Earlier today, the Australian Transport Safety Board released a document entitled “MH370 — Search and debris examination update.” Perhaps occasioned by the recent completion of the towfish scan of the Indian Ocean seabed search area, the document updates earlier ATSB reports and offers some intriguing insights into what may have happened to the plane. Some thoughts:
— The first section of the report expands upon an assertion that the ATSB made in an earlier report: that the BFO values recorded at 0:19 indicate that the plane was in an increasingly steep dive. Indeed, the newly published calculations indicate that the plane was in an even steeper dive than previously reckoned: between 3,800 and 14,600 feet per minute at 00:19:29, and between 14,200 and 25,000 feet per minute at 00:19:37. On the lower end, this represents an acceleration along the vertical axis from 37.5 knots to 144 knots in eight seconds, or 0.7g. On the higher end, this represents an acceleration along the vertical axis from 140 knots to 247 knots, likewise about 0.7g. If the plane were freefalling in a vacuum, its acceleration would be 1.0g; given that the airframe would be experiencing considerable aerodynamic drag, a downward acceleration of 0.7 would have to represent a near-vertical plunge, which a plane would experience near the end of a highly developed spiral dive.
— The second section describes end-of-flight simulations carried out in a Boeing flight simulator in April of this year. These tests were more detailed than others carried out previously. Evidently, modeled aircraft were allowed to run out of fuel under various configurations of speed, altitude, and so forth, and their subsequent behavior observed. Thus, the exercise modeled what might have happened in a “ghost ship” scenario. Notably, it was found to be possible for the plane to spontaneously enter the kind of extremely steep dive described in the previous section. This being the case, the report states, the plane “generally impacted the water within 15 NM of the arc.” This is not surprising, considering that the plane had already lost altitude and was plummeting straight downward. This offers a tight constraint on where the plane could plausibly be if the 0:19 BFO analysis is correct.
— The third section describes the results of debris drift modeling that has been informed by tests involving replica flaperons “constructed with dimensions and buoyancy approximately equal to that of the recovered flaperon.” An important point not addressed by the report is the fact that the French investigators who tested the buoyancy of the flaperon were unable to reconcile its observed behavior with the observed distribution of the Lepas anatifera barnacles found growing on it. So when the French ran their own drift models, they had to run them twice, one for each buoyancy condition. Apparently the Australians overcame this paradox by discarding one of the states.
— The third section notes that, according to modeling carried out by the CSIRO, debris which entered the ocean in the southern half of the current search area would not likely reach Réunion by the time the flaperon was recovered. Meanwhile, debris that entered the water significantly north of the current search area would reach the shores of Africa much earlier than the time frame in which pieces were actually discovered there. Using this logic, the report concludes that the northern part of the current search area is probably correct. However, this seems dubious reasoning to me: one would expect a gap between the time debris arrives in Africa, and the moment when it is discovered. Also, debris can move quickly across the ocean, only to be trapped in a local gyre and move around randomly before beaching. Therefore I think the argument that the pieces couldn’t have originated further north is flawed.
— The fourth section, describing the damage analysis of the flap and flaperon, is the most interesting and newsworthy of all. In short, it makes a persuasive case that the flaperon and the inboard section of the right-hand outboard flap (which, rather remarkably, turn out to have been directly adjacent) were in the neutral, non-deployed state at the moment of impact. Assuming this is correct, this eliminates the IG’s flutter theory, as well as the widely discussed theory that the flap was deployed and therefore indicative of a pilot attempting to gently ditch the plane. Proponents of these theories will continue to argue on their behalf but in my opinion they were dubious to begin with (given the shredded condition of much of the recovered debris) and are now dead men walking.
— No mention was made of Patrick De Deckker’s exciting work with Lepas shells.
— Overall, the thrust of this report is that the plane went down very close to the seventh arc in a manner consistent with a “ghost ship” flight to fuel exhaustion, exactly as the ATSB has assumed all along. There is, however, one very large elephant in the room: the fact that Australia has spent two years and $180 million demonstrating that the plane’s wreckage does not lie where it would if this scenario were correct. Therefore it is not correct. The ATSB’s response to this conundrum is rather schizophrenic. On the one hand, it has recently floated the idea of raising another $30 million to search further—presumably the small remaining area where a plane just might conceivably have come to rest in a ghost-ship scenario, as I described in an earlier post. On the other, it has today convened a “First Principles Review” consisting of experts and advisors from Australia and around to world to scrap their previous assumptions and start with a clean sheet of paper. This implies an understanding that they have proven themselves wrong. I wonder how many assumptions they will scrap. Perhaps, as Neil Gordon mused in his interview with me, that the plane wasn’t really traveling south at 18:40? Or perhaps they’ll dare to go even deeper, and contemplate the provenance of the BFO data… ?
