Homebuilders and Electronic Flight Control Advancement

I believe the new frontier for light airplanes is fly-by-wire, or more likely, electronic enhancement of flying qualities and envelope protection. And homebuilders are the people in a perfect position to pioneer those advances.

Light airplane design and technology is very mature. People have been working to improve the speed, efficiency and safety of airplanes flying from 150 to 300 knots for more than 80 years. As an industry we know how to make good airplanes in that performance range and future improvements will be incremental, a few percentage points at a time, at best.

But electronic technology is still exploding. Nobody can predict what will be possible, and practical, using electronics in the coming years. And it is electronics applied to how airplanes fly that can advance the performance and safety of light airplanes more than any change in design or materials.

It’s already happening in large, fast airplanes where fly-by-wire is the norm for any new design. Fly-by-wire (FBW) uses computers to interpret the control inputs of pilots to then send a command to servos to actually move the flight controls. FBW can be more redundant than conventional mechanical and hydraulic control systems because multiple computers and wire paths can be installed, along with redundant servos to move the controls.

Control redundancy is not a significant concern in light airplanes, but electronic management of how the airplane flies could be an enormous benefit. The FBW computers can be programmed to make the airplane perfectly stable in all flight regimes. That means the airplane would constantly return to straight and level after being upset by turbulence, or maneuvered by the pilot.

An FBW system could also provide envelope protection. For example, let’s say the pilot commands an abrupt attitude change at high airspeed that would threaten to overload the structure of the airplane. The computers would temper the pilot’s input and command only enough control surface movement to take the airframe it its structural limits, but not beyond the limits and break it.

FBW could also prevent the pilot from exceeding both the high and low airspeed limits of the airplane. For example, if the pilot keeps commanding nose up when there is not enough energy available to prevent the airplane from entering a stall, the FBW computers would limit the up elevator travel to prevent a stall no matter how hard the pilot pulls back on the stick.

In its most comprehensive iteration FBW can automatically land the airplane if the pilot were to become disoriented in bad visibility, or if he were incapacitated. Unpiloted UAVs of all sizes make automated flights—including takeoffs and landings—thousands of times a day all over the world. Avionics experts from Rockwell Collins have already configured a Bonanza to land itself, and flown hands off landings. This is not pie in the sky, but technology that is available now and not all that expensive.

True FBW where there is no mechanical link between cockpit controls and the flight controls is unrealistic for a homebuilt at this point, but what is possible, and makes sense to me, is a supervisory electronic system. Computers could generate the commands to enhance stability of the airplane, and to move the controls to protect the flight envelope, while still allowing the human pilot to overpower the servo if the electronic system were to fail.

The FAA is being very cautious when it comes to approving electronically enhanced stability and envelope protection in production light airplanes. Even though loss of control is the single most common cause of serious light airplane accidents, and electronics could help prevent those, the FAA is slow to act. But homebuilders do not have those same FAA certifications restrictions. Builders could include electronic flight control enhancements now because their airplanes are experimental and the electronics can be experimental, too.

Homebuilders are creating a wide variety of attractive and good performing airplanes, and many of those have advanced flat glass avionics. The next step, I think, is to incorporate electronics to enhance the actual flying qualities of the airplane to provide stability not achievable through natural aerodynamic means, and add the safety of protecting the pilot from exceeding the design envelope. Electronic flight control enhancement would always be standing by to right an airplane if a pilot loses control in the clouds, and even automatically fly an approach to a runway.  If builders show the way, maybe the FAA will see the light and help make electronic flying qualities enhancements, and safety, more available to all categories of airplane.

This entry was posted in Mac Clellan's Left Seat Blog. Bookmark the permalink.

21 Responses to Homebuilders and Electronic Flight Control Advancement

  1. Thomas Boyle says:

    It’s a real shame that butt-covering at the FAA is more important than safety. But, as you say, thank heaven there are some corners of aviation, like Experimental and LSA, where the FAA’s butt is covered already, and progress is still possible.

  2. Thomas Boyle says:

    Okay, I thought someone would have jumped all over my last comment by now!

