ENG10IA055

The failure of the left main landing gear (MLG) to extend normally and manually for reasons that could not be conclusively determined.

While the flight crew was configuring the airplane for landing and shortly after placing the landing gear selector in the "down" position, they noted a "gear disagree" warning message displayed on the engine indication and crew alerting system that showed that the nose gear and the right main landing gear (MLG) were down and locked but that the left MLG was in transit. The flight crew followed the quick reference handbook directions to troubleshoot the landing gear issue without success. The flight crewmembers then tried to extend the landing gear by pulling the alternate landing gear release handle; however, the left MLG failed to extend, and they subsequently landed the airplane with the left MLG retracted. During postincident activities, the airplane was lifted, and an examination revealed that the left MLG remained in its full-up position within the wheel well. The alternate landing gear release handle was found in its fully extended position. The incident airplane's left MLG uplock pin exhibited signs of slight wear and flat spotting, consistent with in-service usage, and the uplock mechanism latch had wear marks that were within the in-service maintenance wear limits. After the incident, functional ground testing of the airplane's landing gear system found that, although wear was observed on these system components, the left MLG extended as designed when the landing gear was selected down normally and manually.  Additionally, the incident airliner's left MLG uplock pin was found properly rigged within its mating uplock mechanism.  Therefore, it is unlikely that the worn components alone would have prevented the left MLG from extending during the incident flight. Functional performance testing of the left MLG sidestay actuator showed that it operated within all test specifications. Although an examination of the actuator's restrictor assembly found that it contained two small pieces of aluminum and several particles of debris this contamination did not prevent the actuator from operating during ground tests after the incident.  Therefore, it is unlikely that the contamination within the left MLG sidestay actuator would have prevented the left MLG from extending during the incident flight. The force to operate the lever on the bypass valve was measured and found to be 60 lbs, which exceeded the acceptance test procedure -specified force requirement of 40 lbs. However, during the incident flight, the nose and both MLG uplock assemblies did unlock when the alternate landing gear release handle was pulled indicating that the bypass valve functioned during the flight. As previously discussed, neither hydraulic contamination within the left MLG actuator nor any single MLG system component was identified as preventing the left MLG from extending during the incident flight.  However, because the left MLG failed to extend when the crew used either the normal or alternate extension systems, it is likely that a combination of several factors contributed to the system malfunction. Some potential factors are: uplock pin rigging and/or wear, uplock mechanism latch wear, hydraulic system pressure and contamination, and bypass valve operation.  Even though these items may be within maintenance limits when considered individually, there may be combinations of these factors that result in failure of the gear to extend.  When combined with environmental and flight related conditions such as temperature, humidity, and landing gear component deflection due to in-flight loading, the interaction results of all of these factors are difficult to predict and to demonstrate during testing.

