Saturday, April 12, 2014

Sonar technology used in the Malaysia Airlines flight 370 search

The Southern Indian Ocean is a little traveled location known for it's atrocious weather. The probable search area epitomizes the phrase "in the middle of nowhere" and it's hard to imagine a worse place to look for a missing aircraft.

If the Malaysia Airline's Boeing 777 is indeed at the current search location this will be the second deepest effort attempted to recover the flight, and cockpit voice data recorders on record. The deepest recovery to date was South African Airway's flight 295 cockpit voice recorder at a depth of 16,000 feet in 1988.

The second deepest to date is the 2009 Air France's flight 447 crash whose cockpit voice, and flight data recorders were brought to the surface from approximately 13,000 feet two years later. The underwater locating beacons had long failed before these two aircraft were found. The current Malaysia Airlines flight 370 search area is about 14,000 feet deep. 

Dukane Seacom DK-120 locator beacon

The first thing to do is to sort out all the vocabulary. The term "Pinger" is used by the media to describe an underwater locator beacon (ULB). The pictured DK-120 is manufactured by Dukane Seacom Inc., a division of Sarasota Florida based Radiant Power Corp.

There were two Dukane ULB's on the Malaysia aircraft. One was attached to the flight data recorder, and the other to the cockpit voice recorder.

A "Ping" is more accurately called a tone burst. This is the noise made by your boat's transducer. A transducer's tone bursts are an acoustic signal at a specific frequency and time period. In the case of DK-120 beacon this is 37.5 kHz with a tone burst duration of 9 milliseconds. This occurs about once a second. This frequency was selected because it's not used by other sonar systems and most marine animals.

To put some perspective on these pieces of equipment let's take two off the shelf fish finding systems. On one of them we are going to disable its ability to send out tone bursts, and let it just listen for them. This is now called a hydrophone. On the other fish finder we are going to only let it send out tone bursts and not listen. This is the underwater location beacon. This is essentially the very same technology you have on your boat today, but using a lower transducer frequency.

So why use a low frequency for the locator beacon? Physics tell us the the lower the frequency, the further a tone burst will travel through the water. This is why your fish finder operating at 50kHz will find the bottom in deeper water when your 200kHz frequency signal cannot.

Well if we can hear the tone bursts from the underwater locating beacon why can't this aircraft be quickly found?

The answer to this is they can't consistently hear the tone bursts for a number of reasons. The first is the distance the tone burst can travel is limited by the lithium battery power it has. This beacon has a range of a half a mile, to a mile and a quarter in good conditions. In excellent water conditions it can be just a bit over three miles. But remember the search site is nearly 3 miles down and the water surface is at the edge of the beacon's range. Water conditions would have to be very good, and surface positioned hydrophones would have to be virtually on top the beacon to hear it.

There are other things going on here also. The data recorders are typically located in the aft end of the aircraft. Although these devices are designed to survive an impact of 3400 gees, you don't want them to be one of the first things to arrive at scene of the accident. For this reason the device's overall survival rates are higher if they are located in the rear of the plane.

Depending on the final underwater location of the beacon, parts of the plane's wreckage or higher portions of the seabed may be shielding or reflecting the signal in a direction that doesn't let the hydrophones hear it. Another issue is thermoclines and other forms of ocean boundary layers can reflect some or all of the sonar signals back to down the seabed, or cause them to travel horizontally parallel to the boundaries and never reach the surface. This is a simplified explanation of complex deep water acoustic issues.

US Navy towed hydrophone array.
One of the ways you can overcome these problems is by using a towed array hydrophone. The vessel Ocean Shield has been towing a US Navy provided hydrophone over the most probable site at depths reported to be around 10,000 feet.

Since you're now almost two miles potentially closer to the beacon it makes a signal acquisition statistically more likely. Your also hopefully underneath any ocean boundary layers. The signals that have been heard are likely from one of the beacons, or possible both and they may not be even close to each other complicating the search efforts.

Australian navy vessel Ocean Shield.
But because you can hear the signal, it doesn't mean you know exactly where it's coming from. So the Ocean Shield is trying to triangulate the the signals position from multiple locations.

In theory this is like playing the game of warmer and colder. If the signal is stronger you're closer (warmer), and if it's weaker you're further away (colder). All of this is becoming more difficult to do because the batteries are likely fading, reducing the signal strength, so the colder you hear now might actually now mean warmer.

Australian navy vessel Ocean Shield's AIS track in the search area.
This process is time consuming. The screen shot above is showing the AIS track of the Ocean Shield making three towed array passes over the search area. The Ocean Shield is towing the array at the time of this writing at a speed of about 2 knots.

