Is your airplane out to get you?

It was around the middle of the 1980s that it occurred to me that I just didn't understand my car the way I once did. The vehicle that I owned and attempted to manage was a 1983 Ford Ranger pickup, a car that just never ran right from around 10,000 miles on until I finally gave up on it at 40,000 miles. And that was only after numerous repairs that never did anything to make the truck run better, though they did succeed spectacularly in draining my bank account.

Starting with that Ranger, there were two trends I noticed. The first was, I could no longer see what was going in inside the engine compartment, never mind do anything about it. It was packed. I don't remember the exact details, but I'm pretty sure I had to remove the transmission to change out the headlights.

The second trend was that a computer had taken to running my engine. In theory, that was fine. Like many pilots, I'm an early adopter and at the time was the proud owner of an actual personal computer on which I could write letters and such.

But the computer on the Ford was causing me no end of grief. It used to be that I could, like a country doctor, listen to my car's complaints and then determine the probable cause of the malady. If the engine was running rough, I'd look into things like fouled plugs, a clogged fuel filter or a gummy carburetor. More often than not, I'd get it on the first or second try. But that didn't work with my Ranger, in part, I now know, because the computer was the likely culprit. It was, sadly enough, also the chief diagnostician. To get the bottom of its problems, I'd dutifully bring my lemon into the shop, have them hook it up to their machine, which would tell them which part to replace and, presumably, how much to charge me. This dance continued for a few years until it became clear to me that this was not an improvement over my previous ownership experience in any way. Long story short: I bought a Honda. My car problems were solved.

Computers aren't everywhere yet. In fact, we get emails on a fairly regular basis from readers who take the aviation industry to task for living in the past. And it's true. Airplanes, especially light airplanes, are technological relics in many ways, not the least of which is the design of their engines. Though they've been updated in detail in a hundred ways, their basic design is still very old school. From carburetors to magnetos, they are a veritable museum of 1940s technology.

There are good reasons for this. Certification regulations make it expensive to develop a new-technology engine, for sure, and it piles on after that, making it hard to put that updated engine in an airframe that was previously certified for older technology. For the FAA, it's a classic case of the devil you know. And I do get that. We know what the risks are to the old technology. Why mess with limited success?

As you know, there is one place where new technology has flourished in light airplanes, in their panels. Avionics design has been completely revolutionized by the computer age, and the result is more affordable, capable and reliable options than we've ever had before.

But as was highlighted by the still-mysterious crash of Air France 447 last month, there is always the concern that computers will do too much, that they will do things we didn't ask them to, and once they do, we'll be unable to wrest control back from them. This, in a nutshell, is what some pilots and other observers suspect happened with the Air France Airbus. For some, the general fear is that making our cockpits more and more computerized will eventually, if it already hasn't, remove human pilots too far from the flight control loop. When that happens, they argue, the pilots might be unable to save the day when the computers fail.

And there's no doubt that, despite their designers'attempts to make them as fail-safe as possible, computers do introduce new failure modes into the equation. The question is, do they make up for that by being more reliable overall, by improving safety in other ways, and by allowing us to do things more efficiently than before? Most pilots, me included, would say that they do.

But it also true that we simply don't know our airplanes the way we once did. How does a vacuum powered attitude indicator work? Most pilots would be able to explain, I'd wager. How does an ADAHRS work? They'd probably be at a loss there. And that's without going into the fact that there's code everywhere. What pilot could possibly know their airplane's code inside and out?

And that trend of our airplanes getting more exotic will continue. Within the next 20 years, and that's not a long time, our airplanes will be safer, more capable, more efficient, more reliable and, in some ways, more economical.

But they will almost certainly be more mysterious to us. Make no mistake. Complexity is part of the deal with this thing we call progress.

But it's also the part I like the least.

