Biquad analysis

Abstract

Several sources in the internet shows how to buil the biquad antenna for WLAN application. Various dimensions are suggested. This analysis tries to show the impact of the following dimensions on WLAN biquad antenna performance:

Simulation data

Quick description

Simulation was done with 4nec2 simulator and optimizer. Wavelength at freqency 2439 MHz equals to lambda=123 mm. All dimensions reported in graphs are in wavelength ratio.

3D Radiation pattern

Above figures show typical 3D radiation pattern of biquad

Gain

Combined plot of several reflector sizes with and without lips shows no significant differences in gain. Except some rear impact with lips

Radiator height

SWR shows that the best height is at h=0.1315 lambda = 16.mm. The graph also shows significatn impact of distance. This means that the radiator should be as close to this distance without bends up or down.

Reflector lips height

The optimum height is at 29mm. Note that the lips are essential tool for maintainng good SWR as it is shown in SWR section.

Wire radius

Thiner wire means slightly more gain. But the main impact is in SWR. The best diameter is 2*0.0056 lambda=1.37mm. Best wire area A = πd2/4 = 1.5 mm2

Use this PDF file for wire bending outline.

Impact of lips on SWR



The above SWR plots over WLAN frequency range show, that using lips is an essential tool for maintaining and tuning god antenna SWR and reflection coefficient. It also shows that 110mm design cannot be recommended in any way.

Conclusion

  1. use lips
  2. larger reflectors are better
  3. maintain correct radiator distance
  4. use correct wire

Double biquad

Will double biquad improve gain? Here is geometry for double biquad. I suggest reflector size 123x246mm. Other dimensions should be the same as with biquad.

Simulation data dbiquad.nec was used for center frequancy as well as for VSWR. Geometry of dbiquad with currents is shown below:

Gain

Simulations show 2.7dBi improvement in gain which is enough to start with double biquad instead of biquad. SWR is approx. 1:1.3 at the center of the frequency range.

SWR


It is shown that lisps should be 1/4 lambda or 31mm. We've built one double biquad and tested it. Despite good simulation data, this antenna does not provide additional 3dBi gain as promissed. More testing is needed to verify this.


Author: Leon Kos, feb. 2005, updated: june 2005