Guy Norris/SEATTLE
Boeing is sifting through the data from worldwide inspections of almost 2,700, 737 rudder power-control units (PCUs) and will make recommendations on possible design changes to the US Federal Aviation Administration by the end of the month.
The action follows an alert service-bulletin from the company on 1 November, which was immediately given the force of an FAA airworthiness directive, instructing operators to test the PCUs by 12 November.
The move follows tests conducted by the US National Transportation Safety Board (NTSB) which led Boeing to discover a previously unknown failure scenario during analysis of NTSB data. Boeing describes the NTSB test, which concentrated on the effect of a wide temperature range on the operation of the PCU, as a "red herring", but admits that it was also a catalyst for its own discovery of the potential failure mode.
According to the FAA, the tests reveal that rudder-pedal input can cause "deformation in the linkage leading to the primary and secondary slides of the servo valve of the main rudder PCU".
The tests were part of continued investigations into the USAir 737 crash near Pittsburgh, Pennsylvania, on 8 September, 1994, and the unexplained crash of a United Air Lines 737 near Colorado Springs, Colorado on 3 March, 1991.
The dual servo-valve is made up of two concentric valves, which slide either way to control the flow of hydraulic fluid to rudder actuators. "If the secondary slide of the PCU jams in certain positions, this situation could result in rudder motion in the opposite direction to the rudder command," says the FAA. Chief 737 project engineer Mike Denton explains: "In our tests, we discovered that, in some cases, the primary slide did not cancel out the effects of the secondary, so, even though everyone agrees that the NTSB test was invalid, we feel that there is a potential way a jam could occur."
Boeing analysis of known rudder-related incidents has revealed nine confirmed, or suspected, "jam-type" events in 70 million flying hours "-of which one could have been a secondary-slide binding, which is not as severe as a pure jam," says Denton.
He also stresses: "If a jam occurs, it doesn't result in an immediate problem with the rudder and for normal rudder commands and for normal rates. But, if you ask for high-rate rudder commands in the opposite direction to where the jam exists, that's when you would be vulnerable."
Reports from more than 1,100 737s with, "no fault found" had been received as Flight International went to press. "So far, the data is telling us exactly what we expected, that there are no jam conditions on in-service aircraft," he adds.
Boeing is working on two plans: first, a redesign of the servo itself. "I know we will go forward with some kind of changes to fix that condition," confirms Denton.
The second is a series of "product improvements" for the rudder system. These could include a yaw-damper coupler and a rudder-limiter. A similar limiter has been developed for structural reasons on the next-generation 737 series. Changes to pilot training, maintenance practices and procedures are also being considered.
737 rudders: Pittsburgh needs an answer
Source: Flight International