What AoA Indicators Don't Do That They Should
With the exception of tiny squiggles up and down, the general aviation fatal accident rate remains stubbornly at about 1.0/100,000 hours. That's down a little from a decade ago, when it was 1.28 and down a lot from the early to mid-1970s, when it was more than 2.0.
In GA, we tend to resist the notion of required training and recurrency—the 24-month flight review is a joke—in favor of gadgets that make safety automatic and painless. The latest seems to be the angle-of-attack indicator, of which there are at least now five systems to choose from. All of them perform as intended and in the aviation press, we've fallen all over ourselves reporting on how great these gadgets are and how they should reduce stall/spin accidents. As I've said before, I go a different way on this. I think it's naļve to think these instruments will make a dent in stall accidents just as it was naļve to think the CAPS system in a Cirrus would give the airplane an exceptionally low accident rate.
It took Cirrus the better part of 20 years to learn that you can't just slap a safety system on an airplane and expect it to magically keep pilots from crashing and dying. As the airlines have also learned, these systems need to be tightly integrated into initial and recurrent training; otherwise, they're just another blinking distraction the pilot has to deal with under duress. When Cirrus, along with a devoted owner community, got serious about CAPS integration training, the accident rate finally showed improvement—impressively so.
That's not to suggest that I think AoA indicators are a gimmick or not worth the investment. They're not especially expensive and they have the merit of reprogramming pilot understanding of stalls as an airspeed thing to an angle-of-attack thing, where it should have been all along. And that's why I think the marketing of these devices is off base. They're being sold as stall awareness devices when in fact, they're really performance-measuring instruments that happen to include stall warning and awareness capability. If you limit them to the latter, they're just a visual version of the stall warning horn or aural alert and that almost guarantees they won't be integrated into the pilot's understanding of what the airplane is doing.
About 15 years ago, the Navy invited me for an overnight tour of the U.S.S. Theodore Roosevelt, which was involved in carrier workups off the Virginia capes. Ahead of that, I spent some time at the LSO school at NAS Oceana flying an F-14 simulator to carrier landings. Like all naval aircraft, the Tomcat was equipped with an AoA indicator and an on-speed indexer that summarizes AoA and, through indicator lights on the nosegear, gives the LSO a continuous and instantaneous readout on the approaching airplane's energy state. It took my sim instructor about 30 seconds to explain the AoA workings and it took another 30 seconds to integrate its response to my manipulation of pitch in the simulator. Hey, I got this.
As you can see from the graphic I'm including, the indicator has an approach reference bar and a tape showing actual AoA. It uses arbitrary units rather than degrees, but that matters little in interpreting it. The indexer has colored chevrons and a circle to show slow, optimum or fast speed and corresponds to what the LSO sees on the nosegear speed indicator.
I flew four or five approaches using this system. I crashed every one of them. Hit the ramp twice, took out the island and, I think, flew through the hangar deck. But I was on speed for every one of these disasters because once trimmed up, the AoA makes it easy to stay on speed with minimal mental bandwidth. The rest can be devoted to staying on the ball vertically—not that easy—and adjusting lateral lineup; all but impossible for me. I could figure it out eventually, I think, but that isn't the point. What is the point is that the Navy considers the AoA a critical, everyday tool for pilots that, oh, by the way, has a little crosshatched area indicating that 29 units of AoA will stall the airplane.
With this doctrine in mind and for this video, we did a flight trial in a Bonanza equipped with Aspen's AoA system. The Aspen AoA is a clever piece of design in that it requires no additional hardware, just a software package that uses the system's internal MEMS gyros and an aerodynamic model to infer AoA. Conceptually, the display is not that different from the F-14 design and it's well damped and accurate, graphically projecting the shrinking stall margin as AoA increases. But I'm not interested in that. I'm interested in using it as an on-speed indexer.
So we set up some slow flight and an approach flying on the AoA indications. Aspen says that the AoA flags at the green/yellow intersection should correspond to the aircraft's published approach speed and indeed it does. And it should. When the system is calibrated, you plug in various values from the POH in order to populate the aerodynamic model with data points for the software to compute AoA.
I found that moving the needles well up into the yellow band still gives—at least for me—comfortable stall margin but also four to five knots slower speed, making the touchdown less floaty and a lot shorter. Approaches flown too fast—very often way too fast because so many pilots are terrified of stalls—are a common scenario leading to excursions and overruns. And even if they don't, they use more runway than necessary and chew up tires and brakes. Further, the AoA flags are nicely damped and thus don't jump around nervously like an airspeed digital display or even an analog needle tends to do. It's thus a little easier to fly. It could also be used to index best angle or best rate for best performance departures.
I asked Aspen's developmental team about promoting the AoA system just this way, but they're uncomfortable with that, at least for now. It would require additional aircraft-specific testing rather than the affordable and certifiable one-size-fits-all software package they're offering now. It might also require external sensors, such as a dedicated pitot sensor or a vane. That adds cost and costs narrow the market. For its stated goal of stall awareness, the Aspen AoA and others like it are up to the task outlined for them and that's a broad-brush treatment for stall avoidance. Although I still think the integration training lags and these systems won't have the positive impact some people think they will, they're still a step in the right direction.
Ultimately, I'd like see AoA systems evolve in just the way I've described here—as speed indexers used on every approach to get the most out of aircraft performance and to wrap the sticker puller's head permanently around the concept of AoA. Otherwise, installing an AoA indicator is a little like having a live-in Cordon Bleu chef fix you spam sandwiches for lunch.