    Clearly there’s a need for certification of avionics, for 3 reasons:
    1) so pilots know the kit works
    2) so passengers aren’t put in danger by pilots who decide to “buy cheap” without telling the passengers
    3) so other air traffic isn’t endangered by aircraft that aren’t where the pilots think they are (in any of the 3 dimensions)

    I suggest that 1) can be satisfied by brand, or 3rd party testing, or by the FAA (in an advisory, rather than mandatory way).

    In private aviation, I feel that 2) is not that important, because passengers are putting themselves in the hands of the pilot anyway, and if the pilot is exercising bad judgment about avionics it’s probably not a great situation to begin with. Brand and 3rd party testing, and even non-mandatory FAA testing will be of no help, because passengers won’t know to ask about the avionics. So, there is room for debate about this aspect. A simple solution would be to require passenger notification, as we do with Experimentals now.

    That brings us to 3), where there are really 3 issues. First, 2-D location; second, altitude; 3rd, reliability of key instruments. Certification of altitude encoders will probably be necessary on an ongoing basis, but that’s fairly trivial. As for 2-D location, certification of GPS receivers for location is a solution in search of a problem. There is no epidemic of GPS receivers that don’t know where they are. That leaves reliability of key instruments, where there are two possible approaches: certification of equipment to high reliability (expensive and hard) or what many pilots use, which is redundancy. Two uncertified GPS receivers are a lot less expensive than one certified one. Same goes for (solid-state) gyros.

    I do think the FAA could put its considerable expertise into thinking about getting “out of the box” on avionics by creating more space for, well, experimentation. I suspect the agency may be willing, but too busy to focus on this, so the existing regulations remain in place, inhibiting progress and safety.

  3. Dean Moriarty says:

    FBW could also prevent the pilot from exceeding both the high and low airspeed limits of the airplane. For example, if the pilot keeps commanding nose up when there is not enough energy available to prevent the airplane from entering a stall, the FBW computers would limit the up elevator travel to prevent a stall no matter how hard the pilot pulls back on the stick.

    You mean like AirFrance 447?

  4. Jeff Boatright says:

    Hi Mac,

    Interesting topic. It seems to me that much of what we see in GA glass panels today, I saw at OSH back in the early 90s. Might hold true for EFC, too, though I suspect that The Next Big Thing from experimental aviation will be electric propulsion. Certainly excites me more than electronic control, but then, I’m not a X-ctry guy.

  5. Robert Miller says:

    I for one would like more info, #1 wHY DOES THIS OR THAT DO what?? wHAT SHOULD I DO TO CONTROL THAT. So # 1 I will need more input from Friends>> The Other pilots>> The Controlers>> and Instructors>>>>> I need to have the BACK_UP of people That care about FLYING and SAFTY to STEP UP and PUT ME__ ME In my Place When I Don’t Do RIGHT>>>> We Need to be Our Brothers KEEPER .>> FOR REAL>> AS Pilots, SAFTY With in our Ranks Can and Will MAKE US ALL Better PEOPLE>> and PILOTS only if WE CARE >> And Step UP >> STEP UP >> This is not a JOKE Friends, and Pilots. Talk Is Cheep ACTIONS Speak Out and Save Lives.>> And Us From Our Selves, Because of Either Bad Habits. Or not paying ATTENTION, For some crazy reason. Love To Fly… AGAIN. SAFTY IS ALL OF OUR PROBLEM,, BUT OUR RESPONDABILTY. ALL OF THE NEW STUF F that can help us will only be as good as the PEOPLE behind The PILOT. The Pilot Is In Control FIRST AND LAST..>>>>>> Love to Live> Love to FLY . LIVE To rember it all!!!!
    CB

  6. Bruce Pease says:

    I think Mac’s comments are worthy of some thought and discussion. There is room in the flying population to pick their means of flight without inhibiting the “advancements” others may want. FBW is indeed a very broad concept. Perhaps someone will take this concept and reduce it down to a simple wing leveler device for starters. Electric airplanes are actually a step in the direction of FBW. If EAA members begin to develop suitable acutators, redundant processing, feedbacks, control communications, etc. is this not in the fun of being an experimenter?

