EWE Shortwave Antenna

Home | John | Connie | Publications | Software | Correspondence | NtropiX | NdustriX | NformatiX | NdeX | Thanks


The EWE Antenna was first described by Floyd Koontz in QST Volume 79, 1995, (a reprint of original is available at

Since the antenna has a useful bandwidth of about 2:1, and the short wave bands center around 10 MHz., 10 MHz. would be a reasonable design criteria. 10 MHz. has a wavelength of 100'; the antenna should be 1 / 4 wave length long, 1 / 8 wave length on the top, and 1 / 16 wave length on both sides, or 12.5' on the top, and 6.25' on the sides. The feed point impedance should be about 500 Ohms, and the termination resistance slightly less than twice the feed point impedance. An 820 Ohm metal film/low noise resistor was used for the termination. The antenna is directional (receiving radio waves passing from the termination end, to the feed end,) at lower frequencies in its bandwidth, omnidirectional at higher frequencies, and becoming directional in the opposite direction at frequencies much higher than the design criteria.

Notice how small the antenna is, 12.5' x 6.25', which is adequate for a general purpose, or portable antenna-and works well as an internal antenna.


Figure I. EWE Shortwave Antenna

Figure I, (which is available in larger size jpeg, or xfig, formats,) shows the attention to detail on the antenna lead in, impedance matching, etc. The antenna, (in the upper left of the diagram,) has two 2' copper rods, (purchased from the local home improvement center-they are about a quarter inch in diameter, but any metal rods will work,) driven in the ground 12.3' apart. These rods are connected by a buried bare copper cable, (I used antenna wire, but any copper cable-like stripped lamp cord-will work.) There have been reports that the EWE antenna is sensitive to ground resistivity, and this technique was considered advisable. Each ground rod has a water proof electrical utility box, (available from the local home improvement store,) mounted at ground level, with one containing the 50:500 Ohm balun, the other contains the 820 Ohm termination resistor. The antenna wire runs from the top of the balun up, (for 6.2', across for 12.3', and down 6.2' to the termination resistor. The antenna's listening direction is from the balun to the termination resistor, (i.e., the antenna receives signals from the left, to the right, in Figure I.)

A word about antenna lead in:

  • Always use RG8 for shortwave-its industrial grade, (the PL259 connectors are expensive,) low loss, but worth the money in maintenance; RG8 and PL259's are rugged and durable.

  • If the antenna is mounted in the air, (like a T2fd, or dipole,) the lead in should drop straight down to ground.

  • The lead in should not run in the air, horizontal to the ground-laying directly on the ground is best, (buried an inch, or so, is even better.)

  • Always use a balanced, (NOT an auto-transformer,) balun with two windings, mounted at the antenna, (again, these are expensive, but not that difficult to make-parts are available from Amidon Associates and Allied Electronics, etc.) If the antenna is balanced, (like a T2fd, or dipole,) use a grounded center tap on the antenna winding, (for static electricity bleed off.) If a center tapped balun is not available, add two 1 M Ohm resistors between ground and each antenna terminal.

  • Use a surge protector in the lead in to protect the receiver. These are pricy, but last a lifetime, (unless they fail shorted protecting the receiver.)

  • An optional addition to the lead in, (I always use one,) is a CMR, (Common Mode Rejection,) transformer-it is nothing more than about 3' of RG8 with a PL259 on each end, wound through a toroid core, (I use T-200-2 cores, available from Amidon Associates.) About 10 turns, (its not critical,) is all that is necessary.

  • I also wind the receiver power cord through a toroid, (about 10 turns-again, its not critical, same type of core,) to prevent hash, (there is a lot of it in modern homes, from CRTs, TVs, light dimmers, etc.,) from entering the receiver.

Note that these are general considerations for getting the antenna signal to the receiver, while limiting stray pickup of RFI, etc.

A General Discussion of Short Wave Antennas

Many designers will use design equations for transmitting antennas for receiving antennas, and then via reciprocity arguments, optimize the receiving antenna.

However, it is more complicated for receiving antennas. Specifically because of atmospheric noise considerations.

  1. From "HF Antennas For All Locations," L. A. Moxon, G6XN, ISBN 0 900612 57 6, Radio Society of Great Britain, Lambda House, Cranborne Road, Potters Bar, Herts EN6 3JE, pp 127-128, atmospheric and man-made noise decreases at 20 dB per decade, (6 dB per octave,) below 12 MHz., (and is 13 dB higher at night.)

    1. Daytime noise at 1.75 MHz. = 47 dB above 0.07 uV = 16 uV and decreases at 20 dB / decade to many tens of MHz.

    2. Nighttime noise at 1.75 MHz. = 60 dB above 0.07 uV = 70 uV and decreases at 20 dB / decade to 14 MHz. then at 40 dB / decade to 20 MHz., intersecting daytime noise, and is equal to daytime noise for frequencies > 20 MHz.

