On November 30, 1995, at 0728 hours mountain standard time, a Beech 99, N134PM, experienced a partial loss of directional control after takeoff from Salt Lake City, Utah. The 14 CFR Part 135 cargo flight was operated by Ameriflight, Inc., Burbank, California, and was destined for Boise, Idaho. The aircraft was not damaged and the pilot was not injured. Visual meteorological conditions prevailed and the flight was operating on an instrument flight rules flight plan. The pilot reported that shortly after takeoff he felt a sharp jolt in the rudder pedals as the landing gear was retracting, after which the aircraft would no longer respond to left rudder pedal input. The pilot declared an emergency and returned to Salt Lake City where the aircraft landed uneventfully. Postflight examination of the aircraft revealed that the left rudder control cable had separated in the control pulley radius at fuselage station 385.5. The cable (Beech Aircraft part number 115-524063) was installed 3 months prior to the incident and had accumulated 183 hours in service. Ameriflight, Inc., notified the Southwest Region office of the NTSB of the incident on December 1, 1995. At that time, the aircraft had been repaired at the operator's Salt Lake City base and returned to service. The NTSB did not inspect the aircraft until a later date when it was routed through Burbank. According to the operator, when replacing the cable, their personnel reported that the control pulley, cable retainer pins, and pulley bracket in the area of the failure showed evidence of wire brushing from the loose wire strands, but no evidence that the cable had been misrouted. The aircraft was examined by the NTSB investigator on July 11, 1996. The control cable pulley at the location where the cable failed showed no abnormal wear or misalignment, and the interior faces of the adjacent bracket had many fine wire scrape marks. The two cable retaining pins at the cable tangent points on the pulley bracket showed scrape marks at their midpoints. The failed cable sections were sent to the NTSB metallurgy laboratory in Washington, DC for examination. In his report the metallurgist segregated the failed cable ends into three groups. In the first two groups he stated consisted of failed cable ends had been 1) damaged, or 2) extensively damaged "to the extent that very little of the original fracture features remained." Of the third group, he stated "about 10 percent of the wire ends that were examined in detail contained lesser amounts of damage and at least some evidence of fatigue cracking in the form of longitudinal splits, flat fracture areas (generally with curved boundaries), or microfissures."