Airports across the United States were choked last week with herds of stranded passengers when American Airlines, the country’s largest commercial carrier, called off thousands of takeoffs. The cause of this commuting calamity was a 7 April FAA inspection of nine Boeing MD-80s that revealed that some of the planes’ wiring had not been fastened in accordance with an earlier agency directive. In response, the FAA grounded hundreds of MD-80s and ordered fleetwide inspections of the jets’ electric wiring.
Was it a necessary move? Is there a technological fix that could have prevented such a seize up of the nation’s skies? Spectrum posed this and other questions to Cynthia Furse, a professor of electromagnetics at the University of Utah. Furse was coauthor of ” Down to the Wire,” an IEEE Spectrum feature article about the hidden hazards of airplane wiring, which appeared in the February 2001 issue. Since then she has been developing technologies that would detect wiring faults in flight. IEEE Spectrum’s Willie D. Jones spoke to her on 11 April 2008.
IEEE Spectrum: What was at the bottom of all those canceled flights last week? Were all the called-off departures necessary?
Cynthia Furse: As I understand it, a major reason for the FAA wiring-change order that resulted in the need to inspect the MD-80 planes was to check the bundling and the ties to prevent potential arcing within a wheel well—an area that commonly has faults. The FAA required a change to the wire-tie configuration, but not all the changes were made correctly. The planes then had to be reinspected, resulting in the canceled flights. This was very responsible—even if it stranded a lot of travelers. At least they’re safe.
Spectrum: How would you rate American Airlines’ handling of the problem?
CF: Every commercial and military entity that flies aircraft has a large number of aging planes. And all of these planes, like older cars, are prone to having problems as they age. I am very pleased to see the response of American Airlines and the FAA in making sure that these planes are properly inspected. Inconvenient as it may be for the passengers, just be glad that you are sitting safely in the airport while someone looks out for your safety.
Spectrum: Are we likely to see this again?
CF: There is a possibility of that. I would hope that they would ground airplanes that could be potentially dangerous. But I would also anticipate that in the relatively near future, there will be more-advanced technologies available on the planes that will prevent electrical-wire faults and potential damage and make the maintenance easier, thus preventing mass flight cancellations.
Spectrum: Was there anything that could have been done in the design stage or during regular maintenance to prevent deterioration of the material shielding the wires?
CF: There are certainly things that can be done in the design, and new planes are designed with a lot of improvements over the old ones. But the fact of the matter is that you’re always going to have planes that are older. Even if you [retired your aging planes every year], there is still the potential for wiring problems. So what’s really important is to have a good inspection and maintenance program and then to have the most advanced wire-testing tools available.
The Navy, the Air Force, NASA, and the FAA are all working intensely on developing new test technologies. New stuff is coming online regularly, and over the next 10 years I suspect that we will see some substantially changed methods for locating faults on wires.
Spectrum: What are some examples, and how would they change the state of things?
CF: We have technology, developed since that 2001 Spectrum article, that is capable of locating faults in flight. In 2001, we indicated that this capability was important and that we really needed it, but we didn’t have technologies that could do it. In the 2001�2002 time frame, our group at the University of Utah came up with a method called spread-spectrum timedomain reflectometry, which sends a cellphone signal—a pseudo-noise code—down the wire, gets a reflection back, and tells where the fault is on the wire. But getting something from research to a commercial device that has been tested sufficiently to put on an aircraft takes many years. We are currently commercializing that technology through a spinoff company called LiveWire Test Labs and, with funding from the U.S. Navy and the U.S. Air Force, are in the process of building that technology into a chip.
This is something we would hope could be integrated starting about five years from now and that it would be relatively mainstream within the next 10 years.
Spectrum: Why so long? Surely this problem hasn’t just cropped up?
