Tweet
"NTSB investigators tested one of the rudder control components from the incident airplane, a rollout guidance actuator, at the component’s manufacturer, Collins Aerospace. When the incident actuator and an identical unit from another airplane were tested in a cold environment, the actuators’ function was significantly compromised"
-
I'd be interested in seeing how the 737 rudder actuator design differs from that of the 757. The 737 was not designed for a three-axis autopilot. It became an option on the NG. I wonder if the same overly-complex, crutch mentality that gave us the MAX and MCAS is on display here. -
Isn't this just a twist on the old (and sadly fatal) 737 rudder problem from the 90s? In which case, the underlying fault may pre-date even the NG? Admittedly, that was a problem with un-commanded rudder movement, and this is a situation where the rudder won't move when it _is commanded. But the common cause seems to be the mechanism's vulnerability to moisture ingress and therefore to the heat-thaw-cool-ice cycle. Oversimplification perhaps?Originally posted by Evan View PostComment
-
The original problem (and subsequent crashes) were caused by a jammed secondary slide in the 737 primary rudder PCU. The cause was not moisture ingress but 'thermal shock' from the temperature differential of the hot hydraulic fluid and very cold external temperatures. Once jammed, the internal primary slide could over-travel and cause the rudder to deflect in the opposite direction from the pilot's command. The PCU was (slowly) replaced by a new design that presumably eliminates this threat.Originally posted by flashcrash View Post
The current issue seems to be due to a manufacturing error on a moisture seal (installed backwards?), so the design might be safe if the manufacturing is done correctly. The seal is part of the new Collins Aerospace SVO-730 rudder rollout guidance actuators that allow CATIIIB landings on the 737. But the threat goes beyond rollout: if the unit jam occurs in flight, even though the unit is not in use, it still restricts pilot rudder inputs. The quick fix seems to be to remove the unit, and thus removing CATIIIB capability, until new, properly manufactured units can be installed. That might require some redesign and re-certification of the part, so that might take time. My first requirement would be to modify the design so that the seal can only be installed in the correct manner.
The current issue will not create uncommanded rudder movement. It will only restrict rudder movement. If pilots apply breakout force and the rudder jam is suddenly overcome, the rudder would move very abruptly and overcontrol is the main threat there.Comment
-
I would challenge that. In fact, I am paid for challenging that. If a seal installed inverted can cause an airplane crash, and the seal not being installed inverted depends on an operator not making a mistake, the design sucks. The Japanese invented* poka-yoke some 60 years ago.Originally posted by Evan View Post
(Actually, they didn't invent mistake-proofing but they made it popular by implementing it widely and successfully and make it part of a larger culture of excellence)
[EDIT: I answered the above before finishing reading your post. I eventually got tot he part where you address this too]
--- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
--- Defend what you say with arguments, not by imposing your credentials ---Comment
Comment