Small 80m-10m Trap Dipole

A 50ft Trap 80M-10M Dipole for Small Spaces

An antenna that has caught my eye recently is the trap dipole. Often touted as small and efficient antennas convenient for the small lot, I thought it would be interesting to give one a shot. Before I do that though, I want to take a few minutes to model one and see how it works in theory and in particular, answer these questions: 

What is the estimated effective radiated power at peak gain relative to an isotropic radiator and what is bandwidth of an 80m-10m trap dipole compared to its resonant mono-band counterparts if we shorten it to a measly 50 ft?

Lets find out!

First, a little about what a trap dipole is...

A trap dipole is, as the name implies, a balanced dipole fed at the center usually with coax. However, unlike a regular dipole antenna,  it can be made to resonate on ANY frequency independent of harmonics (though odd order harmonics can be utilized in certain cases such as in the case of the 5 band W8NX multiband trap antennas shown in the 23rd edition of the ARRL antenna book.)

It works similar in principal to a linked dipole. "Switches" are used to shorten or lengthen the antenna to predetermined lengths to bring it to resonance on specific bands except in the case of the trap dipole, the switches are automatic! This is accomplished through what is known as a tank circuit, otherwise known as a trap.

There are two types of tank circuits, series and parallel. We will be using the parallel circuit on the right to create our traps. In the above illustration, you can plainly see that the parallel tank circuit is nothing more than an inductor and capacitor  in a parallel configuration in series with the source.  Since the parallel tank circuit blocks signals at it's resonant frequency, these can be placed strategically along the length of our antennas to act like "switches" that electrically disconnect the appropriate wire without the need for additional tuning or antenna adjustment.

Another neat party trick this antenna can do is depending on the LC ratio you use to bring the trap into resonance, you can shorten the antenna only a little, or a lot! Due to the fact that capacitive reactance has an inverse relationship to frequency, as the frequency is lowered out of resonance with tank circuit, the capacitor begins to look more like an open circuit and thus the trap becomes inductive acting like a loading coil shortening each successive section.

This comes at a cost though. Shortening segments reduces the efficiency somewhat, and reduces the bandwidth a lot. The good news is, like I mentioned earlier, any resonant parallel tank circuit will work. In other words, you can use a small inductor and a large capacitor to reduce loading and maintain bandwidth. Alternatively, you could omit the capacitor entirely and wind an inductor until it becomes self-resonant due to parasitic capacitance. This would result in a very short, very narrow banded antenna on the lower bands. 

If you've ever used an AM pocket radio, or experimented at all with electrically small magnetic loops, you've delt with this already. Any loop of any size receiving by a source oscillating below it's self resonant frequency may be brought into resonance with sufficient capacitance. The same is true here for trap resonance. You can choose the loop size/turns ratio and use a capacitor to "pull" the resonant frequency down.

Below are the values I chose for the traps. In this case I chose 40 turns for 40m, 20 turns for 20m, and 10 turns for 10m on a 1.91" form which in this case is 1.5" PVC pipe. There's no real rhyme or reason for me picking these values other than they made me feel all warm and fuzzy. Additionally, the solenoid length is the wire diameter (18 AWG solid including insulation) times the number of turns and the wire needed to wind the inductor is the form circumference times the number of turns. Appropriate resistance was added in series with the lumped inductors to account for coil loss. A coax stub with an open end can be used to  add the additional capacitance to bring the coil into resonance. These values should create a trap that resonates just above their designed bands.

The reason for tuning the trap to resonate outside of the band is that just like a magnetic loop at resonance, the circulating  currents are VERY high. In this case of the 40m trap inductor could see as much as 20A  circulating current  and the capacitor would be exposed to roughly 6kV on only 100 watts. In other words, while the trap is most effective at blocking currents at it's resonant frequency, It burns most of that current off as heat. Thankfully, the resonant bandwidth of a tuned multi-turn loop is very small so by moving the resonant point outside of the band, we can still achieve sufficient impedance to block most of RF from passing while minimizing circulating currents. in the trap. This will yield the highest efficiency.

I elected to design the traps to resonate above the band due to the fact that a dip probe tends to couple capacitively with the trap. The mere act of measuring a coil changes its resonance ( Schrodenger's trap haha)  by pulling the frequency down up to several 100kHz depending on the probe's proximity to the coil. This means that  trap tuned for 14000khz with a probe  could actually resonate closer to 14300kHz which would be a bad thing if you wish to operate in the phone portion of 20m. By tuning the traps above the band, any rebound from capacitive coupling of the dip probe would only serve to move the resonant frequency further out of band.

After plugging these values in, here are the results. It took a little fine tuning to get the SWR right as EZNEC tends to overstate impedance mismatch when performing complex impedance calculations where significant reactance is present, but aside from everything went smoothly. First let's look at the band coverage and then we'll dive into each band a little deeper.

Above is a full scan of the 3-30 HF bands. As you can see all the band were able to be pulled in nicely except 80m which I suspect is an issue with the NEC engine and not the design.  We'll talk about that in a bit. for now lets do a deep dive on the bands starting with 10m.

Comparing the far field plots both the trap and 1/2 wave dipole, it's clear there is virtually no difference in either gain or radiation. The difference in gain is only 0.06 dBi. and the peak gain angles are the same as well.

Again, the two plots comparing the trap and half wave dipole are nearly identical. The difference in gain between the two is only 0.51 dBi with the edge going to the half wave dipole.

Same story on 40m. Nearly identical radiation pattern out of the two. The difference in peak gain is only 0.03 dBi.

Finally, we see some real performance difference. As you can see in the far field plots above, the half wave dipole outperforms the trap dipole by 2.03 dBi. It's not a lot but it's enough to be noticeable in otherwise poor conditions.

Taking the time to model this antenna has revealed a few things things to me that were not apparent just by looking at the design on paper.

First is that the antenna is surprisingly efficient for its length and I wouldn't hesitate for a minute to put one of these up in the air if I was in place where I could not put up a 135" antenna. The bandwidth on 10m covers the digital and voice portions of the band and the 20m-40m bandwidth is sufficient for single mode operation in whatever portion of the band you wish to operate in. The narrow bandwidth on 80m is sufficient for checking into a nightly net or local ragchew group that operates on a specific frequency and the antenna manages to do so while being shorter than a 40m dipole. Not to mention, if the antenna is mounted in an inverted v configuration, depenting on antenna height, you may have easy access to the ends of the antenna allowing you to adjust the 80m band coverage and gain more band agility.

The second is that the way the inductance is distributed along the antenna, as you go lower in frequency, the inductive reactance moves further away from the feed point, thus retaining more efficiency.

I hope this experiment and analysis has been helpful in understanding the uses, and mode of operation, for this neat little antenna. I think I will model the W8NX 5 band trap antenna next to see and illustrate how it works in theory using only 1 set of traps. Until then, enjoy the airwaves and thanks for reading!