AVweb

« Back to Full Story

Probable Cause #63: Stick To The Procedure

  • E-Mail this Article
  • View Printable Article
  • Text size:

    • A
    • A
    • A
This article originally appeared in IFR Refresher, Oct. 2007.

Probable Cause

On Oct. 7, 2005, the pilot of an A36 Bonanza deviated significantly from the ILS Runway 27 approach procedure he was flying into the Pike County Airport (PBX) at Pikeville, Ky. The pilot lost control of the aircraft and crashed just south of Pikeville, killing all three people on board. The flight began many hours earlier when the pilot contacted the Louisville, Ky., Automated Flight Service Station (AFSS) just after noon (EDT), explaining his intention to fly to Pikeville from Paducah (PAH). Paducah is located in Western Kentucky along the Illinois River. Pikeville is 286 nautical miles to the east not far from the Virginia and West Virginia state lines. The Bonanza pilot asked the AFSS specialist about "weather echoes" in the eastern portion of Kentucky. The specialist explained the returns as light rain over the central portion of the route and at the destination, part of a slow-moving cold front over Eastern Kentucky. The area forecast called for a broken ceiling at 700 feet, an overcast layer at 1,500 feet, four miles visibility in light rain and mist and a north wind at five knots. Two previous automated weather reports from Pikeville showed a broken ceiling at 300 feet with three miles visibility and later 300-foot scattered with 10 miles visibility and calm winds.

No Alternative

Beech A36 Bonanza

The pilot filed an instrument flight plan, but did not file an alternate. He departed Paducah with his two passengers at 1345, and the hour-and-one-half flight to the Pikeville area was uneventful. At 1517 (EDT) the pilot checked in with Indianapolis Center at 7000 feet. Five minutes later, the controller cleared the aircraft to descend and maintain 5000 feet at the pilot's discretion. The pilot responded that he had the weather and that he wanted to fly the ILS approach to Runway 27. The weather at Pikeville at 1513 was recorded as 1700 broken and 10 miles visibility with winds 360 at three knots. At 1525, a Special was issued that indicated 1700 scattered, 2300 scattered and 10 miles visibility. This is most likely the last report the pilot received from the automatic weather system. The controller vectored the pilot for the approach at 4500 feet and cleared the Bonanza for the ILS 27 at 1546. The pilot was told to switch to Pikeville's advisory frequency, with instructions to cancel on the airborne frequency if possible or with Flight Service once on the ground. There were no more radio transmissions received from the aircraft.

Pikeville, Ky., (PBX) ILS 27 Chart (click here for full chart)

The Pikeville approach is about as simple an ILS procedure as they come, with no procedure turn on the approach and radar being a requirement. The airport elevation is 1473 feet msl. Runway 27 is 5350 feet long. The inbound course is 273 degrees and the straight-in decision height is 1664 feet (msl), 200 feet above the surface. Something was obviously wrong when, at 1550, radar indicated the Bonanza was still at 2300 feet but only one mile from the Runway 27 threshold. Shortly after that, radar indicated the aircraft turned off the localizer to the south, where contact was lost as the Bonanza descended below 1,800 feet. The weather recorded at 1544 called the winds light, with visibility of four miles beneath scattered clouds at 300 feet and a broken ceiling at 2100 feet. At 1556, shortly after the aircraft crashed, Pikeville reported 3/4 of a mile visibility, and an overcast ceiling at 200 feet. Several witnesses at the airport heard the aircraft, although no one saw the Bonanza. A Cessna Citation pilot that landed at PBX at 1530 said the ceiling was just above minimums but oscillating during the approach. He told investigators that he heard the Bonanza coming down the approach near the DH. He said it sounded like the aircraft was going south of the airport at approach speed, approach power settings, and either level or at a slight descent until he could no longer hear it.

Pilot Qualifications

The Bonanza pilot held a Private pilot certificate with airplane single-engine land and instrument ratings. The last entry in his logbook was dated July 24, 2005, four months before the accident. The pilot claimed 527 hours of flight experience and 323 hours of actual instrument time, a claim that was suspect early on. A two-page excerpt from the pilot's logbook included in the accident report shows that the pilot had logged almost all his flight time as instrument time, so it is unclear how much actual instrument experience he truly possessed. Investigators reviewed records from the maintenance shop that completed the last annual inspection on the aircraft and found that it was signed off on Feb. 22, 2005. NTSB investigators determined from the wreckage that the landing gear was extended at the time of the accident and the flaps were retracted. The NTSB's probable cause of the accident was the pilot's failure to adhere to the published instrument approach, with the low ceiling being called a factor.

The Whys

PBX Airport High View

A number of interesting questions come to mind following the review of this accident although the answers are not quite as clear. For example, the last weather report the pilot probably received indicated that the PBX weather was still VFR with good visibility. Yet, he asked for the ILS approach. Why? There might be several reasons. Perhaps the pilot planned the approach for currency, or he wanted to show one or both of his passengers how the approach equipment worked. In spite of the good weather report, he likely saw low clouds in the area and realized he might have to fly through them to get to the airport. If the Bonanza pilot was current and proficient -- and we don't know that he was -- why did he drift to the left of the localizer and wind up striking the ground a mile south of the airport? He obviously was not following the glideslope or he would never have been 700 feet above the DH one mile from the airport. Perhaps the pilot kept descending because he thought the weather was going to eventually improve. This is a common gotcha in IFR flying and probably a factor in many accidents. But he still should have realized a full deflection of the localizer and glideslope needles meant it was time for a missed approach. Could the pilot have been so intent on searching for the airport that he paid no attention to his instruments? That's probably a common mistake, too. Going visual too soon out of an approach in real IMC requires discipline to resist. Certainly the pilot had no idea he was losing control of the aircraft until it was too late. In fact, the aircraft did not descend below 1800 feet until it was in the vicinity of the accident location, and that was a mile south of the airport. Weak ILS approach or basic instrument skills could explain this entire chain of events. The Bonanza pilot had good reason to expect better weather than what existed at Pikeville. The forecast indicated that the lowest expected ceiling was 700 feet and the lowest visibility should have been 4 miles. But that is the nature of forecasts. They are not exact by any means and instrument pilots should understand that. You can mitigate unpleasant surprises from blown forecasts by frequently updating current weather while en route, but only if your destination has real-time weather reporting. All instrument pilots need to develop a solid set of personal limitations and stick to them. New IFR pilots with questionable currency have no business in this kind of weather. A cross-check of the weather reported elsewhere would also have helped. One airport, in Wise, Va., 34 miles south of Pikeville reported a 300 foot overcast and 1-3/4- miles visibility half an hour before the accident, while Jackson, Ky., about 35 miles west of PBX, reported a ceiling of 100 feet overcast and 3 miles visibility. Not filing an alternate airport may have indicated the pilot's confidence level that the weather would be fine. However, an alternate was still required. The list of what this pilot did wrong is long and proves yet again that one, or sometimes two, bad decisions do not necessarily court disaster. But as the list grows, the chances of survival grow proportionally smaller. The better -- and sooner -- you're able to recognize that, the safer you'll be.
More accident analyses are available in AVweb's Probable Cause Index. And for monthly articles about IFR flying including accident reports like this one, subscribe to AVweb's sister publication, IFR Refresher.

« Back to Full Story