Aligning a dish without a meter is relatively simple using an analogue system  You can see the picture appearing on your television as you move the dish.
Satellite TV systems for use on boats or moving vehicles, e.g. Tracvision have a builtin satellite tracking giro that automatically seeks the satellite on the mote.
It is a bit trickier with digital channels  since there is a point at which either you see a picture or not. Unlike analogue, there is not a gradual improvement in picture when you get within the correct orientation. Ideally for digital, you should use a satellite finder meter that bleeps or flashes when you have found the satellite.
Alternatively, you can you a trigonometry technique. You will need a decent atlas or a large scale and accurate map with latitude and longitude clearly marked. You must be able to find your exact longitude and latitude to the nearest degree.
Firstly, you will need to locate both true and magnetic south (assuming you are north of the equator). Then, locate the magnetic offset of your location. Ordnance Survey maps will have the magnetic offset marked (check the maps date of publication  the magnetic offset moves.)
If the satellite you are looking for is exactly due south (true) from your location (which is VERY rare) it is dead easy  see figure 1.1 .
figure 1.1 Source: Tele Satellite Magazine France
But if it isn't it gets a little more difficult. The mistake most people make is to assume that the difference between the longitude of the location and the longitude of the satellite is the quick way to calculating the azimuth of the satellite. Not so:
Lets assume you are in the French town of MézièressurSeine, 48.6º North 1.5º East
The longitude of both the satellite and the site from which you trying to receive it are both straight lines stretching from the North Pole (true) to the South Pole (true). But when you are looking at the satellite from any other location, the further you are from the pole, and the further the satellite is from your due south (true) the greater the error you make by assuming the point made in paragraph 7  see figure 1.2.
figure 1.2 Source: Tele Satellite Magazine France
So what you have to do is get out the scientific calculator and determine the true difference between the two lines of longitude, which we will call D .
We need to know the latitude of your own location, which we will call S . And rather than trying to do the whole calculation in one go, (which results in some fairly horrendous calculations), we'll first determine an intermediate value called Y , which we will then use in the second calculation.
Now Y = cos 1 (cos S x cos D ). So using our example of MézièressurSeine on Astra 1 at 19.2º East, the perceived difference between the two lines of longitude is 19.2  1.5 = 17.7: Y = cos 1 (cos S x cos D )
Y = cos 1 (cos 48.6 x cos 17.7 )
Y = 50.9494
Now the elevation E is calculated as tan 1 [(cos Y  0.15116 ) / sin Y ], so: E = tan 1 [(cos Y  0.15116 ) / sin Y ]
E = tan 1 [(cos 50.9494  0.15116 ) / sin 50.9494 ]
E = tan 1 [( 0.63000  0.15116 ) / 0.77658 ]
E = tan 1 [ 0.47954 / 0.77658 ]
E = tan 1 [ 0.61795 ]
E = 31.66
... and the true azimuth A is calculated as cos 1 (tan S / tan Y ), so: A = cos 1 (tan S / tan Y )
A = cos 1 (tan 48.6 / tan 50.9494 )
A = cos 1 ( 1.13427 / 1.2327 )
A = cos 1 ( 0.92015 )
A = 156.94
And, finally, just when you thought it was safe to think you had finished, you need to take into account the magnetic correction, which for our sample site was 3º East in January 1995 moving by 0.114º per annum westwards which in July 2000 gives a correction of 2.373 (or for round figures 2.4) which is added to the azimuth making 159.6
GOOD LUCK!!!
