6m Log-cell Yagi

I have been looking around for a simple way to get onto 6m using a single antenna that would allow me to cover the whole 50MHz - 54MHz band. As usual most published designs are either optimised over a narrow sub-band of 1 or 2MHz, or offer sub-par gain and pattern performance.
The closest design I could find to met my requirements was an antenna built by Steph (Useful Links) based on an article by W4RNL published in QST magazine.

I did build a version of this antenna and it works reasonably well. However NEC modeling showed some areas which might be improved upon, particularly in gain flatness and front-to-back ratio over the band.
So an afternoon of tweaking things in simulation has led to a 4-element version of this antenna. The extra reflector improves matters no end and costs an extra 600mm of boom length. A worthwhile compromise, I felt.


Antenna Design

NEC Listing

For anyone interested in modeling the antenna that I built, here's the NEC listing. Bear in mind that I used the NEC4 engine for the simulations.

log_yagi.nec

Also (in case anyone is interested) I simulated the square boom in NEC by first calculating an equivalent cylindrical radius \(r_e\) based on work from a paper publised by Lo and Jaggard: \[ r_{eq} =n\Gamma(1 + {2\over n}){d \over 2^{1 + {4 \over n}}} \left[ \Gamma({1 \over 2} + {1 \over n})\right]^{-2} \] where \(n=4\) (for a square boom), \(d= \) side length of the regular n-gon and \[\Gamma(z) = \int_{0} ^{\infty} x^{(z-1)} e^{(-x)} dx \] For a square boom of side length \(d=32mm\), \(r_{eq}=18.885mm\). For this particular antenna design, the correction makes no difference whatsoever. But it might be useful elsewhere.

Mechanical Considerations

  • 32mm rectangular cross-section aluminium boom
  • 8mm diameter elements
  • Elements insulated from boom
  • Custom designed phasing line to achieve lowest possible return loss

Feed point Considerations

The Log-cell requires a balanced feed line to connect the driven elements together. My NEC modeling indicates that a phasing line impedance of between 300Ω and 350Ω provides the best impedance match to 50Ω.
Consequently, a 325Ω phasing line has been constructed using 2mm diameter enamelled copper wire spaced 16mm apart: \(d=2mm, D=16mm\).

Other line dimensions may be substituted (keeping the same impedance) using the formula listed below with minimal impact to the return loss of this antenna: \[ Z_0 = {120 \over \sqrt{\epsilon}} cosh^{-1} \left({D \over d}\right) \quad \Omega \]

Do not forget to add a \( 180 ^\circ \) phase shift to the phasing line between the 2 driven elements and make sure that there is sufficient clearance between the phasing line and the boom.

I used several plastic spacers drilled to provide a friction fit to the 2mm copper wire to maintain correct spacing of the phasing line.

The driven elements are drilled to accept M3 hardware and both the coax and phasing line are terminated in lugs and bolted to the driven elements.


Simulation Results

The following results are mixture of a NEC4 free space far field simulation and impedance measurements taken with a VNA. First up is a comparison of impedance between the NEC model and the prototype build. The measured results correspond well enough with the simulation to validate the simulation results. The prototype antenna was tested by pointing it vertically skyward, keeping the reflector 1m above the ground.

The 1.5:1 VSWR bandwidth is an impressive 5MHz, covering the entire 6m band. The far-field patterns are simulated results only, and here I compare results with the W4RNL 3-element version.

Comparison with W4RNL design

Simulation comparison with both antennas in free space.

VSWR and Return loss compare reasonably well
Gain is higher and much more constant in the 4 element version
The front-to-back ratio is likewise better in the 4-element version
E-plane far-field @ 50MHz
E-plane far-field @ 52MHz
E-plane far-field @ 54MHz

The extra reflector does a good job of cleaning up the rear lobes as well as flattening the gain vs. frequency curve.


  • Antenna Resources
    • VK5FQ 6m 3-ele Yagi Antenna
    • 4nec2
    • Stauff Clamps
    • D. Jaggard, "On bounding the equivalent radius," in IEEE Transactions on Antennas and Propagation, vol. 28, no. 3, pp. 384-388, May 1980, doi: 10.1109/TAP.1980.1142336.
    • Y. Lo. “A note on the cylindrical antenna of noncircular cross section.” in J . AppliedPhys, vol. 24. pp. 1338-1339. 1953.