What uses even more time is when a pass is completed over the search area it takes several hours to turn around. About two miles of the towed array cable have to be winched in. The vessel has to be turned around and put on the new course. The two miles of cable have to be then be redeployed to put the array back at the right depth.

Sonogbuoys being readied for deployment on an Orion P-3 aircraft. 
Because time is running out, and the search area has shrunk considerably 1008 sonobuoys have been purchased for the search effort. These are primarily deployed from aircraft and they parachute to the surface where they float.

A hydrophone then descends downward from the buoy base on a cable. The data collected is transmitted via a radio signal. With a large number of these deployed if even a few clearly hear the beacon it will go a long way towards getting a fix on the data recorders and the wreckage.

The downside is they aren't as close to the bottom as the towed hydrophone array. The upside is if you have a thousand of them you may dramatically increase the chances of hearing the locator beacon. These are AN/SSQ-53 DIFAR (DIrectional Frequency Analysis Recording) series sonobuoys. Although they don't deploy to great depths, they are capable of determining the direction the signal is coming from if they hear it. The sonobuoys were sold to the Australian navy by Sonobuoy TechSystems and jointly manufactured by Sparton Electronics Florida and Undersea Sensor Systems Inc.  The hydrophones were specifically modified to hear the 37.5kHz locator beacon frequency. One of these sonobuoys dropped from an Australian Orion P-3 aircraft has already heard a possible signal but it turned out not to be the beacon after analysis.

All sonar based technologies that can be effectively applied to track the locator beacon are being used in this search.

With the impending failure of the locator beacon batteries, hopes are diminishing that the exact location of the beacons can be determined. But the likely location of Malaysia Airlines flight 370 has gone from hundreds of thousands of square miles down to a handful of square miles with likely a much smaller area within that's being considered.

If no more tone bursts are heard over the next couple of days, the next step will be to use the Ocean Shield's sidescan sonar equipped Bluefin 21 robotic submarine to start looking on the bottom for the aircraft.

Garmin GVC 10 side scan image.
The side scan chirp technology that will be used by the Bluefin 21 is nearly identical to the systems you can buy from Garmin, Simrad and others. There is little doubt this aircraft will be found, and the black boxes recovered. The world is eager to find out what happened to this most enigmatic flight, and the sonar technology being used to find it can be installed on your boat today.

The photo of the Dukane Seacom ULB was taken by Wikipedia user Meggar.
The photo of the vessel Ocean Shield was taken by Wikipedia user Hpeterswald.
The screen shots of the AIS vessel position reports and tracks are courtesy of
The Sidevu image is courtesy of Garmin Intl.


  1. Another example of a well-informed specialist blogger taking us far beyond what general journalists can reasonably be expected to do. The best overall explanation of the MA370 sensing challenge I've read. Thanks!

    It seems inconceivable that this aircraft and its passengers might end up as the latter-day equivalent of Amelia Earhart, Fred Noonan and Earhart's Lockheed Electra.

  2. Lucid and yet easy to understand. Thanks for this, Bill.

  3. Very well done description. Wish the networks were this good at explaining this. Thank you.

  4. One perplexing issue is why none of approx. 5 ELTs (the aviation equivalent of an EPIRB) on board the aircraft managed to get even one signal to a satellite. They don't broadcast from underwater, suggesting that the plane sunk before they could be activated or that their was not a satellite in view prior to sinkinh. Some of the ELT satellites are geo-stationary, but only in certain parts of the world. It also appears the most ELTs are old generation without integral GPS, instead relying on triangulation to obtain a fix, a time-consuming process. It also appears that they are not designed to float free of the aircraft to the surface, since they are primarily intended for terrestrial use.

  5. Karl, I did some digging. The plane was originally delivered with four ELT's. Only one of the ELT's is physically installed in the 777 aircraft. It's located in the cabin ceiling about amidships with an external antenna located just above it. The other three were associated with rescue gear like life rafts. They operated on internal batteries and are supposed to actuate with when encountering sufficient G forces. Some versions have a water switch. I think there are a couple of scenarios why they didn't send a signal. As you mentioned if the aircraft sank quickly, is a likely one. Another is if the batteries weren't changed they could have lost their charge. The aircraft was twelve years old, and the carrier seems to have less than a pristine maintenance record. The thing I din't talk about was the SOFAR channel. This is a deep water layer that can act as a sound wave guide allowing sound to travel in some cases very long distances. The Ocean Shield had well over 2 1/2 hours of beacon signal acquisition from its towed array. But this doesn't mean the aircraft was necessarily in that immediate vicinity. I suspect though it's in the general area and hence the now expanded search. The Ocean Shield is still in the area, but didn't show any other vessels hanging around with it. The HMS Echo which was also in the area with Ocean Shield is now just offshore from Perth.