  2. From "Meteors in the Earths Atmosphere," Edmond Murad, Iwan Prys Williams, Cambridge University Press, 2002, ISBN-10: 052-1804310, ISBN-13: 978-0521804318, pp 143, section 6.6.5, cosmic nose temperature, Tcos = 150 * (lambda^2.3), lambda in Meters. The noise power, P = K * Tcos * delF, K = Boltzman's constant, 2 * (10^-21). For 17.5 MHz., lambda = (1000 / 17.5) / 3.28 = 17.4 M. P = (2 * (10^-21)) * (17.4^2.3) * 6000 for a 6 KHz. bandwidth, delF, which is 8.56 * (10^-15) Watts. Finding the voltage, V, in 50 Ohms, (V^2) / 50 = 8.56 * (10^-15), or V = 6.54 uV. Or, the cosmic noise at 17.5 MHz. is 6.54 / 0.07 = 9.34, or 19.4 dB. This agrees reasonably closely, (within a factor of about 2,) with the Moxon reference at 17.5 MHz.

At 10 MHz., the atmospheric noise will be about 2 uV, (daytime.)

Now, assume the characteristics of a good receiver. From "R8 Communications Receiver: Owner's Manual," pp 4, 5.0-30 MHz, (preamp on,) the AM sensitivity for 10 dB S + N / N, 1000 Hz., 30% modulation,) is less than 0.8 uV. 10 dB is a ratio of 3.1623, or the noise floor would be (0.8 uV + N / N) = 3.1623, or N = 0.37 uV.

Note that the atmospheric noise is 5 times as large as the receiver noise-which is the rule-of-thumb for receiver designers-if the noise level jumps when the antenna is plugged in, the antenna is working fine, (this comparison should be made with a dummy load vs. the antenna-but just plugging in the antenna and looking for an increase in noise is a good reference.)

As a tutorial example, suppose a signal level is equal to the atmospheric noise level, and suppose the antenna has a gain of 0.2, (-14 dB,) then the noise level increased by 3 dB, (remember noise adds root-mean-square, and the antenna noise and receiver noise are equal in this example,) so the signal is -3 dB below the noise at the receiver output. Note what happened: the antenna gain was -14 dB, which decreased the signal level, relative to the noise level, by only -3 dB at the receiver output.

The EWE antenna gains run about -14 dB, typically, meaning if lead in losses, impedance matching, stray noise pick up, etc., are well managed, the antenna can exhibit good receiving characteristics.


On 1200-0800 GMT, inclusive, on June 12 thru June 15, 2007, the r8.tar.gz program for controlling the R. L. Drake R8 Communications Receiver via an RS232C serial port, (see: r8.txt for particulars,) was used to scan the frequencies listed in, (from, the Solar X-rays were Normal, and, Geomagnetic Field was Quiet.) The results for world-wide English transmissions were:

  1. 69 countries listed, positively logged 28 = 40.6%

  2. 505 frequencies listed, positively logged 84 = 16.6%

Of those English language broadcasts logged, 43 were from ROW, (Rest of world,) directed to North America, 41 were directed from ROW to ROW. (Broadcasts originating in North America were omitted from the search.)

The EWE antenna was located at Lat: 37 deg., Lon: 121, aligned 29 degrees East of True North.


The EWE antenna was compared with an "A-B" switch with the Terminated Tilted Folded Dipole Shortwave Antenna, and compared favorably. The T2fd averaged about 4 dB lower noise figure than the EWE antenna across 5.75 MHz. to 18 MHz. The EWE was usable down to 3 MHz., the T2fd was not. The T2fd had better man-made noise, (i.e., local CRT RFI, etc.,) performance, (but that's the T2fd's forte.)

Note that the T2fd's performance was slightly, (4 dB,) better, but when ease of maintenance and construction are considered, its doubtful that the T2fd is worth the additional effort, (the EWE is totally in reach while standing on the ground.)

In an informal comparison, the EWE was compared against a random long wire, (10' off the ground, 40' long, jury rigged,) running through a 50:500 Ohm balun, and the signal to noise ratio was slightly better with the EWE. Additionally, the EWE was slightly better with local RFI pickup, (but not as good as the T2fd.)


A license is hereby granted to reproduce this design for personal, non-commercial use.


So there.

Copyright © 1992-2008, John Conover, All Rights Reserved.

Comments and/or problem reports should be addressed to:

Home | John | Connie | Publications | Software | Correspondence | NtropiX | NdustriX | NformatiX | NdeX | Thanks

Copyright © 1992-2008 John Conover, All Rights Reserved.
Last modified: Fri May 16 12:15:27 PDT 2008 $Id: index.html,v 1.0 2008/05/16 19:15:45 conover Exp $
Valid HTML 4.0!