By Joseph E. (Jeb) Burnside
The National Transportation Safety Board (NTSB) has a long list of canned probable-cause categories it uses to keep track of the various ways pilots bend airplanes. Of keen interest to Instrument-rated pilots should be the open-ended category improper IFR, which the Safety Board frequently uses in deciding what happened during an accident involving an aircraft operating on an IFR clearance. Basically, the NTSB is saying the pilot didnt follow some procedure or another and, if he had, the flight would have had a safe conclusion and wouldnt have been on its radar screen.
But what kinds of actions and mistakes does the NTSB consider improper and how can pilots learn from the accident record? What can Instrument-rated pilots do to avoid seeing their mistakes-and we all make mistakes-categorized this way by the NTSB? To answer these questions, we looked at 36 accidents for which the NTSB has issued a probable cause including improper IFR. We looked closely at what happened in those accidents to come up with five basic errors pilots are making. Strap in and hold on-this could be a bumpy ride.
1. Inadequate Equipment
After decades of looking at what can only be described as stupid pilot tricks, this one surprised even us. Basically, what were talking about here is pilots trying to shoot an approach in an airplane lacking a piece of equipment required for the procedure.
On September 28, 1990, the pilot of a Cessna 182 and his passenger died trying to get into the Nantucket Memorial Airport (ACK) at Nantucket, Mass.The flight had been cleared for a second attempt at the ILS Runway 24 approach after the pilot reported an autopilot problem and declared a missed approach on the first try. On the second try, the Skylane crossed the outer marker at 800 feet instead of the published altitude of 1800 feet msl.Shortly after the pilot reported crossing the outer marker, radio contact was lost. The NTSB found that the Skylane was not equipped with a glideslope receiver. Whether the pilot intended to use the localizer-only minimums isnt known. Neither is the weather-the NTSBs report doesnt include that information. But, since there is little difference between the standard ILS minimums of 200 feet and the present localizer-only minimum descent altitude of 372 feet agl at ACK, we can presume the weather pretty much sucked.
Its one thing to leave a final approach fix with everything working and discover, for example, that the glideslope receiver isnt doing its job, or that the DME needed to identify a step-down fix is on vacation. In those circumstances, reverting to the higher minimums or executing a missed approach is both normal and required. But its another thing to knowingly accept a clearance for an approach requiring missing or inoperative equipment and then trying to make it up as you go.
The reason we have flight instruments in the panel and go to great lengths learning how they are used is because its impossible to control an aircraft when theres no natural horizon. Yet pilots seem to regularly forget this and try to get by without using the skills theyve been taught. When theres no discernable horizon, its easy to become disoriented.
Our example accident occurred on November 19, 1995, when an almost-new Beech 58 Baron crashed shortly after taking off from the Burke Lakefront Airport (BKL), in Cleveland, Ohio. The Instrument-rated pilot and his four passengers took off in night visual conditions and turned out over Lake Erie.The airplane crashed into the water approximately six miles from the departure runway. Two passengers survived with serious injuries; the pilot and one passenger are missing and presumed dead. The fifth occupants body was found in April 1996, near Erie, Penn.
According to the NTSB, a Gulfstream IV departed a few minutes after the Baron. Its pilot later stated that the 8000-foot overcast and lack of lights on the ground was very disorienting and he was forced to use the flight instruments to fly the airplane until climbing above the cloud layer. The Baron pilot didnt react quickly enough and never had the opportunity to get above the clouds.
In another example, all four aboard a Mooney M20J died on March 24, 1993, when its pilot lost control while attempting to execute a night localizer back course approach to an airport in Oklahoma. The pilot had requested the approach after attempting to visually land at the airport and was vectored to final and flew through the localizer. He acknowledged that he was on the correct frequency. He was observed turning and flying away from the course line just prior to the loss of control. He had stated that he was below the clouds prior to attempting the approach.
Even when the weather is VFR, it doesnt mean we can control the airplane by looking outside. Pilots taking off can be especially susceptible to disorientation because the physiological effects of acceleration can make us perceive the airplane is climbing faster than it really is. Too, the transition to instruments after a visual takeoff can take some time.
And, in our haste to get visual so we can land without shooting the approach, we often forget that we still have to fly the airplane using the instruments until its on the runway.
We have frequently highlighted the critical impact fatigue and other adverse physiological conditions can have on the safe outcome of a flight. Illness, hypoxia and medications can drastically affect not only a pilots skills but also his or her judgment. Perhaps most pervasive, however, is the impact fatigue can have on pilot performance.
On November 9, 1992, at 0144 local time, a Bellanca 17-31ATC crashed while maneuvering to land at the Dublin Municipal Airport (9FO) in Dublin, Texas.Both aboard the airplane were fatally injured.
