Safety Publications/Articles


Handling broken glass

Response to failures may be simpler than in traditional cockpits

With all the new glass cockpit airplanes, more pilots are speculating on the consequences of electronic malfunctions.

For instance, a ferry pilot taking a Cessna 172 across the Atlantic reported losing all his displays, apparently when an incorrect procedure for fuel transfer set in motion a series of events that caused the Garmin G1000 to continually reboot. The fact that this leg was flown at night added to the drama.

Another suspected display failure occurred in a Cirrus in instrument meteorological conditions when the pilot misinterpreted some ATC instructions that were for another airplane. He told the controller, "I gotta get my act together here," and less than a minute later reported "avionics problems." Within the following 60 seconds, he reported, "losin' it." The airplane descended nose down into the ground. The airplane had a history of primary flight display (PFD) failures.

There are no solid statistics available yet on failure rates of glass panels, but the rate is undoubtedly very low. Still, the specter of "partial panel" (or blank screens, as the case may be) is a very real fear for pilots who have experienced such a failure in conventional aircraft or have read of the unfortunate consequences that can result.

Anyone who has had to deal with computers knows that, while they are extremely reliable, they are not infallible--especially the software. As someone who spent 40 years working with computers, I was quite reserved in my initial thoughts. However, now with more than 100 hours flying and instructing in a Garmin G1000-equipped Cessna 182, I feel confident that truly debilitating failures will be rare, primarily because of the redundancy built into the glass cockpit systems. As my experiences are primarily with the G1000, this discussion will center on that product.

With the Garmin G1000, for example, two independent displays--each powered by a separate electrical bus--and two independent battery supplies are available, making prospects of sudden and complete power failure remote. The ability of the right-hand multi-function display (MFD) to perform as the PFD assures that PFD hardware failures can be readily resolved. Of course, this is at the expense of the enhanced positional awareness that the MFD provides.

The availability of two independent avionics units and duplicate GPS receivers also provides some reassurance against those types of failures. Should everything electronic fail, though, the pilot still has a set of traditional back-up flight instruments--airspeed, attitude, and altimeter in the standby group, and a wet compass. While the scan is somewhat different, the critical information is still there, as opposed to the old "needle, ball, and airspeed" partial-panel failure in the traditional cockpit. Without any doubt, I would rather face any systems failure in a glass cockpit than in a traditional one.

In the case of instrumentation failure, a glass cockpit does some analysis of the failure and alerts the pilot--providing more time to think about the problem and to respond rationally. In the traditional cockpit, on the other hand, the pilot may not even be aware of a failure--say, the vacuum pump--until the airplane is moving toward an unusual attitude that requires immediate correction by a pilot. Moreover, when a glass panel flight instrument or function does fail, the software removes the information displayed and replaces it with a red "X," leaving no chance that the pilot will try to use the erroneous information.

While the glass cockpit provides more assistance to the pilot, it also requires that the pilot better understand the systems, more so than in a conventional aircraft. For example, with certain avionics failures, the comm unit retains the last used nav/com frequencies, while in others, it defaults to 121.5 MHz and bypasses the audio panel.

With failure of the primary battery, the standby battery automatically assumes the responsibility for electrical power, issues a warning message, and powers only those items on the essential bus, providing automatic load shedding. The pilot needs to know the configuration of the electrical systems to understand what capabilities will be lost in such a circumstance. The standby battery should provide power for about 30 minutes.

Pilot understanding of airplane systems is also critical as the autopilot can play a crucial role in helping to handle failures.

Just as the pilot should be proficient with partial-panel operations, the PIC in technically advanced aircraft must understand how to handle broken glass, and periodic recurrent training is vital. PC simulators are invaluable for their ability to fail selected line replaceable units (LRUs are the basic system components), effectively presenting a wide range of malfunctions. Before long, these simulators will be able to present flight plan scenarios in which the computer randomly fails various LRUs. The pilot will then respond to complete the flight with remaining systems--and conceivably the computer could analyze her performance, reminding her of resources she did not use but might have.

Pilots have always had to know how to cope with equipment failure. As far as failures go, glass panels may make life simpler for pilots, not more complex.

Ted Spitzmiller holds an FAA commercial pilot certificate for airplanes, single and multiengine land and sea, with instrument privileges. He has been a Gold Seal CFII for 34 years, logging more than 4,000 hours in more than 62 different types of aircraft. He is a check pilot for the New Mexico Wing of the Civil Air Patrol.

By Ted Spitzmiller

Back to the Index of Instructor Reports