Caution on recovering data from MH370 FDR, CVR

With the news that the Malaysian authorities announced that Malaysian Airlines flight MH370 ended in the Southern Indian Ocean, and the US Navy  is sending a “pinger locator” there to look for the black boxes, we need to raise some caution about assumptions that these will reveal all there is to know about what happened on the flight.


The flight data recorder (FDR) and cockpit voice recorder (CVR) are located in the rear of the airplane, inaccessible to the cockpit or cabin, a Boeing 777 instructor tells us. Unlike the Boeing 737 in which a rogue pilot turned off these devices before plunging the airplane in a suicide dive, the 777’s FDR and CVR power controls are only accessible in the electronics bay and the access is through a floor panel outside the cockpit, in the cabin of the aircraft.


Assuming the FDR and CVR, therefore, were operational right up until the time of the crash of the airplane, there should be data recoverable if these units are eventually found. The FDR, being digital, has a 24 hour capacity and should provide a wealth of information. The CVR has only a two hour capacity and may yield much less, however. Clearly, it won’t reveal anything that happened over the Gulf of Thailand—this will have been overwritten by the end of MH370. But whether there is anything to be revealed on the last two hours for the flight is going to be uncertain.


In the US, by law the cockpit conversations recordings are only in 30 minute increments-the most recent 30 minutes. If this practice is true for other countries, including Malaysia, anything said in the cockpit as to what transpired when the plane originally was “lost” while still over the Gulf of Thailand will be lost to history. But the final 30 minutes of cockpit conversation, and any noise from the cabin within “earshot” of the cockpit microphones, should be retained on the CVR. But also in the US, pilots have the ability to erase the CVR once at the gate—and it’s certainly possible this occurred before MH370 went into the ocean.


Given the success, albeit two years after the crash, investigators had in recovering the FDR and CVR of Air France Flight 447 (the one that crashed into the South Atlantic in 2009, with main wreck recovered from around 12,000 ft), we feel reasonably confident MH370 will eventually be found and the recorders recovered. But manage your expectations about what might be found on the recorders.

31 Comments on “Caution on recovering data from MH370 FDR, CVR

  1. AMEN !! … Now how to get some of the above FACTS and DATA to responsible media to hopefully damp down the out of control theories like time warps, worm holes, black holes, green men, and linking of a few dozen LOW probability theories . . . based on model airplanes, mis information on AF447, etc…

  2. Beware with these “pinger locators”. On AF447 the TPL-25 was used and they didn’t find any signal. In cases like these side-scan sonar is a better choice (AF447 or SA295 i.e.).
    (I think) it is not possible to erase the CVR unless WOW is on ground and parking brake ON so last minutes of audio should be available.

  3. With the technology available today, the CVR should record the entire flight and not just the last two hours/30 minutes. Also, cameras should finally be put in the cockpit. It’s crazy that the pilots union has been fighting this for years.

      • If the pilot was suicidal, then why all the effort turn off transponders? Why were there no cell or sky phone calls from the passengers? What happened to the co-pilot? The disposable phone bought with a fake ID indicates to me a criminal or terrorist involvement of some sort.

        • Like I have explained in a previous post the suicide could be linked with a terrorist act. Few suicides are associated with a terrorist act, but many terrorists acts are the result of suicidal action. I recognize that I don’t have much to support my theory; it is more an inner feeling than anything else.

        • How about trying to prevent shame on the family, life insurance policies? All good reasons to go the length to obscure evidence of suicide.

  4. Seems there are two distinct discussion going on here, one concerning the aircraft itself (which is a huge mystery still), and the other regarding the recording devices.

    In terms of recording, seems that the FDR is still stuck in the last century, when recording a lot of data took up lots of memory that was difficult to store.
    Surely one fairly simple device (and a minimum) would be a GPS signal sender. They have devices that sends out pings when in water, surely they can get the FDR to send out it’s position (if known) if it hits water, or exceeds a certain G-level for instance (and/or other event). I heartily concur on the full flight voice recording, should be relatively easy, and bearing in mind they have been needed in a couple of instances in the past few years, would give great insight into goings on.

    • In terms of recording, seems that the FDR is still stuck in the last century, when recording a lot of data took up lots of memory that was difficult to store.

