When a pilot contemplates an IFR flight, the focus will tend to shift towards the latter part of the trip. This is reasonable, as an approach is arguably the most challenging phase of any IFR flight.
This focus on approaches, enforced during initial and recurrent training, does have its drawbacks: Many pilots tend to dismiss the challenges that come with the departure phase of flight. In at least one case, this dismissal cost both the pilot and his passenger their lives.
On the evening of July 26, 2004, the pilot of a Piper Archer called the Williamsport, Pa., Automated Flight Service Station and requested an outlook briefing for a flight the next morning from the Chester County G.O. Carlson Airport (40N) in Coatesville, Pa., to Portland, Maine. The briefer explained that there would be a stationary front in the vicinity of Coatesville and that the weather was expected to be VFR in the morning with conditions deteriorating later in the day to low IFR.
In fact, a Surface Analysis chart for the period showed a large weather system that included two low-pressure areas and a warm front. The ceiling at 9 a.m. EDT was forecast to be 2000 feet overcast with four miles visibility, and the ceiling at 11:00 a.m. was forecast to be "occasionally 900 feet overcast." The pilot told the briefer that he "could shoot an approach back in if necessary," and he stated he would call back in the morning for another update.
The following morning at approximately 8 a.m. the pilot and his passenger arrived at the FBO where the Archer had been tied down since it arrived on July 24. According to FBO employees, the pilot performed his preflight inspection and ordered that the fuel tanks be "topped." Twenty-seven gallons of 100LL was pumped into the aircraft.
At 8:46 a.m. the Williamsport AFSS received another call from the pilot. He filed an IFR flight plan but did not request any weather information. It is possible that he used a weather computer at the FBO, but there is no official record that he had a briefing that morning before the flight began.
According to the NTSB, another instrument-rated pilot had a conversation with the pilot before he departed. The Archer pilot asked the witness how he would go about getting an IFR clearance on the ground. The witness pointed to a phone on a nearby counter and told the pilot to call Philadelphia Approach Control, which the pilot did. When the pilot hung up the phone, the witness told the Archer pilot that the weather appeared to be "way below standards." The pilot replied that the weather "was nice where he was going." The witness then asked the pilot if he had checked the weather radar. He told the pilot that it appeared to him that he would be flying "in this nasty stuff all the way up," but the pilot shrugged his shoulders and walked away.
When the aircraft departed Coatesville, the same witness was watching. He told investigators that the aircraft disappeared into the clouds approximately 50 feet above the runway. He said the weather at the time was foggy and rainy, and that the trees at the other end of the runway were not visible.
At 9:20 a.m., the pilot made contact with the Philadelphia Approach controller and advised that he was airborne and climbing out of 1800 feet. The controller read the pilot the Philadelphia altimeter setting and instructed the pilot to ident. Then he said, "N91075, radar contact four miles to the east of Chester County. Climb and maintain 4000, turn left [to] heading 090." The Archer pilot acknowledged the transmission and a few seconds later, Approach amended the heading to 080 degrees.
The pilot was then instructed to change frequencies, which he did. He announced to the new controller that he was climbing through 3200 feet. The controller issued a heading change to 050 degrees. A little more than a minute later the controller, seeing the aircraft passing through the 050 heading, queried the pilot. There was no answer, nor were there any more communications with the aircraft.
A review of recorded radar data showed that after takeoff the aircraft climbed on an eastbound heading to 2500 feet. It then turned to the northeast and continued to climb to 4000 feet. Approximately 30 seconds later, the target began a 360-degree turn to the right and descended slightly. It then stopped the descent while continuing a right turn and then returned to 4000 feet. Seconds later the target was seen descending through 2500 feet.
A second witness, who was inside her home at the time, said she heard the "roaring" of an airplane that was so loud that it shook the windows. She looked outside and saw the airplane in a very steep angle, "almost straight up and down," flying between two homes on her street. She said the engine was "revving" as though someone was "accelerating and decelerating." She then heard the sound of an impact.
Several other witnesses heard the airplane prior to impact. They described the engine as sounding "very loud" and running at "full throttle."
The initial impact point was the roof of a private residence where a section of the right wing imbedded itself. The wreckage path extended from the front yard of the impacted residence into the front yard of a second, adjacent residence. The wreckage was strewn on a path oriented on a heading of 095 degrees. It was about 173 feet in length. The 53-year-old pilot and his passenger were killed in the accident.
The private pilot was certified for single-engine airplanes and held an instrument rating. His flight logbook was located in the wreckage and it revealed that the Archer pilot had 560 total flying hours with 11.9 hours of actual instrument time and 114 hours of simulated instrument time. The pilot received his instrument rating on December 31, 2003, seven months before the accident. The NTSB determined that during the six-month period prior to the accident, he had logged four approaches and accumulated 4.5 hours of actual instrument time.
