Also, I've been measuring the (Lunatic) timers again. I'll post my guesstimations when I think I've learned something.
Let me do this now, for the ones I (think I) know.
Stage 7
round 1 - 4:20, first hit at 2:00, second at 3:10, or around there
round 2 - 3:30
Stage 8
round 1 - 6:00, first at 2:20, second at 4:10
round 2 - 4:00
round 3 - looks like 3:20, might be 3:30
round 4 - somewhere between 2:30 and 2:50
Stage 9
round 1 - nonexistent
round 2 - probably around 5:30. First hit seems to be at 2:20, and the second at 3:55. it might be a bit earlier than this however, something like second hit at 3:50, final at 5:20.
round 3 - 4:00
round 4 - 3:30
Later rounds for these stages are unknown because runs involving them are rare, so I have nothing to study. I might estimate, seeing the similarity between round 2 of stage 7, round 3 of stage 8, and round 4 of stage 9, that round 3 of stage 7 and round 5 of stage 9 match the time I have for round 4 of stage 8, but that time is fuzzy anyway, and the round 1 times of stages 7 and 8 don't seem to match up with the second rounds of the following stages, so I wouldn't be too quick to guess. I will note however, when it comes to extrapolating based on patterns, that the time between the first and second hits in a given round seems to be the same as the time between the second and third hits, so if you know the times for any two hits, you can have a reasonable expectation of what the third should be, and if you've measured all three, you can check to be sure that the times make sense, as they should (probably) be equidistant.