CHI01IA073

The Captain reported they were flying a visual approach following the localizer to intercept the glideslope. When flaps 45 were selected, the airplane rolled right about 15 to 20 degrees. Application of left aileron and rudder controlled the roll, and the approach was aborted. After consulting maintenance and flight operations and conducting a visual inspection, the Captain determined they had a failure of the right flight spoiler. A decision was made to land using corrected speeds per the checklist. The approach and landing were uneventful. A post-flight inspection revealed the right flight spoiler was fully deployed and the right spoiler lug had fractured and separated from the hydraulic actuator. A review of the maintenance records indicated the right flight spoiler had acquired a total of 3,987.9 cycles and 4,064 landings. Metallurgical examination revealed the aluminum spoiler lug had failed due to fatigue. Transport Canada issued an Airworthiness Directive on February 22, 2001, that required the inspection of aluminum flight spoiler lugs at 3,000 cycles, and at intervals not to exceed 500 cycles. The AD also stipulated that the installation of redesigned flight spoilers which utilized steel lugs provided terminating action to the inspection requirements. The FAA issued an AD on July 26, 2001, that required the same inspection criteria as the Transport Canada AD.

On January 25, 2001, at 1311 eastern standard time, a Canadair CL-600-2B19, N785CA, operated by Comair as flight 5689, experienced an uncommanded roll while on approach to Kent County International Airport (GRR), Grand Rapids, Michigan. The certificated airline transport flight crew, 1 flight attendant, and 18 passengers were not injured. Visual meteorological conditions prevailed for the scheduled passenger flight that originated from the Greater Cincinnati/Northern Kentucky Regional Airport, Covington, Kentucky at 1105. Flight 5689 was operated on an instrument flight rules (IFR) flight plan under 14 CFR Part 121.

The Captain reported they were flying a visual approach to runway 35 and were following the localizer to intercept the glideslope. When flaps 45 were selected, the airplane rolled right about 15 to 20 degrees. He applied left aileron to correct the roll. The Captain reported he received a "FLT SPOILER DEPLOYED" message on the Engine Indication and Crew Alerting System (EICAS). He reported the flight control synoptic page showed everything normal except the right flight spoiler box was yellow and zero displacement was indicted. The Captain discontinued the approach, declared an emergency, and climbed to 5,000 feet mean sea level (msl).

In a written statement, the Captain reported the following:

"We ran the QRH [Quick Reference Handbook] checklist for the flight spoiler deployed caution message and verified that the handle was stowed. We orbited making left-hand turns while we tried to identify the problem and attempted to contact Comair maintenance control for assistance.

After consulting with maintenance and flight operations personnel, and conducting a visual inspection, we determined that we had some sort of failure of the right flight spoiler. We then ran the checklist for a flight spoiler failure without satisfactory results. By this time we had only 2000 lbs. of fuel remaining and we elected to land using the corrected airspeeds per the QRH. The approach and landing was uneventful with normal GLD [Ground Lift Dumping] deployment indicated on touchdown.

A post-flight inspection revealed the right flight spoiler was fully deployed, probably from the GLD on landing and the left spoiler lug had fractured and separated from its respective actuator. Whether the lug failure was the primary failure or a secondary failure in unknown at this time."

In subsequent correspondence, the pilot reported the following information about the indications and messages provided to the pilots during the incident from the Engine Indication and Crew Alerting System (EICAS):

"In our event, the primary EICAS message was "Flight Spoiler Deploy" caution message instead of the "FLT SPLR L(R)" caution message. We were forced to spend time reviewing the contents of the other checklists to see which one would be most applicable to our situation. In this case, both the FLT SPLR and FLAPS TWIST checklists could have been applicable. We were forced to conduct a visual inspection to eliminate the possibility of a flap failure. Furthermore, when we extended the flight spoilers per the QRH checklist the F/CTRL page indicated a symmetrical deployment of the flight spoilers, even though this was not the case."

The EICAS system relies upon the sensor mounted to the actuator to indicate flight control displacement, and it does not sense the actual displacement of the flight controls.

A review of the maintenance records indicated the right flight spoiler had acquired a total of 3987.9 cycles and 4064 landings.

The right flight spoiler assembly was sent to the National Transportation Safety Board's Materials Laboratory for examination. The inspection of the fractured aluminum lug revealed that one fracture surface contained features indicative of fatigue cracking. Fatigue cracking was on two planes of the lug separated by a large ratchet mark. "Cracking on one side to the ratchet mark emanated from the inside diameter," but the exact origin area could not be determined due to mechanical damage to the fracture surface. The report stated, "The width of the fatigue crack region was approximately 0.2 inch, and the crack extended through approximately 75% of the wall thickness."

The report stated, "The inside diameter surface of the lug contained wear damage that was noted completely around the circumference of the lug, but that appeared more extensive in the area of the fatigue crack. The inside diameter surface also contained a circumferential mark… that extended completely around the circumference of the lug. Portions of the mark exhibited metal flow and steps, consistent with one edge of the bearing outer race moving laterally with respect to the lug."

The airplane's manufacturer had issued a Service Bulletin (SB) A601R-57-027 on April 19, 1999, that specified non-destructive inspections of the aluminum spoiler lugs. The manufacturer also started producing spoilers with steel lugs that could replace the original spoilers with aluminum lugs that would terminate the inspection requirement. The accident airplane did not have the maintenance performed on it that would have complied with the Service Bulletin since a Service Bulletin is advisory in nature.

On June 6, 2000, Transport Canada issued Airworthiness Directive (AD) CF-2000-15 that directed that the aluminum flight spoiler lug be inspected by non-destructive means at a threshold of 7,000 cycles since new. The AD also stipulated that the installation of redesigned flight spoiler which utilized steel lugs provided terminating action to the inspection requirements.

On February 22, 2001, Transport Canada issued AD CF-2000-15R1 that lowered the inspection threshold of the aluminum flight spoiler lugs to 3,000 cycles, and at intervals not to exceed 500 cycles. The AD also stipulated that the installation of redesigned flight spoiler which utilized steel lugs provided terminating action to the inspection requirements.

On July 26, 2001, the Federal Aviation Administration issued AD 2001-12-24 that lowered the inspection threshold of the aluminum flight spoiler lugs to 3,000 cycles, and at intervals not to exceed 500 cycles. The AD also stipulated that the installation of redesigned flight spoiler which utilized steel lugs provided terminating action to the inspection requirements.