Originally Posted by
AlphaZuluRomeo
Joking aside, I also would like to know why the selected V/S varies so much (and so regularly) before AP disconnect. I think (guess) it is a "normal" behavior but would like to have a confirmation by knowledgeable people.
This is an educated guess rather than certain knowledge, but based on what I've learned on the subject of data analysis (which is actually part of my professional remit - lest CONF think otherwise) those negative peaks are at such regular intervals that I suspect that they are unreadable values rather than reflective of the actual setting, as I said here:
http://www.pprune.org/tech-log/46062...ml#post6741729
Given that it looks like they've used Excel to generate the graphs for the interim report, I know that Excel's default graphing behaviour for unreadable values is to render them as the lowest possible value (in this case -5000).
Now, whether that value would be unreadable due to corruption of the memory chips from damage sustained, whether the memory chips were faulty in the first place, or whether there was a pre-existing fault on the FDR bus that meant that the value was incorrectly read is something I'm not qualified to speculate on.
Ultimately though, if we can make a reasonable assumption - based the BEA's statement that the flight followed the ALT CRZ/NAV mode - that the flight was indeed in that mode up to A/P disconnect, then the V/S setting should not have had any effect as, if I recall correctly, the MANAGED mode does not use V/S as a parameter.