Zaharie Shah’s Secret Psych Evaluation

zahrie-before-flight

Yesterday Twitter user @nihonmama released the first two folders from the secret Malaysian police report into MH370. Some parts relating to Zaharie’s flight simulator had been released earlier, but the bulk of this material is coming into public view for the first time. Here is “Folder 1: Pilot” and here is “Folder 2: Co-pilot.”

I was particularly interested in the section containing the psychological evaluation of the pilot, Zaharie Ahmad Shah, found on page 111. As it is in Malaysian, I had to type it into Google Translate to make any sense of it. As I have absolutely no understanding of Malaysian I am copying it and pasting it below without any changes. Corrections welcome!

 

Hon. Datuk Mazlan bin Mansor
Deputy Director (Intelligence / Operations),
CID,
Royal Malaysian Police,
Bukit Aman,
50560, Kuala Lumpur

Hon. Dato ‘

Expertise help the Ministry of Health in Malysia Investigation Missing MH370: The study “Psychosocial and Behavioural Pattern” crew MH370.

Letter from Hon. Dato ‘no. ref: JSJ KPN (PR) 35/3 dated July 3, 2014 and the terms of reference of the assessment panel “behavioral pattern and psychosocial crew of MH370 is referenced.

2. The sub-committee meeting between Kiraja Malaysia Police (PDRM) and KementerianKesihatan (MOH) was held in Room Mesyusarat, Hospital Bahagia Ulu Kinta, Perak on 7 July 2014. The purpose of this meeting was to obtain an independent report (independent) The above assessment.

3. Here is the panel sub-committee has been established.

[The letter lists three officials from the Ministry of Health and six officials from the Royal Malaysian Police]

4. Assessment conducted on flight MH370 pilot Captain Zaharia Ahmad Shah and co-pilot, en. Fariq Ab. Hamid, have been guided by reference listed:

i. Quoting witnesses related conversations pilot, Captain Zaharia Ahmad Shah, total of 40 person which includes 5 members, 20 co-workers, friends WeChat 9 and 6 public witnesses.
ii. Quoting witnesses related conversations pilot, Mr. Fariq Ab. Hamid total of 9 people including 3 members of the family, his girlfriend, and five colleagues.
iii. Quotes clips CCTV video at KLIA’s movement, patterns of behavior and expression on the face (facial expression) Zaharie co-pilot En. Fariq before their flight dated 07.03.2014.
iv. Quotes CCTV video clips KLIA Zaharie on 26.02.2014 before his flight to Denpasar, Indonesia and on 03.03.2014 before his flight to Melbourne, Australia.
v. Medical reports Zaharie.

5. Based on these reference sources, we have studied the background Zaharie including education, personality and coping (coping style), relationship with spouse, children, family members, friends and colleagues, including his interests and hobbies. Attention has also been given to her relationship with her maid. His physical health problems are investigated including asthma and diseases of the spine, which caused him to have to take treatment drugs painkillers “analgesics.” Religious and political tendencies he observed.

6. We also reviewed the background of the co-pilot Mr. Fariq including education, personality, relationships with family members, friends and colleagues.

7. Highlights are as follows:

7.1 In the field of career, Zaharie is an experienced pilot and a competent and respected by peers.

7.2 Available Zaharie not share the same interests with his family members. However, the difference in interest is acceptable. His family was also not reported any change of pattern of behavior (behavioral pattern) before his flight was on 07/03/2014.

7.3 Information from friends and colleagues Zaharie show that he was a friendly, warm and jokes. They are also not reported any change of pattern of behavior before his flight was on 03/07/2014.

7.4 Problems spinal pain he was a fairly chronic physical problems rather than a new stressor.

7.5 Review of comparisons based recording video clips CCTV KLIA on 26.02.2014, 03.03.2014 and 03.07.2014, found him tending to smoke before her flight and movements of his time smoking was similar in all three videos. At KLIA CCTV video clip on 03/07/2014, Zaharie not show any sign of anxiety or depression.

Finally, we have not found, any changes in terms of psychological, social and behavioral patterns Zaharie Ahman Shah before his flight was on 03/07/2014. We also did not find any demolition of psychological, social and behavioral patterns of co-pilot En. Fariq Ab Hamid before his flight was on 03/07/2014.

Thank you.

“CARING, TEAMWORK PROFESSIONALISM AND WE ARE WORKING CULTURE”

I who am following orders,

Dr. HJH. RABA’IAH BINTI MOHD. sALLEH
MMC NO: 25878
Director & Consultant Psychiatry (Forensic)
Special Grade “C”
Hospital Bahagia Ulu Kinta
Perak Darul Ridzuan

I find this to be a truly remarkable document. We’ve been hearing rumors that the investigation found no evidence that suggested Zaharie could have a psychological predilection for suicide/mass murder, but here it is at last in black and white, with details such as the fact that his pattern of smoking before a flight was unchanged before MH370. It is hard to imagine that anyone contemplating his own imminent death could exhibit such sang froid.

Indeed, I don’t think there has ever been a case where someone who is known to have carried out such an act had such an outward appearance of being balanced and well-adjusted. Andreas Lubitz, for example, had experienced years of psychological upheaval trouble, at one point temporarily washing out from Lufthansa’s flight training program, before destroying Germanwings 9525.

In my estimation this psych evaluation must be regarded as powerful evidence that Zaharie did not hijack MH370.

After the jump, the letter in the original Malay, as re-typed by me from the report.

