Photo: Two VC-25s (USAF Photo)
As pilots for the 89th Airlift Wing at Andrews Air Force Base (now Joint Base Andrews), we had hammered into our heads the idea of “Safety, Comfort, Reliability.” We were to do everything possible to provide our passengers a reliable trip, but we were not allowed to sacrifice their comfort to make that happen. Similarly, we were told to ensure they had a comfortable ride, but were forbidden to sacrifice safety. So our unofficial motto became “safety, comfort, reliability – in that order!” This code of conduct should be familiar to most business and commercial aviation crews. But there is a problem with it.
Grammatically, you can think of those commas to stand for a sequential order. Be safe, then provide a comfortable ride, and then look out for the schedule. But each of these factors tends to be interrelated and we sometimes find ourselves thinking about all three but prioritizing the wrong element. As with many things in aviation, a mathematical treatment can lend some clarity. Our motto might be better expressed as:
“Safety > Comfort > Reliability.”
The preceding elements are more important than the subsequent. But theory is often overwhelmed by reality. I cover the various interpretations of this motto in Flight Lessons 3: Experience, the third book of that series. The danger for crews at the 89th and for us in commercial aviation, is that we can be corrupted into thinking of the motto as "Reliability, Reliability, Reliability."
Broken down to its simplest form, the job of any pilot is to takeoff, land, and to do both of these and everything in between as safely as possible. We are often judged as aviation professionals by our ability to leave here and land there as scheduled; and therein lies the temptation to invert the safety > reliability relationship. This pressure exists on air carriers as well as business and corporate aviators; the NTSB database is filled with countless examples. Examining a few can help illustrate how even highly respected pilots can place reliability over safety.
Photo: Gust lock handle, from Eddie's aircraft.
Some pilots have the ill-conceived notion that faster means better in all aspects of aviation, even if that means skipping a long list of required pre-takeoff checks. The 2014 crash of a Gulfstream IV from Hanscom Field in Bedford, Massachusetts left the industry stunned that two professional pilots would intentionally skip a required flight control check prior to takeoff, a step that would have revealed that they had forgotten to disengage their flight control gust lock prior to engine start. The National Business Aviation Association conducted a review of 379 corporate aircraft from 2013 through 2015, examining over 144,000 flights for adherence to required flight control checks prior to takeoff. In 16 percent of the takeoffs pilots did only a partial check. In 2 percent of the takeoffs, pilots failed to perform any check of the flight controls at all. A proper flight control check can be accomplished in less than half a minute in most aircraft; can the need to save so little time really have corrupted so many pilots?
Photo: Air Florida 90, from "AirDisaster.com"
Even when the time saved is much greater, the schedule cannot overrule safety. In the case of the 1982 crash of Air Florida Flight 90 into the Potomac River, the pilot was loath to return to the gate for deicing. That could have delayed them an hour or more. Not only were the wings contaminated with snow, the aircraft’s left engine pressure probe was blocked. That caused the engine pressure ratio indicator to misreport thrust settings and the crew to attempt the takeoff with insufficient thrust. The first officer noted, “ . . . that’s not right . . .” early during the takeoff but the captain convinced him otherwise. Rejecting the takeoff would have delayed the trip even longer; but continuing the takeoff cost 74 lives. Only 1 crewmember and 4 passengers were fished out alive from the icy Potomac.
Photo: Air Illinois 70, from Aviation Safety Net
Even after you’ve made it off the ground, the pressure to achieve the all-important schedule continues. When faced with an aircraft malfunction we often decide we can press on to our destination, overflying viable airports. In 1983, all ten occupants of Air Illinois Airlines Flight 710 were killed when the captain made such a decision. One of the aircraft’s generators failed shortly after takeoff and the first officer erroneously isolated the remaining generator, causing it to fail as well. As this point the captain could have returned to their departure airport in VFR conditions in less than 6 minutes. He elected to continue to his destination in IFR conditions using only battery power. They didn’t make it and everyone on board was killed.