— A postscript: Richard Cole recently posted an update of the seabed search (below). I’m intrigued by the fact that the Fugro Equator has deployed its AUV near the northern end of the search zone. When I interviewed him for my last blog post, Fugro’s Rob Luijnenburg told me that the northern end of the search zone was flat enough that it could be scanned by the towfish alone; there was no need for an AUV scan to infill the craggy bits. So why is the AUV looking there now? Especially given that it’s very close to an area just reinspected by Dong Hai Jiu 101’s ROV. Another MH370 mystery.
UPDATE 11-2-16: I emailed Rob Luijnenburg and he immediately responded: “The AUV is scanning in a section in the north part of the priority search area in the very rugged terrain south of Broken Ridge (the east -west mountain range at approximately the 33rd parallel)… Generally the AUV is deployed in spots of extremely rugged seabed to complete the 120,000 sq km priority area survey.” Worth noting is that if the search gets expanded northeastward, it’s going to be into very rough terrain indeed.
@ROB: Are you replying to somebody else’s post?
@Gysbreght
–“It looks like a high pressure on the lower surface acting uniformly over the span, caused the lower skin to tear along the line of fasteners that connect it to the rear spar. Similar observations can be made for the inboard section of the outboard flap.”
I completely agree. Where we differ is on what that “pressure” was. I find it impossible to believe that that pressure was created by water during a water impact.
When you write “it is not at all like the flaperon was hit by a solid object like an engine pod,” you completely prove my point. A nose-high water impact would ABSOLUTELY have sent the engines directly into the bottom of the wing causing damage longitudinally across the skin. This is clearly not what happened.
In other words, the ONLY way I can explain trailing-edge-only damage to the flaperon is due to a VNE+ / over-G dive with either flutter or excessive loads or both with the leading edge protected by the full chord of the wing ahead of it. Mike Exner agrees with this.
China 006 is a case worth looking at. https://en.wikipedia.org/wiki/China_Airlines_Flight_006#/media/File:Damaged_empennage_of_China_Airlines_Flight_006-N4522V.JPG
In addition, I don’t find it coincidental at all that the relatively intact debris finds have all been control-surface related and the shrapnel sized stuff has been leading edge related.
That says high-speed, near vertical impact.
@Rob
I think you’re actually talking to me. And I’ve explained it many times. So has Mike Exner.
The surfaces detached during a VNE+, flutter/over-G dive.
Like China 006.
@ROB
“My own conclusion is that the aircraft most likely came down within the area already searched, the pink(purple)/black area, but has somehow been overlooked.” Possible but very unlikely IMO. Actually, I agree with the strategy that the ATSB et al. announced after the 22 July meeting “..it was agreed that should the aircraft not be located in the current search area, and in the absence of credible new evidence leading to the identification of a specific location of the aircraft, the search would be suspended upon completion of the 120,000 square kilometre search area.” And I don’t see any credible new evidence yet (but perhaps there is some not in the public domain). So I can’t see a reason to extend the search at this stage.
@ Jeff Wise
So does the latest report about the ghost flight scenario support the ‘loiter’ near Andaman and Nicobar theory?
How else could the perps have escaped?