 

YBhg. Datuk Mazlan bin Mansor

Timbalan Pengarah (Risikan/Operasi),

Jabatan Siasatan Jenayah,

Polis Diraja Malaysia,

Bukit Aman,

50560, Kuala Lumpur

YBhg. Dato’,

Bantuan Kepakaran Kementerian Kesihatan Malysia dalam Siasatan Kehilangan MH370: Kajian “Behavioural Pattern dan Psikososial” krew MH370.

Surat daripada YBhg. Dato’ no. ruj: JSJ KPN (PR) 35/3 bertarikh 3 Julai 2014 dan terma rujukan utama panel pengkajian “behavioural pattern dan psikososial krew pesawat MH370 adalah dirujuk.

2. Mesyuarat sub-committee antara Polis Kiraja Malaysia (PDRM) dan KementerianKesihatan Malaysia (KKM) telah diadakan di Bilik Mesyusarat, Hospital Bahagia Ulu Kinta, Perak pada 7 Julai 2014. Tujuan mesyuarat ini diadakan adalah untuk mendapatkan satu laporan yang berkecuali (independent) di atas penilaian tersebut.

3. Berikut adalah panel sub-committee yang telah ditubuhkan.

4. Penilaian yang dijalankan terhadap juruterbang pesawat MH370 Kapten Zaharie Ahmad Shah dan pembantu juruterbang, en. Fariq Ab. Hamid, telah berpandukan sumber rujukan yang tersenarai:

i. Petikan percakapan saksi berkaitan juruterbang, Kapten Zaharie Ahmad Shah, sejumiah 40 orag yang merangkumi 5 orang ahli keluarga, 20 orang rakan sekerja, 9 orang rakan WeChat dan 6 orang saksi awam.

ii. Petikan percakapan saksi berkaitan pembantu juruterbang, En. Fariq Ab. Hamid sejumlah 9 orang yang merangkumi 3 orang ahli keluarga, teman wanita beliau, dan 5 orang rakan sekerja.

iii. Petikan klip-klip video CCTV di KLIA mengenai pergerakan, corak tingkah laku dan mimik muka (facial expression) Kapten Zaharie bersama pembantu juruterbang En. Fariq sebelum penerbangan mereka yang bertarikh 7.3.2014.

iv.  Petikan klip-klip video CCTV KLIA Kapten Zaharie pada 26.2.2014 sebelum penerbangan beliau ke Denpasar, Indonesia dan pada 3.3.2014 sebelum pnerbangan beliau ke Melbourne, Australia.

v. Laporan perubatan Kapten Zaharie.

5. Berpandukan sumber rujukan tersebut, kami telah mengkaji latar belakang Kapten Zaharie termasuk pendidikan, personaliti dan daya tindak (coping style), perhubungan dengan isteri, anak-anak, ahli keluarga, kawan-kawan dan rakan sejawat termasuk minat dan hobi beliau. Perhatian juga telah diberi kepada perhubungan beliau dengan pembantu rumahnya. Masalah kesihatan fizikal beliau juga diteliti termasuk penyakit asma dan penyakit tulang belakang yang menyebabkan beliau perlu mengambil rawatan ubat-ubatan penahan sakit “analgesics.” Kecenderungan keagamaan dan politik beliau juga diamati.

6. Kami juga telah mengkaji latar belakang pembantu juruterbang En Fariq termasuk pendidikan, personaliti, perhubungan dengan ahli keluarga, kawan-kawan dan rakan sejawat.

7. Rumusan kami adalah seperti berikut:

7.1 Di bidang kerjaya, Kapten Zaharie adalah seorang juruterbang yang berpengalaman dan kompeten serta dihormati oleh rakan sejawat.

7.2 Didapati Kapten Zaharie tidak berkongsi minat yang sama dengan ahli keluarga beliau. Walau bagaimanpun, perbezaan minat ini adalah sesuatu yang boleh diterima. Keluarga beliau juga tidak melapurkan apa-apa perubahan dari corak tingkah laku (behavioural pattern) sebelum penerbangan beliau pada 7.3.2014.

7.3 Maklumat dari kawan-kawan dan rakan sejawat Kapten Zaharie menunjukkan bahawa beliau merupakan seorang yang peramah, mesra dan boleh berlawak jenaka. Mereka juga tidak melapurkan apa-apa perubahan dari corak tingkah laku sebelum penerbangan beliau pada 7.3.2014.

7.4 Masalah sakit tulang belakang beliau merupakan satu masalah fizikal yang agak kronik dan bukannya merupakan suatu stressor baru.

7.5 Kajian berpandukan perbandingan rakaman klip-klip video CCTV KLIA pada tarikh 26.2.2014, 3.3.2014 dan 7.3.2014, mendapati beliau cenderung merokok sebelum pnerbangan beliau dan gerak-geri beliau semasa merokok adalah sama di ketiga-tiga video tersebut. Pada klip video CCTV KLIA pada 7.3.2014, Kapten Zaharie tidak menunjukkan apa-apa tanda kegelisahan ataupun kemurungan.

Akhir kata, kami tidak mendapati, apa-apa perubahan dari segi psikologi, sosial dan corak tingkah laku Kapten Zaharie Ahman Shah sebelum penerbangan beliau pada 7.3.2014. Kami juga tidak mendapati apa-apa perubuhan dari segi psikologi, social dan corak tingkah laku pembantu juruterbang En. Fariq Ab Hamid sebelum penerbangan beliau pada 7.3.2014.

Akhir kata, kami tidak mendapati, apa-apa perubahan dari segi psikologi, sosial dan corak tingkah laku Kapten Zaharie Ahman Shah sebelum penerbangan beliau pada 7.3.2014. Kami juga tidak mendapati apa-apa perubuhan dari segi psikologi, social dan corak tingkah laku pembantu juruterbang En. Fariq Ab Hamid sebelum penerbangan beliau pada 7.3.2014.

