Brock McEwen has released a new reverse-drift analysis of the MH370 debris that has been found in the western Indian Ocean. The executive summary is below.
Broadly speaking, Brock’s new paper supports the conclusion of his earlier work on the subject, and also parallels the findings of GEOMAR and Météo France, as I’ve written about earlier–namely, that reverse drift analysis suggests that the debris did not originate within the current search zone.
In conducting his analysis, Brock has erroneously included objects found in the Maldives which did not come from MH370, but my understanding is that the inclusion of this bad data did not materially change his results.
The Australian is reporting that “Despite finishing his term as the head of the ATSB without finding MH370, [Martin] Dolan said he remained hopeful the aircraft would be found” and believes the search should continue. The full story is behind a paywall but Amanda Rose has provided a screenshot here. Also of interest in the article is the assertion that, due to bad weather, the search might stretch on through October.
Meanwhile the New Straits Times says that “The ministerial tripartite meeting on the Malaysia Airlines Flight MH370 will be held on July 19, Transport Minister Datuk Seri Liow Tiong Lai said Friday… Liow reportedly said that the meeting would deliberate on the next course of action regarding the search for the aircraft, which went off radar on March 8, 2014, with 239 people on board while on its way from Kuala Lumpur to Beijing.” China, Malaysia and Australia have long said that the search will end after the current 120,000 sq km search area has been scanned, but some observers hold out hope that the rash of recent debris finds will encourage officials to press on.
@ALSM
I am not able to achieve a 252Hz BFO match in the vicinity of -37.5 89 for any achievable speed and heading at 00:11. I am merely pointing out that 00:11 could well have an associated non zero ROC. Does not matter whether your graphic was calculated at 00:11 or 00:19. I also do not happen to believe the aircraft was anywhere near -37.5 89 at 00:19, but did not mention that (nice guy that I am) since it is not relevant to the argument you were making.
Your entire purpose in the original three graphic post was to evangelize your end of flight dive bomber scenario. I’m just not buying into for reasons I have articulated. I certainly had no intention to conflate the ATSB’s “could have been” interpretation with your “highly likely” interpretation of what was implied by the 19:29 BFO value.
buyerninety:
The simulation conducted by Geoffney Thomas was poorly planned, and is not fully representative of what might have happened at the end of the MH370 flight. In particular, they simulated the end by switching off the fuel to the engines, rather than allowing the engines to run until fuel exhaustion. I’m not sure what differences in the flight that made, but for starters, I would expect there to be plenty of fuel for the APU.
In our third simulation, we configured the fuel to be equal in both tanks (700 lbs each), and we set the rudder trim to zero. As expected, but engines quit within 1 second. In this sim, the plane started turning after FE, and eventually it turned about 540 degrees in azimuth before impact (heading north). The turn rate became extreme and the the bank angle also became extreme, then there was an abrupt correction that leveled the wings. After that, the bank angle remained relatively level (<30 degrees) until impact. But there were 5 or 6 phugoid cycles. The last one hit perigee at 360 feet and pulled back up to about 1000 feet before hitting the water with the the airspeed indicator pegged at 500 kts INDICATED AIRSPEED. In this sim, we also experience a very brief left engine restart. The APU came on for about 4-5 minutes, but flamed out 5-6 minutes before impact.
Every sim was different. No one can claim to know exactly what happened with MH370, second by second, based on simulations. But all of our sims had periods of steep descent, extreme bank angles, and speeds approaching Mach 1 at various points in time. I provided all of the simulation videos to ATSB and they confirmed that Boeing observed very similar results in all there sims. I know ATSB described the simulated descents publically as
Gysbreght:
Regarding "How did you determine in your simulations that the airplane stalled? What phenomena did you observe that defined the stall?" We observed the stick shaker rattling at the top of most phugoids. Whether it stalls or not, the stick shaker was indicating an imminent stall. The indicated airspeed was very low, below stall speed.
Whit all respect but reading this all about the IMSAT data and all the conflicting interpretations, show to me those data have just too much room for interpretation to come to final conclusions about the flightpath and the end scenario.
It turns out after almost 2 1/2 years still no consensus can be reached on this.
Then I would conclude the limits of those data have been reached and people should take them as such. Those data have been turned inside out and upside down enough now IMO and won’t get any better.
I believe now only the debris will bring real new information and evidence and maybe follow-up drift studies and possibly (hopefully) new reports of officials or others.
I wonder, wouldn’t it be more usefull to concentrate on possible alternative flightpaths that can fit the IMSAT data in general in combination with the new information on debris and drift studies than keep on discussing unprovable details of those data?
