[Elecraft] KX3 Field Ant. for 80/40/30

[donovanf at starpower.net]

Hi Erik, 


Radiation results from RF current flowing in an antenna. An RF 
ammeter is a useful instrument for measuring the relative efficiency 
different types of matching networks feeding similar antennas. 


I'm well familiar with the simultaneous synchronized WSPRlite 
antenna performance analysis techniques, I developed the techniques 
that SOTAbeams implemented earlier this year! They work very, 
very well but a few things must be done to avoid significant 
measurement errors and biases: 


1. The two antennas under test should be located within less than 
one wavelength of each other, otherwise independent selective fading 
becomes a significant source of measurement error. 


2; Horizontally polarized antennas should be oriented end-to-end 
to each other to avoid significant parasitic interaction that washes 
out the other differences in antenna performance 


3, Do not attempt to compare horizontally polarized antennas to 
vertically polarized antennas, independent selective fading 
becomes a significant source of measurement error that takes 
an extraordinary amount of data collection to overcome. 


Enjoy! 


73 
Frank 
W3LPL 






----- Original Message -----

From: "Erik Basilier" <ebasilier at cox.net> 
To: donovanf at starpower.net, elecraft at mailman.qth.net 
Sent: Monday, November 20, 2017 1:51:47 AM 
Subject: RE: [Elecraft] KX3 Field Ant. for 80/40/30 



I am well aware of that, Frank, and in fact I have multiple cores sitting here waiting for such measurements. However, performance of an isolated component is less important than overall system performance, where the matching to antenna impedance as well as counterpoise current routing and losses all play in. After the component testing to qualify different transformers as having low loss, I am planning to test different versions of resonant and non-resonant end-fed antennas against each other in pairs. The resonant (very high impedance) versons will include different transformer ratios. I may even include versions where the impedance transformation is performed with a tapped, tuned parallel-resonant circuit, which is the classic approach from the Zeppelin days. I may also include a center-fed dipole (conventional, K9YC?, sleeve around coax?). To be able to compare antenna systems with potentially small differences, I am set up with a pair of WSPRLite transmitters that let me run both antennas simultaneously in synchronization. In this type of testing one obtains two overlaid graphs representing the two antennas under test, versus time. Each value shows a composite number based on s/n ratios at a number of different receiving stations. Over time, the two curves tend to cross back and forth against each other, but over a few hours one can see whether one tends to dominate over the other. Transmission frequencies will not be exactly the same, but the difference will be very small. When done in my back yard, both antennas in a test will be influenced by all kinds of metal structures around it, including my tower, power lines, metal in the house, and the other antenna under test. I will minimize the latter by erecting the wires at 90 degrees angle, with the feed points close together, so that I can always reach both transmitters at the same time to push the start buttons at the same time. My main method of compensating for interactions with metal objects will be to swap the matching/feeding systems while keeping the radiators and transmitters in place. I like to deploy wire antennas in the field on 24 ft masts, so I will use two of those for the testing and arrange the two wires as inverted vee’s. For reasons of space, I will not include 80 m, so the two wires will be in the 60+ foot range, except for the non-resonant version where 50+ feet are commonly used. For possible tests using center feed, I would use different arrangements that all resonate as ½ wavelength on 40. By comparing two antenna systems at a time, each time taking the winner to compare with the next antenna, I hope to determine an idea of what works best for me in field use. 

73, 
Erik K7TV 



From: donovanf at starpower.net [mailto:donovanf at starpower.net] 
Sent: Sunday, November 19, 2017 3:33 PM 
To: elecraft at mailman.qth.net 
Subject: Re: [Elecraft] KX3 Field Ant. for 80/40/30 


Its easy to measure transformer loss by measuring the loss through 

a pair of identical transformers connected to back-to-back. 



The loss in a single transformer will be half of the loss through the 

back-to-back pair. 



73 

Frank 

W3LPL