Sekian, terima kasih.

“PENYAYANG, PROFESSIONALISM DAN KERJA BERPASUKAN ADALAH BUDAYA KERJA KITA”

Saya yang menurut perintah,

Dr. HJH. RABA’IAH BINTI MOHD. SALLEH

MMC NO : 25878

Pengarah & Pakar Perunding Psikiatri (Forensik)

Gred Khas “C”

Hospital Bahagia Ulu Kinta

Perak Darul Ridzuan

571 thoughts on “Zaharie Shah’s Secret Psych Evaluation”

  1. @Ge Rijn

    The way I understand is that it doesn’t matter much if the flight is bound to NZPG via YPCC/POLUM, but in that case it definitely is a suicide route, and I don’t feel that this was the case. It is more likely I feel that YPCC (for example) could be the final waypoint and afterwards the plane continued on a heading, in which case there would be a difference in true/magnetic heading.

  2. @Ge Rijn

    A lot of people were fooled, including the DSTG and ATSB, including me. Yes, hands up on that one! But as you say, never mind. It turns out, he was more cunning than we ever imagined at the start. We had a bit of bad luck, to say the least. Who would have thought that projecting the initial flight path (the bit between the FMT and the 2nd arc) the only bit reasonably constrained in the east-west direction, would cut the 7th arc at the same place as Boeing’s estimate for the MRC line for an ISO economical cruise? As Dennis pointed out ad nauseam, the BFO is dangerously under-constrained.

    Now, after more than two years of searching, it is becoming increasingly clear that our perpetrator was too smart to allow himself a straight line path into the SIO. He’s laid a blind trail for us. Even the SIM data can now be seen as part of the plan, to lead people to think he was experimenting with great circle tracks to nowhere.

  3. TBill: “I see SK999 argues on Reddit that the smoothness of the BFO data …”

    No, it is the smoothness of the BTO (not BFO) data that was being discussed.

  4. Z made a substantial amount of phone calls to the Defence university, which may indicate he had good knowledge of radar capabilities, and this knowledge may have informed his possible decision to reboot the satellite system at a particular distance from Penang. I wonder if it is possible to gain better understanding of the Sabang radar range. This could in turn help to understand better the distance MH370 was trying to keep when circling around the tip of Sumatra.

  5. @Nederland

    As I understand it now the difference between a True or a Mag heading after FMT would be too small over that distance till ~30S/~40S to tell it was a ghost-flight or a pilot-controlled flight.

    Magnetic variations and wind variations over that distance are too small to make any substantial difference to allow a conclusion one way or the other then, I understand.

    Only the end-of-flight matters then. I also asume the flight was actively piloted till the end.
    But hen what did the pilot do reaching his (chosen?) fuel-exhaustion destination?

  6. @Nederland
    If you go back to Jeff’s Nov_2014 article on why Indonesia missed MH370, I think it is Richard Cole posts the radii of the radar. It (Sabang) encompasses Car Nicobar but it’s on the outside range, so I assume low altitude helps. Most of the proposed flight paths go through Sabang radar circle, some quite close of course. Indonesia suggests they were watching but not recording and did not anything much. Seems like ATSB assumes they were not really watching, if we have a FMT before 18:40, you’d think that would be seen.

    @SK999
    Thank you.

    @Ge Rijn
    If you are going due South 180, from Car Nicobar by the time you get to 30S, a mag heading would shift you approx 3 degrees from 93 to 96 deg. So that’s quite noticeable.

    Where I was seeing less difference (I better check again) is the curved path to NZPG. The curved path I was seeing less difference. And that’s without wind.

    Magnetic headings are calculated from tables which change each year, so the question becomes what mag correction tables everyone is using including MH370 (was it current?).

  7. @Ge Rijn
    The main controversial thing I see, if the “former” search area sort of assumes a straight True path to about 38S, as a ghost flight, but if True heading is not seen after discontinuity (which we still do not understand fully), then those paths are rare or it was intentionally programmed by a non-ghost. Perhaps ATSB understands better than we do when True heading is the default.

  8. @Nederland

    “and this knowledge may have informed his possible decision to reboot the satellite system at a particular distance from Penang.”

    we don’t know if it was him or someone else who entered E/E bay in despair to try to unlock cockpit door and accidentally enabled satcom

    the same person could accidentally affect navigation systems avionics etc. so that might explain why they haven’t reached the destination…

    there is a lot of random scenarios that could happen if you embrace failed hijack scenario which is what most likely happened (statistically 50% of airplane hijackings are (not) successful)

  9. @TBill

    The Nov 2014 blog post gives information on the Lhokseumawe radar in Aceh, but my understanding is that the Indonesian statements are specific enough to suggest that Sabang radar (Maimun Saleh) is the relevant one. Sabang seems to be a smaller base than the one in Aceh.

    And I agree that a true-track ghost flight didn’t make much sense from the start.

    @Ge Rijn
    I could also see the diversion as part of a greater scheme rather than individual suicide, with the pilot passing out at some point (and Z in particular was a smoker in his fifties and suffering from asthma at that). For example, the RMP do not seem to have bothered to dig deeper into the WeChat content, even though China has full access to it. It would normally be among the first things to do. For example, the recent Berlin terrorist sent a message to a friend via WhatsApp, suggesting he is going to undertake an attack (but I wouldn’t expect a message like: “Look, why do you not commit suicide today?”). Otherwise it seems counter-intuitive to spend that much time on social media in preparation for a suicide (I don’t know how often pilot roles are changed immediately before take off).