CF: The development is going just about as fast as it can. Engineering development just takes a certain amount of time to do right. And we are in the midst of that development cycle. So it’s not just saying, ”Oh, someone has to make a decision.” The decision has been made—to improve safety by improving the test systems. Now it’s a matter of time—to finish it, test it out, and get it on board in a sensible manner. The Navy, the Air Force, NASA, and the FAA have clearly made a decision that it’s important to reduce this risk, and they are investing heavily in these technologies. I’m very pleased to see this both from the point of view of someone who flies in planes a lot as a consumer and also as a scientist. They are making a very big difference.
Spectrum: That’s good to know. But is there anything we’ll see earlier than a decade from now?
CF: Some of the changes are happening right now. Arc-fault circuit breakers, which do not rely just on the large current that trips a [regular fuse], actually look for the noise in the current induced by an arc. And they are able to tell when there is an intermittent arcing event, trip the circuit, and prevent an inflight fire. They are already being tested on some military planes and have been shown to be very effective. There is still some development yet to be done, but I would anticipate that the arc-fault circuit breakers will be relatively mainstream within about five years.
Spectrum: An ounce of prevention is worth a pound of cure. So will we ever be able to change wires ahead of time?
CF: The two things I described, spread-spectrum time-domain reflectometry and arc-fault circuit breakers, can diagnose a problem that is actually there, as in, ”there’s an intermittent fault that is happening right now.” There is another set of technologies for prognostics—the ability to predict something and say, ”That fault might happen in the future, so we should do some preventive maintenance.” I think that when the spread-spectrum reflectometry is integrated into aircraft, we will also be able to collect a large amount of data on the vibration signature of the wiring in a plane. Of course, all wires are vibrating, but you would not want to see them vibrate strongly against a metal corner or a thin metal strut, because you know that increases the possibility of chafing the insulation away. In seven to 10 years, it may be possible to use these technologies to predict that a fault is likely. That’s when you really start getting some advantages that let you prevent the things that American Airlines is having to do today.
Spectrum: How big an effort will be required to retrofit today’s planes with these technologies when they do become available?
CF: Installing an arc-fault circuit breaker just means taking out one breaker and putting in a different one. You’ve seen breaker boxes in your house. The breakers in airplanes are similar. In fact, you can buy arc-fault circuit breakers for your house today. That’s relatively easy to retrofit. We are looking at integrating the spread spectrum reflectometry into the arc-fault circuit breakers so that both of those technologies would be in a single unit that would replace an old breaker. So it really just becomes a matter of expense and timing.
Another new technology that we’ve been working on for years is something called a smart connector. It has male and female ends that let you junction it in places where experience has shown that wire deterioration is likely to occur. It has electronics for detecting and even predicting faults already in the housing, so once it’s in, the retrofit is done.
A fourth opportunity for retrofitting old planes is when they upgrade the avionics, the large onboard flight computers. When those are upgraded, these new diagnostic and predictive technologies could be on board those boxes.
Spectrum: With the airlines facing worsening financial straits—mainly due to skyrocketing fuel costs—how likely is it that, barring an FAA order, they will implement these systems?
CF: If they see that it can save maintenance hours, and therefore maintenance dollars, it makes a lot of sense to bring that in. Reducing maintenance is a major factor in being able to bring new things online.
Spectrum: So we can assume that you don’t think it makes sense for the average airline passenger to worry that some short in a wire might cause a sudden massive failure, causing their plane to fall out of the sky.
CF: I don’t know if I’m an average airline passenger, but I fly in planes several times a month, and honestly, I’m not afraid at all. I’ll be on another one next week.
I think that all of us know people who have been injured and even killed in car accidents. Very few, if any of us, know anyone who has been injured or killed in a plane accident. And I think those statistics speak for themselves. Clearly, the airline industry has a good record for safety. And I anticipate that they will maintain that record. Their business depends on it. And people who work for airlines fly around a lot, so they’re looking out for their own safety as well.
To Probe Further
Cynthia Furse’s Web site explains her work on aircraft wiring in more detail.