Occam’s Razor is Overrated

conspiracy theoryMartin Dolan, chief commissioner of the Australian Transport Safety Bureau (ATSB), is plagued by conspiracy theorists. According to an article in the Sydney Morning Herald, since the disappearance of MH370, “conspiracy theorists have been busy trying to solve the mystery themselves. Many have contacted Dolan.”

“You’ve got this big mystery and everyone wants to know the answer and everyone wants to help,” the SMH quotes Dolan as saying. “It’s unhelpful, for the sake of the families more than anything else, in the sense that it has the potential to undermine confidence in what we are doing.”

I feel somewhat guilty for being one of those peanut-gallery denizens who have tormented him. Along with my fellow obsessives in the Independent Group, I’ve been straining my brain for the last eight months trying to make sense of the strangest aviation mystery in history. Yes, I’d like to be helpful; yes, I’d like to know the answers. And yes, I may have unwittingly undermined confidence in what the ATSB was doing, for instance by publicly saying that I thought they were looking in the wrong place. (Though, to be fair, they were in fact looking in the wrong place.)

Nevertheless, I must take issue with one aspect of the article’s characterization of my subculture: the use of the term “conspiracy theorist.” Now, look: I get it. My wife says that I remind her of the Kevin Costner character in “JFK.” I ruminate about the intracacies of a famous case and try to piece them together in a new way that makes more sense. I’m obsessed.

There’s a big difference, however, between true grassy-knoll conspiracy theorists (or 9/11Truthers, or the-moon-landing-was-faked believers) and MH370 obsessives like me. It’s this: there is no default, mainstream narrative about the missing Malaysian airliner. There is no story that officials and all reasonable people agree makes sense.

This isn’t the result of laziness or incompetence. It’s just that the data is so strange.

A lot of people don’t get that. Ever since the mystery began, certain voices have been invoking the principle of Occam’s razor, saying that when we try to formulate a most likely scenario for what happened to the plane, we should choose the answer that is simplest. People who are making this argument are usually in favor of the argument that the plane suffered a massive mechanical failure and then flew off into the ocean as a ghost ship, or that the pilot locked his co-pilot out of the cockpit and committed suicide. However, as I’ve argued over the course of several earlier posts, neither theory matches what we know about the flight.

Instead, I’ve argued that an accumulation of evidence suggests that MH370 was commandeered by hijackers who had a very sophisticated understanding of airline procedure, air traffic control, avionics systems, military radar surveillance, and satellite communications. In other words, what happened on the night of March 7/8 of this year was a intentional act. And when it comes to human schemes, Occam’s razor goes out the window. Instead of simplicity, we should expect complexity, not to mention red herrings and any other form of subterfuge.

Whenever I hear Occam’s razor invoked, I inevitably find myself thinking of something that Sarah Bajc said on CNN. Bajc’s partner, Philip Wood, is one of the missing passengers, and she has been very open minded in considering alternative explanations to what happened that night. “There are 40 crazy stories that you could tell about MH370,” she told the anchor. “And one of them is going to turn out to be true.”

I’ve come to think of this as the Bajc Postulate, which I think should replace Occam’s Razor in situations like this. It goes like this: “When trying to unravel human deception, don’t expect simplicity.”

Remember Operation Mincemeat? In 1943, a fisherman found the body of a British officer floating in the sea off the Spanish coast. The authorities turned the corpse over to German intelligence, who discovered that it carried a number of secret documents, including one indicating that the expected Allied assault from North Africa would target Sardinia, not Sicily, as widely expected. The authenticity of the documents was vouched for by every detail of the body, its clothes, and the accompanying possessions, which included several love letters, a photo of a fiancee, a bill from an exclusive tailor, and a theater ticket stub. Either this man and his belongings had all been elaborately and meticulously forged, or he really was who he seemed to be: Occam’s Razor. Hitler himself was utterly convinced. And yet, of course, the whole thing was a ruse, an elaborate deception cooked up with painstaking care by British intelligence. Hitler shifted three divisions to Sardinia, the invasion landed at Sicily, and the war was that much closer to being over.

I think it’s distinctly possible that MH370 represents a deception crafted at the same level of complexity.

In my mind, the crux is what happened at 18:25. Until that moment, the plane had been on radio silence for nearly an hour. After following a zig-zag path along national airspace boundaries, it had reached the limit of military radar coverage and had disappeared. But then, mysteriously, the satcom system reconnected to the Inmarsat satellite overhead. For it to do this, the hijackers would have had to either climbed into the electronics bay or carried out a complex procedure in the cockpit that few people outside of Boeing itself would now how to accomplish. All this, to no evident purpose: no attempt was subsequently used to communicate via the system.

Other things were odd about the 18:25 logon. The frequencies that the system transmitted over the next few minutes were inexplicable to the scientists at Inmarsat. Though the electronics of the system are perfectly understood by the equipment’s manufacturers, they cannot explain how the frequencies were produced. Investigative efforts within the IG suggest that there was another mysterious aspect to the satcom’s behavior post-18:25: when a pair of incoming calls was received at 18:41, the system was unable to pass the calls through. We’re not sure why, but the most likely cause is that errors in the system’s configuration prevented it from aiming the satellite dish correctly.

By 19:41, the satcom system seemed to settle down and transmit at stable frequencies. If taken at face value, these frequencies indicated unambiguously that the plane was flying south. Yet the ATSB has never able to completely make sense of these values. As I wrote last week, it has proven frustratingly difficult to make the two distinct halves of the Inmarsat data—the timing and the frequency data—match up in a way that makes sense.

