Probable Cause #53: Untimely Distraction

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

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

Probable Cause

Distractions that occur during periods of high workload can create problems for pilots, especially if the event pulling the pilot's focus away from flying the airplane appears to be a major system component failure. With a two-pilot crew, this danger is greatly reduced, since one pilot can continue to fly the aircraft while the other tries to fix the problem. But pilots who fly single-pilot IFR don't have the luxury of having someone else in the cockpit to do the troubleshooting.

Steady To Steamboat Springs

Cessna Caravan

At 4:33 a.m. MDT on May 5, 2001, the pilot of a Cessna Caravan contacted the Casper Automated Flight Service Station and requested current weather for several stations between Casper, Wyo., and Steamboat Springs, Colo. Weather for Steamboat Springs wasn't available, so the briefer gave the pilot the weather for Hayden, Colo., 21 miles west of Steamboat Springs. Hayden reported winds from 270 degrees at nine kts, four miles visibility, scattered clouds at 400 feet, a broken ceiling at 2500 feet and an overcast layer at 4800 feet. The temperature was 4 degrees C and the dew-point was 3 degrees C. AIRMETs called for occasional moderate rime and mixed icing in clouds and precipitation from 8000 to 20,000 feet, obscured mountains and occasional moderate turbulence above Flight Level 180. The pilot was employed by a Billings, Mont.-based cargo company that flew freight under contract for Federal Express. Operating as Airspur Flight 8810, he was scheduled to fly from Casper to the Bob Adams Field (KSBS) in Steamboat Springs, a distance of approximately 150 nm. By 7:36 a.m., the Cessna had been loaded with 270 pounds of cargo and the pilot contacted Casper Ground Control for his instrument clearance. Six minutes later he was cleared for takeoff and, at 7:44 a.m., he contacted the Casper departure controller. The pilot was given vectors to intercept V26 and told to climb to 13,000 feet. At 7:58 a.m. the aircraft was switched over to Denver Center. While on that frequency, a broadcast was made informing pilots of a hazardous weather AIRMET for areas west of the Mississippi River. Pilots were advised to contact Flight Watch or Flight Service for more information. The accident report did not mention that the Caravan pilot talked to Flight Service or received any weather updates while en route. Approaching Steamboat Springs, the pilot requested the VOR/DME-C approach into KSBS with the intention of landing on Runway 32. The flight was cleared for the approach via the Robert VORTAC 280 degree radial 9 DME fix and the 9 DME arc. At 8:39 a.m. the pilot was advised that radar services were terminated and that he was cleared to leave the frequency. There would be no further contact with the pilot.


Steamboat Springs VOR/DME-C Approach Chart Excerpt (Click here for full chart - 124 KB.)

When by 9:23 a.m. Denver Center had not gotten an acknowledgment that the pilot was on the ground, the process was begun to locate the overdue aircraft. This included asking pilots flying in the area to try to contact the aircraft and check for ELT signals. There was no response. At 10:34 a.m. an official Alert Notice was issued for the aircraft. At approximately 9:45 a.m., a pilot who also flew for the company was notified that the aircraft was missing. He checked on the aircraft's last known position and drove to KSBS. He arrived at around 12:15 p.m., where he found a search and rescue preparing to initiate a ground search. The pilot located a student of his and together they took off in the latter's Cessna 170 heading for the area where the Caravan was last seen on radar. At 12:55 p.m. they located the wreckage of the Caravan on the side of Emerald Mountain about half a mile south of the Robert VOR (BQZ). The VOR is located 2.9 nm from the airport and serves as the final approach fix (FAF). According to witnesses on the ground, the Steamboat Springs weather around the time of the accident was a 600-foot overcast ceiling, 1-1/2 miles visibility in misting rain and a temperature of 36 degrees F. The pilot of a Swearingen Metroliner landed his airplane at Hayden around the time of the accident. He reported to investigators that his aircraft accumulated a "dusting" of ice on departure from Denver, but that he did not pick up any ice on the way into Hayden, nor did he hear any reports of anyone else picking up ice in that vicinity. He told investigators that he and his co-pilot heard the Caravan pilot talking to the Denver Center controller and was puzzled as to why the pilot would ask for an approach where the ceiling was much lower than the minimum descent altitude (MDA) of 8,140 feet, or 1,262 feet agl. He was basing this on the ceiling reported on the Hayden AWOS. Another pilot flew the approach into Hayden shortly after the accident happened. He reported no icing encounter during the approach and he told investigators that the weather conditions included a "2000 to 3000-foot thick overcast layer that was improving." The Caravan was equipped for flight into known icing conditions. The wings and tail had deice boots installed and the windshield had an anti-ice system. The cargo pod and landing gear struts were also fitted with a deice system. The aircraft was not equipped with a cockpit voice recorder or flight data recorder, but it was equipped with a power analyzer and recorder, an onboard computer that records and stores engine operating parameters. The NTSB downloaded the data from the computer and found that normal power parameters were recorded for the flight. On-site investigators found that the aircraft struck the ground on a magnetic heading of 145 degrees, which is close to the 172 degree reciprocal of the VOR/DME-C inbound approach course. The angle of terrain impact was approximately 40 degrees, which was measured by studying the damage to the severed treetops. This, along with other damage patterns, suggested that the airplane crashed as a result of a stall. The airplane impacted terrain at an elevation of 7864 feet. There was no evidence of pre- or post-impact fire, but search and rescue personnel reported a strong odor of jet fuel at the scene. They also stated that they saw no evidence of airframe icing.

