[BC] Aligning a satellite dish

[Burt I. Weiner]

Here are a few tricks that I use when setting up a dish:

First: When working at C-Band I always recommend a filter be placed between the LNB and feedhorn.  Get one that's good not only for navigational radar, but also for Wi-Max.  You will not regret the extra expense of $300 to $400 and the chances are high that you will come to really regret not installing one.

Digital signals look a lot alike. To save time, I always try and get a spectrograph from the uplink provider to use as a guide.  You don't want an old spectrograph as things in the satellite world change quite often.  Ask them to use the desired downlink frequency as the center frequency of the span and show 5 to10 nearby signals.  Hopefully this will show you an assortment of wide and narrow signals because that combination will be more obvious than 10 little spikes that all look the same.  Seeing the entire bird or just the desired signal is useless.  I note the analyzer settings from the uplink's spectrograph and generally use the same settings.  At least this way I know what I'm looking for with the spectrum analyzer.  Note that rather than using 10 dB/vertical division they generally use 5 dB/division.  This makes it easier to see weaker signals and also gives better visual resolution for adjustments.

How to convert from C-Band frequencies to L-Band frequencies:  Most C-Band LNB's use a conversion oscillator at 5150 MHz.  So, subtract the desired C-Band frequency from 5150 (MHz)  For example:  The desired C-Band frequency is 4175.5 MHZ.  You need to know the L-Band frequency in order to set your analyzer's center frequency (Because the uplink provider did everything at C-Band and you have a LNB!)  Simply subtract 4175.5 from 5150 and you have 974.5 as the resultant L-Band frequency.

One word of caution, if the uplink did everything at C-Band, the L-Band image will be reversed.  That is, the high end (4200 MHz) will be towards the bottom of the L-Band display and the low end (3700 MHz) will be towards the high end of the L-Band display.  This is because the conversion frequency (5150) is on the high side which reverses the C-Band spectrum as seen at L-Band.

If you're working at Ku-Band, use 10750 (MHz) rather than 5150 (MHz) to determine the resultant L-Band frequency.  This is the typical L.O. for Ku-Band LNB's.

A good source to help you will be found at:  http://www.dishpointer.com/  Enter the dish location address, not your home address (this has happened leading to much confusion!)  This site will tell you all you need to know, including a satellite view of the dish's neighborhood with a line showing the approximate direction to point the dish.  If the line's starting point is not exactly at your location as shown on the map, simply click on the starting point, hold it and move it to your dish's exact location - this is really pretty neat.  It will also tell you all the angle settings.  Remember to use magnetic numbers and be sure your compass is calibrated to magnetic.  Using true north is ok, but unnecessarily complicates things.

At this point don't worry about precise polarization, we'll get to that later.  Just get an approximate polarization.  Remember, polarization is determined by the narrow dimension of the waveguide.  If the narrow dimension is at the top and bottom, the polarization is horizontal.  If it's on the sides, the polarization is vertical.  If the desired bird is due south from you the polarization will be exactly vertical or horizontal.  If the bird is to the west, the polarization skews clockwise when looking towards the bird.  East, it's obviously the other way.  For right now, just set it so it's approximate.

The first thing I do is to set the elevation.  You want to do this as accurately as possible.  On a prime focus dish it's a lot easier.  With offset feed dishes you need to know where on the dish the manufacturer wants you to measure and how much to add or subtract from the actual elevation look angle.  Remember, 0 Degrees is straight up, so you're measuring in degrees down from straight up.

Hopefully you have access to a spectrum analyzer to make life easier.  Set it to the same settings shown on the spectrograph you hopefully got from the provider's uplink facility.  use the downlink frequency for the center frequency on your analyzer.  A hint: some spectrum analyzers show a lot of random baseline spiky type noise, which makes it difficult to see weak signals such as when you get close.  If this is the case and your spectrum analyzer can do averaging, use averaging and set it to about 3 to 5 averages.  This will greatly smooth out the noise and at the same time will be fast enough for you to see what's happening.  Don't try to use a narrow resolution Bandwidth (RBW), this will slow down the sweep speed.  Depending on the bandwidth of the signals in the provided spectrograph I might try and use 1 MHz or 100 kHz RBW and a Video Bandwidth (VBW) of around 1 kHz - you may need to experiment a bit.

Sweep the dish azimuth in the general direction of the line on the picture shown at the above website.  If you've set the elevation correctly, and the LNA/B polarization is not inadvertently set to provide a null at the desired polarization, you should cross the arc at the right position.  If you've set the elevation and spectrum analyzer correctly you should be able to stumble across a display that is identical or very close to what the uplink facilities provided.

Now you can fine tune the elevation and azimuth.  I find it helpful to set the spectrum analyzers amplitude to 1 or 2 dB/Division. But before doing this, set the desired signal to near top line so you won't be surprised when the desired signal disappears off screen - you will need to re-adjust the amplitude when changing the vertical amplitude units/division.  Anyway, this will show you very small changes in amplitude and is good for really fine adjustments.

Now to adjust the polarization.  Hopefully you won't need a cherry picker because the front of the dish is over the edge of a 3-story building!!!  When adjusting polarization, do not look for maximum signal only.  Look for maximum signal and the deepest valleys on each side of the signals at the desired polarization, particularly on each side of the desired signal.  Sometimes I'll rotate the polarization to the opposite polarity just to see what's there and then go back to the right polarity, adjusting for the deepest valleys and make sure the opposite polarity stuffs are gone.

If things have been done right, you should end up with optimum dish alignment.

I hope this information helps and I haven't put you to sleep.  If you have any question feel free to e-mail me directly.

Burt