Probable Cause #57: Out Of Control


This article originally appeared in IFR Refresher, June 2007.

Professional pilots have been hounding the FAA for years to increase their maximum retirement age past 60 for a number of reasons. [And the increase finally happened, although after this article was originally written.] One reason is financial: More time on the job translates into larger paychecks and fatter retirement benefits.The other is that no medical research currently proves pilots past 60 pose any significant threat to themselves or their passengers while in command of an aircraft. Anecdotally, however, even pilots admit their reaction times are slower at 60 than they were at 40 or even 50, not to mention an increase in the overall number of aches and pains they experience.However, the age-60 restriction applies only to pilots flying in revenue service, such as Part 121 scheduled airlines. Part 91 pilots can continue flying as long as they can successfully pass the required FAA medical, an exam that becomes more stringent as the applicant ages. To exercise the privileges of an airline transport pilot, airmen must have passed their medical exam within the preceding six months. At 35, for example, a First Class medical also requires a pilot to successfully complete an electrocardiogram (EKG). After age 40, a First Class medical requires an annual EKG. These tests offer an FAA physician a glimpse of the pilot’s cardiac rhythm at a moment in time. As we age, there are simply more opportunities for our bodies to react suddenly when an internal part breaks down.


On the morning of Jan. 18, 2000, a King Air Model C-90, on a Part-91, single-pilot, IFR flight from Columbus, Ohio, crashed into the guy wire of a 460-foot AGL radio tower on approach to the Somerset-Pulaski County–J.T. Wilson Field Airport (KSME) in south-central Kentucky. The pilot and four passengers were killed in the subsequent crash. Somerset weather at the time of the crash was IFR, but with good surface visibility — not what might be considered a challenge for the PIC who had logged nearly 20,000 hours of flight time during his career. During the post-crash investigation, discrepancies in the pilot’s total flight time logged were noticed, but were not significant enough to most likely relate to the crash.A contributing factor to the accident was that ATC cleared the pilot for a simplified directional facility (SDF) approach at Somerset, a procedure that had not been active for nearly four years before the crash. Placement of the out-of-service notice for the SDF approach would have been difficult for most pilots to find without some research, because the time frame was considerably past the normal 56-day period when NOTAMs are kept in the electronic distribution database. A localizer and two GPS approaches were subsequently certified for use at KSME.The NTSB listed the cause of the accident as the “failure of the pilot to follow his approach clearance and subsequent descent into unprotected airspace, which resulted in a collision with the [radio] guy wire. Factors were the failure of air traffic control to verify the approach he cleared the pilot to conduct was in service and the clouds, which restricted the visibility of the communications antenna.” The radio tower was just over three miles to the southeast of the Somerset airport and clearly marked as an obstruction on all instrument approach plates.Although the sole pilot on board was 72 years of age, he had passed a First Class medical exam the previous April. However, since more than six months had elapsed since the date of that exam, his medical would have automatically been downgraded to a Second Class certificate. This change made the pilot technically illegal to exercise the privileges of an airline transport pilot. But, since this was a Part 91 flight, he would have been legal to fly as a commercial pilot, for which his downgraded First Class medical certificate would have sufficed. The NTSB report did not indicate whether this was the first time the pilot had allowed his medical to automatically downgrade based upon the expiration date. The report also did not note any previous history of pilot medical issues.



The flight began on the morning of Jan. 17 with a leg from North Philadelphia Airport (KPNE) to Ohio State University Airport (KOSU) where the aircraft remained overnight. According to FAA transcripts, the pilot contacted the Dayton Automated Flight Service Station at 0834 local time on the morning of the accident and filed an IFR flight plan in which he listed the aircraft as a King Air C-90/I, indicating the airplane was equipped with LORAN, VOR/DME or INS in addition to a Mode-C transponder. According to FAA microfilm records, an FAA Form 337 was on file indicating that a Garmin GPS155 had been installed in October 1995, but was placarded as “GPS Not Approved for IFR Flight.”The pilot requested the Somerset forecast and was told by the briefer that the airport did not generate a forecast. He was instead given the forecast for London, Ky., 26 nm to the east. It called for 500 feet scattered, occasional ceiling of 500 feet broken, 1000 feet overcast, with visibility of five miles and mist, occasionally dropping to two miles in light drizzle and mist, with winds from 140 degrees at seven kts. The briefer also added, “Of course you’re, I’m sure, you’re familiar with the fact that there’s icing, maybe some turbulence, across that route and occasional IFR conditions.” Just after 10 a.m. local, the pilot rechecked Somerset weather and was told the ceiling was now 300 feet overcast with PIREPs for both light rime and clear ice in the area.No problems en route were reported after the flight took off from KOSU just before 11 a.m. At 11:40 a.m., Indianapolis Center, now controlling the flight, asked the pilot if he had the latest Somerset weather, to which the pilot replied, “Yes.” The center also asked the pilot what type of approach he wanted and the pilot responded with the “SDF.”At 11:45:15, the controller said, “… cleared for the SDF approach to, uh, Somerset, maintain 4000 until you’re established on the approach.” The pilot replied, “OK, maintain four ’till established [unintelligible]. Thank you.” The controller repeated the approach clearance and this time specified the SDF Runway 4 approach, and the pilot again repeated he was to maintain 4000 feet until established and was cleared for the SDF Runway 4 approach at Somerset.Two and one half minutes later, at 11:48:01, the pilot transmitted, “Ah, Indy, [callsign].” The call was not answered by the controller. This was the last recorded transmission from the aircraft. Somerset weather at the time was reported as a ceiling of 700 feet with a visibility of 10 sm.Several witnesses resided in an area northeast of the communications tower. Two witnesses inside their homes went to the door upon hearing a noise and saw the aircraft descend to the ground, after which it erupted in fire. Another witness thought the airplane was on fire before it hit the ground, while another saw the airplane spinning as it descended and thought it was missing a portion of a wing.The airplane was maintained under the manufacturer’s inspection program. According to an avionics shop, the airplane also had been in the shop the week prior to the accident to have a new flight director installed because the pilot reported it would not couple to the autopilot. A maintenance release was signed for the work five days prior to the accident.The pilot completed aircraft specific training at SIMCOM in August 1998 and reported his total flight experience as 18,000 hours. On his April 26, 1999, medical application, the pilot reported his total logged time as 19,200 hours, while a May 12, 1999, insurance application showed 15,456 hours as PIC time. Again, these discrepancies most likely had nothing to do with the accident.