DennisW:
Re “Your entire purpose in the original three graphic post was to evangelize your end of flight dive bomber scenario. ” This is total BS. You only reduce the credibility of your own positions by attacking others this way, exaggerating other’s positions and claiming you know motives you clearly do not know anything about.
For the record, several IG modellers also believe there was a slight descent underway at 00:11. I have seen estimates of 200-400 ft/min, similar to your estimate. Two years ago, ATSB was advising that the right engine flamed out only 5-6 minutes before the left, making a descent at 00:11 highly unlikely. But in Dec 2015, ATSB advised that the first engine may have flamed out sooner, possibly as soon as 00:02 UTC, 9 minutes before the 00:11 BFO value. If it did flame out that early, then it is quite possible that the plane slowed down enough over the next 9 minutes that it was starting to descend by 00:11.
In my graphics, I had no intention of suggesting it was or was not descending at 00:11. I was only pointing out that the BFO values at 00:19 were consistent with a steep descent and accelerating downward.
FWIW, here is what John E. Fiorentino (a former contributor here who was banned) posted on another blog:
“In our third simulation, we configured the fuel to be equal in both tanks (700 lbs each), and we set the rudder trim to zero.”
Why are you still withholding the data obtained during that test? What was done to the simulator to cause the ‘jerk’ you experienced in the cockpit as if someone inadvertently ‘switched on’ the full thrust of one engine on an empty tank?
“We observed the stick shaker rattling at the top of most phugoids. Whether it stalls or not, the stick shaker was indicating an imminent stall. The indicated airspeed was very low, below stall speed.”
The stickshaker operates when the AoA obtained from the AoA vanes exceed a certain threshold, below the stall AoA. That threshold varies with Mach number, therefore it reduces with increasing altitude. The stall speed is defined at 1 g, below 1 g the speed can be below the 1-g stall speed while the airplane is not stalled.
@ALSM
Sorry to assign motive to your posts. Your are correct. I have no way of knowing that.
With the inconclusive end of flight discussions, evasive radar flight manoeuvres, etc. At the northern tip of Indonesia there may have been military action to mh370 which accidentally damaged its Flaperon. Perhaps a warning shot which was not intended on striking mh370 might have hit the Flaperon and maybe the right engine which was totally accidentally. Maybe the coverup is related to this accident. Then mh370 did an emergency landing at Subang, or a military base, etc.
Dennis: I agree with what Dan Daniel O’Malley is said to have reported to Fiorentino. The problem is that you and others that don’t like the picture painted by the data keep misrepresenting how I have described the simulations. I have NEVER described them as a “… “near vertical descent” …”. Never. Yet you and others have repeatedly distorted what I have written with such statements to make it sound that way.
I will add that it was someone much higher up the food chain at ATSB that told me directly that the Boeing sim’s were similar to ours in all respects. They also observed phugoids, according to a person in a position to know more than press guy O’Malley.
End of thread. Let’s agree to work on solutions instead of wasting time on personal attacks and defenses thereof. The debris has the potential to tell us more. The IG is hard at work trying to squeeze a bit more info from the debris. Some of the best debris analysis I have seen is pointing further north on the 7th arc…perhaps as far north as S31-S34. But there are many unknowns for which we currently use drift model assumptions, like windage for each part, the affect of the cyclones, TOA vs. TOD, etc. I’d like to see more detailed work in this area.
@ALSM
Same thoughts here.
Does anyone know what the forward limit of the weight-and-balance envelope is for a B777-200?
If the aircraft ditched at 00:19, it had to have already descended from cruising altitudes and speeds, and would be relatively “low and slow”, preparing for the ditch, at 00:11.
Thus, the 00:11 data needs more intense thought and analysis.
We need to consider all possibilities.
Thus, we need to model the 00:11 BFO for all sensible “below 10,000 ft” combinations of speed, heading and rod along the 6th arc, assuming a pilot with an “intent” to ditch.
@airlandseaman. I return to APU run time only because it is important to the 7th arc log-on interpretation, as you are well aware, and you indicate you, collectively, are reviewing previous assumptions.
From what you say, had the APU fuel pump suction been uncovered during the dive that led to the 00:19:29 5000ft/min descent, that is before the SDU had been rebooted, the APU would have had access to fuel in its pipeline from the APU pump to complete this. The 100 ft line extends aft from the APU fuel pump (at the wing root), the line now pitched down. It would have had to suck fuel through this at a sufficient flow rate. It can indeed access fuel like this on the apron but that is with tank fuel providing head, without pitch and not at the altitude we are talking about (vapour lock?). Since the current ATSB assessment makes no mention of this fuel I assume that this access has not been proved through Boeing testing.
