United States of America

According to the pilot, the airplane '...began moving to the left side of the runway...' immediately upon touchdown. He attempted to abort the landing but the airplane contacted snow at the runway's left edge. The airplane exited the runway's left edge and slid sideways, about 300 yards.

The flight was being vectored for the approach to runway 3R at Detroit Metropolitan Wayne County Airport (DTW) when the aircraft descended and impacted the ground. The aircraft struck the ground in a steep nose-down attitude in a level field in a rural area about 19 nm southwest of DTW. The flight carried 26 passengers and 3 crew members. There were no survivors and the airplane was destroyed by impact forces and a post crash fire. Instrument meteorological conditions prevailed at the time of the accident. The investigation revealed that it was likely that the airplane gradually accumulated a thin, rough glaze/mixed ice coverage on the leading edge deicing boot surfaces, possibly with ice ridge formation on the leading edge upper surface, as the airplane descended from 7,000 feet mean sea level (msl) to 4,000 feet msl in icing conditions, which may have been imperceptible to the pilots. The pilots had been instructed by air traffic control to slow to 150 knots and according to flight data recorder information, the airplane began to show signs of departure from controlled flight as it decelerated from 155 to 156 knots while in a flaps-up configuration. The investigation disclosed that the FAA failed to adopt a systematic and proactive approach to the certification, and operational issues of turbopropeller-driven transport airplane icing. The icing certification process has been inadequate because it has not required manufacturers to demonstrate the airplane's flight handling and stall characteristics under a sufficiently realistic range of adverse ice accretion/flight handling conditions. The aircraft manufacturer had issued a revision in April, 1996 to the approved flight manual which included activation of the leading edge deicing boots at the first sign of ice formation. The airplane operator did not incorporate the procedure, because it was contrary to the company's trained procedures and practices and of the belief that enacting the changes would result in potentially unsafe operation. Investigators' discussion with management personnel at each of the seven U.S.-based operators of the aircraft indicated that at the time of the accident only two of these operators had changed their procedures to reflect the information in the revision. The FAA, at the time of the accident, did not require manufacturers of all turbine-engine driven airplanes to publish minimum airspeed information for various flap configurations and phases and conditions of flight. During Safety Board investigators postaccident interviews with company pilots, there were inconsistent answers on the complex and varied minimum airspeed requirements established by the company for both icing and nonicing conditions. It was also noted that the pilots uncertainty of the appropriate airspeeds might have been associated with the language used, the different airspeeds and criteria contained in the guidance, the company's methods of distribution, and the company's failure t o incorporate the guidance as a formal, permanent revision to the flight standards manual.

After executing a missed approach at the Grand Canyon Airport, the pilots diverted to the Bullhead City Airport. The pilots reported that minimal icing conditions were encountered with about 1/8 inch of ice accumulating on the aircraft wings. The pilots stated they cycled the deice boots to shed ice. They did not observe ice on the propeller spinners, and they did not activate the engines' 'override' ignition systems, as required by the airplane's flight manual. Use of 'override' ignition was required for flight into visible moisture at or below +5 degrees Celsius (+41 degrees Fahrenheit) to prevent ice ingestion/flameouts. Subsequently, both engines flamed out as the airplane was on about a 3 mile final approach for landing with the landing gear and flaps extended. The aircraft was destroyed during an off-airport landing.

The airplane departed the airport and crashed shortly thereafter. Before departure, the airplane was fueled with 120 gallons of 100LL aviation fuel. According to the refueler, the airplane had full fuel tanks. The refueler also indicated the pilot had stated he wanted to be airborne prior to the arrival of bad weather. After the accident, the engines and propellers were disassembled and examined. No engine or propeller discrepancies were noted, except (post impact) heat damage.

During his weather briefing, the pilot was told that his destination weather was not available, and was provided weather for an airport about 10 miles north. He was briefed that low visibility due to fog prevailed. The flight departed earlier then usual because the company business manager was concerned that the weather at the destination airport was deteriorating, and if the airplane was not there earlier they might not get into the airport. At the time of the accident there was a power failure, and lights around the destination airport went out. The airplane had struck power lines and a support tower located on the approach end of runway 1 and runway 5, about 1/2 mile southwest of the airport. The airport had one NDB approach which was not authorized at night. The nearest recorded weather, about 10 miles north of the crash site, at the time of the accident was; '...ceiling 100, [visibility] 1/2 mile, fog, [temperature] 46 degrees F, dew point, 42 degrees F, winds 220 degrees at 5 [knots], altimeter 29.90 inches Hg. Witnesses reported that there was heavy fog at the airport and the visibility was below 1/4 mile.

During the takeoff roll the pilot stated the right engine had an over torque condition and he was unable to control the aircraft. The aircraft went off the runway to the left and crashed coming to rest upright. A post crash fire erupted and destroyed the aircraft. The mechanic rated passenger stated he was observing the right engine gauges during this maintenance test flight and did not observe any over torque indications. When he looked up from the instruments at about the time the aircraft should lift off, the aircraft was drifting to the left. The pilot, who was looking at the engine instruments, looked up, saw the aircraft was about to drift off the runway, and retarded both power levers. The passenger/mechanic (who was also a pilot) reported that the pilot placed the propellers in reverse. Six thousand feet of runway remained at the abort point. The aircraft pitched up and then crashed on the left wing and nose. Cessna Service Newsletter SLN99-15 and AlliedSignal Operating Information Letter OI 331-17 report an abnormality that may affect the model engine in which an uncommanded engine fuel flow increase or fluctuation may occur, resulting in an unexpected high torque and asymmetric thrust. The condition is associated with an open torque motor circuit within the engine fuel control. A system malfunction resulting in engine acceleration to maximum power would produce an overtorque of about 2,288 foot-pounds (ft-lb). This power output is restricted by a fuel flow stop in the engine fuel control. Normal takeoff power is 1,669 ft-lbs; therefore, one engine accelerating to the stop limit while one engine continued to operate normally would cause a torque differential of 619 ft-lbs. The total loss of power in one engine during takeoff while one engine continued to operate normally would result in a torque differential of 1,669 ft-lbs. The Cessna 441 Flight Manual states that at 91 knots indicated airspeed, the airplane is controllable with one engine inoperative (that is, with a torque differential between engines of up to 1,669 ft-lbs). However, if an electronic engine control failure occurs on one engine and the other engine is retarded to idle, the fuel flow to the failed engine will not be reduced, and a torque differential of about 2,288 ft-lbs will occur, at which point the airplane is uncontrollable by the pilot.

