CEN25LA117

During the initial climb when the airplane reached 800 ft above ground level (agl), the airplane shuddered and shook violently. The pilot reduced the throttle, which reduced the shaking, then completed a forced landing in a field. After landing, he noticed that one of the three propeller blades had separated from the hub and was found nearby in the field. The airplane sustained substantial damage to the engine mount. Examination of the propeller assembly revealed that the No. 2 propeller blade likely failed due to fitment issues within the propeller hub, leading to operational vibrations to be transferred directly and unevenly into the blade root, which resulted in fatigue cracking and catastrophic failure. The primary feature that initiated the failure was a likely improperly machined bearing seat in the No. 2 blade slot. The initiation point of the primary fatigue crack of the blade root correlated with the missing region of the bearing seat. The missing bearing seat allowed the blade root fillet to contact the hub. The bearing seat inside the blade slot was the surface that the cartridge bearing radially interfaces with, permitting the centripetal load of the propeller blade to evenly transfer into the hub. In the slots for blade Nos. 1 and 3, this surface was present and appeared mostly consistent around the circumference of the lip. However, the No. 2 blade seat vanished at the 12 and 6 o’clock positions. This would concentrate the centripetal forces generated by the propeller blade to the corners of the housing shells, where cracking was evident in three of the four locations. It is unknown why the No. 2 blade slot would be machined differently than the Nos. 1 and 3 slots. Additionally, the fact that this feature was observed on both the forward and rear shells, which are different parts and would be run on different processes, suggested issues may be related to programming or fixturing of the parts. The fact that two occurrences were not detected before release of the product raises questions about the presence of a quality control system issue for the manufacturer. The propeller hub exhibited several additional features that exhibited undesirable machining characteristics. The internal threaded section of the blade root exhibited chatter marks running along its length. Chatter marks are typically undesirable machining marks that form when running a given cutting tool through a material faster or with more force than its ability to evenly cut through the material. This can lead to resonant vibrations in the tool piece and cause a wavy or uneven pattern in the work piece. In high performance components, these may create unaccounted for stress concentrations leading to fracture initiation and growth. While there was no evidence that these were a primary driver of the failure, they appeared to contribute to the fracture propagation. Additionally, horizontal machined features were observed on several of the slots, consistent with toolingdamage or features occurring during the roughing process of the machining the shell. Similar features were present on both rear and forward shells to varying degrees of severity. It was unclear whether there was an intended purpose for them in the design, but their inconsistency suggested a machining disconformity. While there was no evidence that any of them directly contributed to the failure of the propeller blade, their presence raised questions about the manufacturers machining quality control processes. Machining surface quality control of both the components of the blade roots and the hub assembly was also a contributing factor. Initiation of the primary crack along a single machining mark suggested its coarseness created stress raisers leading to fatigue crack initiation. While no information was provided about what design requirements exist for these specific components, certificated aircraft propellers have strict machining and finishing requirements due to their high vibration environments and critical risks.

On March 10, 2025, about 1240 central daylight time, a Vans RV-7A experimental airplane, N923ZA, was substantially damaged when it was involved in an accident near Plainview, Texas. The pilot was not injured. The airplane was operated as a Title 14 Code of Federal Regulations Part 91 personal cross-country flight. The pilot reported that the accident flight was the second leg of the flight. The first leg of the flight was about 2 hours and 15 minutes and was uneventful. After landing, the airplane was refueled, and he was in the process of departing for the destination airport. He reported that the takeoff was “smooth” until he reached 800 ft agl, when the airplane shuddered and shook violently. He reduced the throttle, which reduced the shaking, then completed a forced landing in a field. After landing, the pilot noticed that one of the three propeller blades had separated from the hub and was found nearby in the field. Postaccident examination revealed the engine mount was cracked. A review of the maintenance logbooks revealed that the propeller was installed on December 19, 2023, and had accumulated about 25 hours of time in service. During the last two condition inspections, the propeller had been inspected and was determined to be in airworthy condition. Examination of blade Nos. 1 and 3 did not find any indications of cracking and both were visually intact. The liberated blade was identified as blade No. 2, and the cracking appeared to emanate from a fillet on the blade root. The primary fracture progressed transversely through the hollow blade root, intersecting the three radial mounting