Which is not to say it wouldn’t be, only that it is an assumption in the meanwhile.
Prima facie, MH370 tried to remain inside of the KLIA FIR. The “IMT-back” at IGARI-BITOD occurred along the NE boundary of the FIR, and the implied “FMT” at (or near) IGOGU-BEDAX @ 18:35-40 locates along the W boundary of the FIR. “Trying to evade other nations’ radar” may be inverted as “trying to remain within its home KLIA FIR”.
Logically independently, an FMT southwards from (or near) IGOGU around 18:35-40 would locate the a/c near ANOKO c.18:43. ANOKO is a waypoint associated with landing at WITT (Banda Aceh), and a request for emergency landing would be expected at (or near) that place & time. Speculating a postulated “cabin disintegrated” scenario would affect fuel models, increasing PDA’s to larger values, which might favor crash sites farther north along the 7th arc. If so, the Dr. Ulich’s very high PDA’s for endurance HOLD modes would then be explicable, perhaps even necessary. Such modes might also imply descent to lower altitude & speed.
@airlandseaman: would you please be so kind as to publish a detailed path – including precise and granular (e.g. at 10 second intervals) speeds, headings, and altitudes – which respects all BTO values, under the assumption that MH370 flamed out at 00:02 UTC?
Thanks in advance.
Not for the first time, you are not extending the courtesy of replying to me. I suppose you NEVER said the above, or that the airplane ended inverted at Mach 1 and rates of descent exceeding 23,000 fpm in “all” of your tests, or this: “at the top of one of these cycles, if the plane stalls in a turn, the low wing will always stall first, tending to roll the plane inverted, in the direction of the turn.”
If in future you refrain from repeating misleading statements like that, we can indeed “agree to work on solutions instead of wasting time on personal attacks and defenses thereof.”
“I will add that it was someone much higher up the food chain at ATSB that told me directly that the Boeing sim’s were similar to ours in all respects. They also observed phugoids, according to a person in a position to know more than press guy O’Malley.”
How about providing the text of what the “higher up” actually said?
with kind regards,
Gysbreght van Aemstel
Aeronautical engineer and pilot, retired.
@all
Still struggeling with this assumption of a major power loss at 18:25 that triggered the SDU reboot there; a possible failure of the left engine just before that time.
Still haven’t found or got a clear answer.
Anyone able and willing to provide a decisive answer to this question?
@MH
For a landing at Subang, or a military base (??? Aceh Province or North Sumatra) you have to assume the BTO/BFO data set is fraudulent. Otherwise if the data set is honest 9M-MRO should be along the 7th arc.
I think Jeff Wise has already commented on this. If you don’t accept the BTO/BFO data the plane could be anywhere (within fuel range anyway).
@Ge Rijn
There is no decisive answer. However as I understand it (and I could be wrong) the French Prosecutors Office consider the loss 9M-MRO an act of terrorism (whatever that means). This is only a hypothesis – or a possible explanation. Its not a decisive answer.
@Steve Barrett
Thank you for answering but I’m not sure if I made myself clear enough.
It’s about the SDU reboot at 18:25:
My more specific question is; while the SDU is powered by the left engine generator, would a shut down of that engine cause an interuption in power supply to the SDU that after this interuption ends would start a SDU reboot sequence.
I understand that the SDU is not backed up by batteries and I assume a (automatic?) switch to power from the right engine will take some time (if this occures?).
I understood also the APU won’t kick in as long as the right engine still operates but on this I have no decisive answer or understanding also.
@Ge Rijn
The SDU draws electrical power from the Left Main AC Bus, which in turn is supplied with power, from either of the main engine IDG (“integrated drive generators”) or the APU.
Presuming that the APU was not operating at that time, then either:
both main engine IDG power sources had been intentionally switched off
or
left main engine IDG had been toggled off, AND the bus-tie-breaker connecting, or bridging, the Right Main AC Bus to the Left Main AC Bus had also been tripped
However, if NEITHER main engine IDG was activated, so that NO primary power was flowing into EITHER L/R Main AC buses, then many communications systems, e.g. SatCom and HF radios, would lose power due to “load shedding”. So, the total radio quiet lack of comm from the a/c all along the military radar track, possibly implies or favors the former scenario, i.e. symmetric loss of primary power to both side L/R Main AC buses, due to both main engine IDGs being toggled off & disconnected. If the APU was on, the it wasn’t connected to the LEFT Main AC Bus… so then it would have been connected to the RIGHT Main AC Bus… but then, why waste the fuel, why not just connect the right main engine IDG primary power source? And, if an acting flight crew was actually in fact actively disabling things, then why not disable the APU ? So I doubt that the APU was on during that time interval. And I currently favor the double IDG disconnected scenario.