Photo: Airborne view of runway 15, from NTSB Accident Brief, Figure 2
The greatest temptation to make it to one’s destination may occur just 200 feet above the runway’s surface; going missed approach to the alternate can delay passengers by hours or even days. While this pressure certainly exists in every airline cockpit, it is nowhere greater than on a business jet with the owner or principle passenger in the jump seat. The crew of Gulfstream III N303GA made many procedural errors during their instrument approach into Aspen Airport, Colorado (KASE) that appear to defy logic. Why would an experienced crew execute an instrument approach into mountainous terrain and then descend below the minimum descent altitude without visual contact with the runway or its environment? There is no doubt the charter customer placed pressure on the captain following their delayed departure; but the captain’s ultimate responsibility is to ensure safety concerns outweigh the need to prove the schedule can be counted upon as reliable.
There is an old saying in corporate aviation, “You pay me to say no.” Anyone can say yes to the pressures of having to make a less than airworthy airplane fly, to make a schedule departure time despite weather and other external factors, and to land at the requested destination despite fuel, weather, and other concerns. But it takes a consummate professional to always keep in mind that safety > reliability. But how does one beat back the pressures sure to come when things don’t go strictly to plan?
While these Reliability > Safety miscalculations have brought down many airplanes over the years, the solution would seem straightforward: follow all published guidance and make commonsense decisions. Pilots often fall prey to a more insidious error, striving to provide the greatest comfort at the expense of safety. Most pilots will tell you they would never do that, but what then explains the tendency to land long?
We professional pilots fall into two categories when along for the ride in someone else’s cabin: disinterested “too cool to care” aviators or white knuckle, “Nervous Nellies.” I was a passenger on a commercial airline flight into LaGuardia a year ago that turned me from the former to the latter. I will never again be able to sleep through an airline pilot’s landing as a result.
Photo: The departure end of LaGuardia's Runway 31, from Google Earth
I was sitting on the right side of the airplane, just forward of the wing for our approach into LaGuardia, doing my best to appear to be a nonchalant passenger. “Yeah, I’m a pilot,” my look was designed to telegraph. “I’ve done this approach a hundred times and it is no big deal.” Every now and then I would steal a glance out the window and deduced we were on the ILS to Runway 31. Gear, flaps, and engine pitch told me we were established on the glide path. When the throttles came to idle my eyes were shut, feigning the sleep of a weary traveller. And then . . . nothing. As a pilot, I shoot for idle thrust at touchdown but don’t mind another pilot getting the throttles a little early. But I would never tolerate a five second flare. I opened my eyes just as the wheels kissed the pavement and I saw the single-bar fixed-distance marker disappear under our wing.
It wasn’t as bad as I had imagined, I thought. The single-bars are 2,500’ down the runway so it was a long landing; it was terrible, but not worth my elevated blood pressure. But then I saw the double-bar fixed distance marker.
We hadn’t touched down just under 2,500’ from the approach end of Runway 31, but with just over 2,500’ remaining from the departure end. The runway is 7,003’ long so our long landing was 4,500’ from the approach end.
Predictably, the pilot gave the aircraft all the brakes he could muster and I could feel the anti-skid system doing its best to keep the tires from turning into a pool of liquefied rubber. The airplane came to an unceremonious stop and we made a 90-degree turn to the left, our wing tracing a line over the opposite runway’s edge and giving us on the right side of the aircraft a close-up view of Flushing Bay. My thoughts ran back to three airliners that ended up in the water here, all due to pilot error. An Eastern Airlines flight in 1945 failed to stop after an approach that was too high and too fast. A USAir flight crashed after the pilot was unable to keep the airplane on the runway during takeoff because his rudder trim was miss-set. In 1992 another USAir failed to takeoff because it was not properly de-iced. And now, in 2015, we almost added another airplane to the list.
As we passengers deplaned the captain stood at the entry door, beaming his major airline smile, ready to accept the accolades for his “grease job” landing. I looked him in the eye and said, “I know where you touched down and I don’t appreciate it.” He diverted his eyes to his shoes and said, “I know, I’m sorry.”
Photo: A Boeing 777 lands at Washington Dulles International Airport, from Eddie's camera.
