Although you hate to admit it, some pilots just seem to be asking for it. They fly like drunken outlaw motorcyclists, always dodging regs and cutting corners. You can usually spot their aircraft by the duct tape on the landing gear. When these guys crash their airplanes, you shake your head knowingly and say, What the hell was he thinking, anyway?
At the other end of the scale, there are those times when a highly trained, proficient, conscientious pilot augers one in. When it happens to airliners the feds spare no effort until they get to a conclusion the experts can live with. But among general aviation crashes, the scorched earth approach to accident investigation falls victim to too little budget and too little information about what happened inside the airplane at the critical moment. Thats the price you pay for not buying a flight data recorder or cockpit voice recorder for your personal transport.
The investigations of these accidents often raise more questions than they answer. The survivors wonder what went wrong. The regulators ponder whether to issue an AD to save face. Other pilots struggle with the question: If that guy wasnt safe, what makes me think I can do better?
The Best Laid Plans
One December evening a King Air pilot was planning a flight from Lewisburg, W.Va., to Concord, N.C. The 149-nm flight would take the big pressurized twin about 35 minutes enroute. The pilots preflight weather briefing consisted of asking for the present weather at Charlotte, about 20 miles from Concord. Shortly after 10 pm he was given the information that winds were calm, mist, ceiling of 400 overcast, and temperature and dew point both 8 degrees C.
The flight was soon airborne, heading almost due south at 12,000 feet. At 10:30 p.m. local time he was cleared to descend in preparation for the approach to his destination. At that point, while on the frequency with Charlotte Departure, he had his first conversation with controllers about the low ceilings in the area.
Departure: I had a couple go in about, ah well, I guess about a half hour ago. Picked it up about four hundred foot.
King Air: OK, well, hopefully well get it there. Calling it a thousand overcast and one and a half now. (The reference to the 1,000 overcast apparently reflected the 400-foot ceiling and the 690-foot field elevation.)
Departure: OK. Charlotte has dropped quite a bit here in the last, ah, half hour, so I dont know if its the same there or not.
King Air: We can only hope.
Less than a minute later, the flight was cleared for the ILS runway 2 approach to Concord and told to switch to the advisory frequency. A few minutes later, the pilot was back with departure, reporting a missed approach. He requested radar vectors to try again. While turning around for the second try, he asked for the latest ceilings and visibility at Charlotte, 21 miles to the southwest, and Greensboro, 53 miles to the northeast. Charlotte reported indefinite ceiling zero and a quarter-mile visibility in fog. Greensboro reported a 900-foot ceiling and six miles, but the controller mistakenly told the pilot 500 and six. Once again he missed the approach at Concord.
At that point, the pilot asked to go to Greensboro and was cleared direct. As it approached 11 pm, he asked again about Charlotte weather and apparently decided Charlotte was much more convenient than Greensboro.
King Air: Ah, Charlotte would be easier for us if, ah, you got some guys getting in there.
Departure: Ah, we got some people getting in. You want to come to Charlotte now, you say?
King Air: You betcha if you can vector us over there. A lot closer.
Departure: Youre cleared to Charlotte via radar vectors, expect runway 36R.
A minute and a half later, the departure controller announced to all aircraft on the frequency that the runway visual range on 36R had just dropped to 1,800 feet and everyone should expect 36L instead, which also reported an RVR of 1,800 feet. The King Air pilot was told he would be No. 2 behind a Boeing 737, with a Learjet landing on 36R in between. As the aircraft got closer, the controller confirmed that the pilot would be landing on 36L.
The King Air was then switched to the tower frequency. He reported in on the frequency, but then vanished.
The tower controllers struggled to locate the airplane in the fog. An hour later, it was found crashed less than -mile from the runway, but in between the two parallel runways. The pilot was killed. The sole passenger, the president of the charter company that owned the airplane, was seriously injured.
The pilot was highly qualified, with 14,000 hours, type ratings in six different kinds of jets, nearly 3,000 hours of actual instrument time and 3,300 hours of night flying. He held ATP and instructor certificates and recently had completed a FlightSafety professional training course for Learjets. The night before the accident, he had flown an ILS in deteriorating weather as part of his continuing training.
The company president, also an ATP, was seated in the cabin of the airplane on its flight back from Lewisburg. After the first missed approach at Concord, he poked his head into the cockpit and offered assistance. He handled radio communications and helped the pilot review the approach. He then returned to his seat. After the second missed approach and the rerouting to Charlotte, the president again made radio communications and helped the pilot brief the upcoming approach to Charlotte. Despite the switch in runways, the pilot assured him everything was set up for the approach.
