NTSB Factual Report

HISTORY OF FLIGHT

On Wednesday, August 3, 1994, at 0533 eastern daylight time, a Cessna 210N, N4875U, operated by Atlantic Aero, Inc., and piloted by Philip T. Sarver, was destroyed by impact, after executing a missed approach at the Eastern West Virginia Regional Airport, Martinsburg, West Virginia. The pilot was fatally injured. Instrument meteorological conditions prevailed, and an IFR flight plan was filed. The flight was being conducted under 14 CFR Part 135.

The cargo flight was scheduled to fly from Greensboro (GSO), North Carolina, to Tullahoma (THA), Tennessee, where the cargo was to be loaded and flown to Hagerstown (HGR), Maryland. After the cargo was unloaded, N4875U was scheduled to return to GSO.

The pilot of N4875U contacted the Raleigh Automated Flight Service Station (AFSS) at 2209 and said:

...centurion four eight seven five uniform departing...greensboro in about thirty minutes ...going to tullahoma...and...i'll be going to hagerstown maryland and i'd like get a weather briefing and file three ifr's with you.

AFSS provided the weather for the flight to Tullahoma, and then advised the pilot that there was no terminal forecast for Hagerstown. The nearest terminal was for Martinsburg (MRB), West Virginia, which forecasted: 10,000 scattered, variable broken, visibility 4 miles, with fog and haze.

The pilot filed an IFR flight plan, estimating time en route to THA of 2 hours, and a fuel load of 5 hours. On his flight log, the pilot noted that the airplane had 89 gallons of fuel at departure.

N4875U completed the flight from GSO to THA, where it was loaded with cargo. Refueling records and an interview with the refueler confirmed that the airplane was serviced with 30 gallons of 100 octane aviation fuel. N4875U departed THA at 0135, destined for Hagerstown.

At 0356, while cruising at 9,000 feet, the pilot of N4875U requested the latest weather report for the Hagerstown area, from Washington Air Route Traffic Control (ARTC). The controller provided the following weather for HGR: sky partially obscured, 300 overcast, 1/2 mile visibility.

The pilot contacted ARTC at 0406 and stated that the weather at Hagerstown was deteriorating, and that he might have to divert to Martinsburg.

At 0417, ARTC advised the pilot that the Hagerstown weather was: partially obscured, 100 foot overcast with a visibility of 1/2 mile.

The pilot of N4875U stated at 0426:

...i've got good ground contact up here...in fact i can...just about pick it out...must be some fog or something drifting through there

At 0432, he said:

...i'm over the airport now and i have the runway...in sight...it looks like it's just a patch of fog...it's close enough to do a contact approach.

He circled the airport attempting to perform a contact approach or ILS, but he was unable to complete landing due to restricted visibility.

At 0457, ARTC asked the pilot of N4875U about his fuel status. The pilot responded:

...i've got two and half hours of fuel up here so i'm in good shape on the fuel...i don't know this stuff doesn't seem to be moving it's just setting right there

At 0502, the pilot requested clearance to Martinsburg. ARTC approved the change of destination and cleared N4875U to "turn left heading one niner zero proceed direct martinsburg...maintain four thousand"

The pilot contacted Dulles Approach Control and was vectored for an ILS approach to runway 26 at Martinsburg. He was unable to land on the first attempt, and he executed a missed approach. He requested a second approach and was again vectored for the ILS. He was unable to land on this second approach, and he conducted another missed approach.

The Flight Service Station Specialist on duty at the airport stated in his report:

At [0528 N4875U] called inbound ILS runway 26; [he] was given airport advisory; at [0531], I heard aircraft coming what sounded like near overhead towards rear of building, pilot called and said this isn't working. I'll have to try something else. Aircraft sounded like engine was running slow...heard a "whomp" sound and silence. I called several times - no answer.

A witness, a commercially rated pilot, stated that he heard the airplane fly over his home. It sounded to him as if the pilot was on a missed approach. He said the engine was at "full power, high RPM...and it sounded good." The witness said the weather was foggy, "near zero, zero, less than 300 feet visibility."

