|
|
|
|
|
Design notes: Omnidirectional
VLF E-Field Receiver This active antenna and receiver was designed as part of my PhD research to receive VLF, Omega and time signals. The antenna is nothing more than a 15cm straight monopole with a very high impedance (JFET) front end and with gain and an anti-aliasing filter. A good ground plane or counterpoise is needed (maybe a 10cm copper or tin disc with a very low capacitance feed-through or hole at its centre). The circuit runs happily at 3V and should draw less than one milliamp. This means it is suitable for running from two or three AA, C or D cells located with the antenna. The circuit uses a TLC27M4CM, which is a surface mounting op-amp. Many other op-amps will do the job, but look for a micro-power, low noise version.
(Click for full size image) Short VLF antennas are inefficient, however a 15cm monopole connected to this circuit and placed in the open (away from buildings or on a roof or mast) will ensure that VLF signals and VLF noise (lightning) are received with at least 10dB margin above the noise generated within the amplifiers and front end. One thing to watch out for is intermodulation caused by overloading the front end amplifier or later stages. This could happen if there is a strong out of band (eg. LW, MW, CB, UHF or mobile) transmission in close proximity. The E-field probe antenna and front end have a very high impedance, so appropriate filtering is difficult to achieve. This is largely because, in this circuit, it is critical to keep the path between the probe and ground as high impedance as possible. Even a few stray picofarads of capacitance will prevent a signal reaching the JFET input. Frequency
response The front end response extends from around 2kHz up to 40kHz, however a low-pass anti-aliasing filter (IC1a) has been included with a cut-off of around 14kHz. This allows signals to be sampled at around 28ksps without higher frequency signals causing interference. If you require a higher frequency response, (for example to receive 60kHz time signals) then it would be wise to change the filter parameters. You may well find that your standard PC sound card can accept signals directly from the output of this receiver. The receiver is certainly capable of driving 10m or more of 50R coaxial cable. Circuit
details Q1 and Q2 form the front end low noise amplifier with a high input impedance. They are connected in a cascode configuration which at high frequencies has the effect of reducing Cgd. We observed a small benefit even at low frequencies because of the input impedance. IC1C is a 20dB buffer. IC1A forms a low pass filter and a split supply is generated by IC1B. IC1D is another 20dB gain stage. R16, R17 and R18 ensure the signal has a preset offset voltage if an ADC is DC coupled. R18, R16 and C9 may be omitted if AC coupling is required (R17 connected to ground). Most of the circuit should be shielded to prevent feedback back to the sensitive input. However – you must ensure that capacitance between probe antenna and ground is minimal, (1-2pf at most). This basically means mounting the first transistors and probably most of the circuit at the base of the antenna assembly. One trade off you may be able to make is that more signal may be found by proportionately increasing the antenna length – and thus lowering its impedance. Just be certain that you are not overloading any of the amplifier stages by observing the output on an oscilloscope. Note that VLF signals are limited by atmospheric noise, so increasing the antenna (probe) length will not increase the overall SNR! What you could
receive This circuit was designed to receive Omega VLF signals from 10kHz to 13.6kHz. Omega now no longer exists, but it was also possible to receive CIS Alpha radionavigation signals in a similar frequency band, 16kHz (and 60kHz without the antialiasing filter) time signals and 19kHz communications signals for places as far away as Argentina, Krasnodar, Norway, Liberia, Rugby and Criggion. These were just a few of the full range of VLF signals we received that were positively identified from their modulation and multiplexing. You will also see large time-domain spikes (noise impulses) generated by atmospheric electrical activity from right around the planet. I regret that I cannot devote much time to answering questions about this circuit, but please do let me know if you build it and receive anything on it. John Bishop Tel: Bath +44 (0) 1225 329 529 Email: john.bishop@bathlabs.com |