  7. Douglas Elting says:

    Thank you, Thank you, Thank you! Finally someone has the courage to say that our airplanes should be as smart as our dishwashers. Our automobiles have over 70 microprocessors connected by a CAN (Controller Area Network). This technology has been in place for almost 40 years. My only comment is that Mac is incorrect in saying that true FBW, where there is no mechanical connection between the pilot and control surface, is not possible for amateur built aircraft. I am building one now with no stick or rudder peddles, only a three dimensional digital controller.

  8. Gordon Arnaut says:

    I have to strongly disagree with this march to computerized flying…

    Picture this…you are descending on short final but while still a couple of hundred feet agl you see that you are going to overshoot the runway …you have to go around…you push full throttle…but the airplane continues to sink…you crash into trees at the end of the runway and are killed or seriously injured…

    What happened…?…the computer in its infinite wisdom pushed the elevator down in order to prevent a stall…

    Couldn’t happen…?…It already did on an A320 flown by Air France chief pilot Michel Asseline…AF296 crashed into trees as he tried desperately to go around…

    So much for the “safety feature” of FBW…three people were killed…Asseline went to jail in what many consider a coverup…we couldn’t fault the much hyped computer now could we…?

    Then we have Turkish Airlines 1951…a B737-800…a faulty radio altimeter that was reading 8 ft below SL as the plane came in on approach…caused the computer to command the autothrottles to idle…

    When the captain saw that the plane was not going to make the runway…he firewalled the throttles…but the computer said no…!…so the computer flew the plane into the ground killing 9 people…

    As a result…Boeing has advised crews to not use autopilot or autothrottle in case of altimeter discrepancies…why not just toss the stupid computer “safety feature” instead of putting more workload on the crew…?

    Then we have Scandinavian 751…an MD-81…the plane took off in ice conditions…it had been deiced but a layer of clear ice on the wings had remained undetected…

    As the plane was climbing out some of this ice broke off the wings and was ingested by the engines…the Pratt JT8D turbofans were designed to take this kind of damage and keep going…provided the engines are throttle down to prevent surge…

    This is exactly what the crew did…too bad they were not aware of the computer which pushed the throttles right back to full power…a “safety feature” called Automatic Throttle Restoration…of which the airline was not aware that their planes had been “upgraded…”

    The ice damage to the fan blades caused them to push less air into the engine…so if you maintain full throttle…you now have a fuel-air imbalance with much more fuel than needed…the engine starts to backfire through the compressor and fan and will self destruct in a few seconds if not throttled back…

    The crew miraculously brought the plane down in a wooded area with everyone surviving…

  9. Mac says:

    The question is not whether computers can fail, because of course they can. The safety question is will electronic systems fail less frequently than human pilots and thus improve overall safety? So far the move to automation has proven itself in airline jet flying. No passenger on a U.S. jet airline has been killed since 2001, a record no one ever believed possible. There are many reasons for this incredible safety record but automated systems clearly play a key roll.
    Pilots who hate computers love to point to the Air France Airbus over the Atlantic where the crew failed to maintain control after the system reverted to “direct law”. That was a combination failure of the humans and automation. But why not point to the Buffalo crash of the Dash 8 turboprop. The human stalled the airplane. An FBW system would not permit that.
    Automation is coming most slowly to general aviation and unlike the jets, GA safety is improving very slowly if at all. I, and many others, believe the decline in GA accidents is merely a reflection in a decline in flying hours, not a true safety improvement. Automation can help change that as it has in jets.
    Mac Mc

  10. bill p. says:

    “The safety question is will electronic systems fail less frequently than human pilots and thus improve overall safety?” Interesting question, and I wouldn’t argue against your apparent position that they probably will, speaking from the standpoint of “overall safety.” So with the new technology, the pilot who incorrectly pulled back on the yoke and stalled in Buffalo would have been overridden by the computer and prevented from making that mistake. But, at least until such time as the new technology is perfected, there will be some pilots who competently command the airplane to do something (e.g., go around, as in the above examples) and find themselves overridden by the computer, with a crash resulting. But since on average more accidents probably come from incompetent pilot actions that should be overridden than from competent pilot actions that the computer erroneously overrides, “overall safety” will be improved. I’ll bet you are right.