If so, then somewhere between MEKAR and NILAM, the acting flight crew (pilots & engineers?) re-engaged the IDGs, re-powering the Main L/R AC buses, and triggering (among many many other things) the SDU reboot around 18:23-24 and eventual login (after a successful reboot completion) at 18:25. I.e. the acting flight crew re-engaged the IDGs between 1:00-2:40 prior to 18:25:27 = 18:24:30-18:22:50, causing the SDU to begin rebooting around that time frame, with reboot successfully completed, and login initiated, at 18:25:27, in round numbers.
That initial login BFO value of 141Hz strongly favors, if not requires, a more NW’ly heading of ~296deg consistent with travel along N571 along MEKAR-NILAM-IGOGU.
If you continue to trust the questionable ensuing values (~270Hz, ~170Hz), then the most obvious interpretation is that the plane was climbing, gaining altitude, at an initially super-fast rate of over +5000fpm, dwindling away by 18:28 (~140Hz again).
Some people apparently attribute the higher values to temperature-related effects, in the crystal oscillators, within the SDU electronics equipment. If so, then no climb. Either way, 18:25 = 18:28 = 141Hz strongly implies tracking along N571 from MEKAR to NILAM to IGOGU.
A +5000fpm climb exceeds anything the a/c did out of KLIA. Hypothetically, it could be achieved with a “zoom climb”, trading airspeed for altitude. And a “zoom climb” up to near 40K’ would slow the a/c down to the ~800kph / 450kts speeds that all the authorities most strongly favor currently. A “zoom climb” might imply problems with standard deceleration methods, e.g. throttling back engines or deploying flaps / flaperons ??
Whatever, the Inmarsat JoN route eventually imputes a “two-turn double FMT” to the a/c, first a turn (~100-deg to port) onto a SW heading of 200-deg around 18:40 from somewhere near IGOGU…
And then another turn (~20-deg to port) onto a near-due-S heading of 180-deg around 19:00, somewhere near Great Nicobar island.
The former is completely consistent with a turn, from either IGOGU or ANOKO, towards BEDAX, south of that island. The latter does not align towards any obvious waypoint nearby, although ISBIX near the equator, and a waypoint for approach to Jakarta, may be the best standard waypoint match. Alternatively, a non-standard waypoint may have been entered, e.g. “copy-paste the BEDAX longitude, and enter latitude 0” (??).
I think that particular point confuses many, maybe even me, IDK, but I understand that the Inmarsat JoN article implies & describes a “double turn FMT”, from ~300-deg heading to ~200-deg heading thence to ~180-deg heading. That “double turn” neatly aligns with MEKAR-NILAM-IGOGU-turn-BEDAX-turn-180 towards the equator passing just east of ISBIX.
trying 2B helpful 🙂
I understand that you (Ge Rijn) may favor a DESCENT after the military radar track. However, the military radar coverage graphic (fig.2), from VictorI’s radar study published on this site (Q’s re: radar), shows that the MH370 radar track coincides exactly, with the “M1” Malaysian Penang Western Hill military radar site. I.e. the IMT @ IGARI-BITOD aligns exactly with the western-most radar-range limit of that site, and the “abrupt” disappearance at 18:22 aligns with the eastern-most radar-range limit of that site. (Especially considering that the a/c was at a slightly lower alitude at 18:22, vs. 17:22 up at 37000′.)
So, the implication is, that the a/c “flew out” of the radar range at high speed, rather than “dropping off” of the radar due to descent. Many users have commented about that verbiage in FI, why that document states MH370 vanished “abruptly” from radar @ 18:22, as opposed to “coasting” for a while before disappearing intermittently along the radar track. But that difference corresponds to “jetting out of radar range” (=> “abrupt” disappearance) vs. flying behind local mountainous terrain partially blocking LOS well within radar range (=> “coasting… off screen… back on screen”). So, actually, the military radar data implies that the aircraft was still travelling at very high speed at 18:22, when it “zoomed” (for wont of worthier words) out of range of the Western Hill, Penang radar site.
So, from 18:22 to 18:25 to 18:28, from MEKAR to NILAM towards IGOGU, the plane was flying high and fast, and, if anything, climbing higher. Conversely, any descent is hard to reconcile with the satellite data.
@Erik Nelson
Thanks for answering so thoroughly.
Still it’s not completely clear to me.
I understand the SDU is powered via the Left Main AC bus and if this one fails the Right main AC bus will take over powering the SDU.