There were five recorded transmissions of conversations from one of the pilots aboard the accident airplane and the Fort Dodge, Iowa AFSS; one on 12/27 and four on 12/28. The weather forecast for icing included wording such as '...moderate to isolated severe icing from seven thousand downward for your entire route of flight....' The pilot stated that he received 'Full Flight Service briefings...' and also indicated that he spoke to flight watch prior to takeoff. While executing the ILS approach to the destination airport, the pilot was unable to maintain the proper glidepath even with the application of full power. The pilot maintained marginal control of the airplane during the descent until impact with trees and the terrain about 10 miles west of the destination airport. The pilot and passengers reported 'vibration' and 'shudder' of the airplane prior to the impact. One passenger reported that she saw ice forming on the left 'rear' wings. Persons on the ground reported severe icing conditions around the time of the accident.

The first officer was in the left seat, flying the airplane, and the captain was in the right seat, for the positioning flight. Approaching the destination, the crew briefed, then attempted an ILS RWY 18 approach. The captain reported not receiving the localizer, when, in fact, the airplane was actually about 5 nautical miles to the left of it. Winds at the airport, about that time, were from 190 degrees true, at 5 knots; however, area winds at 6,000 feet were from 220 degrees, in excess of 40 knots. The crew executed a missed approach, but did not follow the missed approach procedures. The captain later requested, and received clearance for, the VOR RWY 25 approach. The captain partially briefed the approach to the first officer as the airplane neared the VOR, then subsequently "talked through" remaining phases of the approach as they occurred. The outbound course for the VOR RWY 25 approach was 066 degrees, and the minimum altitude outbound was 4,300 feet. After passing the VOR, the captain directed the first officer to maintain 4,700 feet. The airplane's last radar contact occurred as the airplane was proceeding outbound, 7 nautical miles northeast of the VOR, at 4,800 feet. As the airplane approached the course reversal portion of the procedure turn, the captain initially directed the first officer to turn the airplane in the wrong direction. When the proper heading was finally given, the airplane had been outbound for about 2 minutes. During the outbound portion of the course reversal, the captain told the first officer to descend the airplane to 2,900 feet, although the procedure called for the airplane to maintain a minimum of 4,300 feet until joining the inbound course to the VOR. During the inbound portion of the course reversal, the captain amended the altitude to 3,000 feet. As the airplane neared the inbound course to the VOR, the captain called out the outer marker. The first officer agreed, and the captain stated that they could descend to 2,300 feet. The first officer then noted that the VOR indications were fluctuating. The captain pointed out the VOR's continued reception, and the first officer noted, "but it's all over the place." Shortly thereafter, the first officer stated that he was descending the airplane to 2,300 feet. Three seconds later, the airplane impacted trees, then terrain. The wreckage was located at the 2,300-foot level, on rising mountainous terrain, 061 degrees magnetic, 12.5 nautical miles from the VOR. It was also 10.3 nautical miles prior to where a descent to 2,300 feet was authorized. There was no evidence that the crew used available DME information. There was also no evidence of pre-impact mechanical malfunction.

The airplane impacted mountainous terrain while on a post-modification functional evaluation flight (FEF). The pilot flying (PF) had applied inappropriate control column back pressure during the clean stall maneuver recovery attempt in an inadequate performance of the stall recovery procedure established in ABX's (Airborne Express) operations manual. The pilot not flying (PNF), in the right seat, was serving as the pilot-in-command and was conducting instruction in FEF procedures. The PNF failed to recognize, address and correct the PF's inappropriate control inputs. An inoperative stall warning system failed to reinforce to the flightcrew the indications that the airplane was in a full stall during the recovery attempt. The flightcrew's exposure to a low fidelity reproduction of the DC-8's stall characteristics in the ABX DC-8 flight training simulator was a factor in the PF holding aft (stall-inducing) control column inputs when the airplane began to pitch down and roll. The accident could have been prevented if ABX had institutionalized and the flightcrew had used the revised FEF flight stall recovery procedure agreed upon by ABX in 1991. The informality of the ABX FEF training program permitted the inappropriate pairing of two pilots for an FEF, neither of whom had handled the flight controls during an actual stall in the DC-8.

The pilot had received clearance for the ILS Runway 6 approach and was advised that the previous landing traffic reported '...breaking out at minimums.' Radar data revealed that the airplane descended in instrument meteorological conditions to the decision height altitude of 294 feet, approximately 3 miles from the missed approach point. The pilot did not perform the missed approach procedure. The airplane leveled off and continued at or below decision height altitude for approximately 28 seconds, traveling a distance of approximately 1 mile. Four low altitude alerts appeared on the tower controller's display. The controller stated he withheld the alert because '...it was a critical phase of flight and the aircraft appeared to be climbing...' The airplane collided with trees and terrain approximately 1.5 miles from the approach end of the landing runway.