Regardless of these difficulties, most reasonable people share the conviction that, regardless of what particular track the plane happened to fly, it definitely flew south into the most remote reaches of the southern Indian Ocean. I’ve examined the data myself, and come away convinced that, indeed, the frequency data unequivocally supports this conclusion. But no one knows why anyone would do this. One popular notion is that the hijackers had a destination in mind, but something went wrong, they became incapacitated, and the plane flew on autopilot until it ran out of fuel and crashed. This scenario is certainly possible, but as I recently pointed out, a new speed-analysis technique suggests the plane was under deliberate control until the very end.

So if they weren’t incapacitated, why were these very motivated, very sophisticated hijackers flying a perfectly good jet off into the middle of the ocean? As I see it, there are two possibilities:

  • The hijackers were very sophisticated, but for some unknown reason chose to fly the plane off into the middle of the ocean, or
  • They were very, very, very sophisticated, and not only survived, but managed to cover their tracks in a way that has fooled absolutely everybody — and turning on the SDU was an essential part of their plan. This explains why there has been no debris found, why there was no radar track over the southern Indian Ocean, and why Inmarsat has been baffled by the BFO values.

This kind of thinking would have been considered outlandish a few months ago, but the more time goes by without any trace of the plane turning up, the more reasonable it is starting to seem. No less an industry eminence than Emirates CEO Tim Clark, whose airline operates the largest 777 fleet in the world, recently told Der Spiegel: “We have not seen a single thing that suggests categorically that this aircraft is where they say it is, apart from this so-called electronic satellite ‘handshake,’ which I question as well.”

To accomplish a disappearing act, the hijackers would have had to have pulled off a plan that the authorities not only couldn’t anticipate beforehand, few could wrap their heads around it afterward. A plan so devious, it would literally be —


What could such a plan have been? Frankly, there’s no way we can be sure. Until the plane is located, and the black boxes are found, all we can do is speculate. But some speculation runs in accordance with the facts, and some runs counter to it. Over the last few months, I’ve pieced together a narrative that I think matches well the facts we do know, explains some otherwise baffling conundrums, and basically ties together a means, a perpetrator, and a destination. (Which, paranthetically, is something that no one else, official or amateur, has yet attempted.)

In the past, I’ve invited others to share their “conspiracy theories,” and I tip my hat to the very, very few (two) who’ve had the courage to take me up on my offer. For the most part, their efforts were met with skepticism, but polite skepticism, and that reaction has emboldened me to press forward with my own big reveal. I hope that some people will find it thought-provoking, perhaps even convincing. I expect that a great many will find it, yes, inconceivable, perhaps even outrageous or even offensive. Remember, it is speculation, not a statement of fact; but if we don’t risk trotting out our speculations eventually then we will never get any closer to figuring out the truth.

If you care to dive down my rabbit hole, click below:

The Spoof, Part 1: Why (A Speculative Scenario)

The Spoof, Part 2: How (A Speculative Scenario)

The Spoof, Part 3: Where (Not a Speculative Scenario)

The Spoof, Part 4: Motive

The Spoof, Part 5: People on the Plane

The Spoof, Part 6: MH17

And that’s all there is for now.

450 thoughts on “Occam’s Razor is Overrated”

  1. @Myron, @Nihonmama: if Bobby can prove the turn took place PRIOR to 18:40 – AND I can prove the ATSB’s own Oct.8 Report (Fig.3, FL400, et al) says that the fuel-feasible zone should reach 4 degrees further WEST along the 7th arc than it currently does – then Bobby’s suggested impact point goes from “impossible” to being part of a “hotspot” (from east 84 to east 88 longitude) along the 7th arc.

    A hotspot which is, as we speak, (just barely) NOT being searched.

  2. If this new turn takes place sooner and a flies over indonesia (or within their radar range) then does all bfo/bto’s fit strongly as per north path to Almaty ?

  3. @Myron,

    Major take-aways from sawtooth BFO observation at 1840 are:

    1. This is another argument why the final major turn was not underway at 1840. That still leaves 1827 – 1839 as possibilities.

    2. Some of the residual BFO errors after fitting are caused by this effect (up to 4 Hz). No wonder Inmarsat said to allow as much as +/- 7 Hz for agreement.

    I will look at the second satellite phone call BFO data tomorrow.


    Go here to see the sawteeth in the 23:14 (second sat phone call) BFO data:


    In this case the peak-to-peak amplitude is 7 Hz and the period is 22.5 seconds. The sawtooth residuals are less than +/- 1.5 Hz. That is a factor of 2X reduction compared to using just a straight line fit. This provides more evidence that sawteeth are real and probably present in all the BFO data when the aircraft is airborne.

    The presence of sawteeth as seen in the BFO data but which are not included in current BFO models casts significant doubt on the notion that small BFO errors imply a higher probability of being the correct route. I think that any residuals smaller than 4-7 Hz peak are simply a result of over-fitting the BFO data with too many variables.

    It seems Inmarsat was right when they advised accepting BFO errors up to +/-7 Hz.