Sectional Chart of Steamboat Springs Area

Time In Type

The 44-year-old pilot held a multiengine ATP with Commercial privileges in single-engine aircraft. He had a current medical certificate and received his initial Caravan checkout 11 months prior to the accident. With the exception of some incomplete entries in the pilot's logbook, NTSB investigators listed the pilot's flight time as approximately 3041 hours, of which 1834 hours were in single-engine airplanes and 1186 hours in multiengine airplanes. (The NTSB does not explain the 21 hours discrepancy between the pilot's single and multi time and total time, although he did have 21 hours of what the Board called "simulator" time). The pilot also had logged 206.4 hours of actual instrument time, 96 hours of simulated instrument time and 37.9 hours in the Caravan. It appears that the pilot did not begin flying the Caravan exclusively until approximately five weeks before the accident occurred. Between 1999 and 2001, he logged 16 separate icing encounters in his logbook. One of them resulted in a return to Helena, when he noted that he could not get to 8000 feet due to icing in a climb. He was flying a Cessna 402 at the time for a different employer.

Wrong Switch?

Cessna Caravan Switches

Investigators reviewed recorded radar data that showed the aircraft shortly after it took off from Casper until the last radar hit was recorded at 8:56 a.m., when the aircraft was on the approach at 9400 feet. Everything seemed normal as the aircraft turned inbound on the final approach course to BQZ and began what appeared to be a normal descent. At 8:53:03 a.m., the aircraft corrected slightly to the left on the inbound course. At 8:55:39 a.m., the target was 0.75 miles to the northwest at 9,700 feet. On the next radar hit the aircraft was 0.5 miles southeast at 9,600 feet. The last radar target showed the aircraft 0.5 miles northwest at 9,400 feet. The aircraft wreckage was located 0.75 miles east-northeast of the last radar contact. During the investigation, selected bulbs from the aircraft's annunciator panel were examined visually, under a stereomicroscope and by a scanning electron microscope. According to the report, the condition of the tungsten-alloy filaments indicated that the "Generator Off" and "Windshield Anti-Ice" lights were illuminated at the time of the accident. Given the potential that icing conditions existed in the area, it would not have been odd for the windshield anti-ice light to be on. What drew the attention of the investigators, however, was the generator warning. The Cessna 208 Pilots Operating Handbook states that illumination of the generator off light indicates a generator disconnection due to line surges, tripped circuit breakers, or accidental switch operation. The operator's chief pilot told investigators that the aircraft's Before Landing Checklist requires that the ignition switch be placed in the "on" position. The start switch is located next to the ignition switch and inadvertent operation of the start switch would take the generator off line, thereby illuminating the generator off light and the standby power lights. Unfortunately, due to the destruction of the instrument panel during impact, it was not possible to verify the position of the ignition and start switches. Regardless, the loss of the generator would undoubtedly have been distracting to the pilot during a time when he was focusing on flying an instrument approach in IMC. At about the time the pilot would be expected to complete his Before Landing Checklist, the aircraft began its track back and forth across the course. Could it be that the pilot was troubleshooting the problem while trying to fly the approach? The NTSB seems to indirectly back this theory. It determined the probable cause as, "... an inadvertent stall during an instrument approach, which resulted in a loss of control. Contributing factors were the pilot's attention being diverted by an abnormal indication, conditions conducive to airframe icing, and the pilot's lack of total experience in the type of operation (icing conditions) in [the] aircraft make [and] model."

Know Thy Aircraft

So, what can we learn from this accident so that we don't encounter a similar situation? First, make certain you are completely familiar with the aircraft's systems in any aircraft that you fly, especially if you fly in IFR single-pilot operations. We don't know if the pilot realized what caused the illumination of the generator off light, but it does appear that when it illuminated he was distracted from his main duty, which was flying the aircraft. When a system failure follows immediately or shortly after you complete a checklist item, go back and check if that action caused the problem. Check the switch or series of switches you last changed to make certain you didn't inadvertently turn the wrong one(s) on or off. Think about how far along the approach course you are. Can you continue and land with no effect from the system failure? In this case, the aircraft was approximately three miles from the runway, but the weather conditions were suspect. From the reported weather, it did not appear that the approach would be successful. Would the aircraft's batteries provide enough power to get from Steamboat Springs to Hayden for another approach? If you have an electrical failure in instrument conditions and cannot correct the problem by resetting the generator or alternator immediately, reduce power consumption for conservation. If you are in conditions conducive to icing but no icing is present, you probably want to consider shutting of any electrically powered anti-ice equipment. The aircraft's batteries will not power that equipment for very long and you don't want to risk losing all of your instrumentation due to battery failure. The Caravan pilot never had a chance to consider his alternatives because the aircraft stalled and crashed into the terrain. It would appear that he allowed himself to be distracted to the point where the aircraft's speed was reduced to the point of stall, and once that occurred there was no recovery from it. Remember, if you don't fly the aircraft first, nothing else you do is going to make much of a difference.
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.