MEASURING  ANTENNA  IMPEDANCE

 - CONCLUSION -

In the ideal case, these measurements would be made outside, directly below the antenna. 

The purpose of this exercise was to compare the 5 different ways

of measuring, not to actually measure the antenna's feedpoint.

Being February and still very COLD outside, I decided to use a "LOAD" that was readily available inside of my house: the shack-end of my 80m OCFD antenna.   

This has several advantages; first, it is warm and dry in the shack.  Second, it enables me to measure the LOAD directly at the LOAD (see Fig. 2 below), giving me a stake-in-the-ground measurement for comparison to the various measurement methods. 

Fig. 1 (below) shows the antenna terminating in the house into a 1:1 Maxwell choke with about 1500 Ohms of Common Mode Impedance on 80m.  The coax running to the antenna consists of 60m of M&P Airobrne-10 to the mast, and then 14m of Airborne-5 running up the mast.  However, for this exercise, it doesn't matter, because we are measuring the (LOAD) impedance in the shack.

Figures 2 thru 5 show the various configurations used for measuring.  In all cases, if a computer was connected to the analyzer, it was running off of its own batteries, fully floating.

The 5 Methods of measuring remotely:

1) Using an "Active" USB Extension Cable (15m long).  See Fig. 3.

2) Using a BlueTooth connection to the RigExpert AA-230 ZOOM analyzer.  See Fig. 4.

3) De-embedding the Coax.   This is accomplished by re-calibrating the analyzer at the  LOAD-end of .    the coax.  See Fig. 5.

4) Subtract Coax.  This is accomplished by selecting a known coax in the setup menu of the RigExpert .   "AntScope-2" (freeware) software application, and entering its length.   The  antenna is measured as .   usual, and then the  [Subtract Coax]  icon is clicked in the program,  modifying  the displayed curve, .   removing the effects of the coax.  See Fig. 5.

5) Using an exact  (electrical)  half wavelength of coax on the frequency that will be measured.  In .   theory,  this will repeat the impedance at the feedpoint, albeit,  'only'  if the coax is lossless;  which is .   never the case.  However, the results are close.  See Fig. 5. 

HERE  ARE  THE  FINAL  RESULTS:

Note-1:  Because the exact frequency and level of minimum SWR vary slightly from one day to the next, or even on the same day if it rains, each time I began a new method of measuring, I re-measured the LOAD (ANT') before continuing.

Note-2:  Each measurement method was repeated on a different day, as a sanity check that I had made no mistake - except, the De-embedded Coax method could not be repeated because the remote-calculation procedure failed on my analyzer.  I was unable to fix the problem, so this method was only recorded once.   

Note-3: The Half Wavelength Coax method is slightly skewed, because the coax used is an exact (electrical) half wavelength on 3.480 MHz, and I measured and recorded the SWR at the frequency of minimum SWR of the antenna, which was about 3.665 MHz.   When the weather gets warmer, I will measure the antenna in the field, readjusting it for minimum SWR @ 3.480 MHz.

DISCUSSION:

It should be obvious that the BlueTooth method is exactly the same as when measuring directly at the load, because the analyzer is connected directly to the load and has no further physical connectors.

It was thought that the USB Extension Cable might cause a slight skewing of the results, simply because of its physical near proximity to the antenna.  I added 5 clamp-on Ferrite beads to each end of the cable. As you can see, it did not cause any skewing of the measurements.

De-embedding the Coax was found to be the next most accurate method.  Until BlueTooth support for Windows P.C.'s becomes available in AntScopt-2, this would be my method of choice - after I fix the problem with my analyzer's calculation routine.

The problem with the Subtract Coax method is, not all makes and models of coax are listed within the AntScope program.  As long as the coax you are using is listed, the results are fairly accurate.  However, manufacturing tolerances from one production run to the next may cause the results to be slightly off.

Although the Half Wavelength Coax method "looks" fairly accurate here, especially when measuring on the antenna's fundamental band, the inaccuracy increases on the higher harmonic bands.  In addition, on the WARC bands (which are no harmonic bands of my 80m OCFD), the SWR is higher, thus the result skew even more.  IMO, this method is not sufficient for measuring several bands.