The aircraft's logbooks were also located in the wreckage. The aircraft had an annual inspection earlier in the month of July 2004 and no abnormalities were noted. Investigators found nothing in the wreckage that indicated there had been any failures of aircraft components prior to impact.
The National Transportation Safety Board assigned one of its meteorologists to study the weather on the day of the accident. The Surface Analysis chart showed a center of low pressure over extreme southwestern Pennsylvania. A warm front extended southeastward from the low over northern Virginia and then northeastward to another low centered over the northwestern Atlantic Ocean. Also, the chart indicated high pressure prevailed over northern New England. Station plots on the chart indicated overcast clouds, high relative humidity and patchy light precipitation over Pennsylvania, New Jersey, and New York. Scattered to broken clouds were shown in Maine.
At 8:22 a.m. the reported weather at the Pottstown Limerick Airport (KPTW), located on a heading of 012 degrees and 12 miles from the accident site, was reported as 400 feet overcast with 1-3/4 miles visibility in heavy rain and mist. The temperature was 19 degrees C and the dew point was the same. The altimeter setting was 30.05 inches.
At 9:19 a.m., eleven minutes before the accident occurred, the weather at Pottstown was reported as 400 feet overcast with four miles visibility in light rain and mist. The temperature and dew point were still at 19 degrees C and the altimeter setting was 30.04 inches.
Pilot reports for the area indicated that the tops of the clouds were at 4000 feet with higher layers beginning at around 9000 feet.
The Area Forecast for Pennsylvania that was valid until 6 p.m. called for marginal VFR and IFR conditions throughout the day in precipitation and mist. The forecast for Maine, the pilot's destination, included VFR conditions.
There are a number of factors that make a similar accident preventable in the future. First, the pilot was not legally current for the flight. FAR 61.57 states that no pilot may act as pilot in command under IFR, or fly in weather conditions that are less than the minimums prescribed for VFR flight, unless he has met the recent experience requirements. They include six instrument approaches in the past six months, holding procedures and intercepting and tracking courses through the use of navigation systems. In lieu of these requirements, a pilot can also pass an instrument proficiency check (IPC) with an instructor or examiner.
Based on what the NTSB found, the Archer pilot did not meet those requirements.
Besides, just because a pilot barely meets the stated minimums does not mean he is proficient for the flight. It would appear that this pilot was not, given that he had trouble maintaining a heading and climb configuration at the same time. He did reach his assigned altitude twice, but both times he failed to maintain it.
Too many pilots are overconfident in their abilities. Based on the recollections of the first witness, perhaps this pilot fell into that category. When queried by the witness before the flight began, the Archer pilot was flippant. Sure, the weather was better where he was going, but that wasn't the issue. The issue the other pilot was trying to point out was that he was going to have to take off and fly through some serious weather before reaching his destination and better conditions.
The Archer pilot told the FSS briefer the night before the accident flight that he could always turn around and fly the approach back into Chester County, if need be. He may have truly believed that, but in practice it was not going to happen. If he couldn't maintain climbing, turning flight, then it's doubtful he could have flown an ILS or a non-precision approach into an airport that reported a low IFR ceiling.
So, how do you know if you are properly prepared for an instrument flight? First, you have to face reality when it comes to legal currency and real proficiency. Just because you go out with a safety pilot and fly six approaches in VFR weather two days before a scheduled flight does not necessarily mean that you have the proficiency to fly a long cross country trip in the clouds.
Years ago I flew with a pilot in a Baron who, though he had an instrument rating, each time he got himself into the clouds he would turn on the autopilot. Approaching Fort Myers, Fla., one evening I asked him to leave the autopilot off and fly the approach manually. He couldn't do it. The autopilot came back on very quickly. The next day on the return trip to Fort Lauderdale I asked him to manually fly the back-course approach in VFR conditions. He couldn't do that either. He became very confused with the backward movements of the localizer needle even though the runway was right in front of us.
Fort Lauderdale was his home airport, and the back course approach was used routinely. When we got on the ground I asked him what he would do if the autopilot failed and he had to fly that approach manually. He said he thought he should get himself some more dual instruction before he flew in IFR conditions again.
The NTSB cited the pilot's failure to maintain aircraft control in instrument meteorological conditions, which resulted in the airplane impacting a residence. Factors in the accident were the pilot's lack of recent instrument experience.
So, don't fool yourself into believing that nothing will happen on your next flight. Even if you are reasonably proficient, the loss of the attitude indicator or direction indicator could reduce your confidence level very quickly. And you certainly don't want to find out while you are airborne that you are not up to the task of flying your airplane through the clouds. It appears that is the mistake the Archer pilot made.
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.