But I also read elsewhere this takes 60 seconds and and will start a reboot of all equipment powered by the Left Main AC bus (thus also the SDU). See second comment of Sy Gunson:
https://www.quora.com/What-would-happen-to-a-Boeing-777-in-the-very-unlikely-event-of-a-total-electrical-failure-while-in-cruise
Is this correct?
Ge Rijn
It can take 60 seconds for the APU to light up, but it takes only milliseconds for the right bus (powered by the right engine) to take over for the AES if the left bus goes dark. In this case, no reboot would happen.
@jeff
kindly please, some idea why you did some edits and blocking(?) here??
@all
Interesting article linked below which bears directly on the search strategy, Frequentists vs Bayes.
http://www.economist.com/blogs/graphicdetail/2016/06/polls-versus-prediction-markets
The take-away above is how wrong you can be by over-weighting “priors” in a predictive model. The prior in this case is the current search area, derived by a consensus view of what the BFO and BTO Inmarsat data (along with Occam’s razor flight dynamics) is telling us.
As debris accumulates – some ten pieces now (not all confirmed, but all very compelling), the drift models are suggesting a very different terminus. In my view, the consensus modelers have been misled by Occam and the BFO values. I believe it is time to discard BFO other than using it to know that the aircraft turned South sometime after crossing the Malay peninsula.
The BTO data is “golden”, however. There is little doubt the aircraft terminated near the 7th arc. How near depends on the end of flight behavior. I do not wish to stir that pot again, however. My suggestion is to use BTO, the current debris finds, and future debris finds to guide the search area decisions. My personal belief (“eyeballing” the ensemble of drift data) is that the aircraft terminated near or above 20S.
“Eyeballing” ???? How can that possibly be acceptable among the analytical elite populating this site? I offer the traveling salesman problem as an example. If one randomly picks a dozen cities, and you pose the question of what is the shortest route to visit the selected cities, it turns out that a human simply eyeballing a map, and drawing a path does an incredibly good job of minimizing or nearly minimizing the path length. By contrast, the most efficient algorithm for picking a path is still the subject of intense interest, but suffice to say it gets messy in big hurry. (see Wikipedia).
@Dennis
“I believe it is time to discard BFO other than using it to know that the aircraft turned South sometime after crossing the Malay peninsula”.
An interesting statement, provoking a follow up question. Others stated not long ago, that instead of a turn south the BFO at that time also could indicate a rapid climb. When the BFO after the assumed FMT bear no relevance, why should the BFO at the assumed FMT hold some substance?
Instead without the BFO the debris found could be indicative of a turn to south at some time, but not necessarily at the time assumed.
@RF4
The BFO data is unambiguous relative to a Southerly flight direction around the time of the so-called FMT. It becomes ambiguous if one tries to use it to predict heading accurately enough to be useful. It is like riding in a car with your eyes closed, and sensing the car turned left without knowing exactly how far left.
@Ge Rijn
You’re on the right track regarding the electrical bus, flying single engine, and a curving flight path to the southeast.
Consider the following — a known oxygen leak leads to a flash fire in the electronics bay, damaging many of the interface cables. The pilots make their best effort to turn back and descend but the left engine control cable is inoperative. With the left engine defaulting to maintaining power, the only way to descend is to shutdown the engine with about the only means remaining — the fire suppression system.
Now, with the right engine throttled back, a descent is begun but not without a new set of problems. The fire suppression has disabled the left electrical bus, taking down the communications system. What’s worse — far worse — is the left engine is no longer providing air to the flight deck. With the oxygen system already depleted, the pilots’ fate is sealed. (While many theorize the pilots caused oxygen deprivation to the passenger cabin, the opposite happened, with the passengers surviving to the very end. Now THAT’s something that a government might was to suppress.)
Once the plane reaches its target altitude (10k’?), it holds altitude the best it can. The right engine running at partial throttle isn’t enough to maintain level flight so eventually the enters an upset condition. Boeing’s Flight Envelope Protection system begins a recovery that includes increasing power to the right engine and an attempted restart of the left engine. The attempted left engine restart is what brings the left electrical bus back to like, powering on the SDU among other things.
The asymmetrical thrust takes MH370 for a short circuit around the Andaman Sea and then on south/southeast to the Zenith Plateau. More details at http://www.mh370site.com.
@airlandseaman
I’ll literly repeat the following from the site posted earlier and reffering to the ATSB report included:
https://www.atsb.gov.au/media/5733804/Bayesian_Methods_MH370_Search_3Dec2015.pdf
‘Normaly if a genarator failed the other main transfer bus would kick in automaticcaly after 60 seconds, then re-boot all the devices depending on that power relay’.
This is not right you say?
