Alignment Compensation AS1 85 Absolute Arm using probe. . .?

Are the probes freshly calibrated and passing their length bar check? 
If so, then is the base of your Romer aligned with the trihedron as displayed in your program?
The line is controlling one of the axis origins and the rotation, correct? 

That is correct for the axis. The base of the Romer is not visualized in the program. Romer is set on the back side (y+) of the part. Plane on top(zstop), plane on left(xstop), angled plane on front intersecting with left side plane to make theoretical line in z axis(ystop & rotate) which part is rotated to. Only axis which is off to CAD is y, the full length of probe diameter. Alignment has been cleared and redone many times, occasionally it will work, and then not work eventually. And, the calibration for probe tips was done by Hexagon as we have a rental as ours is being calibrated and fixed. Probes and alignments work with other programs, just when making this new one does not seem to be working. 

That is correct for the axis. The base of the Romer is not visualized in the program. Romer is set on the back side (y+) of the part. Plane on top(zstop), plane on left(xstop), angled plane on front intersecting with left side plane to make theoretical line in z axis(ystop & rotate) which part is rotated to. Only axis which is off to CAD is y, the full length of probe diameter. Alignment has been cleared and redone many times, occasionally it will work, and then not work eventually. And, the calibration for probe tips was done by Hexagon as we have a rental as ours is being calibrated and fixed. Probes and alignments work with other programs, just when making this new one does not seem to be working. 

Post your code please. 
Programs "working sometimes" is kinda indicating that your Romer base is not aligned with your trihedron. 

Here is the code for our manual probe alignment. Thanks for your help on this.

A few things to try. 
Transform the trihedron in your program so it matches the reality of how the part is sitting on the table, the vectors in your code are saying its way off. (the Demon dislikes having to guess)
Align the base of your Romer so its roughly in accordance with the trihedron in your program. (this should not matter at all but for some reason it does, we figure cause the Demon hates guessing)

Axis of the Romer base, I assume they are all the same? It's marked on all our machines for this exact reason. 

We will try this.