  5. Any LNAV controlkled eg. following a series of waypoints, is executed with standard rate turn= 180 degrees/min

  6. @Bobby Thats right Bobby, and “If” heading South then it corresponds with a South Easterly Path and Descent with Westerly drop Rate and If headed North and North Easterly Path and looping Back to South with a Westerly Drop Rate in descent as noted and depicted…

    also Data sequence and DuplicationIssuesPgs} 3;4;5;6;7;8;9;10;11;12;14;15;16;18;19;20;22;24;25;26;28;29;30;31;33;34;36;37;38;39;40;41 also…. Phone Data issues out sequence as stipulated……

    also…. TZ Overlaps with corresponding UTC time and sequence Fails to be in order as stipulated for Overlay of Corresponding Flight Path for muilotiple TZ from Take off to incident period and loop back and across malay to Triplicate Offset TZ Local Adjust for Straits to Andaman Sea Area …..

    Keep it going Bobby, You are seeing the forest Now, by cutting those trees down, and there is a lot of crap in the pond to see the Lake too. ;-)>

    Good Work!!!! Thank you, however Chaotic I may be, there is a method to my madness… LOL.. have a good Day everyone….

    @FelinNut The Chaotic Cat!!

  7. BTW…. Everyone, Lets not forget 40d is Double Latitude and Satellite Keyhole 20d is Double Latitude leaving 12N or 51.5d arc, so 40d isnt 40d ;-)> Just a Bone to crunch-on;-)>

    Logarithom base Properties

    49660 = 3hrs Rising = 40d
    Middle time 11505 2 hours rising
    2 hours rising 23063 40 degrees
    1 hour rising = 40 degrees at 11754
    2 hours rising 47.5 degrees 18040

    @FelineNut The Chaotic Cat!!!!

  8. @ Dr. Bobby Ulich:

    Thanks for your post of 14 December at 1:46 AM.

    Are you perhaps stretching things a bit?

    Except for a single point the data points are 218 +/- 2 Hz, and wouldn’t a period of roughly 3 seconds fit equally well?

  9. @ Dr. Bobby Ulich:

    As a possible explanation for ‘sawtooth’ data errors the following may be of interest.

    Extract from the NTSB Aircraft Performance Group Chairman’s Report on the investigation of the A320 ditching in the Hudson river on January 15, 2009 (NTSB#DCA09MA026)

    “Correction of the sample times of altitude data
    Computing rate of climb based on the time derivative of the pressure altitude recorded by
    the FDR results in an unrealistic “square wave” oscillation about the likely true value of rate of climb. The period of this oscillation is about 4 seconds. The source of the oscillation is also apparent in the pressure altitude data itself, which during the climb appears to “step” every fourth data point (see Figure 5c).
    According to the Airbus FDR Group Member, this behavior is a result of the difference
    between the update rate of the altitude data in the Display Management Computer (DMC),
    which is once every 0.8 seconds, and the recording or sampling rate of the FDR, which is once every second. As a consequence of this difference, one out of every 5 DMC altitude samples is omitted from the FDR recording.
    It should also be noted that the recording architecture does not necessarily guarantee that the sampling time associated with a recorded altitude value does in fact correspond to the time the value was sensed. In this case, the data values associated with a given sample time may in fact be up to 0.8 seconds old.”

  10. P.S.

    The Boeing 777 equivalent of the Airbus DMC is the AIMS (Airplane Information Management System) which incorporates the ACARS function and presumably controls the navigation data stream to the SDU.

  11. @Gysbreght,

    No, using a period of ~3 seconds for the 23:14 data set is a noticeably poorer match, especially so during the first 4 seconds after 23:14:00 UTC.

  12. All,

    To summarize, there are now at least 3 theories of the cause of the BFO sawteeth:

    1. The AEX TX algorithm has limited precision in calculation,
    2. the navigation data are limited in precision, and/or
    3. the navigation data are precise but are not timely due to asynchronous generation/sampling rates.

  13. @ Dr. Bobby Ulich,

    In analogy to typical DFDR recording frames, and in consideration of the purpose of the AES frequency compensation, I could imagine that groundspeed and track are sampled several times per second, whereas latitude, longitude and altitude are sampled once every few seconds. An DFDR frame consists of four one-second subframes, i.e. the entire frame is refreshed every four seconds.

  14. Gysbrecht, Bobby,

    Seven ARINC429 words are transmitted from the Data Conversion Gateway Function on AIMS to the SDU:
    310 Present Lat
    311 Present Long
    The above must transfer between 5 and 10 updates per sec. Resolution = 0.000172 deg
    312 Ground Spd
    Updates 20 – 40 updates per sec. Res = 0.125kt
    313 Track Angle
    314 True Hdg
    The above must transfer between 20-40 updates per sec. Res = 0.0055 deg
    324 Roll
    325 Pitch
    324 & 325 must transfer at 50 to 100 updates per sec. Resolution 0.01 deg

    Pitch & roll are sourced from ADIRU while the rest from the Flight Mgmt Compute Functn on AIMS.

    Would I be correct to expect that any BFO variation implied by RoC would not exhibit a sawtooth noise artifact?


  15. @ GuardedDon,

    Thanks for that pertinent information. So I was guessing correctly on two counts: the AIMS is the source for part of the data, and speed data is sent more frequently than position data.

    But the link between the AIMS and the SDU is only part of the chain between sensor-AIMS-SDU-frequency compensation. For example: how frequently does the AIMS update these parameters?

