2
$\begingroup$

I am setting up a 435MHz cross polarized yagi antenna on Yaesu rotor (for both azimuth and elevation). The application is satellite tracking. The beamwidth of my antenna is 20 deg horizontal and vertical.

What is the best way to find the line pointing towards geographical North (or true North, that can be calculated using a formula) from a location for the initial setup? Indeed, compass is one of the options, and I tested locally bought two - three analog and digital compasses. They have almost 4-5 degrees offset with respect to each other. Even the same with my mobile phone compass.

$\endgroup$

3 Answers 3

3
$\begingroup$

So if you pick a value in the middle of your compasses (and don't forget to correct for the difference between magnetic north and true north) you'll probably be within 2 degrees of correct. Two things about that:

  1. With a half-power beamwidth of 20 degrees, a pointing error of two degrees is going to make less than half a dB difference. Probably much less, like 0.1dB, but I won't make that a hard statement without seeing a high-res pattern plot for your antenna.

  2. Backlash and position readout accuracy on your rotator limit the pointing accuracy to no better than 1 or 2 degrees anyway. Getting your mount direction more accurate than that will provide little benefit.

But if that doesn't satisfy you, another solution would be to get a good topographical map (either on paper, or on the computer in some app that has a measuring tool) and determine the heading to some landmark that you can see from the position of your antenna. Then you can sight on that object and use a protractor to find north.

$\endgroup$
2
$\begingroup$

You can get reasonably accurate azimuth numbers just pointing the antenna at the sun, and manually sighting down the boom or looking at the shadow. Try this at various times of day. Use your favourite software to tell you the expected solar azimuth at that time, and compare this to your azimuth encoder readings.

For more precise alignment I suggest taking photos of the stars and submitting them to Astrometry.net. This service compares the stars in your photo to its catalogue. It returns astrometric metadata about the photo - including the exact position of the centre of the image. If the camera was mechanically aligned with an axis of your choice, you can quickly get sub-degree measurements of azimuth and elevation.

The photo doesn't need to be amazing, just needs to have some visible stars in it. You could use a cellphone if you're careful but a DSLR might be easier to aim accurately, and lens distortion will limit the accuracy if you use a wide angle lens. You'll also need to convert from RA/Dec to Az/El; this is easy if you know your location and the time of the photo.


Rather than just aligning the North position, you could use a camera mounted above the rotator near the antenna to take 10 or 15 star pictures, widely separated, and build a pointing model with some more terms in it.

These terms start with the obvious four - Az and El scale factors and encoder zero offsets. But then can be extended to other useful terms: the tilt and tip of the axis of Az rotation; the offset of the El axis from true perpendicular; and the 'box offset' of the antenna from the rotator elevation axis. This all sounds scary but it's routine in astronomy, and essential to be able to point anything better than a degree or so.

(A big well-made equatorial mount will do better of course, but even they probably went through a lengthy alignment process using the stars, and kept a table of pointing errors for different parts of the sky...)

$\endgroup$
2
$\begingroup$

Another way to sight along the boom at a distant point, and locate it on a map (e.g., Google Earth). You can then draw a path on the map and see its bearing to within a fraction of a degree. Even better if you have a camera mounted on the boom and you can do the sighting from the comfort of your shack any time you feel the need.

I agree with the other post about the beamwidth. For all my antennas just estimating that the boom is parallel to the road beside my house is good enough. Google Earth can tell me the track the road is taking, and for local nets "straight south on Bathurst street" gives people a good idea of where I'm aiming.

73, VE3NRT

$\endgroup$
1
  • $\begingroup$ Thanks for this solution. This worked for me well. $\endgroup$
    – enemra
    Commented Apr 26 at 15:44

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .