Super-regenerative Receiver for Scalar RF From: Bob Shannon Cc: freenrg-l@eskimo.com Subject: Re: Tesla's Longitudnal Waves Resent-Date: Thu, 6 Feb 1997 11:33:54 -0800 I'll give you a thumbnail sketch of the design : The sensor itself is an air core coil placed in the field of a fully shielded magnet, much like the Dea/Faretto design. This coil is tapped, and connected to a NTE312 jfet oscillator based on a VFO circuit lifted from the ARRL handbook. A variable capacitor tunes the coil to resonace in the frequency range of interest. My prototype also included a latching relay and a set of varactor diodes, but I digress. This VFO oscillators power supply runs throgh a 1:1 ration audio isolation transformer (a radio shack part, 1k to 1k interstage.) and then to a zener diode regulator that is well bypassed. A trim pot voltage divider allow the DC bias to be set for proper operation. The other winding of the transformer runs to an XR2206 function generator, which modualates the DC supply to the VFO section with a 25 to 30 Khz triangle wave. The output level of the function generator is set to bring the VFO above and below its start up and shut down voltages, so the RF VFO goes into and out of oscillation at the frequency of the function generator. The DC bias and modulation levels must be carefully set so that the VFO both starts and stops as a function of the quench oscillator. This is critical, and some test equipment is needed for checkout and calaibration. A good oscilloscope is mandatory here. This is much like a super regenerative radio receiver design of the 1930's or early 40's, but there is no antenna circuitry. The XR 2206 function generator serves as the 'quench' oscillator. The output of the VFO section is coupled to a transistor detector stage, where it is rectified and integrated. A filter removes any traces of the VFO frequancy or function generator frequancy. The detector feeds an audio amplifier and output driver stage. In operation, the low noise jfet will begin oscillation as the VFO supply increases. The point at which the VFO starts is a function of the noise of the jfet, and the energy present in the coil itself. As the quench oscillator supply drops due to the action of the negative swing of the quench oscillator, the VFO will remain in oscillation for a time also controlled by the jfets internal noise, and the energy in the coil. So the VFO starts oscillation sooner if any energy is present in the shielded coil at near the frequency of the tank circuit, and remains in oscillation longer due to this same energy in the coil. The detector and integrator stages produce a signal whos amplitude is a direct function of the period during which the VFO is in oscillation. Any indicent scalar wave will effect the permanent magnets field in which the coil imbedded, just as in the Dea/Faretto design, as well as the Barkhausen effect designs. This magnetic modulation by an incident scalar wave will deposit energy into the coil, and alter the startup and run times of the VFO, and produce a change in the signal level from the detector and integrator. This will 'detect' several forms of modulation, including SSB, AM, FM, and CW emissions. All electronic circuitry is shielded from the coil and magnet assembly in a chassis bolted to the magnet housing. The varactor diodes are fed with a ramp waveform to sweep them through their range of capacitance. This ramp waveform is also used to deflect the X axis of an external oscilloscope, and the detectors output drives the Y axis. The ramp waveform is in the order of 100 hz or so. The display on the oscilloscope there by forms a primitive scalar spectrum analizer when the latching relay selects the varactor diodes rather than the manully tuned variable capacitor. Early prototypes used a 555 timer as the quench oscillator, but this was far from satifactory, as the 555 timer injected large spikes into the analog ground, which ran through the VFO sections, etc. This is resolved by the use of the XR2206 function generator, but you could also use a XR8038 or other function generator as well. I found the placement of the feedback tap on the coil was very important to getting a spectrally clean signal from the VFO section, and I also needed to place a few ferrite beads on the feedback line to get a clean sine wave. If the oscillator has harmonic content, the selectivity of the detector will suffer accordingly. This design is also rather old today (about eight years now), and I suspect that far better (lower noise) jfets are commonly available. With this description, it should be fairly easy to design and construct a super regen scalar detector without my schematics at all. The circuitry is a bit more complex than some detectors, but the fact that the design is linear, in that is will reproduce the actual scalar signal, and is tunable over a range of frequancies, and can easily perform spectral analisis via the varactor diodes makes it a very good detector design in my opinion. It is also quite sensitive! Care must be taken to prevent any EM leakage into the coil and magnet assembly, which would cause false signal detection. This basic design can be adapted over a wide range of frequancies, covering the whole of the HF spectrum. Care must be taken not to attept to tune the VFO over too wide a range of frequancies in a single unit, as the DC bias and modulation control settings will need to be altered if the VFO frequancy is changed by too large a degree. Of course all conventional design rules for ham radio quality VFO construction must be given proper attention, such as reducing microphonics, etc. I reccomend the use of 'pad per hole' perf board for this reason. Hard core ham equipment builders oftem like 'dead bug' construction on a copper sheet (ugly!) and then pot the finished circuit in epoxy or some similar material. Consult the ARRL 'good book' for this info. Other than this, it's a really soild, reliable dsign. Heat up your soldering iron, and go for it! Just don't try to get any QSL cards from any transmissions you might detect with this thing! ( Especially if they came 'postage due'. ;-> ) Have fun, Bob Shannon.