The president returned to his seat in the cabin and looked back into the cockpit one more time. The localizer was centered and the glideslope was beginning to come in.
Suddenly, I felt the aircraft make a substantial right turn. I looked up in the cockpit and saw the CDI was close to full-scale deflection to the right. Simultaneously I saw the altimeter indicating approximately 900-1,000 feet. I yelled from the back, Go around! he told investigators. I then heard something strike the left side of the aircraft and I remember thinking, Oh my God, were going to crash.
The Plot Thickens
The next day the FAA conducted extensive investigations of the localizer, marker beacon and glideslope transmitters, and concluded that all were operating normally.
The navigation radios and the HSI from the King Air also were examined. Damage from the impact was substantial, and technicians were unable to determine if all of the equipment was operating properly before the crash. Of the defects found, however, none would have accounted for a sudden fly right indication the passenger noted given the aircrafts position. What about the possibility that the passenger, who banged his head during the crash, was wrong about the CDI deflection? The approach controller who handled the flight told the tower controller just after the aircraft disappeared from view, That King Air, right at the end, went toward the right, you know. He kind of dove toward 36R, you know.
That last-minute dive, coupled with his own observations, led the passenger to dispute the NTSBs finding that the pilot was at fault for proceeding below decision height without the runway in sight and failing to execute a missed approach. He tied his opinion to other facts, as well.
I personally saw at least a portion of the approach, and at no time did it appear to be an unstabilized approach, nor did I note any deviation in the CDI until the very last moment, he wrote to the NTSB investigator in charge. Based on everything I know … I do not believe that this accident can simply be written off to pilot error.
In addition, he said that inspectors at the scene of the accident said there had been numerous problems with the ILS ground components in the two weeks prior to the accident, and one person even made the comment that the FAA may have to eat this one. Interestingly, the records maintenance personnel supplied to the NTSB did not reveal any significant problems that may have caused the CDI on board the airplane to suddenly show a full scale deflection.
At the same time, USAirways went on record as stating its pilots had encountered three instances within a few months of the accident where the localizer for runway 36L did not appear to work correctly. The airline only reported the problems internally and, because they did not recur, neither the airplane equipment nor the airport stations were examined.
The Road to Ruin
Without more information, there is no concrete, objective way to dispute the final NTSB finding. From the cynics point of view, the owner of a charter operation would be motivated to try to eliminate pilot error from an incident report. The maintenance technicians who worked on the localizer ground equipment at the airport would not want to be blamed for improper or inadequate maintenance.
Or maybe there really is a mysterious electronic ailment that occasionally crops up there. Some electrical problems are notoriously hard to find. The bottom line, however, is that there are several reasons the pilot of the King Air may have failed at his last landing. While none is a sure-fire recipe for disaster for such an experienced pilot, they illustrate the problem mind-set to which any pilot can succumb.
The first link in the accident chain obviously was the weather briefing – or rather lack of it. The fact that the temperature and dew point were identical with calm winds should have sent up the red flag of fog.
Perhaps it did, but the pilot trusted his equipment and his experience to make the right call once he got to the destination. With a flight time of only 35 minutes, the pilot must not have felt compelled to examine forecasts or weather at other nearby airports.
Even when the Charlotte controller mentioned that weather was deteriorating, the King Air pilot seemed unconcerned about the low ceilings and poor visibility.
Another important factor was the pilots passenger. The pilot had been hired only five weeks before the accident.
Although the president is also the chief pilot and had conducted the accident pilots initial checkouts and training, human nature says you should put your boss on the ground where he wants to be. At the time of the accident, the airplane already had been in the air twice as long as the pilot had planned. His predisposition was to land rather than fly to yet another alternate, passing the initial destination along the way.
As part of the preconceived notion that the approach ultimately would be successful, the pilot perhaps forgot an important element. Recall early flight lessons where, as you approached a VOR, for example, the instructor would say, Stay with your heading as the CDI gets erratic. That means youre close to station passage.
After thousands of ILS approaches, his tendency was to go for the needles rather than trust the heading and attitude that got them centered in the first place. The pilot should have been ready, when the CDI suddenly deflected, to add power, climb and miss the approach, because nothing other than equipment failure could have explained the change in readings.
The King Air accident will likely never be resolved to everyones satisfaction, because the pieces just dont add up. But from the crash comes the important lesson that, as a fictional instructor once told a fictional Top Gun, Its better to retire and save your aircraft than to push a bad position.
-by Ken Ibold
