NavList:

Every GPS that I have worked with has a settings option that allows you to select from several different world models.  Every chart will have marked on it what world model is being used.  Some from the south Atlantic and middle/south Pacific have world models as well as survey dates from the 1800’s.

Every almanac or reference text worth its salt will tell you the world model (for example world model 1996, world model 1994, or world model 2005).  There are minor differences between each of them, especially when they are is years taken fairly close together like the ones that I have listed.  But when you compare the results the that you get from an almanac published with a world model from the mid 1970’s to one using a mid 2000 world model, there is a big difference that would more than account for the discrepancy in your calculations.

The thing is that your discrepancy is with the norm for acceptable error for most celestial LOP’s.  So unless you are using a sextant and horizontal angles to further establish your position, or using some other form of electronic navigation to further refine your position, you are doing quite well.

Me, I would start looking at my chart to determine what (if anything) that I can use to confirm my position (horizontal angles to land points for a fix or vertical angle to a land feature or a structure to establish a COP).  And then keep going.  As long as I am consistent in the chart and reference book that I am using for my calculations, my results will be consistent.  The problem comes in when I attempt to check my results between different types of navigation equipment without checking to see if the underlying assumptions are the same (i.e. world model being used).