Runaway Terror

The Lion Air accident unleashes the dark demons of autopilots run amok. But for light airplanes, the imagining is worse than the reality.

If you ride around Seattle, where Boeing lives, you'll occasionally see a blue-and-white bumper sticker that's been around for decades, with variations of this wording: Unless it's a Boeing, I'm not going. Following this week's revelation that Boeing kept pilots in the dark about an autotrim system on the new 737 MAX, it took the internet 10 seconds to come up with a derisive corollary: If it's a Boeing, I'm not knowing.

As the aerospace giant wipes industrial-grade egg off its corporate face, the company will be hard pressed to explain why it doesn't deserve the drubbing. Meanwhile, the accident investigation that unleashed the opprobrium—the crash of a Lion Air 737 MAX8 off Jakarta last month—continues to focus on what role the airplane's automation may have played in the crash. Press reports revealed that Boeing fitted the MAX with an always-on background pitch trim system designed to improve pitch characteristics at high alpha and high load factors while providing some downside stall protection.

Unfortunately, Boeing apparently didn't document the system, called MCAS, in any of the manuals provided to customers who bought the MAX. Nor was it covered in the airplane's training syllabus, which has been described as minimal because Boeing made that a sales point against its competitor, the Airbus A320neo.

You hardly need George R.R. Martin's rich imagination to conjure the ultimate nightmare out of this: an airplane rendered uncontrollable by runaway pitch trim. Making lemonade from lemons here, that may be the immediate useful takeaway from the Lion Air accident for those of us driving little airplanes. Or, put another way, have you looked at your AFM recently to review how the autopilot and trim system works and how to disable it in a hurry?

For light aircraft, this is a low probability, moderate consequence risk. It's worse for the imagining than the reality, but that doesn't mean you should ignore it entirely. In my flying career, I have had exactly one runaway trim event. It was in a Piper Navajo in VMC 6000 feet over the Berkshires in Massachusetts. I had flown the airplane quite a bit and others flying it noticed that the autopilot had an unnerving habit of hunting vigorously when in altitude-hold mode. It had been to the shop several times, but was intermittent and seemed to defy solution.

The day it went full-up bonkers, it did the little pitch bobble and then twirled toward full nose-down trim aggressively enough to brush my noggin on the overhead. I stabbed the red autopilot release and popped the trim circuit breaker, but not before the airplane dumped 200 feet. I knew exactly where that breaker was because I was a beneficiary of old-school avionics design that had the tech placing the breaker right in view of the pilot, there being no room for it down on the regular breaker panel next to the pilot's left knee.

Some pilots who are nervous about such things or who have seen servos go berserk, have been known to mark them with colored tape so as to be easily visible in a panic. I've seen worse ideas. Part 23 requires a demo that the airplane can remain controllable "following any probable trim system runaway that might reasonably be expected in service." All well and good, but landing one with full-up or down trim would be no picnic.

Of course, if you had a manual trim wheel, you could just retrim and go back to your cookies and coffee. But in an effort to simplify and lighten airframes, lots of new airplanes have only electric trim; no manual backup. Cirrus is one and I can't recall the last light sport I flew that had manual trim. I'm sure there must be a few. Personally, I don't like electric-only trim because no matter how slow the trim rate, for lack of feedback, you just can't tweak it the way you can manual trim. When I flew the Vulcanair V1.0 for this video, I noted approvingly that it has only manual trim and it's surgically accurate.

For saving weight and structure, I have no quarrel with electric-only trim. It's reliable enough not to worry unduly about failures. I did a casual sweep through the service difficulty and NTSB accident reports and there are a few hits, but hardly enough to get lathered up about. With envelope protection now a thing and servos being given ever more responsibility, maybe we'll see more anomalies. But as a potential accident cause, it appears to be a tiny sliver of the midair risk, which is itself quite small. Nonetheless, it's just creepy not to have absolute control of something so basic as trim or to rely on a couple of wires and a cheap hat switch.

I'm not going to make this blog a treatise on the various autopilot and trim system designs and how to disable them. But I will suggest this: Why not haul out the AFM and review all this stuff for the airplane you fly? Can't hurt and could help.