N99APRaytheon Aircraft HAWKER 800XP · 1999

On February 21, 2022, at 1133 mountain daylight time, a Raytheon Aircraft Company Hawker 800XP airplane, N99AP, was substantially damaged when it was involved in an accident at Aspen-Pitkin County Airport (ASE), Aspen, Colorado. The two pilots and four passengers were not injured. The airplane was being operated as a Title 14 Code of Federal Regulations Part 91 business flight.

According to the flight crew reports and the cockpit voice recorder (CVR) audio, prior to departure, the airplane and runway were clear of any contaminants, and all pre-takeoff checks were normal. At 1102:46, the airplane was cleared to taxi to runway 33 with automatic terminal information service (ATIS) information Bravo. The ATIS indicated that the wind was from 170° at 18 knots (kts) and gusting to 30 kts. During taxi, the flight crew changed the takeoff flaps from 0° to 15° to reduce the takeoff length by 800 ft per the crew’s Aircraft Performance Group calculations, and the first officer entered the new airspeeds in the flight management system.

About 1119, the ASE air traffic control (ATC) tower controller informed the flight crew the takeoff would be delayed due to arriving traffic. At 1131:54, the controller provided the takeoff clearance for runway 33 and reported the wind was from 160° at 16 kts, gusting to 25 kts. In addition, the controller provided the “instantaneous” wind, which was from 180° at 10 kts. The captain reported that “at takeoff clearance, constant winds were reported by tower at [180° at 10 kts] which was within aircraft maximum tailwind takeoff limitation.”

According to CVR audio, the takeoff was initiated at 1132:26. The captain performed a static takeoff, and the first officer made all the callouts: airspeed alive, 80 kts, takeoff decision speed (V1) at 111 kts, and rotate (VR) at 121 kts. The captain reported that, at VR, he applied back pressure on the yoke; however, the airplane would not become airborne. The captain reported, “the yoke did not have any air resistance or any pressure on it as we experience normally in Hawkers (the weight and pressure on the yoke felt the same as though…the airplane was stationary on [the] ground).”

After a few seconds without any indication the airplane would take off, the captain called for and performed an aborted takeoff by reducing the engines to idle, deploying the thrust reversers, and applying the brakes. The airplane subsequently departed the end of the runway into the snow (see figure 1). The captain secured the airplane and assisted in the evacuation of the passengers.

Figure 1. Accident airplane following runway excursion (Source: ASE airport operations)

According to the airplane flight manual, the maximum tailwind component for takeoff and landing is 10 knots.

According to the operator, at takeoff, the airplane’s weight was 23,916 pounds (lbs), and the location of the center of gravity (C.G.) was -.02 inches from datum or 17.93% mean aerodynamic chord. The operator reported that the maximum gross weight limit was 25,288 lbs., the forward C.G. limit was -2.11 inches, and the C.G. aft limit was 9.28 inches.

An automated system at ASE reported wind conditions every 5 minutes. Table 1 shows wind information within 30 minutes before and after the accident.

Table 1. Winds from automated ASE weather station

According to the airplane flight manual, the maximum tailwind component for takeoff and landing is 10 knots.

According to the operator, at takeoff, the airplane’s weight was 23,916 pounds (lbs), and the location of the center of gravity (C.G.) was -.02 inches from datum or 17.93% mean aerodynamic chord. The operator reported that the maximum gross weight limit was 25,288 lbs., the forward C.G. limit was -2.11 inches, and the C.G. aft limit was 9.28 inches.

Postaccident examination of the airplane revealed no malfunctions or failures that would have precluded normal operation.

Following the accident, the operator informed its flight crews to no longer consider “instantaneous wind” reports in their decision-making process.

The airplane was equipped with a Universal CVR-30B CVR that recorded a minimum of 30 minutes of digital data stored on solid-state modules. The CVR contained four sources of audio input: one channel for each flight crew, one spare channel (that is, for an observer), and one channel for the cockpit area microphone. The captain, first officer, and spare channels were recorded independently for a minimum of 30 minutes. The National Transportation Safety Board (NTSB) Vehicle Recorders Laboratory completed a summary report of the recorded audio. The captain and first officer audio channels were categorized as good recording quality.

Airplane Performance Study

An NTSB airplane performance study was completed based on automatic dependent surveillance—broadcast (ADS-B) data provided by the FAA. The airplane’s CVR recorded the takeoff roll and accident sequence.

ADS-B coverage of the takeoff roll (see figure 3) did not start until 1132:53, when the airplane was already at a groundspeed of 135 kts. The airplane accelerated to 165 kts by 1133:01.6 when it was about 3,300 ft from the end of the runway, it then decelerated and left the paved surface at a groundspeed of 120 kts (see figure 4).

Figure 3. Aircraft takeoff roll with times and distance to end of runway

Figure 4. Calculated groundspeed versus distance to end of runway.

The flight crew, using a weight of 23,916 lbs and flaps 15°, calculated the airplane’s V1 and VR airspeeds to be 111 kts and 121 kts. The flight crew called out when the airspeed passed V1 (1132:55) and VR (1132:59), and then when the takeoff was aborted (1133:10). The flight crew reported that the airplane did not rotate after VR. Figure 5 shows the groundspeed and three calculated calibrated airspeeds (CAS) based on three wind conditions. The 111 kts V1 speed was in line with what calibrated airspeed would be for a 17 kt wind from 180°, which was the automated wind report for 11:30 (See table 1). The VR callout falls between the CAS for 17 kts and 24 kts (gusting condition), suggesting a possibility that the wind increased between the V1 and VR callouts.

Figure 5. Calculated CAS for three different wind conditions

At VR, the airplane’s pitch control authority should have been sufficient to raise the airplane’s nose and begin liftoff. However, the flight crew reported that the airplane did not rotate. This could imply that when the pilot pulled back on the yoke, that the airplane’s airspeed was insufficient to induce rotation.

The black line in Figure 5 shows the calculated CAS for 35 kt wind from 180°. At the maximum achieved ground speed of 165 kts, a wind of this magnitude would lower the airplane’s indicated airspeed to just below VR. Title 14 Code of Federal Regulations Part 25.107, Takeoff Speeds, requires that the VR be at least 10% greater than the minimum calibrated airspeed at which the airplane can safely rotate and lift off.

Therefore, for a reported VR of 121 kts, the airplane should have lifted off after reaching an airspeed of 110 kts. A 35-kt wind would not reduce the maximum achieved ground speed of 165 kts sufficiently to prevent the airplane from flying, and thus after achieving V1 (111 kts), the flight crew should have had sufficient air load to rotate the airplane. This was not consistent with the flight crew account that the yoke did not have any air resistance when the yoke was pulled back, considering that the wreckage examination revealed no discrepancies with flight control continuity to the elevator system. Even if a tailwind increased to more than the maximum reported gusting of 25 kts after VR or if the flight crew call to rotate was made before VR was achieved, the airplane’s airspeed should have resulted in noticeable air resistance when the yoke was pulled back. These discrepancies could not be resolved with the available evidence.