VK4PK - 70cm 8el Yagi - Build No2 - 438Mhz - Built March 2019

Contents

  1. Choosing a Design
  2. VK5DJ Yagi Calculator by John Drew
  3. Preparation
  4. Construction
  5. Matching Balun
  6. The Final Product
  7. Vector Analysis
  8. SWR Plots 410Mhz to 450Mhz
  9. SWR Plots 430Mhz to 450Mhz
  10. Smith Chart
  11. Mounting the Beam
  12. Pointing the Beam in the Correct Direction
  13. The Feedline Coaxial Cable
  14. Field Tests


70cm 8el Yagi

[Top][Home]


Choosing a Design

I choose the VK5DJ calculator:
http://www.vk5dj.com/yagi.html

I decided to be a little conservative a build an eight element UHF antenna for my second at building a Yagi. The first attempt to build a Yagi antenna was a thirteen element VHF Yaga and is still not complete. It was a little to ambitious but I will get back to it.

I looked around for similar projects and was inspired by Ralph Klimek's 70cm 11 element yagi array project.
http://users.monash.edu.au/~ralphk/70cm-yagi.html
He also chose to design with the VK5DJ Yagi Calculator and has some interesting construction suggestions.

[Top][Home]


VK5DJ Yagi Calculator by John Drew

I plug in these paramaters:



The calculator produced these results:


[Top][Home]


Preparation

I make a printout and label the documents "BUILD W/S", and make sketches to aide the folding of the dipole. I can then take these to the workshop for the construction phase.





[Top][Home]


Construction

Cutting the Folded Dipole:
Cutting the Dipole

The Antenna before Welding:


I purchased twenty 500mm long of Low Temperature Aluminum Welding Solder Rods. They cost AU$13.86 delivered. They are like Wire Brazing rods for aluminium and soften before the alumunium melts. You have to be careful not to overheat the job or the antenna is lost in a puddle of molten aluminium. It was time consuming work soldereing the elements both sides of the center beam. Next time I will buy a canister of argon gas and try my hand at aluminium TIG welding.

You do not need any flux powder. I tried using aluminium flux powder but it did not appear to make any difference.

The application is similar to silver soldering. The welding area must be clean fresh aluminium so I used sandpaper to polish the welded parts. I applied heat evenly moving the flame over the welded parts. The temperature must rise to 360C degrees, for the copper and aluminum welding rod to melt evenly into the joint. Do not apply the flame to the welding rod, only to the job. Remove the torch after the weld flows.

I used a small section og the 25x25 aluminium bead a 10mm tube to test Soldering the aluminium and develope my technique.



[Top][Home]


Matching Balun

To make the copper supporting bracket that I could solder the coaxial cable's shield to, I use a small offcut of copper water pipe and slit it lenghtways, and hammer it flat, before bending to make the angle.

Matching Balun



Soldering the RG405 to the copper was a little tricky. I made lugs from the same copper pipe so I could solder directly to the the lugs and the bolt the lugs to the ends of the aluminium dipole ensuring a good contact.

The RG405 is very stiff and I did somehow get a broken joint between the RG405 inner and one of the copper lugs. I must have snapped it when securing the bolts. I had test it all with the ohm meter, but before bolting. The first VNA test showed a resonate frequency of 2.5GHz. This had me perplexed for a while, untill I applied first principals and tested again with the ohm meter.

Matching Balun

Before sealing with the lid, I applied some silicone cement to secure the balun winding. they must remain in place during vibration and impacts.

Matching Balun

[Top][Home]


The Final Product

To finish the beam I used a 25mm black plastic chair tips and fill the hollow of the 6mm elements with silicone.



[Top][Home]


Vector Analysis

I have a miniVNA Tiny to test antennas. I have it set up to run off my Android phone or my Ubuntu desktop.

How to setup the miniVNA Tiny for testing:

Vector-Network-Analysers - miniVNA Tiny

The miniVNA Tiny Network Analyser attached to the antenna ready for testing:



[Top][Home]


SWR Plots 410Mhz to 450Mhz



[Top][Home]


SWR Plots 430Mhz to 450Mhz



[Top][Home]


Smith Chart



[Top]

Contents

  1. 1.01 Section  [Top]
  2. Choosing a Design
  3. VK5DJ Yagi Calculator by John Drew
  4. Preparation
  5. Construction
  6. Matching Balun
  7. The Final Product
  8. SWR Plots 410Mhz to 450Mhz
  9. SWR Plots 430Mhz to 450Mhz
  10. Smith Chart
  11. Mounting the Beam
  12. Pointing the Beam in the Correct Direction
  13. The Feedline Coaxial Cable
  14. Field Tests
[Home]


Mounting the Beam

Erected

As most UHF antennas a vertical whips I decided to mount the bean with vertical polarisation. That meant that the top section of the supporting post had to be electrically non reactive. I used a meter length of 32mm orange conduit and attached that to a four meter length of 32mm steel pole.

[Top][Home]


Pointing the Beam in the Correct Direction

Google Earth was a good tool to determine a bearing from my location to Cleveland is about 310 degrees "google-true" which I will assume is Grid North. It could be a couple of degrees out but for this application it will surfice. The Magnetic Variation is estimated to be about 11 degrees for this location. The rule is to take the Grid bearing and SUBTRACT the Variation. So we point the beam at apprximatly 300 degrees magnetic.



[Top][Home]


The Feedline Coaxial Cable

12M LMR-400 feedline Coaxial Cable




[Top][Home]


Field Tests

Yet to come.

[Top][Home]


Glenn Lyons VK4PK
glenn@LyonsComputer.com.au
Ver:gnl20190422 - pre published v0.9