  10. TBill asks: “Magnetic headings are calculated from tables which change each year, so the question becomes what mag correction tables everyone is using including MH370 (was it current?).”

    Good question, and one that I have asked previously with no response. This detail is one of many I have looked into to see which of teh assumptions we make might be invalid. Even without knowing, one can bracket the possibilities. One of my routes ends at latitude -33.5 (l is 94.7). Magnetic declination for different years is as follows:

    1980: 20.7
    1995: 19.2
    2014: 17.7

    9M-MRO was built in 2002, so if 1995 tables were loaded and never updated, error could be 1.5 deg at end. However, if the initial heading at the EOR is unknown, it is a degree of freedom that could be tweaked so the endpoint is not particularly affected and the overall flight path is only slightly different.

  11. @DrBobbyUlich

    Have you ever considered that the two log on events are corrupted?
    Bad data in equals bad data out.

    What end location would you suggest if you only use arcs 2 to 6?

    What fuel load is remaining at 0017:30 UTC?

    Start with ACARS fuel load and not incorrect Igari fuel estimate. Assume diversion is Mach 0.84 at Flight Level 350 and at any top of descent the aircraft slows to 272 KIAS.
    When within 40 nautical miles of a destination, speed 240 KIAS.
    At any end of route use True heading and assume Left HGA is unserviceable.

    Keep up the great work.

  12. @TBill

    I have also looked it up on google earth, and Car Nicobar Airport is definitely further away than 200 nm from Maimun Saleh. After the southern turn it would also be possible manually to keep equal distance from Sabang to avoid detection, especially on a reduced flight level.

  13. @sk990 I have read that Magvar table update is done every 5 years 2010, 2015 …

    The differences on this timescale are fractions of a degree so less uncertain than wind vector errors (or altitude, speed, control mode or variation from ISA). Besides a path modeller will simply optimise initial heading for whatever magvar epoch they are assuming…

  14. Paul Smithson,

    FYI – here’s a presentation made by someone at the FAA a couple of years ago with good information r.e. magvar tables and the rationale behind them. (Anchorage and Fairbanks are the big problem spots.)

    http://fsims.faa.gov/Wdocs/OSWG/OSWG%20MagVar%20Brief.ppt

    Update rate is every 5 or 10 years (not very specific), but note that Honeywell can only update ~50 IRUs per month (!), and given that a B-777 has 3, it seems that it would be easy for an airline like MAS (which doesn’t operate in far North or South locations) to fall behind.

  15. @JS
    “Can you remind us how fast the cabin pressure can be reduced in the absence of mechanical damage? We hear “sudden” decompression and “explosive” decompression but how fast can this occur using only the valves?”

    On the FS9 PSS 777-200 model, that Z had used, the max depressure rate seems to be 2500-ft/min in Cabin Altitude. The depressure rate decreases with time, approx. to a rate of about 2000-ft/min by 20000-ft and rate keeps going down. So you’re talking 5 or 10 minutes or more to get up to full altitude.

    I don’t know if PSS model is accurate or if there are ways to speed it up by cutting off incoming bleed air. So this is not explosive, and it is not really rapid.

  16. Does anyone else recall seeing a photograph of Captain Zaharie Shah wearing a t-shirt with a submarine on it? Searched online for it again today, but curiously without success.

    When I saw that photo initially some time back, it occurred to me that someone with an interest in submarines might also be somewhat interested in their primary sensor array, namely SONAR, which is used for detecting various objects underwater.

    Even a rudimentary understanding of basic SONAR systems and their practical limitations wrt to shadow zones is adequate to help pick out the very best places in the ocean to crash a plane such that the wreckage on the seafloor will be as difficult to locate as possible using current U/W search methods.

    Somewhere exceedingly deep with extremely rough seafloor bathymetry would be ideal, which is exactly what you find in the Broken Ridge area of the SIO.

    I agree with @ROB that a nefarious pilot Z would plan ahead and act accordingly beforehand in order to avoid suspicion. Also agree that a predictable straight line flight path down into SIO seems a bit careless for a nefarious pilot with a devious master plan still sitting at the controls.

    I believe the intersection of Broken Ridge & the 7th Arc was the intended final destination, but it may have been a jagged and evasive flight path leading there. This might explain why MH370 is likely farther north along the 7th Arc than the maximum range fuel endurance calculations predicted, in addition to the obvious fact that a live pilot can dump fuel or crash the plane anytime they choose.

    Anywhere along the Broken Ridge area will be exceedingly difficult and dangerous to search effectively with towed array side scan sonar, so it could necessitate more time consuming AUV ops. MH370 will eventually be found someday, but it’s just such a very big ocean to search given our current level of technology and the limited clues available to narrow it down.

  17. @David,

    Good catch, David. Yes, I would expect the “excess” fuel flow at temperatures higher than ISA to become a “savings” in fuel flow at temperatures lower than ISA. The zero slope of the curve fit at ISA was just what EXCEL happened to fit, and it is not accurate close to zero temperature deviation.

    I also expect the Rolls Royce engines to have the same temperature coefficient as the GE engines.

  18. @TBill. Can you say how depressurisation was selected? Did you open both outflow valves having switched them to manual?

    If you turn off bleed air from both engines that should make a difference, supposing the APU is not running. (It will also will cut off other pneumatics.)

    It takes 11 seconds from memory for the outlet valves to open manually. The shutoff of bleed air might also take time: normal shut down of an air conditioning pack has a flow reduction schedule.

    Shutting down the packs will not achieve the same unless trim air is shut off also.

    A pilot pitching down due to depressurisation at fuel exhaustion was raised again recently by Ge Rijn (and by you I think) and I have been delving into that.

