An Attic Coaxial-Cable Trap Dipole for 10, 15, 20, 30, 40, and 80 Meters

John DeGood, NU3E


A coaxial-cable trap dipole antenna installed in the attic provides a surprisingly effective solution to HF operation on the 10, 15, 20, 30, 40, and 80 meter amateur bands at a QTH with restrictive covenants that prohibit outside antennas.

Restrictive Covenants

When we purchased our first home in 1980 amateur radio antenna siting was a top selection criteria. But when a job change in 1995 required relocation, my XYL announced that it was "her turn" to choose our new QTH, and amateur radio was not on her priority list! She chose a beautiful new home in a development with excellent amenities for raising our family, but it came with restrictive covenants that prohibit any outside antenna other than a "small antenna for television reception." I feared my HF operating days might be over.

My early HF operating attempts at the new QTH were not encouraging. The landscaping on our new lot consisted of ornamental trees and shrubs that were barely taller than myself. I tried a full wave horizontal loop of nearly invisible small gauge wire which circled the house hanging below the aluminum gutters, but its performance was disappointing and it caused severe RFI problems, forcing me to limit operation to QRP power levels. I next tried an inverted vee using the same stealthy wire, with the peak supported by the house and the ends supported by ornamental shrubs at corners of the lot. It performed as a classic "cloud warmer" that worked for local contacts but it was a lousy DX antenna. And the low height of the shrubs that served as end supports made mowing the lawn look like I was practicing for a limbo contest!

Attic Installation

One day while staring at our lot I considered the attic as a possible antenna location for the first time. Some of the positive attributes were: But there were negative attributes, too: Our attic consists of 2x4 wood truss construction. The ridge of the main span is approximately 44 feet long with a non-metallic ridge vent. The roofing material is asphalt composition shingles. The siding and soffits are vinyl. The roof is a 12" pitch (i.e. a 45 degree angle) which results in a tall ridge height. The plumbing vent stacks are PVC plastic. The only significant metal objects in the attic are various runs of electrical wiring that service the 5 smoke detectors and 3 ceiling fans installed in the second floor ceiling, and two lengths of flexible ductwork. As attics go, I consider ours is very amenable to the presence of an amateur HF antenna.

After consideration of the alternatives, I chose to construct a trap dipole antenna in my attic using coaxial-cable traps. I desired multiband capability, and selected a single trap dipole over parallel dipoles or a hybrid design consisting of 2 or more trap dipoles in parallel. Parallel dipoles are more difficult to tune due to interactions between the elements. Also, antenna traps function as loading coils below their resonant frequency and result in a shortened antenna: by using a single dipole design with multiple traps I was able to fit 40 meter coverage comfortably along the 44 foot main ridge of my attic roof. I included 80 meter coverage by adding a pair of 40 meters traps and making a right angle bend at each end of the attic, continuing the 80 meter segments down a few inches below and parallel to the slope of the roof. Since most of the current, and hence most of the radiation, comes from the central portion of a half-wave dipole, this is a reasonable compromise.

An antenna tuner could also be used to accomplish multi-band operation in conjunction with a non-resonant antenna. I prefer a resonant trap dipole design for the following reasons:

Coaxial-Cable Trap Construction

The clever use of coaxial-cable to construct antenna traps was first described in the amateur literature by Johns in 1981.[1] Coaxial-cable traps are inexpensive, easy to construct, stable with respect to temperature variation and capable of operation at surprisingly high power levels.[2,3] The traps used in this antenna are based on the "optimized" design graphs derived by Sommer.[4]

Coaxial-cable antenna traps are constructed by winding coaxial-cable on a circular form. The center conductor of one end is soldered to the shield of the other end, and the remaining center conductor and shield connections are connected to the antenna elements. The series-connected inner conductor and shield of the coiled coaxial-cable act like a bifilar or parallel-turns winding, forming the trap inductor, while the same inner conductor and shield, separated by the coaxial-cable dielectric, serve as the trap capacitor.

The resultant parallel-resonant LC circuit exhibits a high impedance at the resonant frequency of the trap and effectively disconnects everything after the trap from the antenna. Any inner traps (which are operating below their resonant frequency) function as loading coils and shorten the overall physical length of the antenna.

cutaway view of trap connections

I constructed my traps using good quality RG-58/U coax scavenged from a discarded 10BASE-2 Ethernet cable. PVC couplings were used for the trap forms: PVC couplings are very inexpensive, readily available in useful diameters, and can be purchased individually, whereas PVC pipe is usually sold only in 10 foot lengths. 14 gauge solid wire was used to form "bridle wires" for electrical termination of the coax and electrical and mechanical termination of the antenna wire elements.