The new ATSB Update provided confirmation of many things I have believed for a long time. I was very happy to see that. But there was one nugget of new information worth noting: the Thales test results on OCXO “warm-up drift”. Granted, Figure 4, without any scales, is disappointing in its vagueness. But the text offers clear guidance in the interpretation of Figure 4. The description on Pg9-10 suggests the following (first order) corrections for warm-up drift:
Warm-up drift at 18:25 = ~100-130 Hz (cold start +1 min)
Warm-up drift at 18:28 = ~30-50 Hz (cold start + 4 min)
Warm-up drift at 18:39 = ~0-10 Hz (cold start +15 min)
Warm-up drift from 19:41 to 00:11 = ~0 Hz
Warm-up drift at 00:19 = ~0-10 Hz (warm start +1 min after power off for 1 min)
Importantly, in all cases the FBO Bias error is reported to be positive. The assumed warm-up error decay rates used in these estimates are based on my experience testing TCXOs and OCXO’s. Although they are only educated guesses, these estimated BFO Bias corrections, at these critical times, may provide new insights, particularly in the quest to better understand what was going on in the Malacca Straight. The jump between the first two values circa 1825 is still odd, but after the warm up bias is taken into account, all except the first value appear to make more sense now.
It should be emphasized that the warm-up bias at 00:19 was almost certainly very small, probably <10 Hz because the power was only off for ~ 1 minute. The oven does not change temperature very fast, so it was very close to equilibrium at 00:19, but not at 18:25. That means the descent rates were as I have been reporting since 2014, about 4,800 and 15,000 ft/min, not the extreme cases ATSB noted if the maximum (130 Hz) error is assumed.
Graph or corrections: https://goo.gl/uTrxMt
@All, If the SAA285, the helderberg is anything to by: there were 3 debris fields at-5000m. It took Oceaneering International 2 days to find the fields. This was 1987. The aircraft, according to ATC logs, was flying at approx FL5 when they lost contact, and subsequently crashed into the SIO. What would this mean, in terms of debris fields, for MH370 considering the aircraft was flying at much higher altitudes and rate of descent was within the ranges Jeff posted?
In answer to reporter questions about the assumptions made in defining a search area in the latter part of 2014 several official references were made to the expectation that the large sunk pieces of wreckage, like the main gear, the engine cores and so forth, would be distributed relatively close together and be readily recognized as probably belong to an air crash debris field. My concern, which I should have mentioned earlier, but did mention in another place is that these major components might not be close together, either measured by depth on a slope, or in horizontal displacement. They may also be partly covered in mud or silt. Note that when the towfish fell to the ocean floor on the side of a mud volcano a part of the supporting assemblage, which acted as a stabiliser, was imaged, partially embedded in mud, by the recovery vehicle. This tells us that there is a risk that heavy objects from MH370 may be significantly covered by silt or mud, which may frustrate ready identification as part of a widely dispersed debris field.
@All
An interesting article on how Bayesian Inference works if your like me and new to this application.
http://www.datasciencecentral.com/profiles/blogs/how-bayesian-inference-works
Does this mean even a part as big as the engine would only be displayed as 4×4 pixels ?? You certainly can’t distinguish a 4×4 pixel spot representing an engine from all the other 4×4 pixel spots representing underwater rocks …
@TBill, Matt’s earlier rebut to Gilbert’s baloney, of solid! frozen eyeballs and eardrums and grabbing a parka from the overhead in 12C and fighting a way back to the cockpit to avoid a crash in a residential area is looking more attractive by the minute. Perhaps the ATSB believes that divine intervention is what manoeuvred the aircraft was to FMT. And they want to burn another 30 mil on their current assumptions? WOW.
@Matt Moriarty:
That is not at all ABSOLUTELY certain. If the engine separated (there are several cases where an engine did not separate), then it quite possible, indeed more likely, that it would pass over the wing, as it is designed to do.
And what do you make of the outboard flap section?
China Air 006 was discussed earlier and I didn’t look up the report again. IIRC, all damage was due to structural overload as the airplane exceeded 4 g in the pull-up. There was damage to the outboard end of ailerons and elevators, no damage to the flaps.
Congrats that Mike Exner agrees with you, but you may have noted that on airplane matters I don’t always agree with Mike Exner.
As to AF447 I agree that we must be careful to make any comparisons, but the angle of attack at impact was more than 60 degrees in that case.
Minister’s Press Conference on current Review.
Nothing much said. Non-committal on new search possibility
http://minister.infrastructure.gov.au/chester/interviews/2016/dci011_2016.aspx
@Tom Lindsay. Bayesian Inference. Thanks, helps.
@Gysbreght
Apologies. Seems like I was replying to someone else’s post. I had rather a busy day yesterday, brain fog must have set in.