Ge Rijn:
Nothing on this subject found in the BM Paper referenced in Sy Gunson’s answer. I would note Sy has not been a very reliable resource. ATSB stated in their first report, updated August 18, 2014:
“Aircraft electrical system
The electrical system on the B777 supplies 115 V AC and 28 V DC power. The main power sources are a left integrated drive generator (IDG) and a right IDG, powered by the left and right engines respectively. An auxiliary power unit (APU) can supply power if either or both of the IDGs are unavailable. The SDU was powered by 115 V AC from the left AC bus which was normally supplied by the left IDG. If power from the left IDG was lost, then a bus tie breaker would close and power would be automatically transferred from the right AC bus. Similarly, if power was lost from the right AC bus, power would be automatically transferred from the left AC bus. This power switching is brief and the SDU was designed to ‘hold-up’ during such power interruptions. To experience a power interruption sufficiently long to generate a log on request, it was considered that a loss of both AC buses or, a disabling of the automatic switching, would be required.”
Thus, I believe my answer was correct. There would be no reboot of the AES unless both the left and right AC buses were dead, or there was manual intervention to prevent the normal automatic transfer.
@DennisW – With the discussion of how to minimize BFO errors I, too, just thought of using them to say the a/c was generally headed south except for the BFO at 00:19 that show a descent. However, I believe the fuel endurance to around 00:17 is also valid (It’s not easy to get ~5.95 hours of flight from 35.6 mt fuel.) If so, that eliminates certain flight paths and speeds. If you add constant altitude and speed profile, many more routes are eliminated.
There are other possible reasons. For instance, damage to the right AC bus
(intermittant or long term shorts, or breaks to its distribution lines) would
cause the Electrical Load Management System (ELMS) to switch to alternately
powering items from the left AC bus. In such a scenario, ‘load shedding’ by
the ELMS as it initially transfers items onto the left AC bus could see the
SDU shed if the ELMS determined there was not sufficient power for items now
fed by the left AC bus. You would be generally aware that electronic items
tend to draw a higher amount of power when starting to power up. When power
draw had ‘settled’, the ELMS could selectively add & repower items according
to their order of priority. The actual priority order in which items are
shed/added does not seem to be made publically available by Boeing.
As a sidenote, this behaviour would act to explain why the ICE never came back
up – simply, it would be reasonable to assume that the ICE, like items such
as the galley and lavatory water heater, would have a very low priority.
@Ge Rijn
Quoting the original “Definition of Underwater Search Areas”
===========================================
SDU power-up Following the loss of AC power on both buses, the SDU would have experienced a power interruption sufficiently long to force a shut-down, the aircraft’s ram air turbine (RAT) would deploy from the fuselage into the aircraft’s slipstream and the APU would auto-start. The APU would take approximately one minute to start-up and come ‘on-line’ after which time it could have provided electrical power to the SDU. After power became available, the SDU would take approximately 2 minutes and 40 seconds to reach the log on stage evidenced in the SATCOM log at 00:19:29. If engaged, the autopilot could have remained engaged following the first engine flame-out but would have disengaged after the second engine flamed-out. By the time of the SATCOM log on message, the autopilot would have been disengaged for approximately 3 minutes and 40 seconds. If there were no control inputs then it would be expected that eventually a spiral descent would develop. In the event of control inputs, it is possible that, depending on altitude, the aircraft could glide for 100+ NM.
===========================================
So, the logon request at 18:25:27 occurred approximately 00:02:40 after power was restored to the SDU, most likely when the acting pilot(s) re-engaged primary power from the main engine IDGs, around 18:22:50.
However, I think that that figure value of 2min 40sec has been revised, downwards, to 1min. If so, then the acting pilot(s) toggled the IDG engage switches only approx. a minute before the logon request, around 18:24:30.
I understand the consensus is, that the a/c reached NILAM around 18:25:45, immediately after or perhaps even during the 18:25 logon sequence.
The 00:19 SDU reboot involves one extra step, namely waiting for the APU to spool up and start supplying power to the L MAIN AC bus, which feeds the SDU. Then the SDU gets power, and begins its bootup sequence which requires between 00:01:00 to 00:02:40 to complete. Whereas, at 18:25, the main engines were already spooled up and providing mechanical thrust, such that they started supplying electricity as soon as the acting pilots flipped the IDG engage switch.
——————————————
@all
Please consider the path
MEKAR-NILAM-IGOGU-ANOKO
That path imputes a FMT to the a/c, at IGOGU, onto a 180 heading, prior to the 18:40 satellite phone call. MAG HDG HOLD after Route Discontinuity @ ANOKO would be subject to winds blowing towards the west which could have pushed the a/c farther west, towards an alignment with the 19:40 ping-arc. Is there any way to make a “single turn FMT” at IGOGU towards ANOKO onto a 180 heading fit the rest of the satellite data ?
