Phased Arrays - Opposite Voltage Fed (OVF)

Opposite Voltage Fed (OVF) arrays were developed by Pekka Ketonen OH1TV. His site contains several examples in different configurations, I've linked them here for convenience:

• 17. Opposite voltage fed array 2009
• 28. 2-el vertical array  for 80m  2010
• 33. 40m OVF stack at OH1NX  2011
• 40. OVF 2-el phased vertical for 40m   2011
• 55. 40m OVF stack at OH2BH  2014
• 64. 40m OVF array at OH1MA  2016
• 67. OVF Tri-bander 20m,17m,15m  2017
• 72. 30m OVF array at OH1AJ  2019
• 75. OVF Dualbander 12m/10m pdf 2021

40. OVF 2-el phased vertical for 40m   2011 is probably the easiest one start with, for 40m a pair of 1/4 wave verticals is one of the easier and cost effective ways to make some gain on this band. I decided to model this one to learn more about it based on the dimensions in the article.

Models: 40m_2ElGP_OVF.ez, and 80m version I scaled 80m_2ElGP_OVF.ez.

I got with-in 0.1dBi of the plots in the article. I've also overlaid plots at 7.0 and 7.2MHz since the article also shows plots at these frequencies. My model favors the lower end of the band slightly, so there is a small detail in there somewhere that is different which some tweaking would fix.

The system maintains a good pattern and SWR over a wide bandwidth, both desirable characteristics. Centered on 7.075MHz, SWR at is 1.2:1 at 7.0 and 7.175MHz, rising to 1.5:1 at 7.26MHz and 1.73:1 at 7.3 MHz.

I'm going to attempt to explain how this thing is wired up in EZNEC. It took "a while" for me to wrap my head around defining the transmission lines, the loading inductor, and lastly the L network for matching to 50 Ω.

How its defined here resembles the article where one 1/2 wave line is used, and the loading inductor 1.66 uH is inserted at the other end of the line where the east element is fed and the system is matched.

Transmission Lines:

• No 1. 1/2 wave line connected to wire 1 (the west element), ends at V1.
• No 2. Very short connection (10cm/4") to wire 5 (the east element), ends at V2.
• V1 and V2 are virtual connections, these can be used to connect to other things such as L networks, sources, and make interconnections.
• The VF and loss figures are for LMR-400.

L Networks:

• No 1. The 1.66uH loading inductor is connected in series with V1. A shunt is required in the L network but not used so I input 1M ohm resister to make it "open circuit". The other side of the inductor is connected to V2.
• No 2. The matching L network "input" is V3 at the bottom, this is where the source is or coax back to the rig would be connected. When the "output" of the L network is connected to:
• V2, puts the loading inductor in series with the west element (wire 1) the antenna fires east (as shown above).
• V1, puts the loading inductor in series with the east element (wire 5) the antenna fires west.

If we look at 40. OVF 2-el phased vertical for 40m   2011 - page 3, all we're changing is which side of the loading inductor we are connecting the output of the matching L network to.

I used this RF Impedance Matching Calculator to calculate the L network values for a 50 Ω match based on the Src Dat (Source Data) in EZNEC.

Other array types OVF can work with include:

• Phased Arrays - 40m 2 Element Horizontal.
• Phased Arrays - 40m Inverted Delta Loops.
• Phased Arrays - 40m Twin Half Square.

What is OVF phasing? 17. Opposite voltage fed array 2009 - page 12, and 55. 40m OVF stack at OH2BH  2014 - page 4 explain how OVF works.

ON4UN's Low Band DXing 5th edition chapters 3.4.9, and 3.4.10.4 cover OVF in detail, and notes that this system had not been published before. The book covers the math and includes spreadsheets to calculate the component values or two and four element arrays in 4-square configuration.

I struggled to make sense of it and emailed OH1TV with some questions about the two element arrays. I got a quick reply which made it simple, use an EZNEC model, and find the values through trial and error.

The elements need to be resonant above the operating frequency, the loading inductor detunes the "rear" element making it a reflector. With trial and error the best element length and inductor values can be found which peaks the F/B, current magnitudes are equal in each element (Src Dat in EZNEC), and the pattern degrades symmetrically either side of the design frequency - the sweet spot is found.

-----

Models are good starting point, and a way to investigate and better understand antenna systems. These tools can also help guide us to and validate the final result, if a good correlation is observed in the real world then we can have confidence the patterns and other information are accurate.

The models I have created and made available may contain errors, or overlook something someone more experienced can see. I don't claim to be an expert or authority on the subject of antenna modeling or phased arrays. I simply want to further my own knowledge and understanding of antennas which I find fascinating. Comments, suggestions, discussion are welcome - lonney@gmail.com.

This post is one of several on Phased Arrays.