“You’ve got to know why a thing works on a starship.”—James T. Kirk, Star Trek II: The Wrath of Khan
I have watched the many and various incarnations of the Star Trek universe, and this line has stuck with me more than any other. Kirk is fighting a space battle against an old enemy who has hijacked another Federation starship, and while he knows how to fly and fight with it, he doesn’t really understand it at the engineering level. He knows how to work it, but not how it works.
Kirk, on the other hand, the ultimate master of everything he does, knows that starships are controlled through computer commands and that the first part of every command is a code that designates if a command is “legal” for that particular vessel. This is to prevent an enemy (who does not know the code) from using captured weaponry. But in this case, it allows Kirk to send a command to the other ship, lowering its shields and making it vulnerable. He won the engagement because he knew more than his opponent—he knew how the systems worked.
Here on Earth
Forgive the lesson in fictional space travel—it is simply a way to introduce the topic of understanding your aircraft’s systems. Many a pilot who has gone through the process of getting type rated on a complex aircraft has bemoaned the need to learn countless, apparently useless facts about every detail of the aircraft, items such as the nitrogen gas pressure in the nose strut, or how many quarts of oil the auxiliary power unit takes, or the allowable temperature of the fuel servo when it is not operating. It all appears to be some weird hazing ritual by those in the know, perpetrated upon those who would like to be members of the elite club. And to be truthful, many of these arcane pieces of trivia truly are useless, at least most of the time. The average pilot mostly wants to know how to get the engine started, how to tune the radios and what speed he needs to get the aircraft flying—forget all that mumbo-jumbo about how it is built. And many times you can get away with that level of knowledge.
Sometimes, however, knowing the details about how an aircraft system works can be a real lifesaver. Let’s take a simple problem in a typical light homebuilt with an electronic engine monitor and EFIS. The airplane has fuel gauges, but it also has a fuel-flow transducer and the capability to account for fuel flow over time, thereby keeping track of how much fuel has actually been burned. As long as the pilot resets the counter when the tanks are filled, this can be (and generally is) far more accurate than the typical gauges in a GA airplane.
All of the fuel data is displayed on the pilot’s EFIS—the data from the float gauges as well as the totalizer values. But what is this? The numbers don’t agree! Wow, something must be off—the fuel gauges say I have 11 gallons in one tank and 15 in the other, but the totalizer says I have 34 gallons left. That’s a difference of 8 gallons, which can be an hour of flight.
OK, quick: Is this an autopilot head? No, more like a prop control! The difference is important, and a label might help.
Which Is Correct? And Why Are They Different?
The undereducated pilot might spend a great deal of time brooding over the situation and make poor decisions based on a misunderstanding of it. But the pilot who knows the aircraft’s systems will understand that the float gauges are there mostly for show—because of dihedral and the way the tanks are shaped, they say the tanks are full until several gallons have been burned away, and then they decrease in a non-linear fashion. The indicated amounts remaining do not accurately reflect the amount of fuel on board. The only time they are accurate, in fact, is when the tanks read zero.
The totalizer, on the other hand, is accurately reflecting fuel burned—with one exception. It actually shows slightly more fuel having been consumed than actually has been (giving the pilot an unknown, but positive, reserve). This is because the fuel-flow transducer reads high whenever the boost pump is active. The pilot knows this because he has seen the flow jump when the boost pump is on, even though the engine is putting out the same amount of power (and therefore burning the same amount of fuel). Knowing all of this, the educated pilot has a much better idea of just how much fuel is on board and can make better decisions. (I should point out that some engine monitors with fuel-level systems allow you to quantify a tank’s dihedral effect or non-cube-ness to get much more accurate level readings. But they’re still not likely to be as good, or repeatable, as a flow-based accounting system.)
Do you know how this simple mechanical fuel gauge works? Does a lever or a worm gear drive it? Knowing could tell you how it might lie to you.
They’re Quirks, Your Quirks
While it is important for all pilots to understand their airplanes (and the quirks of behavior and instrumentation that can affect their operation), it is vitally important for those flying Experimental aircraft. Because homebuilt aircraft vary widely in their construction and systems, and because by their nature they are frequently proving grounds for new ideas, all sorts of different and unusual things might surprise an unsuspecting pilot. In a certified aircraft, the engineer designs and builds things a particular way. The test department proves the work and writes procedures that go into the checklists. The pilot merely has to operate the aircraft within the bounds of the documented procedures, and things should go well.
But in the Experimental world, the builder is working with unknowns, especially when he begins to modify designs or is designing from scratch. Naturally, the designer is well versed in the design’s capabilities and its potential quirks, and it will be unlikely that he gets surprised by these behaviors in flight. But if a new pilot, unfamiliar with the design, comes along, and something odd should appear, all bets are off.
Experimental test flying is an art that combines engineering and aviating. The best test pilots are engineers, intimately acquainted with the new designs they are testing. Whether the envelope is being pushed in aerodynamics, propulsion or avionics, they have been part of the design process (or have learned as much as they can about the design goals and implementation), so that they can not only fairly evaluate the results of the experiments but also handle off-nominal situations or failures.
Do you push the green button or the red one? And what will happen when you do? Complicated stick grips make the piloting task easier if you understand them and much more dangerous if you don’t.