Erik Nelson: The “2 minutes and 40 seconds” estimate was later revised to ~ 60 seconds, so FE occurred around 00:17:29, not 16:09 as originally estimated.
@Ge Rijn In regards to your SDU reboot log on request question. Neither ATSB or Immarsat could explain the log on request just prior to MH370 making its final turn south into oblivion. Some experts,pilots with very detailed knowledge of how the 777 ACARS, transponder and other instruments on how the sequence for pulling circuit breakers to disable ACARS etc to avoid radar detection. Have suggested this was a deliberate act from someone with very detailed of how to do this from within the cockpit.Or access to the avionics bay. Also the hatch to avionics bay is in forward gallay and can be accessed by a passenger or crew. But if not the pilots, who has knowledge to pull circuit breakers etc etc
Correction to last ALSM post: 16:09 should read 15:49
@ALSM
So, the acting pilot(s) re-engaged primary power, from main engine generators, to the L(/R) main AC buses, at approx. 18:24:30, initiating an approx. 00:01:00 SDU bootup process, inducing logon at 18:25:27 ? Thanks much for that definitive clarification.
If the aircraft reached waypoint NILAM at almost exactly precisely that time, then is anyone implying, that the acting pilot(s) intentionally timed the main engine IDG generator re-engagement (@18:24:30), so as to re-power the SDU and/or associated systems (@18:25:30), in time for some act and/or maneuver at NILAM ? Is the timing not particularly conspicuous ?
@airlandseaman and others
Thanks for making this clear.
Conclusion to me now is the SDU re-boot could not have been triggered by only the loss of left engine power or only disabling power feed to the Left main AC bus. The right engine IDG will kick in almost immediately with the SDU staying on line without interruption.
Only a temporarily power loss of both engines or temporarily interruption of feed to both Main AC busses could have caused the SDU reboot of 18:25 is how I understand it now.
In the case of a temporarily loss of both engines at around 18:24 the APU would kick in and the log on sequence would start 60sec. later. With at least one engine coming on line again somewhere between 18:26 and 18:39.
In the case of an interuption of feed to both main AC busses not caused by engines (temporarily) shut down I assume the APU would not start up automatically and the log on sequence would start after someone manualy put at least the left main AC bus on line again at 18:25.
I also understand the plane must have flown on two engines before 18:22 according to the speeds and altitudes estimated before that time. So I regard it as very unlikely now an event which caused both engines to shut down temporarily after 18:22 triggering the SDU reboot could have occured.
To me than this leaves only the possibility of human intervention causing the SDU reboot at 18:25 as most probable bringing both main AC busses on line again or at least one of the main AC busses.
While I assume the IFE must have been still on line at 23:13 with the second SATCOM call it’s hard to imagine it was not anymore at 00:21. But there is still the possibility the IFE was switched off manualy after 23:13 which is easy to do in the cockpit.
If this happened the not recieved IFE log on at 00:21 could mean it was not due to impact and the plane was still flying at that time IMO.
I think it’s very hard to completely rule out this possibility at the moment.
Survived another 4th of July.
http://tmex1.blogspot.com
Re Egyptait 804:
“One of the pilots on board EgyptAir flight 804 tried to put out a fire before the crash, according to data from one of the plane’s black boxes”.
http://www.independent.co.uk/news/world/africa/egyptair-crash-plane-black-box-fire-on-board-wreckage-paris-flight-a7120386.html
Hard to argue that logic Dennis. The lat cluster graph is a nice way of collating debris finds to date.
Butthurt report – PMSL…..
Think a few that are now MIA need a copy of that.
@DennisW
IMO a key problems with this GEOMAR drift study remain unexplained:
-still no debris is found on the densily populated islands of Sumatra and Java and other islands in the region.
-the great majority of most possible starting areas are far outside the 7th arc. Only a small portion of it in front of the coast of Java is around the 7th arc.
So I think it requires to dismiss the Inmarsat-data for the most part of it.
-a big part of most probable start areas are in a zone where the currents move from west to east. So if debris started there it had to drift first to the coasts of Sumatra/Java etc. (where nothing is found) and then turn around to the west. Probability of debris landing in Mozambique and South Africa are not present in this GEOMAR study and IMO highly unlikely.
How would you explain these problems?
The Geomar study provided above is out of date. There is a new study available now. It also takes into account (in general terms) at least two more recent finds in east Africa. As a result, MH370 could have crashed in the area first predicted in the preliminary report. So, it seems somewhere west of Exmouth would be consistent with both drift modelling and satellite data. It also makes sense in theory (two islands with runways and the east coast of Australia as possible targets in the vicinity).
http://www.geomar.de/fileadmin/content/service/presse/Pressemitteilungen/2016/MH370_Report_May2016.pdf
@Nederland
This study is in essence the same as their previous one. It’s still only based on the flaperon.
