Anonymous

The FAA confirmed in NPRM 2007-NE-37, issued on April 4, 2008, that at least twenty-four Superior Air Parts, Inc. (SAP) Parts Manufacturer Approval (PMA) investment cast cylinder assemblies have already separated, or cracked through the cylinder wall in the area of the exhaust valves. The NPRM states that some of those in-service failures of these critical engine components resulted in emergency forced landing of the aircraft. According to the NPRM, all of those cylinders reportedly failed between 823 and 1,985 hours time-in-service. The FAA stated in the NPRM that the failures are due to a SAP cylinder design that is significantly thinner than that of the original equipment manufacturer (OEM) design in the exact location where the cylinders are failing. The FAA stated in the NPRM that all cylinders produced to that design have the aforementioned inherent design deficiency. The above listed NPRM statements raise serious questions relative to the wisdom of allowing all of these under-designed cylinders to remain in service until their normal TBO. The FAA proposal that allows them to remain in service and only be visually inspected and subjected to a standard compression test with application of a soap water solution every fifty hours time-in-service until those cylinders reach their normal TBO seems ill advised. A soap inspection might be a useful technique for locating holes in bicycle or automobile tires, but it hardly seems appropriate for locating cracks in aircraft cylinders, which are critical aviation parts. One wonders if anyone would even consider this proposal to be an acceptable corrective action to address a similar deficiency with their personal automobiles, which clearly are not at the same level of risk as aircraft in flight. Such an inspection technique would not even be able to detect a cylinder crack that has not yet propagated through the entire wall thickness. It is requested that the FAA provide answers to the following questions- • Does the FAA really believe that the soap inspection technique will identify all cylinders that are cracked, including those that are nearly cracked all the way through the cylinder wall? If yes, please provide a detailed supporting explanation. • Did the FAA perform controlled sensitivity engine testing to quantify the probability of properly identifying a failed cylinder in the first place? • What is the (engineering) basis for the fifty hour inspection interval? • Did the FAA confirm the validity of this inspection interval by conducting actual FAA engine testing to determine crack propagation rate under controlled test conditions with a cylinder that has a predetermined crack length, but has not yet cracked completely through the wall in the failure location? • Or, did the FAA substantiate by a legitimate engineering crack growth analysis that a cylinder that is nearly cracked all the way through the cylinder wall would not fail within the next fifty hour inspection interval, when subjected to the expected internal operating pressures and temperatures that are experienced by those cylinders in service? If the answers to these questions is no, then the FAA has no real engineering data to justify the proposed corrective action, nor does it have any rational engineering basis for allowing the cylinders to remain in service beyond the lowest time-in-service at which any of the failures has occurred. Since the FAA itself states in the NPRM that the lowest time-in-service of a failed cylinder was 823 hours, then all cylinders of this design should be removed before they reach 823 hours time-in-service. This should be further reduced by application of appropriate knock-down safety factors, which are typically applied for such situations to account for the uncertainty in the data. Finally, the proposed NPRM also provides inconsistent instructions relative to the area that is to be inspected. In paragraph (f) (1), it is stated that each cylinder should be visually inspected around the exhaust valve side for cracks, but paragraph (f)(3)(ii) refers to leakage at the head- to-cylinder interface, which is a completely different area of the cylinder assembly. In conclusion, instead of acting appropriately to correct the documented inherently unsafe design of these cylinders, the FAA has proposed in the NPRM a course of action that does not address the fundamental safety concern associated with these cylinders and it perpetuates the risk of a catastrophic accident to the flying public and to the public on the ground under which these aircraft fly. This NPRM is inconsistent with the FAA’s safety mission. The FAA should therefore revise this NPRM to require removal of all of these under-designed cylinders at less than 823 hours time-in service in the final rule.