    But it doesn’t make you feel too good if you fancy yourself a competent pilot who doesn’t think he would have pulled back on the yoke in Buffalo and doesn’t want his decisions overridden by an unperfected computer.

    Reminds me of the time I tried to pull out smartly from a side street into a gap in oncoming traffic on a main road. When one of the rear wheels descended into a small depression and bounced slightly the Anti-lock Brake System (ABS) computer apparently sensed the wheels were spinning at different rates in a manner indicative of a skid and shut the power down. Here I’m asking for power with a Mac truck bearing down and the stupid computer is refusing to give it — to make sure I don’t go into a skid. Thank you very much. It’s a lousy and angering feeling, I can tell you. If what I have heard is correct, you won’t find ABS on race cars and motorcycles. Those guys want to make their own decisions.

  11. bill p. says:

    P.S., that system is probably not properly termed part of the ABS system, but hopefully you get the point.

  12. Frank Giger says:

    The computer commands the plane to take off. The computer lands the plane. The computer controls the plane in flight.

    Why bother having a pilot?

    Of course I’d love to see a computer land a Champ with a 90 degree 10 knot crosswind with 5 knot gusts thrown in for good measure….oh, and some thermals just for fun to deal with.

    At least we’ll have a new acronym…CGGL (Computer Generated Ground Loop).

  13. Mac says:

    Landing a Champ in a gusty crosswind would not be all that difficult for an electronic system. Small, lightweight UAVs without even the stability of a Champ takeoff and land automatically every day.
    Electronic systems have a huge edge over humans because of the processing speed. We humans need a significant amount of time to see a flight path deviation, calculate what is going on, determine what control insputs to make, and then direct our muscles to move the controls. The entire process takes a fraction of the time when it is done by electronics.
    But an electronic system could putt better than any golfer, or roll a ball better than any bowler, which is beside the point. The same is true in the Champ which is used for fun and recreation. The challenge of getting a Champ down and stopped on a windy day is much of the fun of flying a light airplane. But when the mission is to travel in the air safely in almost any kind of weather and stay on schedule the challenge becomes to arrive safely and that’s where electronics can provide the most benefit.
    Mac Mc

  14. Gordon Arnaut says:

    “Electronic systems have a huge edge over humans because of the processing speed.”

    One thing electronics do not have is judgement…

    The problem is that the computers are only as good as the scenarios that they are programmed for…can the human programmers think of every possible combination of events that could occur…?

    You’d better hope so…but I would say that is a fantastic leap…and that is where computers flying an airplane become dangerous…as the above examples show…

    A bad reading from a malfunctioning instrument…engines swallowing some ice…even a go-around at the last second…and the computer’s logic turns to mush…

    I’m don’t agree with the computer overriding the pilot’s flight control inputs under any circumstance…avoiding stalls is a matter of training and should never happen with a competent pilot…

    Warning devices, yes…stick shakers and pushers, yes…but a computer overriding the pilot…?…well…use your judgement…

  15. Frank Giger says:

    “But when the mission is to travel in the air safely in almost any kind of weather and stay on schedule the challenge becomes to arrive safely and that’s where electronics can provide the most benefit.”

    I guess that’s the crux of the matter – how one views what a GA aircraft is best used for.

    Personally, the only thing in that statement that even comes close to what I consider my aviation mission statement are “travel in the air safely” and “arrive safely.”

    If I absolutely have to be there regardless of weather I’ll either drive or call Delta. For me it’s about learning, seeing, and having fun; when an airplane becomes just another form of personal transportation it becomes work.

    Adding tens of thousands of dollars of electronic gizmos that do all of the work for me (and don’t allow me to actually pilot the aircraft unless it approves of any control input I put into it) defeats the whole reason I’m into aviation to begin with.

