VLF receiver for Solar flare detection

This receiver was built to monitor the 21.75KHz French naval transmitter at Le Blanc. At this low frequency, world wide communication is available because of the suitability of the waveguide effect of the ground and the ionosphere (in this case the 'D' layer). Since the ionoshere is caused by the interaction of the solar wind and the earths magnetic field, it is sensitive to solar flares, and the effect of this on the waveguides characteristic can be used as an indication of the suns activity.

The spectrum analyser response opposite shows what was picked up on a 44 turn loop antenna (200 x 500mm), operating indoor. A span of 50KHz (5KHz/division) is shown - the verticle scale being 10dB/division. Several signals can be seen between 15 and 25KHz in addition to the French one.

To differentiate between the various signals, six tuned circuits are incorporated into the receiver, these being salvaged 10mm 455KHz IF tranformers with extra capacitance added to reduce the resonance to 20KHz or so. Despite all of these items originating from differant equipment, it is interesting to note that 100nF was the right value of extra capacitance to add in each case. In fact, the loop antenna is also resononated by the receivers input capacitance, so additional selectivity to that shown below is obtained.

Left:
Selectivity provided by the six tuned circuits.

 

 

Right:
The effect of this selectivity, when inserted between frame antenna and spectrum analser.

Circuit:


Three TL071 op-amps are used, giving voltage gains of 10, 67 and 200. Given that the selectivity networks are quite lossy, this actually results in a power gain of about 80dB. A simple forward biased diode is used as a detector, and the output drives a 50uA meter to indicate signal level. From the same detector output, a comparator is added which drives a LED to act as an alarm, indicating that signal strength is higher than normal ( ie, indication that a solar flare has probably just occured.

Surface mount components are used since this greatly reduces the number of holes that need drilling on the pcb...

'Sunrise effect':

The following Excel plot of detector output level, taken over a 24 hour period, shows the 'D' layers characteristic daily fluctuation. When refered to the noon time level, signals are about 3dB higher throughout the hours of darkness. More interesting, though, is the short-term dip of about 10dB in level just after sunrise.