NASA’s Dryden Flight Research Center lost an Experimental airplane, the X-31, a few years ago because the pitot tube froze up in thin, icy clouds. Like most accidents, the cause was a chain of events, not one single massive failure. It began with the design of the aircraft—a unique, fly-by-wire jet intended to fly with vectored thrust at extremely unusual angles of attack and conditions of sideslip. The pilot moved the stick, which told the computer what he wanted it to do, and the computer figured out what combination of control motions were required given the specific flight regime (altitude, airspeed, dynamic pressure, etc.) in which it found itself.
Airspeed was vitally important for the software to make the right control motions at a given time. Airspeed, of course, was taken from a pitot tube. Because this was a test airplane, it was never intended to operate in anything but clear VFR conditions. Even so, the original pitot probe was equipped with a heater, as are most air data probes for jets. Shortly before the final test flight, however, the probe was replaced with an experimental unit, one that did not have pitot heat—something that was not a requirement for the test program.
The day of the final test, a pilot newer to the program, one who was not familiar with the history of the pitot probe replacement, was flying. When he got into the thin ice clouds and sensed that there might be a problem with the software (because of an inaccurate airspeed value) he turned the pitot heat switch in the cockpit to ON. (It was still there, though it was not connected.) This did no good because the probe had no heater. The probe iced over, the computer lost its airspeed value, and the result was a loss of control. The pilot ejected, the aircraft was lost, but everyone survived (including, I believe, the engineer, who changed the probe without labeling the switch INOP). If the pilot had known that the pitot heat had been disabled, he might have tried harder to stay out of the ice clouds. His lack of knowledge of his aircraft systems was a contributing factor in the mishap.
Engine and fuel systems are places where builders frequently experiment, and there is nothing wrong with that (it is how aviation advances) as long as they fully understand what the implications might be. This understanding must also be passed on to any pilot who will fly the machine, so that they are not operating on any false assumptions about how things will work. Many an aircraft has been lost due to fuel mismanagement, and a fair number of those incidents stemmed not from carelessness but from a lack of knowledge about how the aircraft was plumbed, where the fuel was and how to get it to the engine. When a builder does something new and unusual with flight-critical systems, he owes it to others to document the potential differences from the so-called norm.
The last topic in this vein is probably the fastest-growing segment of homebuilt-aviation technology: avionics. The pace of advancement and change in modern Experimental avionics is breathtaking—new models and even entirely new concepts come out every six months, and no two instrument panels are alike. For many years, any single-engine GA pilot could get into just about any single-engine airplane for which he was rated and fly it with little difficulty, because the panels were much the same. Steam gauges and basic radios all worked the same way; it was easy to move from one airplane to the next. Enter today’s world of Experimental EFISes and engine monitors. Not only does it take many hours to learn how to work the various modes and controls, but no two systems work exactly the same way. While displays are becoming more standardized through natural selection and survival of the best ideas—which are then copied by other manufacturers—the methods for pilot interaction are widely varied and numerous. It takes hours to master your systems at the operator level, and that is just the beginning.
That orange exclamation point is almost certainly trying to tell us something, but without understanding the system, how would you know what it is?
EFIS technology has grown up with GPS navigation and moving maps. Some attitude reference systems rely on GPS to provide a good solution for “which way is up.” Others rely on airspeed to do the same thing. And still others don’t need external assistance, but they do need to be stationary when they are powered up. All systems have limitations that need to be observed to make them accurate and dependable. But rarely do the various systems operate all by themselves—they have to trade data back and forth. A typical GPS/EFIS/autopilot system may have three different manufacturers, and pass position and flight-plan data from the GPS to the EFIS and autopilot, or from the GPS through the EFIS to the autopilot, with the EFIS performing some modifications to the data before the autopilot receives it.
Moreover, the system might use flight-plan data derived in the EFIS instead of the GPS, so which one is the autopilot going to listen to? A builder who actually wired his own avionics will probably understand the data-flow paths well, whereas the builder who paid someone else to wire his system will only know it if he has taken the time to study it. I frequently help people set up their new EFISes (some software configuration is always required), and I see a broad spectrum of understanding. Some know every pin and wire combination, and which channel is connected to what. Others ask, “What’s a channel?” Needless to say, when you’re bumping along in the clouds on a stormy night and some of the lights go out, or the boxes disagree, that is not the time to suddenly wonder where the truth is and how the connection schemes could be failing.
Have You Heard the One About…
There is a new joke making the rounds about the three most common phrases heard in modern glass cockpits. The first is, “I didn’t know it could do that!” The second is, “What do you think it’s doing now?” And the third is, “What the heck do you think it’s going to do next?”
None of those is good news when you’re in the clag, especially while shooting an approach. This is why it is vitally important to the serious pilot to know not only how to work his equipment, but also how it works. Experimental avionics are not yet at the point where any good pilot can sit down and instantly understand how to use them. Likewise, builders need to spend time during the design process understanding how the various components are wired and how data flows from place to place. It is during this design phase that an understanding of the redundancy and backup plans is built up. Skipping the process by having someone else do it is acceptable if you have another way of obtaining this knowledge, but skipping it altogether is not a good idea when your life depends on the functionality you have in the airplane.
Captain Kirk knew what he was talking about when he said that you have to know how things work on a starship (and, I would add, an airplane). As another Star Trek character asked in yet another old movie: “Who am I to argue with the captain of the Enterprise?”