The pilot did not contact Dulles Approach Control, nor did he respond to several radio calls. The wreckage of the airplane was located on the runway 35 by airport and Air National Guard personnel.

The accident occurred during the hours of darkness, at latitude 39 degrees, 24 minutes North; longitude 77 degrees, 59 minutes West.

AERODROME AND GROUND FACILITIES

The published missed approach procedure for the ILS to runway 26 called for a "climb to 1000' (feet), then a climbing left turn to 3000' direct to the MRB (Martinsburg) VOR and hold." The airport elevation was 557'; therefore, an airplane would be 443' above ground level upon reaching 1000' indicated altitude, where a turn could be commenced.

The Federal Aviation Administration conducted a flight inspection of the runway 26 ILS system on August 4, 1994. The results of this flight inspection were "satisfactory."

PILOT INFORMATION

The pilot held an Airline Transport Pilot Certificate, with airplane single and multi-engine land and instrument ratings. In addition, he held a Flight Instructor Certificate with single engine and instrument airplane ratings.

Mr. Sarver was issued a First Class Airmen Medical Certificate on March 1, 1994, with limitations that he wear corrective lenses while exercising the privileges of this certificate.

His pilot's flight time log book was not recovered during the investigation. Based upon information received from his family, company and FAA records, it was estimated that he had a total flight time of about 4800 hours, with 38 hours in the last 90 days. In the last 30 days, he had flown this make and model airplane 14 hours.

Mr. Sarver was employed as a part-time pilot for the Atlantic Aero Company. He was employed full-time in a family business, which involved working during the hours from 0800 to 1730.

METEOROLOGICAL CONDITIONS

The recorded surface weather observations, at the Martinsburg Flight Service Station (FSS), for the time period prior to and after the accident, were as follows:

0150: Estimated 8000 overcast, visibility 7 miles

0247: Partial obscuration, estimated 8000 overcast, visibility 2 1/2 miles, with fog

0352: Partial obscuration, estimated 8000 overcast, visibility 2 miles, with fog

0447: Partial obscuration, 8000 thin overcast, visibility 1/2, with fog

This last observation was provided to the pilot by Dulles Approach Control, as he was being vectored to Martinsburg. At 0532, the FSS Specialist recorded the following observation, indicating in his log, "Aircraft mishap:"

Partial obscuration, estimated 8000 overcast, visibility 1/2 mile, with fog

The next observation, at 0550, reported:

Indefinite, 400 obscured, visibility 1/4 mile, with fog

WRECKAGE

The wreckage was examined at the accident site on August 3, 1994. There was an impact mark with 3 slashes, 24 inches apart, on an airport road about 1400 feet south of runway 26. Wreckage was scattered over a distance of 1100 feet. The left door was located 40 feet away from the impact mark. Ground scars along the wreckage path were on a magnetic heading of 273 degrees. Approximately 150 feet from the first mark was the oil cooler, the outer portion of the right wing tip and one propeller blade.

Continuing on the 273 degrees heading, the remainder of the wing assembly was located 400 feet from the first impact mark.

The fuselage, including the engine and empennage came to rest on runway 35. This was about 775 feet from the first impact marks and 375 feet from the wing.

The starter and the battery were located in a grassy area approximately 300 feet beyond the main wreckage.

In the fuselage, aft of the two forward seats, there was cargo consisting of plastic bags containing developed film. The load manifest listed the total weight of this cargo as 820 pounds. Investigators weighed the bags and the total weight was 838 pounds. The airplane weight and balance was checked and found to be within center of gravity limits.

The two remaining propeller blades were located to the right of the wreckage path, 290 feet and 600 feet respectively from the first impact mark. All three blades showed evidence of leading edge marks, gauges and were twisted and curled.

The wing flaps, landing gears and spoilers were retracted.

The fuel tanks were intact, and there was evidence of fuel spillage on the asphalt ramp.

