VHF Optimized Yagi Antenna for the 6-meterband (50 Mhz)
RE-A50Y3
By Guy, de ON6MU

Click to enlarge (Hernan, LW5DTZ)

Schematic fig1

Highlighted


click to enlarge (Greg, SP5LGN) click to enlarge (Greg, SP5LGN)

 

 

Parts list for the long yagi antenna RE-A50Y3

Note: there are many ways to build your antenna and I'm sure some can come up with better mechanical designs then described here although the design and material used here is cheap and easy to find. Also, the diameters of the tubing described here is not too critical.

Links of interest:

Click to enlarge (Greg, SP5LGN)
Gamma-match

When the dipole is the driven element of a Yagi parasitic array, the impedance that appears at the center of the driven element will usually be quite different from that of the isolated dipole. The reason for this is that the Yagi antenna is equivalent to a number of resonant circuits tuned to different frequencies and coupled together. The self-impedance of the parasitic elements and mutual impedance between parasitic elements and driven element cause resistance and reactance to be coupled into the driven element.
A gamma-match tube act's like a capacitor. The "capacitor's" sole purpose, is to oppose unwanted inductive reactance. It is a series-tuned LC circuit between the coaxial cable's center conductor and the gamma's connection point at the driven element hence getting a pure Ohms resistance. So, a gamma match taps a point on one side of the center of the driven element and connects the feedline to that point through a capacitor. In this design a tube with a copper wire inside acts as air-spaced capacitor. Another advantage is to allow the center of the driven element to be directly grounded to the antenna's boom, making driven element assembly easier and no static build-up on the driver.

To tune the gamma-match to a desired frequency for best SWR you will need to experiment with the placement of the strap on the driver (feedpoint) and the position of the strap relative to the feedpoint.
If you build the yagi antenna according to above specs, then you should be pretty close (if not spot-on) to a ideal match.
Click to enlarge (Hernan, LW5DTZ)
How Hernan, LW5DTZ did the Gamma-match

Specifications ON6MU 50 Mc wide-spaced yagi antenna RE-A50Y10

Be sure to seal everything up to avoid moisture, corrosion etc...

Today's specials:

.ON6MU Homebrew projects
.Radioamateur related projects

.ON6MU Ham mods
.Modifications of transceivers


Please take a look at my 50mc vertical antenna

 


Users Feedback

This is how Greg, SP5LGN made my 50mc long-yagi:
click to enlarge click to enlarge Click to enlarge
please click the images to enlarge

Take a visit to Greg, SP5LGN his website:
http://www.sp5lgn.ampus.pl

Thanks Greg for the photo's!!
---------------

 

This is how Hernan,LW5TDZ made it:
Click to enlarge (Hernan, LW5DTZ) Click to enlarge (Hernan, LW5DTZ)
Thanks Hernan for the pictures!
---------------

 

This is a webpage on how Eric KB3CNH made the antenna:
http://impbarn.blogspot.be/2016/10/amateur-radio-building-3-element-yagi.html
He needed to shorten the driver element.
Nice webpage, great review!
---------------

 

This is how Kevin, VE7ZD made it:
Click to enlarge (Kevin, VE7ZD) Click to enlarge (Kevin, VE7ZD) Click to enlarge (Kevin, VE7ZD) Click to enlarge (Kevin, VE7ZD)
"
I built the antennas mostly out of surplus junk from my basement, and it is working very well. I have it on a 6 metre pole in the back yard, fed with about 20 metres of LMR400.

I am new to 6 metre work and the "step up" to a Yagi from my mis-matched 20m dipole has made an amazing difference. I am regularly hearing stations over 1600 km away, this without a significant band opening.

I used solid 9.5 mm aluminum rod for the elements and an old 38 mm aluminum boom from a retired 2m Yagi that I used for EME in the 1980s.
My gamma match uses the centre conductor and foam dielectric from an old length of RG8/U "foam" type cable.

Weatherproofing is critical as we get a high volume of rain here on the west coast of Canada.
I also had to shorten the driven element. My resonant frequency was too low so I cut off first 0.8, then 1.0, then 0.8 cm from each side of the driven element. This brought the low SWR point to 49.982 MHz.
I was aiming for 50.313 MHz for FT8 so I decided that this would be good enough, as I thought that the SWR may change further than the antenna gets moved from 1 m to 5 m elevation on the mast.
(Kevin, VE7ZD)
It was a good decision and the antenna at height is at SWR 1.1 to 1.

You’ll note that I used 9.5 mm solid aluminum rod for the three elements, rather than tubing. This was cheaper than the tubing and only a bit heavier.
My antenna boom is a re-purposed 38 mm aluminum boom from an old long-boom 2m EME Yagi. I was able to use the element supports from that antenna as the solid rod was the correct size and fit the holders perfectly.
My three elements are spaced off of the boom using Delrin spacers, but ARE connected to the boom through stainless steel hardware.

My gamma match was made of some old 3/8 “ tubing, and the centre conductor was from spare RG8U “super flex foam” coaxial cable.
I used the centre conductor and the white foam covering, just removing the cable insulation and braid.

The antenna works very well so far, and I note good directivity and gain in the chosen direction. I am using a manual rotation method. The antenna is near the top of a 7m aluminum pole.
"
Thanks for the feedback and pictures, Kevin!
---------------

 



Home