pins, and final fracture occurred on the opposite side at approximately a 120° angle to the origin. Progressive growth was observed across the majority of the fracture surface, consistent with high cycle and low nominal stress loading. Adjacent to both the fracture initiation and final overstress regions, a portion of the machined fillet surface was found intact, exhibiting heavy wear marks indicating sliding contact against the hub housing. Surfaces adjacent to the wear path exhibited coarse machining marks, from which the crack appeared to originate. A single ratchet mark was observed towards one end of the region, while the rest of the origin appeared to emanate from a common plane. Several secondary thumbnail cracks appeared to have initiated from the corners of radial retention pin holes, often progressing until terminating at the intersection with the primary fracture progression or region of final overstress. Black greasy witness marks indicating the position of the blade root were observed within the pin holes suggesting that the top cartridge bearing inner diameter ring had slid at some point for an unknown reason. The No. 2 blade root fracture surface exhibited a feathery appearance with tear ridges, consistent with high-cycle fatigue. Numerous fine secondary cracks oriented perpendicular to the progression direction were observed going into thefracture surface. The origin sustained some fracture rub but appeared to largely emanate along a single machine mark spanning the region. Marks from other machine passes were observed parallel to origin. Numerous ratchet marks were observed midway through the progression on the inner diameter of the root, indicative of multiple fracture origins that quickly coalesced with the primary fracture. Electron microscopy revealed these origins emanated from machining chatter marks observed on the adjacent threads. It was unclear whether these thread features were permissible according to the design of the blade root. Portions of the No. 2 blade root were missing. Some secondary fracture propagation appeared to initiate from the fillet region and progress within the region of the first cartridge bearing interface, resulting in the separation of blade root fragments. The bearing interface surfaces that were still present appeared oxidized and had patches of galling, consistent with fretting and suggesting an amount of vibratory movement at the inner diameter (ID) bearing ring/blade root interface. It was unclear if an insecure bearing fitment at this interface was permissible by design. The propeller hub exhibited damage consistent with the radial ejection of the No. 2 blade on the No. 2 blade hub slot. Large cracks were observed on three of the four corners of the clamshell No. 2 blade pocket. One of these cracks on the forward shell exhibited an outward deformation, revealing a thumbnail progressive fracture emanating from the ID corner of the shell. The orientation of the crack corresponded with aggressive contact wear observed on the inner lip of the blade opening. Regions of the blade root fillet radius that exhibited aggressive contact wear corresponded to wear observed on the inner lip of the hub blade opening, indicating contact between the blade root and hub. No manufacturer documentation was provided to indicate if this was an intended design of the assembly. Clearance was observed between the hub opening and the blade root of No. 3 blade, allowing the blade pitch to rotate freely without contact. However, this clearance was not present in the assembled No. 2 blade, permitting contact between the blade root and propeller hub opening. Measurements between the bottom of the blade root and the fillet radius appeared consistent across all three blades and no significant dimensional deviation was observed. However, examinations of the No. 2 blade hub slot indicated a discrepancy compared with the slots for blade Nos. 1 and 3. In both blade Nos. 1 and 3, a full-circumference seating surface is machined into the lip that interfaces with the outer ring of the cartridge bearings. However, in the No. 2 blade slot, this surface blended into the machining of the slot opening lip, starting at the hub parting plane and disappearing altogether at the base of the slot (see figure 1). When properly configured with the fractured blade root, the initiation pointof the primary fatigue crack, as well as the adjacent contact wear on the fillet, correlated with the missing region of the bearing seat. Similar bearing seat features were also observed on the rear shell No. 2 blade slot. Figure 1. Annotated images of the inner lip surfaces of the forward shell Nos. 2 and 3 blade slots. The bearing seat surface (contacting the top cartridge bearing) radial boundaries are identified via dashed lines, with the inner boundary denoted with orange and the outer with red. While both the Nos. 1 and 3 blade forward slots exhibited a continuous flat seat surface, the No. 2 blade seating surface disappears along the middle portion of the lip (Source: NTSB Materials Laboratory Factual Report 25-089). The NTSB investigated a similar AXSport propeller failure, CEN24LA038, in November 2023. Images of the blade root in this investigation exhibited similar coarse machining and chatter markings on the subject blade root. In both cases, the blade root of the propeller failed due to transverse fatigue progression through the component. While the primary cracks did not initiate in the same location as this investigation, the presence of similar undesirable machining features as well as similar progression influence in the same component was evident.