Investigators accounted for all flight control surfaces.

The airplane was equipped with a dual vacuum system, which was disassembled by investigators. The insides of the pump housings showed impact marks, and the carbon vanes were shattered. The shear shaft couplings were intact.

The flight attitude indicator was disassembled and no discrepancies were noted.

The fuel tank selector was in the "Right" tank position.

The engine case was fractured near the propeller flange, and the oil sump was crushed. All engine mounts were fractured. The propeller flange and ring were bent aft. The engine driven fuel pump was examined and the drive shaft was intact and rotated.

The fuel manifold was opened and no contamination was evident on the screen.

Cockpit communication navigation and radio switches and instruments were damaged and provided no useful information.

MEDICAL AND TOXICOLOGICAL INFORMATION

An autopsy of the pilot was performed by Dr. James L. Frost, Deputy Chief Medical Examiner for the State of West Virginia, on August 8, 1994, in South Charleston, West Virginia.

Toxicological tests were performed by the State of West Virginia Office of the Chief Medical Examiner. The results of these tests were negative for alcohol, carbon monoxide and drugs.

ADDITIONAL INFORMATION

In an interview with Mr. Ralph Caldwell, the pilot's half- brother, he described a telephone conversation with Mr. Sarver, which occurred about 2000 on August 2nd. One of the items discussed at this time was Mr. Sarver's scheduled flight for that night. Mr. Sarver made no mention of any plans to sleep prior to reporting for the flight at 2200. Mr. Caldwell stated that Mr. Sarver arrived at work, on August 2nd, at 0800 and worked until about 1730.

Interviews with the pilot's family members indicated that he would often work all day at his normal, day-time position, and then fly all night, in his capacity as a part-time pilot.

The following information was excerpted from an NTSB report of an accident that occurred on August 18, 1993.

SCIENTIFIC EXAMINATION OF FATIGUE

Cumulative sleep loss: Scientific literature has established that people require a certain number of hours of sleep each day to be fully alert, typically between 6 to 10 hours depending on the individual....there is evidence that only 2 hours less sleep than is usually required by an individual can create major degradations in alertness and performance.

Issues of sleep loss have been cited by the Safety Board as issues in previous accidents.

Continuous hours of wakefulness: In a recent Safety Study in which the Board reviewed 37 major aviation accidents in which flightcrew performance was determined to be either a causal or contributing factor to the accident, it was found that one factor related to performance and judgment errors was the time that a pilot had been awake....flightcrews...whose time since awakening were determined to be elevated...made more errors overall, specifically more procedural and tactical decision errors. The study adds to the scientific evidence that fatigue problems can increase simply with lack of sleep.

Circadian disruption (time of day): Scientific literature has established that there are two periods of maximal sleepiness in a person's usual 24 hour day. These are determined by physiological fluctuations regulated by the brain, and occur between roughly 3-5 every morning and 3-5 every afternoon. During these periods, the body is primed to sleep. Individuals can remain awake during these periods, but the physiological pressure to sleep is maintained and may affect waking levels of performance and alertness. Failure to sleep during these periods, or efforts to sleep when the body is physiologically primed to be active, are labeled circadian disruption.

The Airman's Information Manual addressed the subject of pilot fatigue with the following:

1. Fatigue continues to be one of the most treacherous hazards to flight safety, as it may not be apparent to a pilot until serious errors are made. Fatigue is best described as either acute (short-term) or chronic (long-term).

2. A normal occurrence of everyday living, acute fatigue is the tiredness felt after long periods of...lack of sleep...

The airplane wreckage was released to Kyle D. Moore, Claims Manager for the United States Aviation Underwriters, on August 4, 1994.

NTSB Probable Cause

FAILURE OF THE PILOT TO MAINTAIN ADEQUATE ALTITUDE AND PROPER CLIMB DURING THE MISSED APPROACH. FACTORS RELATED TO THE ACCIDENT WERE: DARKNESS, FOG, AND CONDITIONS CONDUCIVE TO PILOT FATIGUE.