On September 21, 2002, at 1655 central daylight time, a Beech V35 airplane, N9486S, was substantially damaged when it impacted a fence and a ditch during a forced landing following a loss of engine power during takeoff at the Del Rio International Airport, near Del Rio, Texas. The airplane was registered to and operated by the pilot. The private pilot, sole occupant of the airplane, was not injured. Visual meteorological conditions prevailed, and a flight plan was not filed for the 14 Code of Federal Regulations Part 91 personal flight. The cross-country flight was originating at the time of the accident, and was destined for Scottsdale, Arizona. According to the pilot, he departed Brownsville, Texas, on Labor Day for Scottsdale with a planned refueling stop at Pecos, Texas. About 30 minutes past Del Rio, while in cruise flight at 10,500 feet msl, "the engine quit without warning." He immediately made a turn back to Del Rio, and then switched fuel tanks, cycled the boost pump, mixture control, etc., and set up a best glide speed of 105 knots. "After some fiddling [he] determined that the engine would produce some power using the aux pump on the high boost position resulting in a very low power setting." He continued to the Del Rio International Airport and landed the airplane without further incident. After conversations with a local aircraft mechanic, it was determined that the likely problem was the failure of the engine driven fuel pump. The mechanic removed the fuel pump, rebuilt it, and the same fuel pump was reinstalled on the engine. On the day of the accident, the 758-hour pilot completed a preflight during which the fuel was sumped. After the fuel tanks were topped-off, the pilot started the engine. He stated "that the start-up was a bit different from what [he] was used to, in that generally after the engine starts, if there is any stumbling or roughness, a quick pulse of the aux pump on low usually clears any vapor remaining in the fuel line. However, when [he] pulsed the aux pump on low this time, the engine almost died. Operating the engine at about 1,500 RPM for 5-10 seconds seem to cleared it up, and it seemed smooth and operating in a normal manner." The pilot reported that he taxied the airplane to runway 13 and completed a "normal" run up. After performing the before takeoff checks, he taxied the airplane onto the runway, held the brakes, and applied full throttle. "It took a few seconds for the engine to reach 32.5 inches of manifold pressure (takeoff power) as it usually does, and [he] observed the fuel flow to be not at, but very close to the red line pressure (where it is supposed to be). The fuel flow was in roughly the same position on the gauge as it has been in this airplane for many years, so [he] deemed that it was acceptable." At full power, he held the brakes for 30-40 seconds, "checked the cylinder heat-temperature and exhaust gas temperatures and fuel flow, everything was within normal limits" so he started the takeoff roll. About 50 feet agl, the pilot retracted the landing gear and established an 90 knot climb. "Just as the gear had cycled up, about 100 feet AGL, the engine quit completely." He switched fuel tanks and turned on the boost pump, but it had no effect. He selected a field to land in and extended the landing gear and flaps. The nose landing gear "clipped the airport fence, and [he] made a gentle left turn (about 20 degrees) to aim for the longest part of the field." The main landing gear touched down and he held the nose up as long as possible, but when it came down, it started skidding along the ground. The airplane came to a stop upright in a ditch. Examination of the airplane revealed that the low-pressure pump was deteriorated and came apart internally. The pins that hold the fuel pump vanes in place in the drive wheel were broken and a quarter-inch piece of the tiny metal shafts flowed towards the engine and wedged in the mechanical engine-driven fuel pump vanes shearing the shaft at the "fail point."