Some pilots pride themselves on the smoothest possible touchdown – what else does the passenger have to judge a pilot on? – while paying lip service for the need to always fly a stabilized approach and to land in the touchdown zone on speed. But they often end up much further down the runway and much slower than a minimum safe flying speed.
But this touchdown roulette isn’t the only gamble some pilots make in an effort to provide that extra level of comfort. An excessively slow rate of takeoff rotation can completely negate obstacle clearance planning. Routinely selecting “half-bank” autopilot turns can place an airplane outside of protected airspace. Choosing a lower than optimal altitude to improve cabin pressurization puts an airplane in denser air traffic. Forgoing reverse thrust after touchdown to keep noise levels down not only wears brakes, but also gives them more work to do when they are needed the most.
Pilots who came of age fighting for that smooth “grease job” landing will argue that touching down at a slower speed actually helps the brakes and reduces component wear from the jolt of a firmer arrival. They neglect to consider that an airplane is at its most vulnerable in the flare with the engines at idle and the angle of attack high and getting higher. One gust of wind can drop a wing in an instance and the engine spool up time can make an escape questionable. But won’t passengers object?
Let’s consider a more earthbound example to fully understand what smoothness really is. If a motorcyclist is riding along a marked course and passes the 336 foot mark 2 seconds after the 168 foot mark, his progress can be plotted by subtracting distance markers and dividing the result by the elapsed time. In our example, the biker is doing (336 – 168) / 2 = 84 feet per second. On a graph this is known as the slope of the plot. This “distance divided by time” is the pure definition of velocity.
Now if it took the biker 10 seconds to go from a standing start to that velocity, and if the biker went from a standing start to that velocity smoothly, his increasing velocity can also be plotted graphically. If we were to take a snap shot of the bike’s speedometer at two instances and divide that by the elapsed time, we will have the acceleration. In this case, the bike is accelerating (34 – 17) / 2 = 8.5 feet per second per second. (His velocity is increasing 8.5 feet per second, every second.)
But our debate is over smoothness; wouldn’t it be nice to have a similar mathematical description akin to velocity or acceleration? We have just such a thing and it is called “jerk.” (Note: deceleration is negative-acceleration, but it is acceleration nonetheless.)
We can take snapshots of the bike’s acceleration and divide that by the elapsed time yet again. Let’s say our biker spots an obstacle on the road and decides to slow down a little at first, and then a lot, and then rethinks it and eases up on the brakes again. This on and off braking will end up with a jerky deceleration that can even be seen in the motorcyclist’s heading bobbing fore and aft.
There is a direct correlation with our motorcyclist’s braking and that of an airplane. You can set the airplane down and apply a constant brake pedal pressure to start the deceleration. As the brakes heat up and become more effective they will increase the rate of deceleration. From the airplane’s cabin there will not be any “jerk” because the rate of increase in the deceleration is constant.
The micro-lesson from this brief sojourn into physics is that you can have a firm landing followed by a smooth deceleration and still have the passengers think nothing of the experience. (In the end, isn’t that what we want?) Or you can gamble on your touchdown point and risk having a jerky braking effort to follow.
The macro-lesson applies to all facets of aviation, not just the landing phase. Your efforts to make the act of flying seem smooth also risks making the act of defying gravity more risky. A safer approach is to learn to fly the aircraft as safely as possible while understanding where smooth works in your favor or only to increase the risks.
While I was a pilot for the 89th Airlift Wing we had a cynical motto for those pilots who lost sight of “Safety, Comfort, Reliability.” These pilots, we said, believed in “Reliability, Reliability, Reliability.” It is a mindset we are all susceptible to adopting. If in a long career you have never canceled a flight due to a mechanical issue or have never gone missed approach in the weather, it could be that you have been very lucky. Or it could be that your math is wrong. It is never too late for a refresher: safety > comfort > reliability.
Much of this concept is covered in the book Flight Lessons 3: Experience.
NTSB Aircraft Accident Brief, AAB-02/03, Avjet Corporation, Gulfstream III, N303GA, Aspen, Colorado, March 29, 2001