All images and schematics may be enlarged by clicking them
Receiving VLF amateur stations is not that difficult once one gets into the mindset that we are dealing with very weak signals buried in noise. Forget trying to hear stations. The only way of detecting these signals is by patient monitoring of accurately known frequencies over periods of many hours or even days in extremely narrow bandwidths. Thankfully this is now possible thanks to software freely available on the internet. You don't need a very high spec conventional receiver costing thousands of pounds! All the kit (apart from the PC) will cost you just a few pounds to make.
Any of the circuits shown below fed into the soundcard of a PC running software called Spectrum Lab (running a suitable configuration file) will allow you to copy amateur stations hundreds of kilometres away.
It is recommended that the active antenna should be mounted well in the clear and away from the house feeding the power up the coax). This is to avoid noise pick-up from mains wiring, switched mode PSUs and TV line timebases.
It works because the FET has a very high input impedance. The output of the FET is converted to a low impedance using an emitter follower. Although the signal level may be lower than a large outside antenna so is the noise and overall the S/N may actually be better if the antenna is located carefully. Overall, if you are unable to erect a decent LF antenna then this may be a solution which will allow you to hear signals on 8.97kHz, 136kHz, 500kHz and 1.8MHz. For VLF reception there may be some advantage in adding a low pass filter between the FET and the transistor to reduce the levels of LF and MF broadcast signals that might otherwise overload the unit. I've not tried this.
Roelof PA0RDT has tried the active probe antenna at 8.97kHz and says it works well with his Perseus receiver. Gain at this frequency may be increased (along with the noise though) by increasing the capacitors to 1uF and increasing the power supply feed choke to 8.2mH. This is the schematic from PA0RFT's site. Power for the unit is supplied along the feed cable. It is advisable to separate the ground from the house ground and use a suitable isolation transformer feed. More details in the link above.
Note that the MOSFET first stage can be remotely mounted with signal and power fed to it via a twisted wire pair.
---- A RECOMMENDED CIRCUIT -----
This video describes my current set-up for VLF radiated DX reception
The same basic design can be converted into a tuned E-field probe by using the FET in
There are basically 3 frequency response determining elements in the design. The first is the loop inductance, the shunt input capacitance C8, and the input resistance of the preamp (about 300R, determined mostly by resistors R1, R2 and R4, which also set the gain of the input stage). These together form a singly-terminated, second-order low pass filter with a roughly Butterworth response and cut-off frequency something over 20kHz. The preamp stage gives about 30dB gain, and a reasonably low noise level - evidently below the external band noise. This was arrived at after some back-of-envelope calculations and tests of likely noise levels, induced EMF in the loop, and so on.
The second element is the op-amp gain stage IC1A. This is configured as a "lossy integrator", with a gain that rolls off at 20dB/decade in the operating frequency range. This compensates for the loop output, which rises at 20dB/decade below its cut off frequency, due to the loop EMF being proportional to d(phi)/dt. It isn't essential to do this, but I felt this would be useful in comparing noise levels, etc., and reduce the likelihood of overloading by VLF utilities. At 9kHz, the gain of this stage is about 30dB also; the overall 60dB gain gave a good signal level for the audio line input socket on my lap-top, determined by trial and error.
The final element is the bandpass filter. This was a straightforward ladder filter designed with a n=3 Butterworth response, with the component values fiddled a bit to suit off-the-shelf inductors. I aimed for an impedance level that could easily be driven by normal op-amps. The filter is terminated by 1200ohm resistors R10, R11, which are necessary to give the correct response (actually, the initial design values were about 900ohm or something, but I increased the value to "absorb" the loss resistance of the rather low-Q 4.7mH inductors). The IC1A stage has a high input impedance that does not load the filter significantly, but the input preamp stage has a fairly high output impedance, and its gain would also be affected by connecting the filter as a load. So the IC1B buffer stage was included, which also provides 9dB gain to make up for the 6dB loss between the terminations, plus approximately 3dB insertion loss of the filter.
So all the parts of this circuit were designed to work together - if you wish to build a circuit with a different antenna, or use the filter in a different system, you need to take similar considerations into account - there are a near-infinite number of feasible solutions that will perform as well or better, but all will have differences.
Cheers, Jim Moritz
73 de M0BMU
A later version of M0BMU's preamp was created which included a direct conversion VLF receiver. This is useful for checking local signals, for example when doing earth mode tests. Jim writes:
Regarding portable receivers for 9kHz, I have now added a BFO/product detector/AF output to my 9kHz loop/preamp/filter circuit. This effectively converts it into a small, self-contained fixed-tuned upper sideband receiver for headphone aural reception around 9kHz. The sound card preamp function is still available of course. The main idea of this was to have a portable receiving system for investigating noise sources at 9kHz, but it would also work well for near-field transmission experiments that people are trying. I also have an up-converter which can be used in conjunction with an FT817 as a portable 9kHz RX, but the dedicated circuit is much smaller and simpler, and more convenient. It only consumes about 20mA or so from the 12V supply, so only a small battery is needed.
The attachment shows the complete circuit - The 7.5kHz BFO and product detector uses a 4053 CMOS analogue switch IC as a combined mixer and RC-tuned oscillator. There is a bit of an explanatory diagram at the bottom of the schematic as to how this works... The BFO frequency stays within about 100Hz, which is fine for aural reception - for narrow band modes needing higher stability, down conversion can instead be performed by a PC and sound card of course.
The overall audio gain between loop antenna input and headphone output is very high, and when the "wide band" 3 - 22kHz bandwidth is selected, oscillation can occur due to internal ground loops, or external coupling between loop antenna and headphones. But for listening purposes, the wide band setting is not very useful anyway, and feedback is not a problem when the bandpass filter is selected since then input and output are in different frequency ranges. The complete circuit is contained in a small diecast box - however, I found this is not very effective as a shield against magnetic fields at 9kHz - best to keep it some distance from any lap-top or PC which it is connected to.
It is quite interesting to walk around with loop and RX listening to the wierd buzzing and whirring man-made noises and mains hum that is superimposed on the background QRN around 9kHz, and a good way of checking the suitability of a receiving location for future 9kHz tests. I have found that if you stand near overhead power lines, the noise present sounds quite different depending on the orientation of the loop with respect to the overhead wires. I guess this is due to different levels of 50Hz harmonics being present in the differential-mode and common-mode currents flowing in the lines. If you try this, you can expect very strange looks from passers by ;-)