Somerset Airport


In January 2000, four instrument approach procedures were being regularly published for Somerset, which included the SDF Runway 4, an NDB Runway 4, a GPS Runway 4 and a GPS Runway 22. According to the Airport/Facility Directory (A/FD) and the Notice To Airmen Publication (NTAP), the SDF was listed as “Out of Service (OTS) — Indefinitely.” Neither note was published on the instrument approach charts, nor were they required to be.The SDF Runway 4 approach called for a minimum safe altitude of 3600 feet within 25 nm of from the Cumberland River NDB, which was about 4.2 nm southwest of the airport. The approach called for a procedure-turn altitude not lower than 3000 feet. Straight in MDA on the inactive SDF approach would have been 1460 feet MSL or 600 feet AGL.

Final Approach

Radar data provided by Indianapolis Center tracked the airplane as it approached Somerset Airport from the northeast. At 1154 local, the King Air passed about half a mile abeam the airport to the northwest at 4000 feet. The aircraft began a descent, which would have been normal to set up for any of the approaches.As the aircraft passed just to the northwest side of the Cumberland River NDB, things turned odd as the King Air descended through the minimum procedure-turn altitude of 3000 feet and began a left turn that appeared to be taking it back toward the airport. The aircraft had completed about 180 degrees of turn — but was aligned with no known navigational signal or approach profile — and passed through 1,900 feet before the target disappeared from radar a few miles southeast of the NDB and just seconds before it sheared the guy wire from the tower.During the crash investigation, the selector handle showed the landing gear had been in the down position at the time of the accident. Dropping the gear prior to the beacon outbound is unusual in some cases, but not unheard of if the aircraft may have been approaching Somerset a bit fast. The aircraft’s radio panel was destroyed so it was impossible to learn what frequencies the pilot had tuned in.The NTSB findings tell us what happened, but not why. Was this crash simply the final broken link in a long chain of other unexplained events? We’ll never know for certain why the King Air descended with no approach guidance of any kind. But the paragraph that should have been added to the NTSB report was that no experienced pilot flying in the clouds would have allowed his aircraft to descend like that aircraft did if he was fully in charge of the aircraft.Certainly the pilot should have known the approach was not available, but how many times have you picked up a weather briefing and asked for the NOTAMS? Have you ever asked the briefer to check the NTAP to see if any had slipped outside the 56-day notice period that you should be aware of? Most of us would have failed on that test. A larger question is why the Indianapolis Center controller seemed to be completely at ease clearing the aircraft for an approach that had been shut down four years earlier. When asked about the accident, the controller said he’d forgotten the approach was decommissioned.Did the 72-year-old pilot experience some sort of sudden incapacitation issue such as a heart attack or stroke? With only a single pilot in the cockpit, in the clouds, the people in back would be unable to tell what was happening as the aircraft descended in an ever-tightening turn, much less do anything to help. The pilot at one point before the descent obviously had something he wanted to tell ATC, but that communication was never completed.Did the change in flight directors play a role? Did the autopilot uncouple unbeknownst to the pilot, who might have been trying to figure out why he could not seem to hear a signal from the SDF? Did the GPS play a role? It wasn’t certified for IFR, but the pilot most likely knew how to use it because it had been installed for five years. Was he hoping to ask for the SDF approach and fudge a bit by really shooting the GPS anyway, assuming no one would notice? Would a second pilot in the cockpit have been able to help prevent this tragedy?Certainly ATC was of very little help to this pilot by approving an out-of-service instrument approach procedure. But the PIC is still responsible in the end. That leaves only a few options. The pilot was unaware of what the airplane was doing because he was distracted by something else, a physical problem, or perhaps a mechanical or navigational problem that overloaded the pilot to the point he was no longer able to maintain control of the airplane.

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.