  16. @Bobby, BFO lovers,

    A possible fourth theory for the BFO saw tooth is the 1hz resolution of the frequency compensation as follows:

    The actual doppler changes continuously as velocity, ROC, heading, etc. change. The correction algorithm will keep the offset constant while the actual doppler delta remains within the current 1Hz bracket. Once the actual doppler delta moves into the next 1Hz bracket, the algorithm adjusts the offset to that next 1Hz bracket.

    If the above model is close to the mark, we can draw the following important conclusions:

    1. Both the saw tooth frequency and slope are accurate measures of the instantaneous rate of change of doppler.
    2. The slope is the residual of the continuous actual doppler change.
    3. The higher the slope and frequency, the higher the rate of change of the doppler contributors.

    This of course would indicate that the 18:40 saw tooth is indeed an indication of a large rate of change of something, possibly the heading change and associated altitude change during a turn.


  17. @ gysbreght

    It depends on which mode of operation the SCU and SDU and AIMS may be running in at that Data Point.

    All information provided appears from the SDU in the Mode of Operation LIMITED in tranfers Rate andf speed by the ATN FANS baud and Bit Rates mandated by ICAO, initiated by the FAA, Designed by USA and Has been a Redundancy Failure Operation since the SDU inception.

    The issue of confusion is beneath the HSU as the SDU / SCU Brain being the Areo is the Brain for the HSU as a Redunacy Failure.

    This SDU Brain controls its Functions for these Event, being the 1st of its kind in that mode of operation.

    The private Coroporations tried to phase the ATN FANS out of the SDU, and ICAO refused in 2010.

    This mode mode of Operation has several Funtions, all limited in Bandwidth and Rate on specific Channels and Freq. and they are CAPPED in Rate Intentionally, not to overload the Emergency system of the ATN Circuit Mode Data packages.
    The Data can be sent via Sat-Phone also with-out answer on the opposite end on a different Baud and Bit Rate, still capped.

    It also allows for a GPS multi-site source code and logarithom base properties to be sent.

    The Problem is the system was Never updated in 1997 when the FAA and Boeing alloweded inmarsat a 20B$ Grant to update and incorporate the system, and Inmarsat Failed and Blew the Money, and the system slowly faded away, but the operational Functions still existed, and we have MH370, the very reason that system was kept by the FAA; Boeing; and ICAO.

    So, the answer to you question is in the the designed Functions of the ATN Limited Band Rate Cap, No matter what the channel is capable, it only recieves what is allowed in
    Rate by the Baud and Bit rate sequence, thereby inflating the Value in disquise on some lines of Data as the measured Data INCLUDES these Data Transmission, appearing as Distance, but is nothing more that Data in the structure of that Streaming Transmission.

    The Chaotic Cat!!

  18. @ Bobby In re: @ MuOne

    Bobby: Will has a great point on that issue, and I would take it under advisement and see where it fits into line, as I already have done that on this issue, Im saying “its feasible and founding” along with the function operations of the SDU/SCU.

    @FelineNut The Chaotic cat!

  19. Got back home from ranch sooner than expected. Storm was actually pretty mild. Dug up the Xmas island scenario. Been awhile since I even looked at it. It was simply an end point feasibility quick look. I send the spreadsheet as is without proofing since creating it some time ago. It does answer your question about arc crossing, speed, heading, etc.

    See link below:


    Sorry, too cheap for dropbox.

    A little explanation since I organize things a bit differently than other stuff you have seen. The upper part of the chart is comprised of data/analytics that would be true for any route – BTO rings, times, and various Doppler calculations unrelated to the flight path. Headings are pretty self-explanatory (but described below).

    “SV->GS” is the Doppler produced by satellite motion relative to the ground station at Perth.

    “BFO off” is the BFO offset I calculated months ago from the KL pre-takeoff data. Not sure what the latest thinking or refinement is.

    “Table 4” refers to the Perth corrections supplied in the ATSB reports. Page 57 of 13 August report.

    “Dop Resid” is the difference between the measured BFO and the rest of the values already described. This residual must be accounted for by any valid set f aircraft dynamics.

    The bottom part of the table is the data for the last two arc crossings (nominal positions, heading, speed, ROC, and altitude). Apparently, I got bored after that. The heading change at the 00:11 arc is for a North turn toward the island from the Eastern heading from the 22:40 ring.

    The take-away is that aircraft Doppler is quite sensitive to ROC changes and adjustments can be made to support just about any non_AP flight mode.

    Column headings in the lower chart area are for the AES Doppler Correction, the actual aircraft to satellite Doppler, and for the Doppler produced by SV motion relative to the aircraft.

    Speed and altitude were kept low consistent with the hypothesized motive to avoid radar detection.

  20. @MuOne,


    I discussed the 1 Hz BFO quantization effect in my previous post at:


    The BFOs during the two satellite phone calls have sawteeth with amplitudes significantly larger than 1 Hz. Therefore the 1 Hz quantization is certainly present but does not dominate the BFO variations.

    With respect to your statements:

    “1. Both the saw tooth frequency and slope are accurate measures of the instantaneous rate of change of doppler.

    2. The slope is the residual of the continuous actual doppler change.

    3. The higher the slope and frequency, the higher the rate of change of the doppler contributors.”

    I agree with all these statements. Item #3 is the same as #1.

    You also said “This of course would indicate that the 18:40 saw tooth is indeed an indication of a large rate of change of something, possibly the heading change and associated altitude change during a turn.”