Flight Data Recorder (FDR) information from the airplane's previous landing shows that upon gear down selection, both the left and the right MLG "up and locked" discrete's transitioned to "not up and locked". Within approximately 12 seconds, both MLG "downlock" discrete's indicated "down and locked". The nose landing gear (NLG) "uplock" discrete transitioned from "up and locked" to "not up and locked" about 3 seconds after gear down selection and required approximately 9 seconds to reach down and locked. Because small pieces of aluminum about 1mm by 3mm in size were found in the left sidestay actuator restrictor manifold, radiographic studies were done on July 21-26, 2011 in Chicago, Illinois to examine and document the internal configuration of the pressure and return hydraulic filters. The filters were documented using a combination of computed tomography (CT) scans and digital radiography. The filters were imaged using a total of 6 digital radiographs (DR) (3 DR's for the pressure filter and 3 DR's for the return filter) and 8,985 CT slices (4,841 slices for the pressure filter and 4,144 slices for the return filter). Review of the images indicated that there were a substantial number of particles (ranging from 0.28 mm to 1.59 mm long) in the pressure filter. The debris inside the pressure filter was composed of both high density (bright) particles and lower density material. Some of the particles were found in clusters, while others were individual particles. There were no particles found in the return filter, however there were several cracks noted in the epoxy material in the end caps of the filter, and there were some high density areas noted within the filter material itself. A review of the CRJ Series Regional Jet Quick Reference Handbook Volume 2, model CL 600-2B19, revision 58 dated October 31, 2005 found that the GEAR DISAGREE checklist included two parts, a "Gear UP Disagree", followed by a "Gear DN Disagree" procedure. Interviews with the flight crew indicated that the procedure to be used was determined by the position of the landing gear selector lever. If the "GEAR DISAGREE" EICAS message was displayed when the selector lever was in the up position, the crew was to use the "Gear UP Disagree" procedure and if the message was displayed when the selector lever was in the down position, the "Gear DN Disagree" procedure was to be used. The captain indicated that he may have initially used the "Gear UP Disagree" procedure but eventually used the "Gear DN Disagree" procedure for trouble shooting prior to landing. According to interviews with the flight crew, upon reaching step 8 "landing gear manual release – stow" of the "Gear DN Disagree" procedure, the captain took over pilot flying duties, handed the QRH to the first officer, and asked him to interpret a note contained within the procedure that states: "In some cases, stowing the landing gear manual release handle may cause the main or nose gear to retract. If this occurs, the landing gear manual release handle must again be pulled to full extension and left in this position until the gear pins have been installed." After discussing their interpretation of the note, the flight crew decided to leave the landing gear manual release handle in the extended position and proceeded to land. The captain stated that the note in the checklist implied to him that the landing gear would come back up if he stowed the manual extension handle. Following this note, was a procedure (step 9) to perform the "Landing Gear Up / Unsafe" landing procedure. The crew did not recall if they had completed this procedure and the captain stated that he was not aware that step 9 referred to another checklist until he reviewed the procedure after the event. An NTSB review of the "GEAR DISAGREE" QRH procedure indicated that it had been revised a number of times since it was originally issued. An earlier version of the checklist included a step to leave the landing gear manual release handle in its extended position after use. The procedure was later revised and included a step to stow the handle after deployment of the landing gear. Revision RJ/98 dated, April 12, 2007, was then issued which maintained the step to stow the handle, and added notes indicating that stowing the handle could cause the landing gear to retract. After the incident, Bombardier Regional Aircraft Division's Flight Operations and Engineering groups initiated a review and evaluation of the GEAR DISAGREE procedure. Based on their review, several editorial changes were made to the GEAR DISAGREE procedure in the AFM and QRH to clarify the procedure. An additional note was included to inform flight crew that they may have to reposition the seat and stand up prior to manual release handle deployment. The actual force to operate the landing gear handle of greater than 40lbs was also included in the note. The procedure has been unified in a single procedure. There are no longer two separate procedures for up and down based on the handle. The current version in the QRH is dated 30 Nov 2012 (Rev. 60). After the incident, SkyWest Airlines published an informational article in an internal company flight operations newsletter available to flight crews. The article provided guidance on the use of the gear disagree checklist procedure and emphasized the identification of the proper procedure, the possibility that the procedure may need to be repeated multiple times to achieve success, and the direction to stow the manual extension handle after the gear has been extended. SkyWest Airlines also modified their CRJ training program to include a discussion of gear disagree events during recurrent ground school, and a gear disagree scenario during flight crew recurrent simulator training. Subsequent to the incident with N498CA, Bombardier revised the contamination limits in reference to hydraulic fluid in the aircraft maintenance manual. The manual was updated on June 9, 2011.The class 9 contamination level per NAS 1638 was previously identified as a "permitted" contamination level. The AMM Task 29-10-00-750-801, Hydraulic Fluid Sample Check was updated to restrict contamination level of the fluid as follows: Class 8: Perform a repeat sample check within 3000 flight hours (FH) Class 9: Within 30 calendar days or 200 FH W.C.F flush system as per AMM Class 10 and above: Flush system as per AMM The AMM task for the ACMP replacement after a pump failure was also updated. It now includes requirements to examine associated pump filter elements (pressure, case drain and return) for contamination and replace if required. In addition to that it requires to take a sample of the hydraulic fluid from the affected system and to do analysis for contamination. Prior to SkyWest's September 2010 and June 2011 (Reference NTSB CEN11IA379) gear up landing events at the General Mitchell Airport (MKE), Milwaukee, Wisconsin, Bombardier required freedom of movement inspections at the 'C' checks (18 mos.) on each MLG uplock roller. After the landing gear events, SkyWest developed a task independent of Bombardier due to a Continuing Analysis and Surveillance System (CASS) board mandate which required the checking of the MLG uplock roller for freedom of movement at each service check (every 5 days). This task is still in effect. During the period of December 5th – 17th, Bombardier conducted a landing gear system study on an exemplar CL600-2B19 airplane in Tucson, Arizona. Present during this testing were representatives from the National Transportation Safety Board, Bombardier Aerospace, Messier-Dowty, Tactair and SkyWest Airlines. The study was comprised of a series of simulation tests that induced failures into the landing gear and hydraulic system to generate data for use in a safety analysis of the landing gear system. To obtain an initial understanding of the operation of a normally operating landing gear