Since 9FO does not have an instrument approach, the flight was cleared to cruise at 3000 feet; while at 2100 feet, the pilot reported he was still in instrument conditions. Shortly thereafter, witnesses reported hearing the airplane descending at what they called maximum powerand then striking a telephone pole. The pilot operated a 24-hour, seven-days-a-week veterinary clinic and had worked a consecutive three-day shift from 0800 to 2400 before leaving the clinic for the departure airport.
Fatigues effects can be insidious. For one thing, being tired can force us to rationalize our way into doing something we might not otherwise do, like failing to update our weather information or forgetting to perform a checklist item. But the main effect of fatigue is to sharply reduce pilot performance, meaning our decision-making ability and our motor skills are not as sharp as they should be. This can result in shading a minimum descent altitude, for example, or failing to closely track a localizer or glideslope needle. Another result can be failure to understand what our instruments tell us-that the airplane is not in the proper climb attitude, for example-and translate that information into the appropriate control movement.
Of course, another effect of fatigue is quite literally falling asleep at the wheel. The NTSBs accident database contains more than a few reports of crashes for unknown reasons involving pilots who were known to take catnaps at the controls. Too often, we try to fly after a long day at the office and, for one reason or another, whether IFR or VFR, end up in the weeds.
In this day and age, theres no good reason for a pilot to be surprised by unforecast weather, whether en route or at a destination. While we frequently grouse about busted forecasts, often were really complaining that the promised 350-foot ceiling dropped to 200 feet without our approval. At the same time, our skills at evaluating what those reports and forecasts say-reading between the lines-hasnt kept up.
On May 16, 1992, a Cessna 172 crashed after running off the end of the runway at Caldwell, N.J. The pilot and passenger were seriously injured.Shortly after taking off from Caldwell, according to the NTSB, the pilot told controllers he was having control problems and wanted to return. On the subsequent localizer approach to Runway 22, the Skyhawk deviated left of the final approach course. When the pilot sighted the runway, it was off his right side, and about 600 to 700 feet below. He banked steeply right, lowered flaps 10 degrees and sideslipped the airplane to lose altitude. Witnesses reported the airplane touched down with only about 200 feet of the runway remaining. According to his logbook, the pilot had obtained his Instrument rating more than six months before the accident and had not logged any actual instrument flight time since.
While this accident involves a botched VFR maneuver-landing-it has its roots in the pilots inexperience. Well never know if there was a problem with the airplanes control system but the smart money is on an inexperienced pilot who failed to properly plan for his first real-world IFR excursion after getting the rating and got scared. The NTSBs accident database has many more examples of pilots who either werent current or simply didnt have enough experience to handle the challenges they faced.
5. Descent Below MDA/DH
Improper IFR often can be distilled down to failure to follow all or part of a prescribed procedure. When the failure involves descending below the minimum descent altitude (MDA) on a non-precision approach or the decision height (DH) on an ILS, bad things can happen. Simply put, theres a reason-obstacles, navigation signal, altimetry, aircraft performance-that approach procedures have a specified altitude below which we shouldnt descend.
One example came on December 2, 1993, when a Piper PA-44-180 was substantially damaged at the culmination of an approach to the Hopkinsville-Christian County Airport, Hopkinsville, Kentucky. The pilot and two passengers were not injured. Weather reported by the fields AWOS included a partial obscuration and 300-foot broken ceiling, a 1000-foot overcast ceiling and -mile visibility. The published MDA for the approach at the time was 500 feet. While being vectored for the localizer approach to Runway 26, the airplane struck power lines and trees, and came to rest 2481 feet from the runways threshold. The distance to the runway from the severed power lines was 4441 feet.
Another, more recent example involved the solo pilot of a Beech A36 Bonanza. On March 7, 2003, he died while making his third attempt at the GPS Runway 10 approach to the Baldwin County Airport in Milledgeville, Ga. The aircraft was 6.6 miles out, at 1200 feet msl, when radar contact was lost.Nearby weather reports included overcast clouds at 300 feet agl. The MDA was 900 feet msl, or 516 feet agl.
Most flights we conduct under IFR are in severe clear. A few of them involve a takeoff or landing through a low overcast to VFR on top. Rare is the several-hour IFR flight in IMC from takeoff to the missed approach point. Is IFR more demanding than VFR? When the weather is bad, unquestionably. But the procedures and predictability of IFR usually make it the safer bet.
Until we are tired, cut corners, get into a situation over our head, dont have the required equipment aboard or just plain ignore the rules and procedures, thinking Im good enough to bend a few rules and handle this little bit of weather.
As the accident record shows, there used to be a lot of IFR pilots out there who thought it couldnt happen to them.
-The author is Editor-In-Chief of Aviation Safety.