@Ben Sandilands said;
“Note that when the towfish fell to the ocean floor on the side of
a mud volcano”
Although I assign a somewhat higher likihood of probability that the
sonar search is likely to detect MH370 debris, I won’t quibble with
you over that – but the incident as reported is that the towfish
collided with the mud volcano, therefore isn’t your
example a poor supporter of your viewpoint? It is just as likely that the collision acted to partially bury the assemblage, particularly
if you figure that the assemblage may have been dragged for a short
period against the mud volcano before the tow cable snapped.
@Gysbreght
“If the engine separated.”..”it would pass over the wing”
Haven’t you made a descriptive error here? “over” seems incorrect.
@buyerninety: In my understanding of what is not my native language “over” is the opposite of “under”. I don’t understand your question.
@Seeker, I should acknowledge that I got the scale of the engines wrong. The core is more like 1m x 1m x2m (sorry, relying on memory here.) So the issue you raise is even more germane.
I don’t really have an answer for you, lacking expertise or special insight on the topic. Fugro and ATSB have expressed full-throated confidence, and are soon to release the full data set, supposedly. Perhaps someone would like to look deeper into this topic?
@AirLandSeaMan. Further about what exactly the Report says about separation, its rationale does not support its final conclusion, “The right flaperon was probably at, or close to, the neutral position at the time it separated from the wing.” It makes the case for the flap being housed at separation but based just on its discussion the flaperon could have been damaged by the separating inner flap part striking it on departure, itself separating later at other than neutral deployment. From the damage to both it is the more likely to me that it hit the flap than v.v. but the Report leaves that to conjecture.
The ATSB had a rep present during the flaperon investigation and I notice they have a detailed BEA photograph, fig 21., so they have information. The shame is they do not have physical accesss so cannot match the indentation and witness marks of the two by “joining” them, disclosing relative angles. They may well be bereft of failure characteristics also of the four main flaperon attachment points, which would well contribute a deal to sequence as could the crushing of the flaperon at its outer front.
If they have this information already, their report is shallow. If not, the investigator, Malaysia, should have insisted on it. This is tying your own hands behind your back.
@ALSM, @Dennis W, @JW.
The new report indicates that it is highly probable that the BFO readings from 1825 to 1840 were affected by “warm up drift”. As ALSM states, it is frustrating that Fig.4 gives no axes labels for magnitude/time on the decay function of oscillator drift.
Prior to this revelation, BFOs during this period were judged to be consistent with northwesterly flight direction at 1825 and a turn to the south by 1840. Where does this leave us now? ALSM has provided a thumb-suck idea of magnitude of drift correction that may be applied to BFOs during this period, ranging from ~ -100 to -130 at 1825 to ~0 to -10 at 1839.
ATSB tells us that “The maximum OCXO drift value observed in the previous data of 9M-MRO was around 130 Hz”. This, presumably, after an extended powered-down period, on the ground, at whatever the ambient temperature was.
Can any of you suggest how much greater the drift correction factor might be if the ambient temperature in this case was ~ -40 centigrade rather than +20?
Perhaps it is time to begin considering the possibility that the “possible turn” we learned about in late March 2014 might not have occurred at all; that the trajectory south arose from initial turnback, and that the “radar data” that many hold so dear arose from “looking for what we ought to see [per initial ISAT analysis]”.
@Gysbreght wrote;
“If the engine separated (there are several cases where an engine did not separate), then it quite possible, indeed more likely, that it would pass over the wing, as it is designed to do”.
The engines are located under the wings. The paragraph above, as read,
suggests an engine would pass above a wing. It is difficult to
understand the circumstance as to how this could happen, especially
in the case of a 777 striking the sea surface in a slight, or
substantial, nose-high attitude (i.e., tail, or tail + rear underbody,
being the first part of the aircraft to impact the sea surface.
@buyerninety: Yes, you read my comment correctly. You need to consider not only the attitude but also the flight path angle. Angle of attack is roughly the sum of attitude and flight path angle. It is the direction of the air coming towards the airplane, and ultimately the sea water.
The engine pylon is attached to the wing essentially at three points, two in front at each side if the pylon, and one point at the rear. In a wheels-up crash landing the rear attachment bolt is designed to fail, allowing the engine+pylon to swivel upwards about the front attachment points, and to depart rearwards above (over) the wing, to avoid rupture of the fuel tank.
