PPRuNe Forums - View Single Post - AF 447 Thread No. 5
View Single Post
Old 11th Jul 2011, 13:15
  #72 (permalink)  
HazelNuts39
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
TE plots

mm43;

Thanks for your superb graphics work enhancing the crude EXCEL graphs I posted here, and for posting your graph on this thread.

Originally Posted by BOAC
One further comment - I am concerned that these 'graphs' produced after hours of hard work by folk like MM and A33 will pass rapidly into the 'factual' world - as pseudo FDR traces - when in fact they are supposition - guess-work - based on extremely limited information. We just do not know enough at this time.
Of course you are quite right. My 'pseudo FDR traces' are to some extent 'supposition - guesswork', but I don't understand your concern. I thought that, while we're waiting for the publication of the real FDR traces and many posters discuss AoA's, pitch angles, airspeeds etc., this was better than nothing.

Perhaps it would help to explain in some detail how the data shown on these graphs have been derived. In this 'computer simulation' (an EXCEL spreadsheet) the airplane manoeuvre is driven entirely by an assumed variation of acceleration (az/g) along the earth vertical that is shown here. At first, that driver function is just a wild guess. But then the resulting v/s, altitude, airspeeds, AoA, etc., is compared to the data in the BEA Update, including the time line, and the input function is 'massaged' to improve the goodness of fit, and the same process is repeated. After a large number of iterations I felt that 'version 4' is as good as it will ever get until we get the real data.

The calculated parameters are derived from the 'driver' function az/g by strict application of flight mechanics (i.e. Newton's laws) and A330 aero data published by Airbus (where available(*)). (**). The calculation then goes as follows: V/S is calculated from (by integration of) az/g. Altitude is then calculated similarly from V/S. Subtracting altitude from total energy yields kinetic energy and hence TAS, CAS and Mach. V/S and TAS together yield FPA. Next lift coefficient CL is calculated from weight, normal load factor n, Mach and altitude. CL and Mach yield AoA from the aerodynamic data (*). Finally, pitch angle is calculated as the sum of AoA and FPA.

For what it's worth, until we get the real FDR traces ...

(*) The available data cover the operating envelope between Mmo and buffet boundary. There are no data close to the stall and beyond. For these conditions I have assumed a pitch angle of between 15 - 16 degrees and calculated AoA from Pitch-FPA. Whenever I quote a stall-AoA, it is based on what I think is a reasonable extrapolation of the available data.

(**) This means that, if the airplane is subject to the assumed az/g, then it must follow the derived data shown on these graphs. In other words, the data shown are mutually consistent.

Last edited by HazelNuts39; 11th Jul 2011 at 13:30. Reason: Footnote (**)
HazelNuts39 is offline