Coaxial-cable traps must be "tuned" before use. The coax turns were spread slightly until the desired resonant frequency was reached, as measured by a dip meter whose signal was monitored on a nearby calibrated receiver. After adjustment the coax turns were secured by coating with lacquer (I used Deft® brand left over from a furniture finishing project.)

Here are several important points to keep in mind if you attempt to build coaxial-cable traps:

  1. The coaxial cable length specified is the active length in which the shield is intact. The overall length of coaxial cable required, including the prepared ends, will be approximately 2-3 inches longer.
  2. The outside diameter of the trap form is a critical dimension. I used Schedule 40 PVC couplings as trap forms (NOT Schedule 40 PVC pipe!) Note in Table 1 that the nominal size of the PVC couplings represents the PVC pipe size the coupling is designed to join, which is significantly smaller than the outside diameter of the coupling.
  3. Coaxial-cable traps have a relatively high Q, which results in a relatively sharp frequency resonance. You must adjust (i.e. tune) the traps or the antenna will not work properly, as traps can't do their job if they don't resonate (i.e. become a high impedance) at the correct frequency.
  4. If you don't have a dip meter, you can use an HF antenna analyzer, such as those made by Autek or MFJ, to adjust the trap resonant frequency using the procedure in the section titled TRAP FREQUENCY MEASUREMENT at With either a dip meter or antenna analyzer you will get the most accurate result by using the minimum coupling between the trap and the measuring instrument which produces a discernable dip.
  5. Unlike traps made from a discrete inductor and capacitor, coaxial-cable traps at resonance, i.e. in their high impedance state, exhibit a different amount of end loading depending on which end faces the center of the antenna. Either orientation works, but to maintain dipole symmetry trap pairs should always be installed symmetrically. I use the easy to remember rule, "Connect center conductor of trap coax toward center of antenna."

Table 1. Specifications of the traps used in this antenna
design frequency
trap form
coax length (shield intact)
coax turns (approximate)
actual frequency
10 meters
28.85 MHz
1.375" OD (3/4" PVC coupling)
28.5 MHz, 28.7 MHz
15 meters
21.225 MHz
1.375" OD (3/4" PVC coupling)
21.1 MHz
20 meters
14.175 MHz
1.625" OD (1" PVC coupling)
14.2 MHz
30 meters
10.125 MHz
2.0" OD (1.25" PVC coupling)
10.12 MHz
40 meters
7.15 MHz
2.25" OD (1.5" PVC coupling)
7.15 MHz

10 and 15 meter trapsThe 10 and 15 meter traps, wound on 3/4" PVC pipe couplings.

20 meter trapsThe 20 meter traps, wound on 1" PVC pipe couplings.

30 meter trapsThe 30 meter traps, wound on 1.25" PVC pipe couplings.

40 meter trapsThe 40 meter traps, wound on 1.5" PVC pipe couplings.

Trap Connections

Pigtail connection to traps I used this simple method to connect the traps to the antenna wire elements. I soldered a short (approximately 2") wire pigtail to the bridle wire on each end of the trap. Then the antenna wire was looped through the trap bridle wire and secured to the pigtail using an electrical wire nut. This made trimming the lengths of the antenna elements easy, as the connections could be readily disassembled and no soldering in the attic was required. When the antenna trimming was complete I used a nylon cable tie to secure the antenna wire loop to the pigtail to strain relieve the connection.

I used 14 gauge stranded household electrical wire for the antenna elements. This wire is very inexpensive when purchased in 500 foot spool quantities at home centers. The insulated jacket causes the wire to have a velocity factor somewhat lower than that of bare copper wire. This is a beneficial attribute for an antenna intended for limited space use such as in an attic, as it results in a shorter overall length.

Center and End Insulators

Antenna center insulator The antenna center insulator was constructed from a piece of scrap Plexiglas® stock[*]. The center of a half-wave dipole is a current feed point so just about any insulating material will work here. Plastic cable ties are used to secure the antenna elements and the RG-58/U feedline to the insulator. A rope attached to the topmost hole is used to support the antenna center. The rope is approximately twice the height of the attic. It passes through a screw eye secured at the peak of the attic which functions like a pulley, allowing the antenna center to be easily raised and lowered.