@Gysbreght
I am surprized that is (how you state) it is designed – I believe
other readers may query this assertion, so perhaps you could have
a reference ready to cite to them, if they request a reference.
(I was aware of only one aircraft where an engine moved as you
describe over the wing, this was not the designed intention, however.
That was American Airlines Flight 191. Its engine(s) support
structure was designed {in the case of a frontal impact upon the
engine} to allow the engine to separate from the wing without damage
to the wing. Due to faulty reinstallation of an engine, which caused
stress fracture or cracks to the support structure, the engine
support structure failed at its rear, but the front of the support
structure remained attached to the wing. Result was the engine
went forward of and above the wing, damaging wing controls {slats}
and hydraulic lines {which Boeing had poorly routed all along
the front interior of the wing. All on board were killed.)
@all
I realise at this stage many more people will be hopefully finding the forum so I apologize in advance for this fantasy-fuelled post but I’ve thought up a chain of events which I’m eager to put out there.
(Warning: yet-another-wacko-theory ahead…)
So, we learnt yesterday of a straight vertical death dive followed a few months later by a mid-air explosion… (okay, here goes…)
MH370:
On 8th March 2014, for whatever reason, MH370 suddenly goes dark, does a u-turn, travels over Malaysia and then out to the Andaman Sea. Taking no chances, the US military shoots it down; not a brash direct hit, more like a carefully crafted ‘surgical strike’ (a la Wazir Roslan), just enough to blow a large enough hole in the fuselage so it nosedives almost vertically and leaves no debris field…
Senior Malaysian officials know what happened (Hishamuddin – Four Corners “…the Americans would…”), but word also trickles out to local intel – senior officials in Thailand, Indonesia, Australia etc – who simply close ranks ‘saw nothing, heard nothing’ and move on.
But – one way or another – an ‘outsider’ nation completely detached from the event manages to learn of the shootdown… yep, the Russians!
MH17:
MH370 massively plays into the hands of the Russians. Concurrent to the unfolding mystery, the Russians suddenly find themselves in the middle of a war versus the Ukrainians, the EU, and the West at large. Russian intel learns of MH370 relatively early on, but having no real use for this ‘info,’ ignores it and moves on.
As war rages and the EU and West become more belligerent, the Russians want to teach them a lesson. This is when that earlier info comes in handy. They craft a plan to shoot down an airliner with majority EU citizens on board, yes, quite simply, cold-blooded murder – ‘this is what we’re capable of… DON’T MESS WITH US!’
But also an indirect two fingers up to the US and its allies – ‘You’ve Chinese blood on your own hands… and you need our silence… SO STAY AWAY FROM OUR YARD AND WE’LL STAY AWAY FROM YOURS…!!!’
@Sajid, I agree that the first part is crazy but the last bit — MH17 as “DON’T MESS WITH US!” signal — is eminently plausible, IMO.
Paul Smithson: If the rear cabin was at -40C at 18:25, I don’t think the initial BFO Bias error would necessarily be larger than 130 Hz. The SDU firmware prevents transmissions until the temp stabilizes to a point near equilibrium (130 HZ?). The power may have come on sooner than 18:24, requiring more than the reported 1 minute for a boot, but I don’t think the logon BFO Bias error was > 130 Hz.
@buyerninety:
American Airlines Flight 191 was a McDonnel Douglas DC-10-10. Boeing was not responsible for it at the time. The Wikipedia article on the accident cites two ‘similar’ accidents. I’m familiar with the ElAl accident at Amsterdam which was a B747 where the engine/pylon separated from the wing in a similar manner. Among the references of the Wiki article you’ll find the NTSB report on AA191. It has a drawing of the pylon attachment structure and says in paragraph 1.17.4 on page 39:
@airlandseaman
Do you have any theory on how could the BFO of 18:25:27 be around 130 Hz less than the subsequent one 6 seconds later?
Specially as you say “The SDU firmware prevents transmissions until the temp stabilizes to a point near equilibrium”!
Does that mean that the real value at 18:25:27 is around 130 Hz too high too? IE it should be 12Hz instead of 142Hz? Would that also mean that the plane was in a steep dive then? And the fact that this value aligns with the stabilized ones a few minutes later is just an other coincidence?
