I use a variation of Linear interpolation between two known points to locate a new sat position. This works if you already have some sats set in your positioner box, if you are starting from scratch it won't help much. Lets say I have 72w and 83w set in my box and I want to find 77w. I move to 72w and note that the positioner box count is 159. Now I move to 83w, the count is now 206. The count difference between those sats is 47. The orbital difference is 11 degrees. Divide 11 into 47 and I have 4.27 counts per degree in this part of my sky.
Now there are 6 degrees between 83w and 77w. 6 times 4.27 gives 25.6 counts from 83w to 77w. My dish is still pointed at 83w with a count of 206. I back up 26 counts to 180. At this position I am pointed at or very near 77w.
Do a blind scan then peak up the position and set it in the positioner.
Since the counts per degree are not linear across the sky, there is a slight error using this method. But you will be close enough to find signal every time. You should re-calculate the counts per degree between the two nearest known sats for each search.
Here is a link to more information on Linear interpolation.
http://en.wikipedia.org/wiki/Linear_interpolation
Now there are 6 degrees between 83w and 77w. 6 times 4.27 gives 25.6 counts from 83w to 77w. My dish is still pointed at 83w with a count of 206. I back up 26 counts to 180. At this position I am pointed at or very near 77w.
Do a blind scan then peak up the position and set it in the positioner.
Since the counts per degree are not linear across the sky, there is a slight error using this method. But you will be close enough to find signal every time. You should re-calculate the counts per degree between the two nearest known sats for each search.
Here is a link to more information on Linear interpolation.
http://en.wikipedia.org/wiki/Linear_interpolation