I used "real" pulleys at the ends of the attic where the 80 meter segments were bent 90 degrees to fit within the available space. The insulated 14 gauge wire rolls easily over the "real" pulleys, allowing the antenna to be easily lowered for adjustment. I supported the pulleys from the top of the attic walls with a length of plastic rope, which also serves as an insulator. The antenna end insulators (not illustrated) must withstand high voltage in operation, so a bit of care must be taken with their design to insure that you don't start a fire in your attic! I fashioned mine by drilling holes at the ends of lengths of scrap plastic rod stock. A generous length of rope was attached to each end insulator, and screw eyes were used as pulleys to allow the antenna to be easily raised and lowered.

[*] The original sheet of Plexiglas® was purchased to replace a pane in the shack window so that holes could be easily drilled for the purpose of bringing cables into the shack. When I move from this QTH I can replace the original glass pane, leaving no trace of my antenna installation.

Choke Balun

I constructed a choke balun near the antenna center insulator by wrapping approximately 6 feet of the antenna coaxial-cable feedline as a single layer winding on a scrap polyethylene food container that was approximately 4 inches diameter. I used cable ties through small holes drilled in the container to secure the coax winding.

Some amateurs argue that a balun is not necessary when feeding a dipole with coax, but the proximity of this antenna to other objects and the physical constraints of attic installation make antenna symmetry unlikely in this situation. The simple choke balun used here is trivial to construct, and I do not feel it is worth the risk of feedline radiation problems to omit it.

Antenna Dimensions

The final dimensions of my antenna are shown below. If you try to duplicate this antenna you should start with longer lengths and then trim as necessary, as the lengths will be affected somewhat by height above ground, and in an attic installation by proximity to the building. An antenna analyzer, such as the MFJ-259 that I used, greatly speeds the trimming process.

6-band antenna dimensions

If you are not interested in the 30 meter WARC band, here are the dimensions of the antenna without the 30 meter traps. You may note that the 80 meter end sections are significantly longer in the version without the 30 meter traps: much of that difference may be due to the larger percentage of the 80 meter section length that had to be bent to fit my attic in that version.

5-band antenna dimensions

Electrical Measurements

One of the most often quoted disadvantages of trap antennas is reduced bandwidth. But the useful bandwidth of the coaxial trap dipole described here is sufficient for no-tuner use over much of the 6 bands. As the measurements in Table 2 illustrate, the antenna performs with better than 2:1 SWR over the entire 10 and 15 meter amateur bands. Almost all of 20 meters is usable with less than a 3:1 SWR. The 40 and 80 meter bands were trimmed for operation within the CW band segment.

Table 2. 2:1 and 3:1 SWR Bandwidth (Measured with MFJ-259 Antenna Analyzer)
amateur band
2:1 SWR
3:1 SWR
10 meter (28.0-29.7 MHz)
2.2 MHz
4.23 MHz
15 meter (21.0-21.45 MHz)
640 kHz
1.04 MHz
20 meter (14.0-14.35 MHz)
190 kHz
330 kHz
30 meter (10.1-10.15 MHz)
100 kHz
190 kHz
40 meter (7.0-7.3 MHz)
50 kHz
110 kHz
80 meter (3.5-4.0 MHz)
60 kHz
200 kHz

Table 3 contains the resonant frequencies and SWR, 2:1 SWR limits, and 3:1 SWR limits of the antenna as measured after the final trimming of each of the elements.

Table 3. SWR vs. Frequency (Measured with MFJ-259 Antenna Analyzer)
10 meter band
15 meter band
20 meter band
30 meter band
40 meter band
80 meter band
27.17 MHz
20.64 MHz
14.00 MHz
10.05 MHz
7.06 MHz
3.56 MHz
27.70 MHz
20.83 MHz
14.07 MHz
10.09 MHz
7.09 MHz
3.64 MHz
28.65 MHz @ 1.0
52 ohms
21.14 MHz @ 1.3
54 ohms
14.16 MHz @ 1.3
44 ohms
10.13 MHz @1.6
82 ohms
7.12 MHz @1.8
35 ohms
3.67 MHz @1.9
39 ohms
29.90 MHz
21.47 MHz
14.26 MHz
10.19 MHz
7.14 MHz
3.70 MHz
31.40 MHz
21.68 MHz
14.33 MHz
10.24 MHz
7.17 MHz
3.76 MHz

On-The-Air Performance

I finished installing this antenna on a Saturday. The next morning I connected my Heathkit HW-8 QRP rig and answered the first CQ I heard, which was an SM5 (Sweden) station on 15 meters. He responded to my call and we had a nice QSO, with solid copy on every exchange. I was running 2 watts output. I've had similar results on the other bands as well.

