Cessna 560XL N91GY

The crew was conducting an on-demand charter flight with eight passengers onboard. As the flight crew taxied the airplane to the departure runway, the copilot called air traffic control using his mobile phone to obtain the departure clearance and release. According to the pilot, while continuing to taxi, he stopped the airplane short of the runway where he performed a rudder bias check (the last item in the taxi checklist) and applied the parking brake without verbalizing the parking brake or rudder bias actions because the copilot was on the phone. After the pilot lined up on the runway and shortly before takeoff, the flight crew discussed and corrected a NO TAKEOFF annunciation for an unsafe trim setting. After the copilot confirmed takeoff power was set, he stated that the airplane was barely moving then said that something was not right, to which the pilot replied the airplane was rolling and to call the airspeeds. About 16 seconds later, the pilot indicated that the airplane was using more runway than he expected then made callouts for takeoff-decision speed and rotation speed. The pilot stated that he pulled the yoke back twice, but the airplane did not lift off. Shortly after, the pilot applied full thrust reversers and maximum braking, then the airplane exited the departure end of the runway, impacted a ditch, and came to rest 1,990 ft beyond the departure end of the runway. The airplane was destroyed by a postcrash fire, and the crew and passengers were not injured. Postaccident examination of the airplane and computed tomography scanning of the parking brake valve revealed it was in the ON position at the time of the accident. Calculations based on airplane performance data and airport surveillance video indicated that the accident airplane was able to accelerate to the speed required for the airplane to rotate and lift off as configured for a normal takeoff. However, the airplane’s performance was substantially degraded during the accident takeoff roll by an unexpected retarding force acting at the wheel/runway interface. Based on the parking brake valve examination and the performance of the airplane during the accident sequence, it is likely that the engaged parking brake produced the unexpected retarding force at the wheel/runway interface, and at a magnitude and direction that adversely affected the airplane’s acceleration and rotation capability during the attempted takeoff. Review of the airplane’s before takeoff and takeoff checklists revealed no explicit item that directed the flight crew to release or check the parking brake. The operator and manufacturer’s taxi checklists direct the flight crew to check brakes but do not specifically refer to the parking brake, and the operator’s flow diagram for that checklist did not point to the parking brake pull knob. Additionally, the position of the parking brake lever was not evaluated as part of the conditions that trigger a NO TAKEOFF annunciation. Review of the CE-560XL parking brake’s certification revealed that the parking brake was designed to prevent the airplane from rolling if one engine is at takeoff power and did not require a cockpit indication if the parking brake was not fully released. These conditions met the parking brake standard at the time of certification. However, the standard was updated in 2002 to require an annunciation if the parking brake was not fully released when takeoff power was applied, and the CE-560XL parking brake was not required to be updated to that standard. The manufacturer did not test the parking brake with both engines at takeoff power and was not required to as part of certification. The head of training where the pilot and copilot received their CE-560XL training stated that no airplane flight manual, checklist, or training curriculum provided a caution or warning that takeoff speeds may be achievable with full or partial pressure applied by the parking brake, and the pilot reported that he expected the airplane not to move with takeoff power applied. Available evidence, including continuous, heavy rubber deposits on the runway consistent with the main landing gear wheel spacing throughout the acceleration segment of the takeoff roll, indicates that it is likely that the pilot did not release the parking brake after setting it to perform the rudder bias check. Because the parking brake pull knob was located on the pilot’s left side and was obstructed from the copilot’s view, it is unlikely the copilot could have seen the pull knob before or during the takeoff. Although the crew was aware of the airplane’s slow acceleration, it was not clear at the time that it was related to an unexpected retarding force or the unsafe condition of the engaged parking brake. Had the parking brake lever position been incorporated into the conditions that trigger a NO TAKEOFF annunciation, as required for airplanes certificated after 2002, the pilot and copilot likely would have identified that the parking brake remained engaged and corrected the unsafe setting before attempting takeoff, just as they did for the unsafe trim setting. Despite the flight crew’s non-adherence to standard operating procedures, a checklist item that directed the pilot to fully release the parking brake before takeoff could have also served as an important redundancy to an annunciation, but none appears in the before takeoff checklist.

Quoted verbatim from the NTSB record.