    The simulator might help there too, though I suspect a double engine failure and APU autostart might be beyond its ken.

    As yet I am unsure whether the APU would supply bleed air for the packs and trim air after the minute it needs to get started (according to the ATSB, though not me) following engine fuel exhaustion.

    Even if it did, as I suspect, akin to on the ground before engine start, at high altitude there would be doubts as to air quantity and pressure being sufficient for pressurisation – it cannot supply sufficient for engine restart above 22,000 ft. from memory.

    Whether it could manage this and simultaneous pneumatic drawdown of air powered hydraulic pumps etc is doubly problematic, while also delivering the electrical demand (aided a bit by the RAT) which includes reboot of the SDU.

    Of course it would run for limited time only anyway.

    So, to the point. At fuel exhaustion we have a pilot conscious of potential depressurisation, presumably unaware the APU would do the ATSB autostart.

    Since the outflow valves will have been in auto they will now close and reduce depressurisation rate to slow, which he will know. Thus there is no urgent need to descend, that is if the pressure has not dropped markedly during outflow valve closing (presumably not; that is what the automatic closing is for).

    If he did put his nose down for a descent consistent with the ATSB interpretation of the BFOs it would be near enough at two minutes after fuel exhaustion when he did, the nose down being abrupt to get his speed vector pointing downwards in 8 seconds.

    That would require an urgent reason well after fuel exhaustion.

    Therefore it is unlikely that depressurisation would been the instigator of the BFOs and an APU contribution to pressurisation is immaterial.

  19. @Oxy,

    I’m not sure which log-on data you are suggesting may be corrupted. I don’t use the 18:25:34 and 00:19:37 BTOs or BFOs because Honeywell has never explained why those BFO values are different. I don’t think they are “corrupted”, but they are at least processed in a different way in the SDU. I think they are deterministic, not random. I think in some way the calculated frequency compensation term is different from the other messages.

    In the past I tried fitting 19:41 to 00:11 data only, and that was useful to find which routes were possible because the fitting routine would converge more quickly with fewer variables to optimize. Then I went to fitting the 18:22 – 18:28 data first followed by a second fit for the FMT to 00:11 data. That works fairly well and the processing time is still reasonable. Finally, what I do now is fit 18:22 to 00:11 all at once. It is slower but more reliable, and it allows accurate fuel calculations.

    As an example, if you fit the 18:22 to 18:28 data, you get a lateral route offset of about 12 NM. If you then fix that portion of the route and fit the FMT to 00:11 part, you get a route that passes right through ANOKO. However, when you fit both of the data sets together, you get a very good fit at 18:22 to 18:28 with a 10 NM offset and a route that passes 10 NM to the right of ANOKO. I was actually surprised that the result was a slightly better fit overall than the one I did in two parts.

    I didn’t notice any obvious bias in the end location when fitting starting at 19:41.

    At 00:17:30 the main tanks are dry with both main engines flamed out. The APU apparently had a few minutes of fuel available then.

  20. @TBill,

    The default navigation method triggered by EOR and Route Discontinuity is apparently not a selectable option. I don’t think even Honeywell knows what it does.

    @GeRijn,

    The NORM/TRUE switch is usually kept in the NORM position for all flights except in the polar regions, where TRUE is recommended. If Honeywell is to be believed, when this switch is in the TRUE position and an EOR error occurs, the resulting path will be constant TRUE heading.

  21. @Gysbreght,
    @VictorI,

    Thank you for your comments on the temperature dependence on fuel flow.

    For the Enroute Climb data, I believe Gysbreght’s point is that there should be a change in slope above ISA+10C because the thrust is reduced there to avoid overheating. That makes sense, but it would not necessarily apply to level flight in cruise mode, which is not done at maximum thrust, and so in that case there may not be any change in slope.

    Since the Enroute Climb data are actually the total fuel required to make the climb, and not the fuel burn rate during the climb, it increases because a greater height in feet must be climbed (taking longer), and above ISA+10C the maximum thrust is reduced (again making the climb take longer). So as Gysbreght pointed out, both effects increase the time to climb and therefore the fuel needed to climb. That accounts for the rather large upward turn of the graph at high temperatures.

    It seems the temperature dependence of the cruise fuel flow (the fuel burn rate) therefore won’t follow the same sort of nonlinear curve as the Enroute Climb fuel. As Victor pointed out, the fuel flow for a given thrust varies as the square root of the absolute temperature. For small temperature changes of 10-20C, a linear approximation will work just fine. I don’t know the functional form of the drag versus temperature term, but again it is probably fairly well represented by a linear equation for temperature excursions of < 20C. So the linear correction factor in the GE engine tables is probably still accurate even somewhat above ISA+10C, and the same line can be used to estimate the fuel flow reduction at temperatures below ISA.

  22. @DrBobbyUlich

    Thank you for your additional comment on the TRUE/MAG switch.

    But do I correctly see a kind of contradiction between your comment to @TBill and to me?

    To @TBill:
    “The default navigation method triggered by EOR and Route Discontinuity is apparently not a selectable option.”

    To me:
    “when this switch is in the TRUE position and an EOR error occurs, the resulting path will be constant TRUE heading.”

    Doesn’t your comment to me mean the navigation method after EOR/Route Discontinuity is a selectable option by switching the TRUE/MAG switch to TRUE?
    Resulting in a constant TRUE heading path if an EOR error occurs?

    Assuming the pilot knew he was going to fly to an EOR or Route Discontinuity and wanted to maintain a constant TRUE heading after that, all he had to do to be sure of this was setting the TRUE/MAG switch to TRUE before this EOR or Route Discontinuity.
    Is this correct?