    I disagree with the notion that the sawteeth at 18:40 indicate a “heading change and associated altitude change during a turn.” First, if there was altitude change, there would be significant vertical speed and unmodelled ROC Doppler in the data. There is no clear evidence of this. Thus it is more likely that the altitude was steady at 18:40. Second, if there were large bearing changes (say 35 degrees during the one minute of data), the average BFO would drop by about 10 Hz (see Victor’s post). Obviously there is no evidence of this large slope in the BFO data. Therefore there is no clear evidence of an ongoing turn at 18:40. In addition, the mere presence of the sawteeth in BFO data does not imply a turn. Straight and level flight will also produce sawteeth. The only requirement is that the aircraft be moving. When it moves, the position changes, and the calculated frequency compensation is also continually changing. Since the sawtooth period does not appear to be a constant, asynchronous timing cannot be the only cause, although it could be a minor contributor.

  21. @Bobby @ Will

    While I agree with Bobby, and understanding Will’s analogy…

    It’s the Triggers that interest me… {Triggers}

    ….and my attention as well as Where MH370 nwas actually at during the Local TZ in relation to the UTC comingnled and intertwined UTC data with The Actually Path inside the Local TZ corresponding with that UTC correctly….

    for Example……

    2.2 Emergency Detection Criteria
    The group looked into what criteria would detect a maximum number of accidents and
    incidents with the greatest warning times, while at the same time limiting nuisance triggering for normal flights. The criteria developed by the WG use a limited set of parameters (see appendix 3 for the list of parameters). Many more parameters would be available for actual aircraft systems that might be available in the future.

    2.2.1 Binary logic approach
    The BEA elaborated the following set of criteria, based on an estimation of what constitutes an emergency situation. The approach is binary in the sense that a condition is either true or false. If one condition is true, then it is considered that an emergency has started. Otherwise,
    if all conditions are false, then the flight is considered normal.

    Excessive Bank {|Roll|>50°} OR
    {|Roll|>45°AND|Roll rate|>10°/s} 2 sec
    Unusual attitude
    Excessive Pitch
    {Pitch>30°} OR {Pitch20° AND Pitch rate>3°/s} OR
    {Pitch<-15° AND Pitch rate<-3°/s} 2 sec

    STALL STALL Warning=TRUE 1 sec
    Low CAS
    (*) 60 kt for DHC-6
    2 sec
    Vertical speed
    {|V/S|>9000 ft/min}
    2 sec
    {IAS>400kt} OR {OVERSPEED Warning =
    TRUE AND Alt2.6g OR nz0.25g} 2 sec
    Excessive roll
    {|Captain Roll cmd|>50 OR |F/O Roll
    cmd|>50 } AND {IAS>80 kt} Control 2 sec
    inputs Excessive use of
    {|Rudder position|>6° AND IAS>240 kt} 2 sec
    TAWS warning TAWS warning/alert = TRUE 1 sec
    Too low altitude
    (poor altitude
    gain after takeoff)
    {40<Radio Altitude80%
    AND Eng2N1>80% } 10 sec
    TCAS TCAS RA = TRUE 1 sec
    Others Cabin Altitude
    The number of times each criteria is true when run against the 68 accidents and incidents is
    as follows:

    6 Number of consecutive seconds for which the condition has to be true…

    @FelineNut The Chaotic Cat!!!

  22. @GuardedDon,

    You are correct. The contribution to the BFO caused by Rate Of Climb will only suffer the 1 Hz quantization at the GES. It will not have the sawtooth function applied to it (i.e., it will not be “modulated”).

    If we had a bunch of MH370 BFO values close together in time when a climb or descent took place, it might be possible to separate the “unmodulated” ROC contribution from the “modulated” component produced by the aircraft electronics. Of course, the BFO components due to satellite velocity (toward the aircraft and toward the ground station) are also not “modulated”.

    There is only one BFO component that is calculated in real time in the AES, and that is the component of the aircraft horizontal velocity that is toward the nominal satellite position over the equator. This is the component of the total measured BFO that is “modulated” by the lack of precision in value and possibly timing to create the sawteeth.

    I believe the shape of this modulation waveform would be a classical sawtooth if the cause is purely a lack of precision. More complex waveforms could result from a combination of precision and timing errors. However, I can’t think of any error of this general type that would produce a square wave in BFO.

  23. Is the much that can be found out about the Inmarsat data just before and-or after IGIRI / BITOD and the last big turn. More or less wondering if it gives out any hidden clues that could be of use later in determining which way mh370 went … And such.

  24. @Bobby,

    Thanks for your response.

    I overlooked the 4Hz amplitude when responding previously, which somewhat negates my proposed 1Hz jump model.

    Your 18:40 graph is certainly intriguing and appears to be a close match to what is actually going on.

    Your 23:14 saw tooth graph, does not match quite as well. However my eyeballing/gut instinct also sees a near 4 Hz amplitude in that plot. If you were to reduce the slope of the saw tooth and superpose a non-linear mean, e.g. a kind of rise than fall as per a region around a local maximum of a function graph (say a sinusoid or inverted parabola), the 23:14 BFO saw tooth may well fit such a 4 Hz amplitude saw tooth overlaid on some other function.

    One thought I had was, maybe the frequency correction algorithm is not stepped at the resolution level (1 Hz), but may apply a check against a maximum allowable error (e.g. 4Hz) before adjusting frequency to the next best fit to maintain centrality to the permissible frequency band.