Robert Goyer

Comments

[Michael Grant]

Not only are airplanes getting more complicated, all the new electronics are making them more expensive. How did we manage to get from point A to point B before the advent of electronic panels? We learned to navigate; Pilotage (remember that skill?); VOR navigation (when was the last time you tried that?); IFR (I follow roads). Okay GPS makes it a lot eaiser I will admit. And a good GPS doesn't have to cost an arm and a leg. I worked for Piper Aircraft setting up dealers and flight schools back in the late 70's and early 80's. We sold Tomahawks for $25,000, Warriors for $35,000, Turbo Lance for $100,000, the Seneca twin for $150,000. Cessna equivalent models were about the same price. Have you priced a new Cessna 172 lately, $200,000! How about a new Seneca, $1,000,000! Now there are a lot of reason for the increase in price. But how many of us want to spend $50,000 and more for a glass panel when the old steam gauges work just fine? How many of us fly IFR and need a glass panel? The industry needs to bring the cost of aircraft ownership down to reality before they price themselves out of existence. Oh yeah...back to that glass panel; when it goes dark you will need to dust off those old navigation skills to get where you are going and pray that you are not in the clouds when it happens. And if you are in the clouds...pray that you have backup steam gauges and know how to use them.

[Lee]

I rarely fly, anymore, merely because analog guages communicate more useful information, usually trends, thereby increasing one's awareness.
Seems as if pilots now are expected to fly computers, not aircraft; the digital age makes the experience far less intuitive and much more expensive, both of which deprive a pilot of the belief he or she is involved in the experience.

[Terry Van Blaricom]

Anyone who thinks that computer electronics are reliable has never had anything to do with computer electronics. The FAA is making a mistake in not requiring vacuum system instruments as a back-up. The big Garmins and Avidynes have failed more times than anyone could have imagined.

The FAA was in denial about he necessity of an telectric attitude indicator back-up (to vacuum instruments)for many years, and many pilots died because of this. A bum off the street could see the need to mandate this kind of (inexpensive) back-up for IFR operations, but the people that we pay to be smart about these things were clueless.

[Clayton Shrewsbury]

I see a lot of ballyho about fly by electionics...but see no real benefit. My Beech P35 has a 1984 installation of HSI and RNAV (KNS80 - stores 4 waypoints). For 25 years I have found them quite adequate for IFR navigation and approaches to minimums. If one is not competent to hand fly approaches to 200-1/2, he is foolish to fly IFR regardless of the computer power aboard.
Clayton Shrewsbury

[michael Bellis]

Yep-
Most of this is technology just for the sake of technology,
I fly a rental 172 with all the fancy gear and hate the gizmos.
'Course I learned to fly with steam.

[Henry H]

There I was, an airline pilot, all glass qualified, 777s, 767,757, 737 getting checked out to rent a local 172 (which I instructed in for a couple thousand hours 30 years ago). Checkout wasn't too bad, flare high, plop it on. Then the GPS? How do it work? Never mind, I just want to fly locally, VMC.
All checked out, another day, I rent a 172, different plane, different avionics stack. OMG, the switches. I never could figure out how to transmit and/or receive. This things was too complex. See and be seen today. End of flight, 500' AGL, I see a King Air fly over top of us, on a go-around. Oh Oh. I haven't PIC'd a small plane since.

[Mark A.]

1) Computers are here, better get use to it.
2) You are free to choose your aircraft and instruments.
3) Flying is expensive for the average U.S. Joe. Ever fly in England?
4) Test drive the new Honda Insight for an automotive computer experience.
5) Can you spell HAL?

(C170 owner)

[Richard Magnan]

I have been in the computer business for the past 37 years. Since then, the hardware has come from very unreliable to very reliable now. For example, if a disk drive fails another one takes over, thanks to the RAID system. You remove the bad one and replace it without interuptions. The software though is still scary. That is still a big problem: There are and will alsways be "bugs". When thousands of lines of code are involved, it's inevitable. No matter how thorough the software is tested, "bugs" will be missed (For example Microsoft). And these, in an aircraft, can, is and will be deadly.

[Edd Skoglund]

Those old rangers were good pickups except as you say, the computer. But then the always failing beast was mounted beneath the passenger's floor board where it could get but not take much stomping. You piece reminded me of guys I've known that have a few planes dismantled in the barn and can successfully reassemble them at any time and can fix most anything with a pocket full of tools that will last an eon.