  23. @sk999. Thanks for sharing that very interesting brief. It seems to me the “default” assumption for 9M-MRO’s magvar tables is that they date from time of aircraft delivery or soon (a year or three) before that?

    If so, we’d expect those tables to be ~10 years vintage.

    This will make no practical difference in the tropics, especially if the “movement” of declination is in the the same direction over our area of interest. You model the same path, only with a marginally different magnetic heading.

    It could conceivably make a greater difference in southern latitudes (say, south of about 30S) where rate of change is much greater – affecting which path model offers superior fit.

  24. …and (of course) makes no difference at all to optimised great circle, true heading or true track path models

  25. Here is a video that might interest some of you. It shows the EICAS messages generated by isolating the Left AC Bus by opening the Left Generator and Left Bus Tie breakers. The displayed messages indicate that SATCOM DATALINK and SATVOICE are lost. A pilot would be very aware of the consequences of isolating the Left AC Bus.

    In the video, the synoptic display of the electrical system shows that the Left AC Bus may be fed by either the Left Generator or through the Left Bus Tie. This again shows that two breakers must be opened to isolate the bus.

  26. @David
    For the depressure event, you are correct, I set the outflow valves to manual and then I opened the valves. It took me 20-minutes to get the Cabin up to FL350 altitude, and I had ascended to FL410 during the process to see if I could speed it up. Turning off bleed air did not seem to matter (in the PSS simulator).

    Bringing pressure back up was even much slower, with the PSS 777 suggesting 700 ft/min was the fastest you could pressure back up.

    So not sure this is accurate, but the implications are a fairly slow process to depressure and repressure back to normal. The depressure rate is slow enough that is does not get too cold, although it would probably be uncomfortable. I didn’t see a message about masks popping down but I get an alarm around 8500-ft regardless of the landing altitude which can be set up to FL140.

  27. Paul Smithson,

    Here’s another document on magnetic variation and the World Magnetic Model (WMM):

    https://www.faa.gov/about/office_org/headquarters_offices/avs/offices/afs/afs400/parc/parc_reco/media/2013/130617_PARCMagVarRecommendations.pdf

    WMM is updated every 5 years, but according to Boeing, its MagVar tables are updated every 10 years. WMM contains time-dependent predictions into to the future, but they are too complex to use onboard an aircraft, so what this suggests is that MagVar tables are a static snapshot of WMM projected 5 years in the future, so that the tables should be good for 10 years. Thus, a table issued in 1995 will have tables for epoch 2000, and these values will not match exactly the actual WMM update in 2000. So if this is correct (and I am just “reading tea leaves”), 9M-MRO tables will be for either epoch 2000 or 2010, depending on MAS update policy. One thing I am more certain of is that one should not use WMM Mar 7, 2014 values!

  28. @PaulS @SK999
    I see it in the FS9 flight sim the orig older mag headings are more curvature probably 1 deg more than 2017 headings.

    I try to calibrate off of Victor’s 180 S Mag path paper, to get closer to what Victor showed. I like the fact that Victor provides “with wind” and “no wind” paths so I can try to match the “no wind” to calibrate.

    Victor, how did you calc the Mag path in the paper?

  29. TBill. The differences (2000 model vs 2015 model, for longitude 95E are:

    0.3 at 0N
    0.8 at 10S
    1.52 at 20S
    2.28 at 30S
    2.60 at 40S

    In each case, magnetic variation was greater in 2000 than in 2010. So a magnetic track in 2000 would be more “bent” than a 2010 one.

    At any given point along the path there will be only max 2.6 degree difference in instantaneous azimuth (at 40S in this case). But because of the progressive difference along the path they will diverge quite a bit in end position.

    Unless you are into modelling magnetic paths, it doesn’t matter though!

  30. @GeRijn,

    TBill’s question, and my answer, concerned whether or not an airline could specify/select which default nav mode they wanted. The answer is no, it’s not an option the purchaser can specify. All the units are the same (whatever they do, they all do it).

  31. @all

    Direct quote from Kahneman and Tversky below based on reams of experimental data:

    “There is much evidence showing that, once an uncertain situation has been perceived or interpreted in a particular fashion, it is quite difficult to view it in any other way.”

    Humans, including professionals of all persuasions, are notoriously incapable of outperforming simple clinical algorithms – even algorithms they have created themselves. We have good days and bad days. Unfortunately the bad days far outnumber the good.

  32. Update on CTH Route Lateral Offset

    It turns out there is a very good fit to a route like this:

    1. 10.0 NM lateral offset to the right of N571 initiated at 18:25:35
    2. Holding pattern initiated at 18:38:25 with ANOKO as the fix (FMT begins)
    3. FMT toward ANOKO completed at 18:40:15 (during first part of 18:40 phone call)
    4. “END OF OFFSET” warning displayed at 18:41:30, but offset was not subsequently set to zero
    5. FMS automatically decelerates from LRC to Best Holding (257 KIAS) prior to reaching ANOKO
    6. At 18:43:31 the “ANOKO + 10 NM Right” position is reached
    7. First error occurring then is “END OF OFFSET” which cancels the holding pattern because of the non-zero route offset (but leaves the speed at 257 KIAS)
    8. Second error is “END OF ROUTE” because there are no other entries after the ANOKO + OFFSET fix
    9. EOR error produces constant heading route at current heading
    10. Constant heading and constant KIAS continue until fuel exhaustion

    For a detailed timeline, go to :

    https://drive.google.com/file/d/0BzOIIFNlx2aUbFJwQ2gxUjJFVEU/view?usp=sharing

    In my opinion, this “cancelled holding pattern” is a more likely scenario than the route fit that passes directly over ANOKO since in that case the following actions are needed:

    1. ANOKO would be selected as a single waypoint in the route,
    2. the route offset would have to be set to zero before reaching ANOKO,
    3. the Best Holding speed would have to be displayed on the CDU, and
    4. the Best Holding speed would be entered manually in the MCP.

