On August 10, 2011, about 1900 central daylight time, N3659N, an Air Tractor AT-400, N3659N, sustained substantial damage when it landed short of a private airstrip near Nelson, Nebraska. The commercial rated pilot was not injured. The airplane was registered to and operated by the pilot. Visual meteorological conditions prevailed and no flight plan was filed for the local aerial application flight conducted under 14 Code of Federal Regulations Part 137. The flight originated at the private airstrip approximately 1830. The pilot stated that when he reduced power to land, the power would not decrease. So he shut the engine down by pulling the fuel cut-off lever. The pilot misjudged the altitude he needed to glide to the runway and landed short resulting in substantial damage to the airframe. The fuel pump, fuel control unit, and the propeller governor were removed from the Pratt &Whitney Canada (PWC) PT-6 engine for further examination. The governor was examined at Woodward, Incorporated, Rockford, Illinois, under the supervision of the National Transportation Safety Board (NTSB) and no anomalies were noted that would have precluded normal operation of the unit. The fuel pump and the fuel control unit were sent to PWC and examined under the supervision of the Transportation Safety Board of Canada. No anomalies were noted with the fuel pump when tested; however, the fuel control unit did not meet test specifications during the functional check. The unit was disassembled and it was noted that the spool cap bearing and its associated hardware were damaged. The damaged parts, which included the cap containing the bearing outer race, the post with the bearing inner race, a retainer, a shield, a second retainer and six ball bearings, and six of seven ball bearings, were sent to the NTSB Materials Laboratory, Washington DC, for metallurgical examination. Examination of the ball bearings revealed they were smaller than their original dimensions and exhibited wavy lines or a frosted appearance. The #5 ball bearing was metallurgically encapsulated, ground down to its diameter, and polished for micro-hardness testing. The bearing’s hardness was within tolerances. The two retainers were thin circular ribbons made of metal. Both retainers were mechanically damaged. One retainer exhibited fractured ends with visible bulges, consistent with the original location of a ball bearing. There was also a folded area that displayed a smooth bulge, consistent with it being the location of the seventh ball. The second retainer was also fractured at each end and exhibited defined bulges, consistent with the original location of each ball bearing. The shield was intact and undamaged. A wire clip used to retain the shield was not present with the cap. Examination of the inner surface of the bearing's cap (outer race) revealed circular impressions and a rectangle-shaped smear of material. The smeared material was consistent with a rolling contact path produced by the fuel control arm bearing during governor operation. The inner and outer edges of the cap’s annulus were not well defined; however, measurements of each diameter were consistent with each other. Examination of the post, with the inner race of the bearing still installed revealed two parallel impressions on the surface consistent with normal contact of the counterweights. The depth of the impressions was found to gradually taper becoming deeper at the opposing ends. Sections of the bearing's inner and outer races were examined using a scanning electron microscope (SEM). Colonies of aluminus particles were found embedded in both sections. Other sections of the races were metallurgically encapsulated and polished for hardness testing. Both sections tested within specifications. Honeywell, who manufactured the fuel control unit, had a history of premature spool cap bearing failures in fuel control units installed on their LT-101 engines. These failures were attributed to high diamond content and other hard particle contamination. Honeywell took immediate corrective action to determine the root cause of these failures and implemented new manufacturing procedures with the spool cap bearing supplier. A Federal Aviation Administration (FAA) Airworthiness Directive was also issued. The NTSB’s metallurgical findings on the spool cap bearing from the PT-6 engine were compared to examination data of the failed spool cap bearings on the LT-101 engines, and a direct link between the failures could not be established. Though hard aluminus particles were found embedded in the inner and outer race of the bearing on the PT-6 engine, there was insufficient historical data to support that these particles had attributed to any previous bearings failures. The fuel control unit was overhauled and re-installed on the airplane in November, 2010. It had accrued a total of 197.4 hours at the time the spool cap bearing failed. Failure of the spool cap bearing will result in power/torque fluctuations, either higher or lower than the set metered fuel flow conditions.