
The signals from the antenna/earth system obviously need to be connected to the receiver, and the wires responsible for this are known collectively as the feedline. You could use coaxial cable for the antenna feed but the relatively high capacitance of around 100pF per metre for this type of cable will shunt a fair bit of the signal, possibly leading to disappointing results. I decided to make some ladder line for just this job, and with its very low capacitance between the conductors, it transfers the signals efficiently.
Making some spreaders |
Some home-made ladder line |
A length of ladder line isn't too difficult to make; I used some cable ties cut into 50mm lengths, and then carefully drilled a 2·5mm hole in each end. This happened to be just the right size for me to thread the spreaders onto the wires, the fit being tight enough that the spreaders didn't need gluing in place, or fixing in any other way. Of course, the size hole you'd need depends on the wire you plan to use, and a dab of silicone rubber would secure the spreaders if necessary. The spreaders were spaced at around 100mm intervals, but this is by no means critical; just use enough to support and spread the wires.
The length of ladder line is then run from the antenna and fork assembly, passing over grass and brambles, in through the car window and into the receiver. Due to the very high input impedance of my receiver, it was interesting to find that just rubbing the 'live' conductor with my finger produced a rubbing sound in the speaker; it's a very sensitive piece of equipment!
Notes:
It is worth mentioning that I now drive a separate earth stake into the ground beside the car door, and earth the car body (via the seat mounting bracket) and the case of the receiver, linking all 3 points. This for some unfathomable reason reduces the level of mains hum.
MAKING RECORDINGS
To record the received signals I use an oldish laptop running Windows 7, dedicated to recording duty only; older laptops sometimes use the cold-cathode type of screen illumination which creates a lot of hash, but this one uses LED illumination which is a lot quieter electrically. I discovered that having a USB mouse plugged in (I dislike those touch pads) caused noise to be radiated so that was removed, forcing me to use the touch pad.
My isolation transformer lash-up
As a final exercise in noise reduction, I fitted a 600Ω : 600Ω audio isolating transformer I'd scrounged from a computer internal modem card between the receiver's record output and the laptop's input which helped isolate more laptop interference.
Ultimately you may not need all of these measures, but this gives you an idea of things to try if you experience interference problems.
Back to top
What Can Be Received?
If the equipment is working correctly, the first thing you should hear is a crackling sound, the sound of the living planet, the atmospherics; this sound forms a permanent backdrop to the VLF spectrum in which other sounds may be detected. These crackling sounds are caused by lightning stikes around the globe, presumably the louder ones being closer in origin. The sound of the crackles range from a low popping through normal-sounding crackling to higher-pitched almost squelchy sounding components, which together cover the entire audio range.
I use Sonic Foundry's Soundforge 6 for my recording and audio editing; it's old software but does the job very nicely, as it always has done since I bought it. I discovered during editing that I could apply a notch filter at 50Hz 0·2 of an octave wide, which removed all of the residual mains fundamental. Interestingly, the only harmonic I could hear was the fifth at 250Hz, which I could also notch out using a width of 0·1 of an octave.
All my sound clips have been rendered in ogg format which is supported by all the major browsers; quite why people are fixated on the inferior mp3 format I don't know, maybe it's the result of agressive marketing back in the day when a licence fee was payable? In any case, mp3 performs poorly on transients, which these recordings consist almost entirely of; ogg format performs considerably better for a given file size, and is good ol' open source too, so there.
Slightly 'tweeky' atmospherics
This is a 75 second recording of the usual VLF background noise which is always present to some degree, although some days are more active than others. This was recorded at dusk in September 2025 and if you listen carefully, some tweeks are starting to become noticeable among the normal noise, making their characteristic 'pew pew' sound.
A pair of passing cars
If you are parked beside a road on a dry day, every time a car passes you will most likely receive the sound of the static electricity built up on the car's tyres. Not particularly interesting, but it does illustrate how sensitive this equipment is.
Insects flying close to the antenna
Here is the sound of a couple of insects flying past the antenna, the static on their wings making them audible in the receiver. This tends to occur mainly during warm days when the insects are more active, and while not related to VLF, a nice example of a fly-by is worth recording in my opinion; I'm still waiting for the perfect capture.
As of August 2025 I plan to add a lot more recordings to this section; I'm just waiting for the right conditions and phenomena to occur. I did record some whistlers recently, but the interference from the laptop spoilt the recording; I've now resolved that issue.
Back to top
This page is currently being added to (as of late 2025), so if you have the patience, do check back in a few weeks to see what else I've added.
INTERNET LINKS
The Inspire Project
"Since 1989, INSPIRE has provided radio receiver kits to over 4,000 students and other groups within the United States and internationally to make field observations of VLF signals."
I can be contacted at this address:
Copyright © Andrew Westcott 2003 - 2025
I'm happy for anyone to use this material for private, non-commercial or educational purposes, but credit to the author must be given. For any other use please contact me for permission.