    In the cancelled holding pattern route, the FMC picks the speed and automatically sets it before the EOR error occurs. All the PF needs to do is set up a holding pattern with ANOKO as the fix.

    There are at least two issues with this scenario. First, why use a ~10 NM lateral offset instead of 1 or 2 NM, which are the ICAO standards? Second, why fly a holding pattern? I have some ideas, but it seems the answers people come up with (including me) are tied to their view of why the diversion occurred in the first place. See Dennis’ quote above. This is a perfect example.

  33. @DrBobbyUlrich

    I did understand your answer to @TBill.
    Now I doubt if you got my comment and my question the way I meant it.

    My point was if you set the TRUE/MAG(NORM) switch to TRUE before an anticipated EOR or Route Discontinuity, you will create a default navigation mode being a constant TRUE heading after this anticipated EOR/RD.

    Is this correct or not?

  34. Something else off topic but in a sence it is on topic.
    Dutch media state today Trump declared the downing of MH17 will never be resolved and he excluded Russia as being possibly responsible on forehand.
    We truly get another psychopath added to rule the World .
    God bless America…

  35. @Ge Rijn

    Yes, Trumps’s remarks fly in the face of all the progress the Euros have made on the issue.

  36. @Dennis @ Ge Rijn
    One could assume, that SVR has landed its greatest succes, to place an agent into the white house.

    God bless the world.

  37. @TBill
    @Ge Rijn
    @StevanG

    Further to the Indonesian radar riddle, I have now undertaken an exercise to try and determine a viable flight path.

    The aims and objectives of this exercise were to investigate an example flight path that satisfies the following criteria:

    – the flight path is not at odds with official statements that Maimun Saleh radar was unable to spot the plane within Indonesian airspace, but may have detected it in the Car Nicobar area
    – the flight path intersects the 19:41 ping ring (red bow) and the distance covered amounts to a reasonable ground speed, taking into account suitable waypoints en route.

    Methodologically, I assumed Maimun Saleh has an approximate radar range of 200 NM (red circle) consistent with both primary military radar capacity and the extent of Indonesian FIR at the intersection of waypoints AKINO and TOPIN. Underpinning this approach is a hijacking scenario in which the person in control of the aircraft wanted to make sure that he is not identified by Indonesian radar as a potential threat.

    I came up with the following example flight path that satisfies the criterica outlined above.

    https://www.allmystery.de/static/upics/4ac7186f85_earth5.png

    The distance covered is around 660 NM in 80 minutes and this amounts to an average ground speed of 495 kts. This is slightly lower than the groundspeed oberserved in the early flight leg by primary radar in the FI and may have to do with a speed reduction due to the posited temporary descent at 18:40. Note that this descent comes at a time when MH370 was approaching the maximum range of radar detection and could therefore be explained as an attempt to duck out of sight at this critical moment.

    I have no flying experience but have been told it is feasible to replicate this route. The pilot can proceed from MEKAR to VOKC and, at a distance of 200 NM from Maimun Saleh (WITN) swap over to AKINO, manually directing the plane via heading and observing equal distance to WITN. He or she may eventually proceed to Cocos Is waypoint.

    Limitations: I have not yet crosschecked BFO values, but I assume the above scenario can be aligned with the early ATSB “northern hook around the tip of Sumatra” flight paths. That would be p. 27-28:

    http://www.atsb.gov.au/media/5243942/ae-2014-054_mh370_-_definition_of_underwater_search_areas_18aug2014.pdf

    Thoughts?

  38. @Nederland

    The latitude of your 19:41 range ring point is ~2N. IMO, this is too far South to enable flight paths terminating in the area of 33S and above. I think the interesting paths at the moment have a 19:41 latitude ~8N. JMO.

  39. @Nederland
    I agree with staying out of “view” of the Sabang radar. OKINO/TOPIN mark the west boundary of Indonesian air space. Note that the ZS home simulator case also curves outside those waypoints {DOTEN or 10N90E to NOBEY is an approximation of home sim route in SkyVector}.

    The curved flight path around the radar is probably not natural pilot thinking. I’d be tempted to fly further west or north and switch back. You are quite a bit further south than Victor at 19:40 so you have “time”. Of course hitting the BTO/BFO is the game.

  40. @Nederland,

    I think it is generally accepted that the flight path has to cross the 19:41 range ring from outside to inside, and also that the point of closest approach to the sub-satellite position occurs at about 19:52.

  41. Article by Geoffrey Thomas in The West Australian newspaper.

    I have covered the tragic loss of MH370 almost from the moment it disappeared on March 8, 2014 and there have been many moments of sadness and even tears when meeting some of the relatives last year in Perth.

    But nothing has affected me as much as flying the route of MH370, as we understand it, in a Boeing 777 simulator at Jandakot in April of that year.

    As I nestled into the captain’s seat I felt a chill up my spine. It was dark outside — inky black, in fact.

    This was going to be as real as it gets without flying the plane.

    The simulator cockpit is an exact replica and I had a former 777 check and training captain as my co-pilot.

    Simulators are not toys and can replicate any incidents that cannot be trained for in real planes.

    The take-off was effortless and the lights of Kuala Lumpur International Airport disappeared below. I called for undercarriage retraction, then flaps up as the speed of the 777 increased.

