I'm sorry it's such a struggle, bit that isn't enough of the end to have any confidence of where it fits. If it is really the same signal and it does overlap with the starting learn, then there is only one way it could overlap, so we could construct a macro and a single version of the full code and see if either work (but I don't have time tonight).

But since that end learn is so short, if it didn't work, we wouldn't know why. With a longer end learn we would have a better understanding of how it really fits together.

Did any of your other attempts yield interesting results? (Something other than matching the start or missing entirely).

Any chance of getting the "end" again but starting a twentieth of a second sooner.