Originally Posted by
tdracer
No, the rotor burst risk did not change due to moving the engine forward - the rotor burst risk changed because the potential rotor burst debris got larger due to a larger diameter engine.
Big Pistons, did you read the NYT link? The ODA didn't approve rotor burst, the FAA manager did, overriding the concerns of some FAA personnel in the process.
The engine certification including containment is not dependent on the size of the engine, the outcome for the blade separation is constant. For a ruptured disk, history has shown that no design is able to be protected from the release of the disk, and no real protection exists beyond avoiding having disk failure. Control system redundancy is nice to have, no doubt, but the A380 burst was looking at a wing fire as much as any other outcome, in an aircraft that had redundancy, which was mostly lost from the destruction wreaked by a single failure...
SWA 1380 lost #13 fan blade on the CFM56-7B just outside of the dove tail. That liberation resulted in the subsequent failure of the nacelle, and liberated components of the nacelle caused the decompression and fatality of the passenger. The earlier SWA3472 liberated a blade forward of the the nacelle, and the nacelle was destroyed by the out of balance loads. As devastating as that outcome is, the fan blade was technically contained in both cases. Had a disk failed, then all bets are off, disk rupture usually results in 3 and occasionally 4 major sections of the disk being released, and each one has catastrophic kinetic energy levels, as in it doesn't matter what flavour the plane is, you don't want to be any where near it. The LEAP engine fan blade is required to be mitigated by the cowl, as the -7B was, a disk, is mitigated by clean living, religion and good maintenance inspections.
As frustrating as the MCAS shambles has been, and the manual trim issue, I would have thought that the manufacturer was justified in the position of the control issues notwithstanding queries by the FAA TAD that may have occurred.
The probability of a disk failure is expected to be an extremely remote event. The disruption of the rudder cable would require the part of the disk, (say 3 items liberated...) with about a 1' arc at about 20 radius hitting the cable/cable supports etc. that ends up being 3 x (1+1) /126 or about 5% likelihood of impact, if you assume that the release is able to be of 3 parts at the same time. The physics of the release don't permit that to be the case, they are released with a slight latency, which reduces the likelihood of a specific part being hit. The rough likelihood of hitting the rudder control is less than 5% on a liberation, between 1.5-3.5%... of an extremely remote initial failure.
The CFM 56 is approaching 1 billion hours of engine operation, and a liberated fan blade caused a single fatality in that time, and no failure of a rudder control has occurred from any disk failure. Halve that time for the aircraft flight hours, and it is still well outside of the area of interest as a risk factor for a certified design.
The rudder control has a risk from being severed by a meteor strike that would be similar levels of likelihood.
Assuming that a failure occurs, then the catastrophic outcome that has caused gnashing of teeth, wailing and renting of cloth is not certain to lead to a hull loss unless it occurs at high speed on a take off roll, and in the early stages of the takeoff flight path up to around V2 +40. Above that speed, control can be maintained effectively with aileron alone, but at the cost of an increased stall speed, well above a normal "in balanced flight" stall, due to the vagaries of yaw-sweep-spoiler rise and the resultant AOA increase required for a given flight path. (the "40kts" is the amount that the actual stall was increased in the RAAF B707 accident off Sale in the late 80's, where a double asymmetric demo was undertaken with rudder boost inhibited... the resultant roll rate was enough to throw an engine off the pylon). A disk failure surprisingly doesn't have a propensity to occur on a takeoff, they happen at idle, cruise, power reduction, and more or less anywhere the 600lb gorilla wants to sit, they are random in nature, and therefore the probability of their occurrence on any takeoff where a loss of a rudder control would then be catastrophic is a function of exposure alone, and that is about 0.25%-0.5% of the total flight time of the B737... multiply that by the overall likelihood, and the likelihood of a consequent rudder problem and you are much more likely to get taken out by the meteor.
I don't object to control systems that have redundant and separated control architecture, that is just good design practice particularly coming from a military background, but I don't think that amongst the other serious issues that have been raised that hysteria is helpful to successfully getting Bill Boeing back in the air with an adequate product that does a fair job for the industry.
The risk following a fan blade failure is that the stabiliser is severely damaged by the subsequent loss of the nacelle, shades of foam shedding on the SST.