Further MH370 Debris Analysis

2015 August 05
Updated: 2015 August 06, 07 and 08

1. Preamble by Duncan Steel

In a previous post I provided a link to a paper by Mike Exner of the Independent Group (IG) regarding drift analysis of debris from the crash of MH370. Two other IG members have now prepared independent papers on this topic, which are available for download and inspection by any interested party: please see items 2 and 3 below.

At item 4 a link is provided to a recent ATSB update on the drift modelling.

Two new items 5 and 6 below have now (August 7) been added. The first (item 5) contains some observations on the apparent damage to the flaperon based on the photographs available on the internet; these comments come from a consulting engineer who needs to remain anonymous for employment-related reasons. The second (item 6) is a further set of observations regarding the implications of the flaperon damage, from Brian Anderson, a member of the IG.

Something that perhaps needs stating, and which has not been mentioned in most of the media reports so far regarding the flaperon found in Réunion, is the fact that the item in question may not have left the rest of the aircraft structure when it hit the ocean. It is feasible that the flaperon became disconnected whilst the aircraft was at high altitude, and then fell independently into the sea. An indication of this comes from: (a) The ragged trailing edge of the flaperon (with 40 per cent of its total surface area being missing), this being consistent with rapid ‘fluttering’ as would likely occur if the aircraft lost all control surface power whilst travelling at high speed, as would occur when the fuel was exhausted from the two engines and the APU, and before the RAT fired up and provided a new (short-lived) source of electrical power; and (b) The lugs holding the flaperon to the main wing structure appear to have been torn off in a manner that may be consistent with such violent fluttering (at 10-20 Hz) at high aircraft speed. If the flaperon did fall from high altitude, perhaps having left the aircraft at around 00:15 UTC, then it would have entered the ocean at some distance from the crash site of the rest of the aircraft.

This (free fall from altitude) might also be regarded as being consistent with the fact that the flaperon remained otherwise intact. An entry into the ocean at a speed around 500 knots/900 kph by the aircraft might be anticipated to cause gross fragmentation of the aircraft in line with what occurred in previous jetliner crashes of this type. However, if the flaperon tore loose from the aircraft at altitude then it would fall to the ocean under gravity but be impeded by air drag, reaching some terminal speed which would be anticipated to be lower (perhaps only 150-200 kph) despite the fact that it is aerodynamically-shaped, due to its large surface area to mass ratio (remember that it floats!). That is, the flaperon might well have dropped in a tumbling, spinning manner, and so hit the sea surface at a relatively low speed, limiting the impact damage.

This scenario – the flaperon being ripped off the aircraft by aerodynamic forces at altitude and then free-falling into the ocean – is supported by observations of the damage to the flaperon as given in items 5 and 6 below.

2. Paper by Henrik Rydberg

The paper by Henrik Rydberg is entitled MH370: Finding the Debris Origin, and is available by clicking here or here or here (252kB download; note that Henrik’s paper was updated on August 05 with a slight rewording, and the links above are to this updated version).

3. Analysis by Richard Godfrey

The analysis by Richard Godfrey is entitled Combined Inmarsat, Drift and Fuel Model, and is as follows.

I have attempted to combine the information available from the Inmarsat (i.e. satellite-derived data), oceanic drift modelling and fuel consumption/limitation models into one summary table, as below. A brief explanation of the content of the table is as follows. 

I have covered the latitudes from 10S to 43S in terms of start locations near the 7th arc. 

I have made use of the BFO error at 00:11:00 UTC assuming an aircraft altitude of 35,000 feet and a ROC (rate-of-climb) of zero.

I calculated the point reached on the 7th ping arc by assuming a continuation on the same track at a ground speed of 432 knots.

I then used the closest integer (latitude, longitude) point to the 7th arc location as the March 2014 starting point in the drift model of Erik van Sebille (see this paper). His data provide the time in months to reach Reunion together with a relative probability of doing so. 

I also made an estimate of the remaining fuel range at 00:11:00, assuming MH370 did not fly around in circles at some point.

There is only one place on Earth which meets the Inmarsat (BTO and BFO) data, matches the fuel range, and is likely to result in debris washing up on Reunion after 12 months. That location is 34S, 94E. 

This table is also available within a PDF document (rendering better legibility) available by clicking here or here (36kB).

4. Revised account of drift modelling on the ATSB web site

There is a new post on the ATSB web site regarding debris drift analysis; here is a link.

5. Observations on the flaperon damage from an anonymous consulting engineer

i. If one strut were to fail the other strut would likely follow (doubled dampening duty) if similar flaperon action/forces continued. Once struts fail and disengage the flaperon can rotate at high speed about the forward/bottom hinge points without much, if any, resistance. The struts could fail last, but because the hinges have structural design to hold the flaperon on the wing while the struts function to control movement, this makes the hinges the strong feature, and so less likely to fail first.

ii. The two forward/bottom hinge connectors appeared to have failed due to lateral movement of the flaperon in the inboard/outboard directions. The curvature of the remaining inboard hinge metal is indicative of the hinge developing a fatigue (yield) line prior to failure.

iii. Once that inboard hinge failed (if indeed it were first) then the outboard hinge would fail in a short instant at its weakest point. The inboard hinge indeed seems to have broken at its weakest lateral point: If you were strong enough to laterally wiggle the connector back and forth (like a spoon), it would break at the surface as shown in various photographs.

iv. This lateral movement could be generated from rapid load reversals created by a spiraling/turning aircraft.

v. If the flaperon forward/bottom hinges experienced a massive tensile direction forces (i.e. pushing the flaperon straight up until the point of failure) then they would likely fail at the hinge’s hole connector, not at the flaperon bottom surface connection point as in  the photos. In addition the flaperon body would likely be severely deformed due to the extreme forces required to ‘tear out’ the connector from the flaperon.

6. Further observations on the flaperon damage by Brian Anderson

i. The relatively large missing training edge (TE) piece has sheared off immediately behind the rear spar of the flaperon. It is at this point that I would expect the bending moment to be greatest, and hence the most likely failure line for at least two failure modes:
(a) aerodynamic flutter, and (b) impact from the engine or parts of the engine passing aft immediately after impact. The media suggestions of the damage being caused by the TE “dragging” in the water in a low speed or controlled ditching attempt, I think, ignore the fact that there is a massive engine hanging way lower immediately in front of the flaperons.

ii. The shape of the deformity around the attachment point for the outboard [NB: altered Aug 08] Power Control Unit (PCU) attachment points suggests that it was the inboard [NB: altered Aug 08] attachments that failed first, followed immediately afterwards by the outboard [NB: altered Aug 08] .

iii. The tearing of the whole flaperon away from the PCU attachment points is much more indicative to me of (a) above. I think an impact to the underside and TE should result in compressive damage to the leading edge (LE), and would be more likely to shear at the mounting points, i.e. hinges and PCU points, leaving part of the hinge brackets still in place.