    Whatever the case may be, i have the feeling that your and Gysbreght’s discovery of saw tooth behaviour may be very significant and provide a new avenue of extracting information from the sat log.

    It may well be worth analysing the BFO at the gate and on the runway to see what saw tooth if any are present there.

    One thing, I didn’t quite follow you on was your assertion, that the only thing required for a saw tooth effect is that the airplane is moving. Why does a straight level flight produce a saw tooth and why would the frequency correction noticably change? I would have thought that straight level flight would have minimal doppler change, i.e BFO change, over a short time interval.


  25. @2424giger posted December 14, 2014 at 2:11 AM :

    “Any LNAV controlkled eg. following a series of waypoints, is executed with standard rate turn= 180 degrees/min”

    Rubbish – see B777 Operations Manual:

    Rotate – sets the AFDS commanded bank limit when in the heading select (HDG
    SEL) or track select (TRK SEL) roll mode as follows:
    • AUTO – varies between 15 – 25 degrees, depending on TAS
    • 5, 10, 15, 20 or 25 – the selected value is the maximum, regardless of
    airspeed. (B777 Operations Manual, 4.10.10, Bank Limit Selector on MCP)

    The FMC computes holding patterns with constant radius turns based on current
    wind and FMC commanded airspeed. The pattern size is limited to FAA or ICAO
    protected airspace. In LNAV, the AFDS tracks the holding pattern using up to a 30
    degree bank angle. (B777 Operations Manual, 11.43.26, Holding).

    LNAV BANK ANGLE LIMITED – before entering or while flying a curved path
    or holding pattern, the FMC predicts the LNAV roll command will be limited by
    thrust or buffet based roll limits. (B777 Operations Manual, 11.60.3, FMC Alerting Messages)

  26. @2424giger:
    I wonder if you have tried to calculate the bank angle required for a B777 to execute a standard rate turn while flying at say Mach 0.82. I think you might be surprised.

    In fact the FMC will determine an appropriate bank angle for the turn required while in LNAV, and it will not always be the same.

  27. @MuOne,


    The reason you will see sawteeth in straight and level flight is because the calculated Doppler frequency compensation toward the satellite is constantly changing as the aircraft position changes. Even with no change in speed or bearing, at ~18:40 the horizontal aircraft velocity compensation term w.r.t. the satellite is roughly -300 Hz and rising rapidly toward zero as the path crosses the equator, where the aircraft is moving perpendicular to the line to the satellite (and the aircraft Doppler then crosses zero and becomes negative (= + compensation correction) for the remainder of the flight). Recall that the compensation term is defined as additive, so velocity toward the satellite, which produces a positive Doppler shift (higher frequency), results in a negative compensation term (so the sum of the actual Doppler shift and the calculated transmitter shift is nearly zero at the ground station).

    At 18:40 the AES compensation term is about -300 Hz, at 23:14 it is about +600 Hz, and at 17:07 it is about +500 Hz. At 17:07 the sawtooth amplitude and period are unclear. There are 3 channels involved, and even after applying the FFBs for each channel suggested by Mike Exner, the pattern is not clear to me. It does appear that the data are affected by some kind of waveform with amplitude roughly 4-5 Hz in amplitude. This shows up more in Channel T8 than in the others. A square wave does not seem to reduce the residuals at all. A sawtooth is better but still not very satisfying for this data set.

    Here is a summary table of time, approximate sawtooth amplitude, approximate sawtooth period, and the amplitude of the AES compensation term:

    18:40 4 Hz 3.3 s -300 Hz comp.

    17:07 4-5 Hz? ~24 s? +500 Hz comp.

    23:14 7 Hz ~23 s +600 Hz comp.

    Looking at these very rough estimated parameters, there may be a correlation between the magnitude of the compensation term and the magnitude of the sawtooth. That is the behavior one would expect if the cause is limited precision in trig look-up tables in the AES. This happens because you have a vector calculation of projected magnitude that involves a number times a trig function (sine, cosine). If look-up tables are used, the steps in value are likely to be fairly uniform. The sawtooth magnitudes we see are about 1% of the magnitude of the compensation term. You would expect that if you had look up tables with 1% steps. The actual algorithm is much more complex, but this example demonstrates the principle.

    In general, I would expect to see the sawtooth period to be quite steady over 1-minute periods if there is only straight and level flight. Turns would cause the sawtooth period to fluctuate fairly rapidly. In other words, turns will frequency-modulate the BFO data. The lack of sawtooth frequency modulation in the 18:40 data set demonstrates that a turn was not in progress.

    I tried your idea on the 23:14 data for adding a parabolic baseline with the sawtooth added on. Here is what that looks like:


    It is definitely a better fit, but there are 2 additional parameters. All residuals are within ~1.3 Hz. The sawtooth amplitude is 6.5 Hz. The data are sparse, so it is unclear to me exactly how meaningful the parabola is. It certainly fits better.

    Your idea about quantization to 4 Hz steps (at the end of the AES algorithm) is intriguing but would be inconsistent with the variable sawtooth amplitudes that we MAY be seeing in different data sets.

  28. @ Dr Bobby

    My guess is the frequency steps you are seeing result from the way the variable frequency is implemented, numerically controlled oscillator (NCO), in the mobile terminal. It does not have to be very fine for communication purposes. That is also most likely the source of the Inmarsat disclaimer. Google NCO for a description.