@Sajid UK
There seems to be little reason why the U.S. would shoot down such
an aircraft – the Andamans are about 1500nmi from Diego Garcia.
Also, I never found any indication on the internet that there was
any military sea exercises taking place in the near sea area at that
time (although Kate Tees thought there maybe were sea convoys to the
south of her).
Disregarding Australia or Thailand, it is unlikely that Indonesian
‘officials’ would remain quiet about an aircraft downing by the U.S.,
especially when that aircraft had 7 Indonesian citizens on board.
@Sajid, @JW – while I agree mh17 may have been a signal I think it was delivered to Malaysia as a reminder to honor or demand payment on MH370. As Nazak immediately payed up whatever seemed in owing. I can see connection to mh370 if there was a significant value in the cargo but was directed away from getting into Russia’s hands.
@Paul Smithson
re: SDU oscillator
I am glad you brought that up. I have been scratching my head over that for some days now, but decided to just gag it down, and move on with my life miserable as it is.
The BFO logged at 18:25:27 corresponds exactly to an aircraft track of 296 degrees at a ground speed of 510 knots at a location where the airway from Mekar to Nilam crosses the 18:25:27 range ring. The 144Hz is not just close. It is perfect. (The airway from Mekar to Nilam is at a bearing of 296 degrees.) This path is the path most analysts including the DSTG believe the aircraft followed.
That is truly amazing to me given that our working hypothesis is that the SDU oscillator had been off for some time. In fact, it falls in the category of too good to be true. Possible, yes. Probable, no. I would ask Exner to ponder other possible scenarios in which the AES was unresponsive but with the SDU oscillator in the “on” state.
Certainly this measurement would rule out extreme cabin temperature changes associated with a depressurization event. An oscillator is a very good proxy for a thermometer.
@ALSM. Your suggestion sounds sensible. However, we don’t actually know what max drift was for this SDU under non-normal ambient temp conditions, do we? Only that max warm-up drift observed for this unit (flights pre-accident) was -130Hz. It seems we don’t have sufficient information to know whether a larger drift is feasible/probable or not if ambient temp was significantly lower on power-up. I hope that other commentators with expert knowledge on OCXO might contribute their views.
sinux: I do not have any solid idea why the 18:25:27 BFO value is what it is. Speculating, it might be that the cabin was at -40C at 18:25, and the OCXO warm-up characteristics were different from what Thales measured in the lab, due to the extreme temperature difference. This is only one of many possible explanations.
@Paul
BFO above should read 142Hz not 144Hz.
Regarding OCXO error vs. ambient operating temp…I did turn up some data on that question. The OCXO is very stable over the full range from -40 to +55C. But it is not “perfect”. In one test, the frequency (at L band) was about 20 Hz low for -40C. It was essentially constant (no error) from about -10C to +30C.
@Gysbreght
(I beg your pardon, yes, that was a McDonnel Douglas, not Boeing, designed
aircraft.)
Although the reference you gave described the designed intention of the
(engine) holding points to fail so that the fuel tank is not damaged, it
does not follow that the designed intention is for the engine to then
rotate forward and upward to pass over the wing top. My understanding is
that the designed intention for the underwing engine support structure
(applicable to all aircraft manufacturers) is that it allow the engine to
separate from the wing so as to cause no damage to the wing – the
unspoken implication being that if the engine remained attached to the
wing, that the wing underside could be ripped away (and given that the
wing fuel tanks are integrally formed by the enclosed space of parts
of the wing boxing structure, ripping away part of the wing underside
is likely to cause fuel leakage).
The main circumstance they are planning in case of, is a frontal impact to
the engine, and the designed intention in that case, is that the engine
separate from the wing without damaging the wing. The unspoken understand-
ing is that the engine will simply drop downwards and away.
Your reference supports my understanding, but it does not provide any
evidence for your viewpoint that the engine is meant to then rotate
forward and pass over the wing (top). This is why I don’t embrace your
assertion on this matter.
Cheers
@ALSM
OCXO stability depends on how hard you look. Back when I was working for a living I would have a room full of GPS disciplined double oven oscillators for CMDA base stations under test. The double oven was far better (cost ~$700 in high volume) than the single oven TCOCXO used in the AES. It was easy to see oscillator frequency changes logged against a cesium when the air conditioner in the room cycled on an off. The frequency will always change with temperature. The only question is how much. The oscillator cut makes a big difference as well, and very small variations result in significant variability among units intended to be identical. In fact, we had a 20% reject rate on the $700 double oven.
