FTW03LA173 · Apex Aircraft CAP 10B

The aerobatic airplane was destroyed when it impacted terrain following the in-flight separation of its left wing. The airplane was observed traveling southeast, at the bottom of an established aerobatic box (1,500 feet msl), when the left wing separated from the airplane. The wing spar failed in upward bending, as indicated by the flat compressive failure zones near the upper surfaces of the spar caps and the fibrous tensile fractures near the bottom surfaces of the spar caps. Both spar caps showed the same type of bending failure, which suggests that the shear web interconnection between the spar caps failed before the final failure of the spar caps themselves. The failure was almost certainly initiated by the progressive compressive failure zone at the top surface of the upper spar cap, which was visible as a band of lighter color on the fracture surface extending approximately 12 mm to 15 mm from the top surface of the spar cap (through the top layer and part of the second layer of spruce in the spar cap). The discoloration and/or material loss at the upper forward corner of the fuselage piece of the spar could have played a role in the failure of the shear web, but the discoloration and damage is more likely a result of impact with the ground. The materials used in the spar construction were as specified, and the adhesive bonds used in the construction and previous repair were sound. There is a long history of ADs and SBs requiring inspections for compressive crack-like features on the top surface of the spar. The representatives from the DGAC and BEA indicated that the first accident resulting from this type of progressive compressive damage occurred in 1968, and that there had been approximately 10 similar failures in total. A significant portion of the upper spar cap of the accident aircraft was replaced in 1980 as a result of an inspection detecting such damage. It appears from the maintenance records that the appropriate ADs and SBs were signed off as having been complied with in the time since 1980, but the compressive damage that led to the accident was not detected. It appears that the position of the inspection opening on the left wing was not strictly in keeping with the instructions with SB 16, but there are other problems with the structure and the inspection procedures. The highest compressive stresses would occur at the top of the spar, just at the outboard edge of the fuselage attachment reinforcement block, which is where the fracture occurred. This location also coincided closely with the edge of the wing-walk support, which would further increase the stress concentrations in the area. Furthermore, the upper spar caps were damaged during preparation of the inspection openings, leaving small steps in the spar cap material at the edges of the wing-walk supports. These steps would cause additional stress concentrations, making progressive damage more likely (particularly as the step sat almost directly above the outboard edge of the fuselage attachment reinforcement block). The steps probably also obscured the appearance of the crack-like features that formed at the base of the step. The step on the right wing was much smaller than that on the left wing, and the right wing had only a small area of progressive compressive damage compared to the damage on the left wing. Based on the circumstances of this accident, there are several problems with the service bulletins intended to prevent such failures. The position of the inspection opening is described in relation to rib 1, but this position is unreliable. Of interest is the top surface of the spar at the location above the edge of the fuselage attachment reinforcement block. Once the box beam has been completely assembled, the reinforcement block is no longer visible and the position of rib 1 may not correlate with the edge of the block, particularly after a modification or repair involving rib 1. In addition, the SBs do not address the location of the wing-walk support. In this case, the wing-walk support appears to have possibly limited the size of the inspection opening, led to an additional stress concentration in precisely the wrong place, and obscured the progressive damage.

Verbatim NTSB analysis and probable cause from the NTSB dataset