Clarification on "Equate Alignments"

You don't need the exact same features, you need SIMILAR features. If you touched the granite before flipping, then touch the top surface after flipping. The trihedron needs to look in the SIMILAR direction, i.e. up before and down after (if you flip the part) or vice versa. If you end up with the trihedron into the granite, that's no problem, your free to rotate your trihedron any way you want after the equate. Don't forget the clocking of the part!

Easiest way to understand it is if you took your finger and held it on the part that you want to use as your initial origin.
Move the part as necessary (while holding your finger on that same spot).
Now create a new alignment to match how the part sits on the machine (following your finger).
Create the equate align afterwards, detailing what you did different.

As it applies to your specific situation:
--Initial alignment is a simulated part surface (by leveling to the surface of the machine), which is an inverted vector. you need to match the part's intended 6dof (finger on the spot).
--Your part's plane is actually a Z- vector, but you are probing in Z+ and leveling in Z+. This is the issue.
--When you flip the part and measure that true surface, the vector of that part's plane is flipped.
--You should construct that initial cmm surface plane, then construct a "flipped" vector plane of that plane. That way your initial alignment is relative to the PART.
When you do this, you won't need to flip the X vector between alignments.

So:
Measure cmm plate (Z+).
create flipped Z- vector plane, from that plane
create your rotate line
create your origin circle

align/level Z- to the flipped Z- plane
align/rotate X+ about Z
align/translate x y & z

move part.

create top plane
create same rotation line about the same vector direction, using the same face of the part as before
create the same origin circle

align/level Z+ to the top of the part
align/rotate X+ about Z
align/translate x y & z

equate align between the two, and you are set.

Normally when I use a Manual (A1) and DCC (A2) iterative alignments. When I have to use an Equate alignment I will measure the 1st end of the part, then when I'm ready to move the part (either flip or in my case usually in the 'Y' axis) I will move the part recall the Startup ailignment and do a new Manual (A3) and DCC (A4) iterative alignments. I will then equate the DCC alignments (A4=A2). I can then recall anyother alignments I might need or in the case of the latest equate I had to use finish measuring the -A- & -B- surfaces so I could create my A|B|C alignment.

Oh my god. I just realized that the reason it wasn't working the way I thought it should was because I didn't adjust the trihedron after the equate alignment! Thanks!