Comments about technological history, system fractures, and human resilience from James R. Chiles, the author of Inviting Disaster: Lessons from the Edge of Technology (HarperBusiness 2001; paperback 2002) and The God Machine: From Boomerangs to Black Hawks, the Story of the Helicopter (Random House, 2007, paperback 2008)

Thursday, May 29, 2014

MH370 Had a Satellite Phone: New information

I had mentioned in previous MH370 posts that I'd like to know more about what, if any, satellite phones the Boeing 777-200ER airplane carried, and whether there's any evidence of use during the flight. (A satellite phone looks something like a large cellphone, but it relies on satellites rather than cell towers, so a satellite phone is usable worldwide, whether you're calling from mountaintop, ocean, or desert.) 

This obscure subject never got attention from commentators discussing the many mysteries of the flight; rather, the TV air time focused on ACARS messages, speculation about possible cellphone use by passengers if hijacked, and eventually got around to the satellite pings revealed by Inmarsat. 

Of those, Inmarsat's "hand-shake satellite pings" provide the best info we have about the last hours of flight, if sketchy:

Now, we have some glimmerings that go beyond the ping question. Buried in the otherwise tedious 47-page report from Inmarsat listing many hundreds of signals is a page showing satellite-phone log entries. (For those who have had trouble locating the full ping log, here it is.)

Page 40 shows two attempts from the ground to telephone the aircraft, using Inmarsat's satellite-phone service. The first call to MH370 came at 18:39 Coordinated Universal Time, or UTC, or 2:39 am Malaysia Time, and the second came at 23:13 UTC, or 07:13 am Malaysia Time. 

Here's the page:

What's it mean? Maybe not much, but when the mystery is so opaque, just about anything can be interesting. Even the Wikipedia page on MH370 doesn't reference the satellite-phone info.

A few points:
  • Note that the two calls are shown as "not answered," rather than "terminal unreachable." From a lengthy discussion about the ping log on the DuncanSteel website, that suggests the satellite phone had power from the airplane's main circuits; it rang for an extended period; but no one picked up the phone.
  • I feel confident that the calls were directed at a hard-mounted satellite phone intended for priority uses by the crew, rather than a mobile sat-phone carried by a passenger, or one provided for passenger use by the airline. (Note that Malaysia Airlines does advertise that satellite phones are available for business class travelers on this type of airplane, the Boeing 777-200.) 
  • Finally: we are left to guess that the only sat-phone calls made to or from MH370 are the two attempts cited in the log, but we don't know that for sure.
There's solid information from the ping log that the airplane spent another hour in the air following the second call. The airplane then experienced a temporary power failure as the fuel supply ran out. 

All these bits of evidence lead me to think that after a couple of months' pause, the next deep-diving autonomous-submarine searches will center where they should have been all along, the southern terminus of the most likely flight path (graphic by Reuters):


Saturday, May 17, 2014

Costa Concordia Salvage: Trouble at the Big Dents

Since the successful righting of the hulk Costa Concordia by Titan-Micoperi last fall, it's been said often that the most difficult part is over (photo, Parbuckling Project):

That world-famous parbuckling job was accomplished using strand jacks, cables, anchors, temporary steel supports on the seabed, and giant ballast tanks, called sponsons, mounted on the port side.

Though the most spectacular part is complete, I continue to believe that the work to get it off the rocks and into a breaker's yard in one piece will be very difficult and maybe more so than the parbuckling itself.

I'll explain the reasons below, but first this obscure, yet worrisome news item: Last week an 800-ton sponson tank (numbered S13) intended for the starboard side came loose from its fastenings shortly after being placed. Starboard means right side; that's the side facing the shore of Giglio, which was underwater for months until  the ship rolled upright.  

Sponson S13 had positive buoyancy at the time, and in rising out of control, collided with an adjacent sponson. Divers were underwater at the time but apparently no one was injured. 

The heavy-lift Conquest MB1 crane has since loaded Sponson S13 back onto the pontoon barge MAK and now it's back at Genoa for repairs (crane photo, Concordia Group):

As I explained in previous posts, the two sides of the ship present different engineering challenges. Most of the port-side hull was relatively easy to work during the parbuckling preparations, because it was completely exposed to view, and because it wasn't heavily damaged. At the time, the port was the "uppermost side" of the wreck and riggers could work on it while harnessed to safety lines.