    I quickly engaged the autopilot once on track and the 777 soared to 10,600m for the almost boring flight to Beijing.

    But this plane and its 239 passengers and crew were not going to Beijing.

    The climb was routine and the scene surreal.

    We burst through the layer of cloud, which was lit up by a half-moon, and the view outside was peaceful, belying the events about to unfold.

    At 1.07am, 27 minutes after take-off, I switched off the 777’s aircraft communications and reporting system. ACARS sends hourly data reports on the health — or not — of the 777.

    It was too easy, merely requiring the press of a button on the flight management computer to my right.

    Just 15 minutes later, after a sign-off to Malaysian air traffic control, I turned a knob and killed the plane’s transponder that sends our signature to air traffic control. I have now vanished from air traffic control.

    If I had wanted to get rid of my co-pilot, as some have suggested may have happened on the ill-fated MH370, it was incredibly easy.

    When he left the cockpit for a break, I selected deny on the cockpit door access switch and he is not coming back. I don’t even have to move from my seat.

    Back to the flight, a simple twist of the heading select button on the autopilot and depressing the same button put me on a new course south-west towards Malaysia.

    There were contradictory reports about the 777’s change of altitude at the time.

    To test the capabilities at our given weight, I coaxed the 777 up another few thousand metres.

    Getting it down to a lower altitude as reported was much easier. I just dialled in the altitude and vertical speed required.

    The 777 is certified to descend at a maximum 1524m a minute. I changed course to due west and headed across Malaysia just below Penang.

    Once over the Strait of Malacca, I made another turn north-west.

    Again, it was easy to dial in a new course of 330 and push select. The 777 obeyed my every wish.

    As I was effortlessly commanding this 777 my mind wandered to the early hours of March 8 and what was happening on the real MH370.

    I was going to finish this assignment and drive back to my office — the passengers on MH370 could not.

    My co-pilot reminded me it was time for the next turn if I am to reach the accepted final resting place of MH370.

    I dialled in 180 due south for the heading and 10,600m for the altitude.

    I pressed select and the 777 turned slowly but assuredly to meet its fate.

    Many in the aviation industry suggest that by daybreak on March 8, the passengers and crew on MH370 were already dead — quite probably from induced hypoxia designed to mercifully overwhelm them and put them to sleep.

    While passengers would have run out of oxygen in 45 minutes, the pilot had four hours.

    In my simulator, the sun was rising in the east as I contemplated the mystery of MH370.

    What would it be like at the end? I was about to find out and it was chilling.

    The serenity of the moment was shattered as I cut the fuel to one engine.

    The 777’s systems reacted with some urgency and compensated for the lack of power from one engine as the flight continues.

    But all hell was about to break loose as I switched the fuel off to the remaining engine.

    The result, I was told, is unpredictable. And it was. It is also utterly terrifying. We tried it three times.

    The first put us nose-up into an aerodynamic stall and then a dive followed by another nose-up. We were on the rollercoaster from hell. Bells and alarms rang out. It was chaos and suddenly we hit the sea.

    On another attempt, I found myself in a flat spin with alarms shattering the eerie silence of engines without power. The control column was shaking violently, the altitude read-out was a blur and the forward speed non-existent. I forgot I was in a simulator — this was real and I was sweating.

    The third was a spiral dive at near supersonic speed, which as it turns out, was the most likely scenario, according to Boeing.

    It took seconds to bring the flight to a ghastly halt.

    The drive back to Seven West Media took on a different perspective. I was able to go back to work, back to my darling wife and back to my wonderful life.

    The 239 passengers and crew of MH370 could not.

  42. @TBill. During climb the cabin altitude rate is controlled to a maximum (normally) of 500 ft/min according to the AMM. This might be influencing simulation during re-pressurisation. Depressurisation rate during descent is also controlled (varies with circumstances), though with air bleed off and outlet valves closed this should have no bearing on the simulation.

    Correction to my earlier, not that it matters much. Outflow valves close in auto when the cabin altitude exceeds 11,000 ft so de-pressurisation rate to that would be faster initially.

    Incidentally I notice that the higher rate of descent described by DrBobbyUlich for the first BFO (9,100fpm or 56 knots vertically) would most likely bring forward the dive initiation.

    About a glide (and for others), if lack of IFE connection at the second log in was caused by the aircraft crash this would require a descent from the time of the first BFO at say 30,000ft to sea level in 1min 37secs (maximum), an average of 185 knots minimum vertically. This would rule out any glide even if from a lesser height.

    However as the ATSB noted, the APU running out of fuel would also cause this as would loss of antenna line of sight to the satellite; and besides a pilot might have selected IFE ‘off’ after the previous log-on.
    I add a (speculative) possibility that it could be the result of load shedding, the APU being overtasked (pneumatics and electrics) at altitude. The load shedding list includes “electronic seat equipment”, presumably IFE.
    The IFE can require power of up to 22kVA (ref available) – quite substantial when compared to the aircraft’s generation; B/U Gen 20kVA; APU, IDG 120kVA ea.

  43. @ed
    Re: Geof Thomas simulator case
    Is was a moonless night, so either they did not set the correct day, or the simulator does not have the Moon correct. Couple other little things.

    @Nederland
    …maybe Brian’s rule suggests MH370 went below the radar beam, instead of around it?

  44. @David,

    You said: “Incidentally I notice that the higher rate of descent described by DrBobbyUlich for the first BFO (9,100fpm or 56 knots vertically) would most likely bring forward the dive initiation.”

    Perhaps you misunderstood. The descent rate at 00:19:29 is about 4,600 fpm based on its BFO. I am guessing the descent rate at 00:19:37 was about 9,100 fpm.

Comments are closed.