The performance of this attic coaxial-cable trap dipole doesn't compare to the 10-15-20 meter Yagi and 45 foot tower I enjoyed at my former QTH, but it continues to surprise me with just how well it does work. I have found the SWR bandwidth adequate for no-tune operation with my transceiver across the entire 10, 15, 20, and 30 meter bands, and the CW segment of 40 and 80 meters. I experienced no RFI problems at QRP power levels, but I did experience serious RFI problems with our stereo receiver at QRO (100 Watt) output power on 40 meters that I remediated by wrapping its power and surround speaker cables around split core "snap on" filter chokes (Radio Shack 273-104).

Your Mileage May Vary

Although many hams succeed with attic antennas, I know several who have tried attic dipoles and were disappointed with their performance. Perhaps my attic is more "antenna friendly" than theirs, or perhaps other factors conspired against them. I do hope that this story will inspire others with restrictive covenants (or restrictive spouses!) to not give up. This antenna has made it possible for me to operate a satisfying HF station in spite of the restrictive covenants imposed on my dwelling. Good luck, and I hope to hear you on the air soon!


[1] R. H. Johns, "Coaxial Cable Antenna Traps," QST, May 1981, pp. 15-17.
[2] G. E. O'Neil, "Trapping the Mysteries of Trap Antennas," Ham Radio, Oct 1981, pp 10-16.
[3] D. DeMaw, "Lightweight Trap Antennas -- Some Thoughts," QST, June 1983, pp. 15-18.
[4] R. Sommer, "Optimizing Coaxial-Cable Traps," QST, Dec 1984, pp. 37-42.
[5] J. Grebenkemper, "Multiband Trap and Parallel HF Dipoles -- A Comparison," QST, May 1985, pp. 26-31.
[6] D. Kennedy, "Coaxial-Cable Traps", QST, August, 1985, p. 43.
[7] M. Logan, "Coaxial-Cable Traps", QST, August, 1985, p. 43.

Frequently Asked Questions

Since posting this web page, dozens have written me with questions about my antenna or to report that they successfully constructed their own trap dipole after reading this paper. Below are the most frequently asked questions I have received:

Appendix - Why Aren't 12 and 17 Meters Supported?

I did not include 12 and 17 meters in the series trap dipole described above because the loading effect of the 10 meter traps when operating on 12 meters would require the length of the 12 meter elements to be negative in order to achieve resonance. Similarly, the loading of the 15 meter traps when operating on 17 meters would require negative length 17 meter elements.

If operation on the 12 and 17 meter WARC bands is desired, one could construct a second trap dipole for those two bands using a pair of traps resonant at 12 meters. The inner elements (approximately 112 inches each) would form a full size half wavelength 12 meter dipole. The length of the outer (17 meter band) elements would be reduced by the loading effect of the 12 meter traps. In Table 4 below I include the dimensions of 12 meter traps that could be used in the construction of such an antenna. Table 4 also includes dimensions for 17 meter traps. These would not be used for a 12/17 meter dipole, but are included for completeness in case one is interested in constructing another band combination, e.g. a trap dipole covering the 12/17/30 meter WARC bands which would require a pair of 12 meter traps and a pair of 17 meter traps.

Table 4. Specifications of traps for 12 and 17 meter amateur bands
design frequency
trap form
coax length (shield intact)
coax turns (approximate)
12 meters
24.94 MHz
1.375" OD (3/4" PVC coupling)
17 meters
18.118 MHz
1.625" OD (1" PVC coupling)

A 12/17 meter dipole could be fed with a second feedline or alternatively, it could be connected in parallel with the 10/15/20/30/40/80 dipole described in the main section of this paper. I have not tried or modeled the parallel connection so I do not know what interaction, if any, would occur between the two antennas.

Revision History

29 Jun 2010 Clarified that the listed coax length is the "active" length in which the shield is intact.
08 Sep 2003 Clarified trap connection illustration and added FAQ regarding trap frequency relative to band of interest per suggestions by Will W1ZRV.
26 Aug 2002 Add inverted vee testimonial and trap construction hints. Reverse trap form dimensions to emphasise OD rather than nominal size of PVC coupling.
24 Aug 2002 Add appendix regarding 12 and 17 meters, cutaway illustration of trap conections, paragraph comparing a trap dipole to a non-resonant antenna plus antenna tuner, paragraph on trap orientation, FAQ section, and other minor changes.
30 Nov 1999 Restore dimensions of antenna without 30 meter coverage.
06 Apr 1999 Added 30 meter coverage.
03 Jan 1999 Added 80 meter coverage.
10 Dec 1998 Original version for 10, 15, 20, and 40 meters.