IMO it’s out of date anyway for the latest Godfrey study includes 5 pieces of confirmed debris and 4 unconfirmed but likely pieces of debris.
The only update of this GEOMAR study consists in the suggestion to include more debris as they are found and more is known about the arrival time of the two other pieces mentioned for that would probably lead to a more precize area.
Both things happened in the meantime.
The second picture of the Mosselbay piece (the engine cowling) with the barnacles highly suggests it arrived in- or shortly before december 2015.
Several other pieces were found since then.
Up to 9 were incorporated in the latest Godfrey study leading to a more precize area of origin of this debris.
This makes this Godfrey study the most accurate study till now IMO and this GEOMAR study out of date and virtualy obsolete IMO.
I also think by using reverse drift modeling like this GEOMAR study and the latest model of @Brock McEwen you add chaos in the wrong direction calculating from west to east as those models clearly show in their enormous potential areas of origin.
This is just not what happened. Debris did not start drifting from their locations of arrival. The debris started drifting somewhere near the 7th arc and with drifting from the west (cq north then west) to the east chaos gets added what shows in the enormous area debris is found.
So IMO forward drift studies which take a beginning somewhere near the 7th arc are more (perhaps the only) usefull in finding out which areas on the 7th arc can explain all the found debris best.
@Ge Rijn
“So I regard it as very unlikely now an event which caused both engines to shut down temporarily after 18:22 triggering the SDU reboot could have occured.”
Unlikely but possible…
BA38 experienced such an event when water in the fuel formed ice that clogged the oil-fuel heat exchangers. General Electric and Pratt & Whitney engines were not affected by the problem. Rolls-Royce developed a modification. We assume of course MAS installed it….
@Ge Rijn
It is obvious that the Roy find in Mossel Bay is the single most important one to rule out the current search area and to suggest a more northerly crash site.
The Godfrey paper on the other hand is based only on a website (adrift.org.au). I’m not saying it’s therefore wrong but surely this is not cutting edge research. The methodical flaw I see is its assumption “that the time of arrival was the fastest possible given by the Adrift model.” Obviously, this favours a relatively southerly starting point.
You can try adrift.org.au yourself to verify that in standard setting a starting point roughly intersecting with the area(s) proposed in the preliminary report best fits the evidence hitherto available, but I’m not saying this is therefore right.
At any rate, the flaperon find is still the most valuable in the sense that it was the earliest in time. According to Geomar and collaborators it is consistent, too, with an impact roughly west of Exmouth.
Ugh. Do yourself a favour, the CSIRO webpage here;
http://www.marine.csiro.au/~griffin/MH370/index.html
has drift studies back to ’97. Not hypothetical particles or
rubber ducks but physical drifters, drogued and undrogued,
from the Global Drifter Program. From that data, indeed,
crash site west of Exmouth as posted above, entirely possible,
from there southwards down the Arc to above the current searched
area, more probable.
Crashing off ‘the coast of Sumatra’? – a far lesser probability.
And as to the liklihood of any debris making landfall on the
west coast of Western Australia – virtually NOTHING! We’re
talking what, 5, looks like no more than 10 since ’97 out of
hundreds of drifters, and certain of those drifters came from
below the current searched (seabed) area. Amazingly, some
drifters got to within a hundred, and one to almost 50km,
from the west coast of Western Australia, only to then be
swept back out into the central Indian Ocean. So lack of
debris at Western Australia=Indicative Of Nothing.
Problem with the Griffin/CSIRO website is it has not been updated since the NO STEP find in Mozambique which it claims “can not provide particularly strong support for [the current search site].” This is despite the fact that more debris was found just after and that this debris is most inconsitent with the search site.
P.S. via the ATSB feedback page, I’ve suggested they update
their 1/12/2015 MH370 searched area graphic (that would be
DD/MM/YYYY).
Sorry, I disagree with your ‘most inconsistent’ – as the webpage
says, “the trajectories of drifting items are so chaotic”.
Therefore we are disagreeing on shadings of probabilities (or
perhaps we agree but your choice of words is poor) – I would say
that the area to date currently searched is reasonably consistent
with debris findings, but is less probable as a origin site than
areas further up along the Arc (note; for purposes of your/my
discussion, we are disregarding the ‘steady state engine thrust
until fuel exhaustion’ assumption that led to the reasonable
choice of current search area on the Arc, and considering only
the drift data.)