      24 hours’ worth of data isn’t all that shabby, I would say.
      Now – the CVR is a different story. I do understand privacy concerns by pilots, and I do think that there should be an obligatory erasure X hours after a flight has successfully landed, but I would also say that its recording capacity should be much more than 2hrs/30mins. 24h seems reasonable, to be honest.

      • I do understand privacy concerns by pilots

        Why should the pilots have any rights to privacy? Come on now, people in working in retail stores are videotaped all the time. Why not tape what a pilot is doing when hundreds of lives are on the line?

        • Why should the pilots have any rights to privacy?

          Because 99.9% of the time, there’s no hijacking, and no rogue pilot action, but they’re still being taped for later analysis/review.

          And once you do start recording something, you have to keep a very close lid on the circumstances under which these recordings can be used, otherwise the main use additional CVR info will see is for 24/7 surveillance of pilot’s workspace and actions.
          That’s against labour laws in a lot of countries, and there are good reasons for this.

          Also, if you want to carry your argument further, everybody should be taped/monitored 24/7 at their workplaces, from the cleaning lady to the CEO, to make sure that nobody does any wrong. If you take it even further (I mean – where would you stop?), you have to bug every home because people could be plotting homicides in their own homes.
          As much as I disagree with some of the decisions some people make in their private or work lives, I don’t think that mass-surveillance of everybody all the time is the way forward. That’s more like a dystopian vision to me.

          Mind you, I *am* in favour of extending the recording period of CVRs, but there have to be very strict rules under which these can actually be used.

        • If the desired “live flight data streaming” is achieved simply by re-programming the existing ACARS uplink protocols to expand in sequence and scope (as has been suggested is a cost-effective avenue ?) then – ipso facto – two of the specifications itemized as ‘musts’ for the new system will not be achieved : (a) exemption from tampering and (2) prevention from routine distribution to operator management … ?

          Conclusion : what we are seeking is NOT a modified ACARS !

    • There are several active aircraft tracking systems available on the market that use the “bread-crumb approach” to SAR. Rather than relying on an emergency locator transmitter to transmit upon impact, the next generation of emergency locating devices are active tracking devices that send position reports at regular time intervals. If the unit stops transmitting upon impact, the historical transmissions will give the last known location of the aircraft, its speed, direction and altitude. Tracking as an alternative or complement to current technology has recently been encouraged by the Coroner in New Zealand.

  5. Sadly, what many had feared and suspected all along (including yours truly) has been all but confirmed thanks to pioneering analysis carried out here in Britain. Britain’s Inmarsat working alongside Britain’s air safety agency (AAIB) used “a type of analysis never before used in an investigation of this sort”.

    What I find completely unacceptable in this day and age, is that we may never really know what actually happened to flight MH370. Even if the FDR and CVR are recovered (and that may take an awful long time), significant amounts of data have been lost forever and that makes a mockery of the available technology. Changes are surely due now for the sake and reputation of the aviation industry and the safety of the flying public.

    • @ Gustiewind;

      Don’t forget to thank Boeing as well. After all an Inmarsat VP said Boeing had contributed and of course built the satellites for Inmarsat. 😉

      • Inmarsat-3 series are based on the Lockheed Martin Astro Space Series 4000 sat bus.
        Inmarsat-4 are based on the EADS Astrium Eurostar-3000GM sat bus
        The payloads appear to be from Matra Marconi Space.

    • Gustiewind, when the DFDR and DCVR from AF-447, after about two years under 12,000′ of salt water, they still had all their recorded data within their memories. That A-330 had modern recorders, just like this B-777 does.
      But the challenge is finding them first. The pingers are good for about 30 days, and it has already been 16 days since the airplane went missing. Challengers to finding the wreckage will include the depth of the water where the wreckage came to rest, water density and salinity in the ‘water column’, number and amounts of temperature differences within the water column, undersea mountains and valleys, etc.

      • A challenge is also the amount of data the recorders can hold – particularly the CVR. The FDR should be fine regarding the period covered. But it’ll only show you how the plane flew, and whether it was controlled by auto-pilot or by somebody in the cockpit.
        But – assuming the most significant control inputs came from the cockpit – there’ll hardly be any hint on the FDR regarding why the course was altered, and by whom.

        I would concur with Scott’s earlier analysis that a technical malfunction that incapacitated both pilots is only a very remote possibility. Which means it would be all the more essential to find out who brought the plane off course and why.