  16. Thomas Boyle says:

    Yes, the computers will make mistakes, and even so they will do better than the pilots do. And, even so, there will be a long period of time where pilots worry – on occasion justifiably so – about the computer suddenly doing something “weird”. Yes, it will, and pilots will have to learn the systems awareness to allow them to identify that there’s a problem, and know what to do about it. We learned these things about our engines and our gyros and our radios and our transponders; it’s a new form of systems awareness, and it will come more easily to generations that have grown up with computers doing weird things.
    I do think pilots of my generation have all sorts of trouble accepting that a computer is a better stick-and-rudder jockey than we are, and in a certain narrow sense has better judgment than we do, too. This is from the self-same generation who preach the gospel of “trust your instruments, not your instincts” in IMC, and who learn how to identify a failing gyro so as not to follow its erroneous output all the way to the ground. Sadly, that same mantra should apply on the turn to final, too – and a computer will pay heed to it.
    I have no difficulty accepting that a computer would do a better job of takeoff, climbout, cruise, descent and landing than I would do by hand. I would welcome it as a significant safety upgrade – at least 75% of the work of getting, and maintaining currency of an instrument rating is in learning to hand-fly on instruments, to a standard any computer would find unimpressive. In fact, down the road I could see an argument for a “autopilot only” instrument rating. I also think it would be fun to try to build a computer that could out-thermal me in a sailplane (it’s a different kind of challenge).
    But none of that need detract from the joy of flying by hand. Think of it as being like music: I get a lot more fun out of playing Beethoven on the piano myself than listening to it on my iPod – but if I really want to hear Beethoven played well, I have my iPod do the playing.

  17. PhD_aero says:

    So Mac -
    1. You talk about enhancing stability of homebuilts and envelope protection. Which homebuilts are you talking about, and how many hours do you have in each? Or are you going to duck the question again because you’ve never flown any homebuilts… No excuses, answer the question.
    2. Since you’re advocating fly by wire, why don’t you explain to the readers what DO-178B is, what Levels A and B mean, and what you think would be an acceptable alternative for homebuilts.
    3. Of all the homebuilts you’ve flown, which ones had stall warning systems? Which ones didn’t? Name those in each category that you’ve flown.
    4. Why do you think homebuilts should get envelope protection when they don’t even have stall warning systems?
    5. What statistics indicate a need for FBW in homebuilts? And don’t quote the Nall report, its taxonomies are obsolete.

    • Frank Giger says:

      I think that’s a bit harsh.

      While I think fly-by-wire is a bit of overkill for most homebuilts and auto-magical buttons that could make excuses for bad pilot decisions aren’t a grand idea (“it’s okay to fly into IMC, I can just flip the Jesus lever and be saved”), there’s a difference between stating that the homebuilt arena is perfect for innovating inexpensive, effective tools for safety and saying that it should be mandatory (which Mac isn’t).

      Langewiesche was probably poo-poo’ed when he wrote the future of safety in aircraft was tricycle gear instead of conventional ones (which has been widely adapted) and mechanically coordinated rudders to prevent spins (which weren’t) at the end of Stick and Rudder.

      My just-over-ultralight homebuilt would not be a good platform for all that stuff for a host of reasons, but I can see some of the heavier, faster types out there that could benefit from it.

  18. PhD_aero says:

    Overly harsh or not, Mac ducked every question. If he knew what he was talking about, he could have given reasonable answers to every one of them. His failure to respond is damning.

    EAA has for years been the home of the passionate and the knowledgeable. EAA publications should serve such people, not insult their intelligence and drive them away. Mac, you are in the wrong job. Please find something else to do, and soon.

  19. Mac says:

    Hi Mr. PhD,
    I am surprised that you demand answers from any authority. Homebuilding is about freedom to develop your own solution. I do know what software levals of certification mean, and other aspects of FBW, but they do not apply to AB. So, I suggest that builders, as they have always, find answers to enhance the stability and safety of their airplanes using electronics without waiting around for the government, or me, to tell them how to do it.
    Bests,
    Mac Mc

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>