I present here a regenerative loop antenna which supplies signals to a high frequency "table top" receiving system based on the Ten Tec RX320D software defined receiver, the Gateway E2500S desktop computer,and the ACER 15.6 inch LCD monitor. This system is one approach to reception of HF signals from a variety of sources without an outside antenna. Reception of high frequency signals without a wire antenna in the open is considered to be a distant second best option. It is rapidly becoming a necessity. The system presented here is very much a work in progress.

The tuning range of this system as of September 18, 2009 is from 7 mHz to 15 mHz.

Sustained, controllable feedback is achieved by connecting the feedback winding as the entire drain load of the tuned JFET stage. Circuit details are shown in the schematic diagram. The signal winding is bent from a single length of number 6 AWG solid copper wire 11 inches on a side. The feedback winding, wound with number 22 AWG enameled magnet wire is 1.75 inch(4.45 cm) inside the signal winding. That's 2.5 inches(6.35 cm) measured along the X-frame diagonal. The X frame support for the antenna windings was made from two 3/4 inch by 1 1/2 inch strips of pine. LTSPICE modelling has shown that the overall system performance is relatively insensitive to coupling coefficient between these two windings, but that does not say anything about the relationship between signal capture and winding separation nor does it say anything about the relationship between winding geometry and coupling coefficient. Control of regeneration is achieved by variation of applied drain voltage using an LM317 variable voltage regulator.

The antenna electronics enclosure is connected to the antenna by three parallel runs of cable, two of 50 Ohm coax and one of 300 Ohm twin lead, each 21 inches(53.3 cm) long. During operation the antenna enclosure and the antenna are separated as much as possible.

There is an additional control function of this regenerative loop antenna that should be elaborated upon: An H11AA1 optical coupler is used to vary the capacitance from the top or hot end of the feedback winding to system common or ground. In other words the resonant frequency of the feedback winding is controlled. As the resonant frequency of the feedback winding is increased, it having initially been lower than the resonant frequency of the signal winding the frequency response curve of the signal winding circuit becomes broader. As these two peaks are brought closer to one another the effective Q of the signal input circuit is lowered thereby making the onset of regeneration more gradual, and overall gain is reduced. And, overall gain can be increased by moving the two resonances further apart.

There is a somewhat novel output circuit that should be described. The feedback winding is wound together with a separate output winding of number 22 AWG enameled magnet wire. Each of these windings has the same number of turns. The output winding drives the gate of the output JFET directly. This output JFET is connected as a source follower with the source resistance capacitively coupled to the Ten Tec RX 320D 50 Ohm input. The antenna output impedance should be relatively independent of frequency. The tuned input jfet circuit runs on a separate 9 volt battery from the jfet output circuit which has its own 9 volt battery. This configuration is something of a luxury, but it works well; it's very stable. Battery life in this application is still an open question.

I strongly recommend the use of small diameter coax as shown on the schematic inside and outside the antenna electronics enclosure. I built my electronics on a small aluminum cake pan. The output jfet circuit is built on a separate circuit board near the output end of the enclosure.

Throughout the development of the regenerative loop antenna system LTSPICE has been employed to optimize tuning range,output signal levels and control function. Values of inductance for the antenna windings were chosen iteratively with LTSPICE and the online rectangular inductance calculator provided by the EMC Laboratory of the Missouri University of Science and Technology. Tuning is, as a result right where it should be.

To achieve optimum performance, it is necessary to try different combinations of signal, feedback and output winding polarity. Of course there is the basic problem of getting the feedback winding arranged for positive rather than negative feedback,but,further,it is necessary to experiment with the two possible polarities of the signal winding and the two possible polarities of the output winding.

This is a very entertaining system to operate. One can enter 9.750000 mHz, peak the antenna there,then enter 9.50000 mHz and proceed to tune up the 31 meter SWBC band in 5 kHz increments by clicking on > to the right of the digital frequency display. Interesting signals can easily be saved for later listening. Computer noise is there, but not oppressive. The spectrum display featured by the RX320D is very useful; it can be watched while tuning. Strong signals show up as obvious peaks, and they usually fall on the standard 5 kHz shortwave BC frequency assignments.

Actual use of the regenerative loop antenna with the RX320D receiver has clarified operation of the controls of the regenerative loop antenna: 1.) Turning the knob of the signal circuit gate bias potentiometer doesn't have an obvious effect on anything. Set it about 30 degrees clockwise and leave it there. This is a development or breadboard unit. 2.) The feedback winding tuning knob is used to peak the received signal. 3.) The drain voltage should be set for maximum gain without oscillation. Oscillation is more or less obvious if you seem to be tuning in signals with the regenerative loop antenna main tuning knob. Back off the drain voltage just a little counterclockwise. Note also that the second small knob from the left on the antenna electronics front panel is not at present serving as an active control. It may be utilized for fine tuning in the future.

My regenerative loop antenna is separated from the PC controlling the RX320D by about 9 feet, and it's about five feet above the floor. I can receive more SWBC signals this way than I could with an earlier receiver connected to an outside antenna of about 30 foot length, however this system is not optimized for DX. Some frequencies are unavailable because of noise from the PC and/or the monitor. I may have the option of using some coax installed in our home for TV purposes to "remote" the regenerative loop antenna to a closet over 20 feet away. That would not be difficult, The regenerative loop antenna output can drive 50 Ohm or 75 Ohm coax. Coordination of receiver and antenna would then require me to walk between rooms to retune the antenna. But for now I have a good SWBC receiving system to listen too while I am doing other things. As I write this I am listenening to Deutsche Welle in the 25 meter band in English a little past 2100 Z. The signal strength is not what it will be 3 hours from now, but the signal is there,with QSB. It's intelligible on peaks. Some computer noise is heard on this particular frequency when the DW signal fades. I have peaked the antenna on DW. So far, the best time for SWBC listening here is between 2300 Z and 0200 Z on 31 meters.

There are, of course commerciallly available active antennas. I don't know how they compare with this regenerative loop antenna.

Please note that on the schematic diagram "#1", "#2" and "#4" refer to three diagrams also numbered 1,2 and 4.

One important improvement that could be made is the inclusion of protection for the tuned circuit jfet gate from lightning induced "spikes". This is not a hypothetical problem. I have already had to replace one jfet so destroyed.

Bill Young WD5HOH