  6. I am sure any changes to the CVR & FDR require massive amount of certification to change, if it is simple to change the memory chip for the CVR to get say 2 hours or 4 hours of conversation without a huge expense that would be nice.

    I would think a GPS location & altitude pinged out every minute along with the most vital aircraft data that happens anytime the engines are running (say 10 min past shutdown) would provide location of the plane always. Separately, when better communications are available a full download of the CVR and FDR on a burst mode . OK so the CVR tapes are not used except in an accident.

  7. As in the aftermath of the AF447, I’m seeing a few calls in the industry for deployable black boxes.

    Here’s an article dated 2009 regarding the matter, noting that the technology existed even back then.

    “These deployable recorders, first developed in Canada, don’t replace the existing black boxes. Instead, they serve as a back-ups. If the main boxes can’t be recovered, these would still be available.”

    5 years on, I’m surprised that not a lot of progress has been made on this matter and unfortunately, we now face a recovery operation worse than that of AF447.

    Considering the fact that, going forward from now, there are going to be more and more long open water flights far from land, there should be a great sense of urgency now for the industry to improve its communications and tracking systems.

  8. Regarding erasure of the CVR in flight, the park brake has to be set in order to erase it. To be honest I never tried setting it in flight but I’m pretty sure that wouldn’t work.

  9. Whilst MH-370 was struggling with its fate somewhere in the Indian Ocean in the night 7th to 8th March, the invasion of Russian troops into Ukraine’s Krimea had commenced :

    Interesting parallel, pure coincidence ?
    It would be pertinent to fact-find the “Breaking News” alerts released early on March 8th re MH-370 from official Russian news agencies : exact timing, scope … plus Media echoes from allegied agencies in other parts of the world ?

      • Plausible script for a Hollywood “Mission Impossible” or far-fetched/wild speculation ?

        The word “parallel” – I beg your indulgence – was misleadingly used to express “more or less concommittantly” … as if the events on one side conveniently fitted in time-wise with the action on the other size, ie intended as a “DIVERSION” for Media and state agencies to look “elsewhere” whilst Putin is making some personal arrangements for a Dacha in Crimea ?

    • Mr Motts – I note your analysis was on the 18th- and at that time- other data re doppler effects and a detailed analysis by Immersat folks plus others pretty well cancel out most of your suppostions. It then is a puzzle as to why – one week later- you have elected to post you analysis .

      Even so a few major flaws are evident.

      1) : ELT do not always work – and do not work under water.
      2) For your and other similar analysis to be probable requires a significant linking of ***Many *** low probability events at critical times. that does NOT mean they absolutely could NOT have happened, but given the tracking for 5 to 7 hours should be evidence that a gross electrical failure you postulate is a non starter.
      3) Trim tabs are normally used to move large control surfaces such that manual- cable controls do work- which the 777 does have for backup. I’m sure you would be amazed at the very low power requirements ( electrical or hydraulic ) needed to move the horizontal stabilizer, even in flight.

      4) There are about a dozen other issues – for which while I am generally aware of- do not have the expertise to discuss in detail.

      5) Why not leave it to those who do have or can acquire the facts and data to rule out most other scenarios ?

  10. As an assist for many who read this blog- I will post a link to another normally respectable blog which has been somewhat overloaded with partially reasonable theories- but which also has some professionals posting . . . In this case, the particular post to look for is by


    Join Date: Mar 2014
    Location: US
    Posts: 2
    And STARTS as ” A few thoughts from a man who designed bits and pieces of the missing airplane
    Let me share a few thoughts from a man who designed bits and pieces of the missing airplane, and probably bits and pieces of half the jets you folks fly on.

    You should be highly suspicious of these stories about fires. Have any of you folks ever seen a cargo compartment smoke test? Modern jetliners will detect a burnt napkin in a space the size of a living room in under 3 minutes.

    You should be highly suspicious of stories of large volumes of smoke propagating out of the cargo compartment. That’s because after we supplier types detect a teensy puff of smoke in all that big space, the airframer types goes back and flood it with smoke so dense you can hardly see and makes sure not one bit of smoke comes up into the passenger compartment.

    Those folks at Hamilton aren’t sitting still, neither. Once we find smoke, they turn off the air conditioning fans and turn up the packs to keep smoke downstairs. . . .

    +++ It is quite comprehensive re the fire and electrical issues postulated

    The current page number is 409 . . . which may change

Leave a Reply

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