While I'm at it, I noticed some odd behaviour of my CVT Unit/Extractor setup:
I tested a setup where I would input the redstone signal to the CVT unit from below using Project Red red alloy wires. Since the wire doesn't visibly connect to the CVT from below, I didn't know if this would be working. In the isolated test (with no machine actually attached to the CVT but a dynamometer instead) it appeared to switch correctly. However, when using the same method on my production setup which runs the extractor, stage 2 and 3 didn't visibly speed up when the switch was activated. I then changed the setup again to input the redstone signal from the side of the cvt unit, and things worked as they should again.
Needless to say, I find this very odd. I wouldn't have been surprised had things not worked at all with the red alloy wire not visibly connecting with the CVT, but the dynamometer said it was working. I'll test a similar setup again with the latch realized in a modular socket with direct redstone output to the top, and anyway I have a way that does work but I'd rather like to know the reason for this odd behavior.