DennisW: I also have extensive experience with these oscillators, mainly for Goes DCPs, GPS and spacecraft applications. And I have also witnessed the AC cycles in test data. But we have hard test data on the specific OCXO used in the Honeywell AES. See this graph: https://goo.gl/amxQbs
@Jeff
Sorry to get so animated on this oscillator issue, but it is troubling. It is almost as if someone left a “breadcrumb” the size of a Buick for us to find.
@Keffertje
Not following you exactly, but I still do not see a mechanical failure accounting for the accident as yet.
@buyerninety: ” The unspoken understand-
ing is that the engine will simply drop downwards and away.”
How does the engine “drop downwards and away” in a wheels-up landing on a low-wing airplane? Perhaps you should rethink your understanding.
@Jeff You said “Also, debris can move quickly across the ocean, only to be trapped in a local gyre and move around randomly before beaching. Therefore I think the argument that the pieces couldn’t have originated further north is flawed.”
To me this is typical government bureaucratic ignorance from ATSB. They’ve only selected certain data to fit into their projected final end point of MH370 and have ignored what I believe is other plausible calculations made from some pretty intelligent people outside of the ATSB..Like IG…Therefore ATSB data calculations as to where MH370 may have impacted in the SIO is flawed…
@Sajid @Jeff
what if what you think was exact wish of someone to be considered in strategy against well known devil from scratch, as perfect media coverage followed by call to action – call to war (I dont mean US, no) – may be it all can stay hidden for another 100 years or so to not worse it all for regular people; even buicks may be seen only be some parties
@ALSM. Thanks for sharing the freq stability vs temperature chart. It’s not clear to me how exactly we bring this information to bear on the issue at hand. This, presumably for a unit that is already warmed-up/stable, and freq change with incremental ambient temperature change?
ATSB clearly shows a “drift” that decays over time. Magnitude of initial drift value presumed contingent on how much the unit had cooled down. Slope of the “decay” presumed contingent on how quickly it warms itself up. Both factors clearly dependent on what the ambient temp was. “Max drift” as the unit temp reaches operational threshold unknown; possibly -130Hz, possibly more?
@Dennis W. Yes, odd, isn’t it? Odd also that we just happen have a solitary “radar” detection within 3 mins of first BTO (which happens not to align very well with Pulau Perak re speed/position).
Bear in mind that what we know as the “radar narrative” didn’t take shape until the ISAT data was out. Up until that point, all we had was a turnback, heading unspecified (reciprocal course, KLIA and Subang all mentioned early on as possibilities). Only on 15th March did the PM assert that post-Penang radar traces belonged to MH370 – with direct reference to the ISAT data. And *still* the specifics (where/when) of those Malacca straits detections kept changing…
@Tbill, I apologize, mechanical failure is last on my mind: I was being sarcastic as it relates to ATSB and the fact that theyare ignoring events between 18.22UTC – 1940UTC and pilot intervention.
@DennisW, To clarify, it seems that there’s a BFO value at the 18:25 that no one, not even Thales, has any kind of explanation for.
@airlandseaman
If the cabin and the SDU were -40C at, thus before the 18:25 log-on, then the cockpit was also -40C. And the cabin and cockpit would be depressurizes also.
Would a pilot be able to fly the plane and perform a FMT at or after 18:40 under those circumstances? Or even survive during this time?
I doubt it.
And then indeed the flutter-issue was dicussed before regarding the 747 exceding 4G after a high speed dive pulling out that caused the damage mainly on the horizontal and vertical stabilizer. No (outboard) flaps were damaged.
If the flaperon and outboard flap were damaged and seperated by flutter it’s very hard to imagine they could have damaged eachother in neutral positions. They both would have fluttered violently in such a case. The coïncidence of hitting eachother exactly on those spots would be very slim IMO.
And if you look closely. The pointed damage in figure 21 on the flaperon is on it’s bottom-side. And the correspondencing damage on the outboard flap is on the top-side.
This suggests to me that if the flaperon was in a neutral position the outboard flap had to be in a slightly lower position of about 5 to 10 degrees.