Both sides of the ship need to have sponsons installed so the ship will float enough to get it off the rocks and temporary platform; therefore both sides must serve as supports for these giant tanks.

Even now that parbuckling has set the ship upright, the rusty starboard side is much more difficult to work on than the port side had been. For one thing, the starboard side of the hull lies underwater, so work must be done by divers. (The parts in view above the waterline are the superstructure).

Another reason for the current difficulty is that the starboard side sustained enormously more damage than the port side. That side supported the entire deadweight of the ship for months, and much of that stress was concentrated into two zones, fore and aft, where the ship lay on its side, supported mostly by a pair of underwater pinnacles. 

I call those damage zones on the starboard side the Big Dents. Here's a Reuters photo of the starboard side, after parbuckling:

Sponson S13 has to attach onto the dent to the left of the picture, near the aft end of the ship. Here's a Parbuckling Project photo of S13 being lifted into place:

Here's a side view, also from the Project:

The lower diagram in the sponson map below shows the starboard side. Look at the long green rectangle near the stern: 
Sponson S13 is the long green object, horizontally oriented, and it's the one that had the problem earlier this month. From what I gather, the hull at this spot is so dented that a vertical tank can't be secured, so it has to be horizontal instead. 

The official explanation of the recent breakaway of Sponson S13 is that a chain passing under the hull wasn't tight enough, and this slack caused one end of the sponson to lift and damaged another one nearby. There might be more to it; certainly the Big Dent is a very difficult area in which to work.

I hope this post gives a little context to the latest news reports from Titan-Micoperi. To summarize: the ship is severely damaged due to wave motion over the months since the grounding, and because of how the ship came to rest; sponsons need to be placed there anyway to get the ship off the rocks; and the attachment points to that irregular, weakened zone are problematic. The whole ship may be so weakened that it won't hold together under the strain.

So once the hulk lifts off the reef and begins moving (whether under tow, or being pulled onto a semi-submersible recovery ship), it's sure to be a tense time for the salvors.  

Here's my information request of the week: imagery showing the results of an underwater sonar or photogrammetric survey of the starboard side of Costa Concordia.

Tuesday, May 6, 2014

Wish List for the MH370 Report: Did the aircraft have a QAR?

If someone is keeping a Wiki of items that people would like to see covered in the eventual report from Malaysia's accident investigation board, here's one for the list.

Many 777s, but not all, carry a Quick Access Recorder in the E&E Bay. When British Airways Flight 38 crashed on landing in 2008, investigators tapped into a QAR on board. 

A QAR is sort of like a Flight Data Recorder in that it stores flight data, except that, unlike the FDR, it offers easy removal of information. Airlines buy and use this optional gear to improve operational quality over time. 

Here's a photo of a Teledyne Controls QAR:

QAR downloads and analyses are vital to a good quality program. Here's a snippet on QARs from a report by Canada's Transportation Safety Board:

"Initiatives undertaken by airlines, such as the development and implementation of increasingly complex flight operational quality assurance programs, require that an increased number of data sets be recorded. Quick access recorders (QAR) were developed because information in FDRs was not easily accessible for routine maintenance and monitoring of aircraft systems. This type of recording has been done on QARs, which are not required by regulation. Most QARs in use routinely record far more data parameters, at higher resolution and sampling rates, than do FDRs."

If the MH370 aircraft (9M-MRO) had a QAR installed, then downloads from previous flights are in the hands of Malaysia Airlines. Even before the airplane turns up, these downloads could have information of real interest, such as error messages connected to past maintenance work. Further, then the airliner's wreckage is found, a QAR holding data from the last flight might be of enormous value if the tail section (which holds the FDR) broke up and the FDR can't be located. While QARs aren't built to be crash-rated, their contents -- even when only partially recoverable -- have been helpful to investigators of several recent crashes, Swissair 111 being one.

An electronic data-processing device analogous to the QAR, called the accumulator, was of great value in the Costa Concordia investigation, because the Voyage Data Recorder malfunctioned on its last trip. Fortunately, the accumulator served as an emergency backup.

Note: If the airframe didn't have a QAR installed, that would make me question the airline's attention to quality. Just saying!