  29. @Rand

    re being a pilot:

    And it is apparent to me the vast majority of commentators here are not and never have been pilots and do not understand the levels of discipline and self discipline on a flight deck.

    What is preposterous is the suggestion that Zaharie got out of his seat at 1:06am walked back to the floor hatch to the Avionics Bay and lowered himself below to disconnect the ACARS system.

    First Officer Fariq Hamid spoke to Lumpur Control at 1:07:55 (virtually 1:08am.

    So all you people who subscribe to pilot suicide suggest the co-pilot witnessed this bizarre behaviour and said nothing at 1:08am and then 9 minutes later at 1:19am with Zaharie allegedly still below in the avionics bay still said nothing during the final radio communication.

    That is not credible.

    Here is what is credible. The ACARS, VHF radio and Transponder are all controlled by the same Rockwell Collins CMU-900 transponder and we know that it was giving ACARS signals with a doppler shift equivalent to a speed of 80-90 knots whilst the aircraft was parked at the gate.

    Only overheating of the AFC oscillator would explain this.

    Once MH370 decompressed and that is the conclusion of the ATSB, then temperatures in the cabin would reduce to -53 degC and therefore the ACARS would have returned to normal behaviour.

    These wide fluctuations in temperature would also influence the ACARS signal delay by altering the AFC oscillator frequency.

    A difference of 1Hz = variation of 300km. Nobody knows the actual signal delays. These had to be estimated from an average of 17 different signals prior to take off.

    That is why the currently accepted BTO ping rings are totally unreliable.

    They will not find MH370 on the seabed in the current search because the calculation of the 7th Arc is totally flawed and ignores the effect of these temperature fluctuations in the avionics bay.

    Instead of chasing satellite calculations they should have started with the large debris field sighted by satellite drifting from 44S,90E to 46S,89E (20-24 March) and reverse calculated the debris drift pattern.

    None of these debris were ever recovered or examined at the specific request of Malaysian Authorities.

  30. Given that vessels engaged in the Seabed survey along the so called Southern Arc not only towed a sidescan sonar, but also a hydrocarbon sniffer, the sniffer would have detected some indication by now but hasn’t.

    So my question now to those who have insisted their satellite based calculations can’t be wrong, time to practice your excuses…

    I am all ears how you will explain the failure to find MH370 where you have predicted?

  31. @ Bobby


    Hey Bob, Please DM me in Twitter when you have a few extra minutes…

    I have something of interest for you…

    Im already following you, so when You can…



    The Chaotic Cat!!!

  32. @Simon Gunson

    Your defense of ‘pilots’ (for that it what it most certainly is) is particularly disingenuous.

    The scenario that you depict (knowing indeed that YOUR scenario does seem preposterous) is NOT the favored scenario by many who have an absolute conviction of Zahrie’s culpability. Nice try. There are multiple other scenarios with a high degree of plausibleness that you of course fail to mention…Sigh. Shame on you.

    Furthermore, your post above is riddled with misstatements (I’m being kind here. They are really gross misrepresentations in an effort to spin an ‘accident’ scenario). So as to enlighten you, some examples of your spin:

    You claim the ATSB has CONCLUDED a decompression took place. NO, THEY HAVE DONE NOTHING OF THE SORT.

    What they did do was use an ASSUMPTION (NOT a CONCLUSION) of a hypoxic event, while STILL allowing for the possibility of control inputs. Nice try.

    You offer that Fariq Hamid spoke to KLATCC at 17:07:55. Yet, you have no way of knowing this. The FACT of the matter is that you don’t have a clue as to who’s voice made that transmission. What IS interesting about that transmission is that it is a redundant FL350 confirmation, but of course you fail to mention this peculiarity.

    In MY narrative, this redundant transmission id indeed made by Fariq, who is by now keenly aware that he has a BIG problem sitting in the seat next to him…he is trying in the most subtle way possible to alert KLATCC to looming disaster.

    Your entire post is just spin and convolution. You don’t control the narrative.

  33. @ spencer..

    I like that, Not that I am crushing the Pilots experience, I must say 1 thing and many pilots argued many points to deaths end on thier own behalf, having me prove it to them, so I did on many functions, however I didnt expect the pilot to know or understand those parts, although they stressed many normal operations, I showed what the actual behavior was in numerous adverse behaviors, so lets not smash the pilots, lets show them so they understand too….

    For example… Turning the Transponder Switch OFF…

    OFF=Standby and the TCAS & Altimeter also is Standby, NEVER OFF, so a call sign may dissappear but the plane doesnt and the shadow would NEVER shadow in a Civillian Aitcraft as OFF=Standby and all redundant Behavior Over-rides and prevents that collision or Terrian close-up intentionally..

    As for ATSB and decompression…. Let me give you a Hint… The Data Logs have an Eviromental Hazard Alert Code embedded within, and that includes Fire~Tempature~Decompression~All Enviromental Scenerios under Context, so it may very well be a part of the incident…

    Just saying, and this has been long known…

    okay.. thanks for the good writes above, as your analysis is warranted and needed… Thx again..

    The Chaotic Cat!


  34. @ all

    While I thought Simon’s remarks were a bit harsh, I think the science and math community (of which I am a part) has some serious egg on their faces right now.

    1> Chasing the sonar pings was incredibly stupid.

