[Home]

Circular and elliptical polarisation and propagation on shortwave

Tks to Eric, KL7AJ, I got interested again in the effect of circular polarisation on shortwave.

See: QST December 2010, "Gimme an X, Gimme an O", by Eric Nichols KL7AJ

I agree with Eric, that the ham radio community wrongly has neglected these aspects of the propagation.

After building the measurement setup and doing some measurements I needed to make an overview for myself.

In the EXCEL sheets I have summarized the different aspects, consequences and conclusions as far as I have learned until now.

I am not a specialist on this field, so please read it as such.

In the EXCEL sheets you can find the following topics:

- Introduction to X and O (Extraordinary and Ordinary Mode Propagation)

- Calculation of polarisation by summing X and O                                          

- Calculation of ellipticity and orientation                                            

- Basics Frequency Selective Fading (FSF) and Multipath Propagation over an Ionospheric - Path

- Frequency Dependent Polarisation (FDP) and Frequency Dependent Polarisation Mismatch (FDPM)

- Polarization Diversity and signal to noise ratio

- Using single antennas and FDP

- Using intended or unintended circular polarised antennas      

- Measuring setup using 2 K2's and DSP

- Measuring results examples  

I can strongly recommend reading NM7M’s book on:

NM7M The Big Gun's Guide to Low-Band Propagation

Measurement setup

The hardware used for the measurement:

2 phase coherent Elecraft K2 receivers

2 small orthogonal active receiving loop antennas (1.33 by 1.33 meters)

The digital signal processing is programmed in Matlab/Simulink.
See also the EXCEL sheets.

Fig.1. Screenshot of the simulink model.

Fig.2. Screenshot of the measurement results on 80mtr.

On the left are the “real-time” results in the frequency domain (0-4kHz).

Top to bottom: phase difference, amplitude, orientation and ellipticity.

On the right (black scopes) are the results over time.

Top to bottom: orientation, ellipticity and the signal levels of both antennas A and B and of the Left and Right hand circular components.

In between are the two S-meters for the antenna signal levels of antenna A (magenta) and B (blue) and of the Left (red) and Right (green) hand circular components. The levels are in 6dB S-points.

The measurements shown are typical for 80m NVIS communication during day-time (in PA-land, magnetic inclination about 60°).

The ellipticity shows almost full circular numbers. The signals are almost circular polarized.

That is why the orientation is noisy. However the drift (between X and O) is still visible.

The level of the Left (O) is 3dB higher than the signal levels of the antennas A and B.

The Left (O) is very dominant. The signal level of the Right (X) hand circular component is 24dB lower than that of the Left (O) hand circular component. This is not unusual, 12-24dB is common.

The corresponding ionogram (thanks Belgian Dourbes) is:

Because the NVIS signals are never exactly vertical, a somewhat lower effective frequency must be selected.

Drift becomes better visible when both the Left (O) hand and the Right (X) hand signals have comparable levels. Frequency Dependent Polarization is than possible.

Also a dominant Right (X) hand circular polarization is sometimes possible, mostly during night time.

Last update: February 21, 2012

© PA0SIM