    2> BFO is very unreliable. Anyone with practical real world experience would know that holding a chain of three Xtal oscillators, mobile terminal, satellite, and Perth to better than a few Hz over several hours is simply unrealistic in the extreme. A reasonable expectation for accuracy is on the order of +/-5hz, and perhaps as much as 1000 nautical miles. At best, BFO can be used to rule out impossible places, not as a guide for probable places. The plane really could be anywhere on the 7th ping arc limited only by fuel range if one does not invoke fixed AP mode assumptions.

    3> The technical community has not really come together on this issue. We have teams working independently, not sharing data, sniping at each other’s work. Is it any wonder that an outsider views it as comical and evidence that we are not as smart as we think we are (which is almost certainly true).

  35. 1> Chasing the sonar pings was incredibly stupid.
    I see this sentiment a lot, and apparently people hold this belief because the frequency was low by about 10%. But the president of the company which manufactured the devices stated that the frequency could be off by this amount. The reason, I believe, is that these pingers are designed primarily to make a loud signal very efficiently and place little importance on the frequency. In consequnce, the piezo-ceramic sonic transducer is itself the resonant component, and its resonant frequency will be affected by anything which increases the effective mass of the resonator, e.g a glob of mud, or a piece of metal leaning up against it.

    What else should the team have been doing with their sonar equipment?

  36. Simon,

    When one apparently quotes facts it’s reasonable for the reader to expect that the facts are correct.
    However, the B777 does not employ a Rockwell Collins CMU-900: the data comms function is performed in its (Honeywell) AIMS system. The VHF radios & transponders are entirely separate Line Replaceable Units.
    See http://bit.ly/1moDFtW

  37. @ Dave

    Simply not possible. The most inexpensive oscillator imaginable could not possibly be pulled that far off frequency by any combination of stresses – temperature, pressure, aging,.. or battery voltage. It was just plain dumb.

  38. I’m not referring to stress. Instead think of the mass of the resonator: f=1/(2*pi)*sqrt(k/m), where k is the spring constant, and m is the oscillating mass. Adding mass to the exposed resonator won’t change its k, but it does change m. Needs about a 20% increase in effective mass to get a 10% reduction in frequency. The ceramic tube is around 100 grams in weight, I believe, so a little less than 1 ounce of mud would be enough.

  39. @David Reed,

    Mud is not stiff enough to resonate coherently with the quartz oscillator at ~30-35 kHz. It will damp the amplitude of the response, but it won’t shift the resonant frequency noticeably.

    To shift the resonant frequency by 3 kHz, one would need to effectively glue a piece very stiff material to the exposed oscillator face.

  40. @ Bobby @ David @ Dennis…..

    While I agree with the analogy of Bob & Dennis, Dave has a much deeper unstated point..

    Referene USN Studies for past 90 Years & AUS Study of Indian Ocean about 80 Miles from Ocean Shield SAR 2 NW & Same Bottom Seabed..

    As for the AUS Study of The Idian Ocean in similar Depths and Nearby Enviroment, The study showed the unexpected from all studies elsewhere, being….

    It that General Location and Similar waters in the IOR, the signal may be Directly Below you, however, unless you are directly on top of that signal at the bottom ot just dropping the reciever in the straight above the transmitter, showed that the signal waves will arc up to 450 miles away at its Radial Center point of the arc propagation and bending Back topside above the transmitter in shallow depths, so the propagation there is different, by a long shot and still being understood…

    As for the USN studies, Its long been known that signal recieved in the manner that Ocean Shield had recovered for propagation Shall Not be EQ’ed for signal clarity and propagation, as the area is Not surveyed and the actions of the signal from the terrian below it needed in order to EQ the signal data and when they did so, It kept reducing the Singal Freq. & which is normal, but NOT to be done in cases such as the manner it was collected in MH370, therby making your Data Collecting Worthless as your underwater propagation is no more accurate than The velocity still misunderstood, and I have offered to share that algarithom to Ulich so he may calculate the limited Velocity Shift in Data time sent, accuratley, however, Everyone seems to know it all, just like AUS didnt look at there own Study and the USN went against thier Studies, as Everyone is eager to creat a new math, and the Math has been part of our foundational structure for years, and Like Proffessor DS would say, do your homework, but tell me every thing from your studies so I may write a paper on your your work and you will get credit from DS, by lose your deserved credit and reference from your reports.

    Anyways, The Simplicity Factor, KISS ~ Keep It Simple Stupid~ and just get chaotic sometimes, or the chaos will get you….


    The Chaotic Cat….. ;-)>

    {needing HD Structure Sat Scan for SAR Zone(s)}

  41. @ Dave Reed:

    A pinger frequency is controlled by a very small piezo ceramic device mounted on a circuit board with a few other components. The board is potted in some resin like material and then sealed inside the aluminium which constitutes the external housing. So it is the stability of the oscillator controlled by the piezo device which is important. Such devices are stable to within a few Hz at the operating frequency of 33Khz.

    The external aluminium tube becomes part of the sounding device to transmit the pulsed 33Khz into the water. Any mass attached to the outer casing may affect the amplitude of the signal, but not the frequency.

    A Google search will show a pinger being dismantled to illustrate the construction and the components.

  42. You lost me when you posted the picture of the freshly bulldozed area at Baikonur. I found the spot on Google Earth, albeit a year earlier (Oct 2012) and my issue is how to get a honking big 777 from the runway to that spot. Not that it couldn’t be done but there was only an hour and a half until daylight. That and no real motive.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.