HISTORY OF FLIGHTOn August 1, 2014, at 1028 mountain daylight time, an experimental amateur-built Gerald Spittler (Rans Inc.) S-16 Shekari, N116GS, lost engine power following takeoff from Fort Bridger Airport, Fort Bridger, Wyoming. The airplane was registered to, and operated by, the pilot under the provisions of 14 Code of Federal Regulations Part 91 as a local flight. The private pilot and passenger were not injured, and the airplane sustained substantial damage during the forced landing. Visual meteorological conditions prevailed, and no flight plan had been filed. The pilot reported performing an uneventful preflight check and engine run-up. He taxied onto the runway and applied full engine throttle control, and the airplane accelerated appropriately. Having reached a speed of 65 mph about 1,500 feet down the runway, he initiated a climb to 30 feet above ground level (agl), remaining over the runway as the airplane accelerated. Having attained a speed of 100 mph, he applied aft elevator control and the airplane climbed to about 200 feet agl as he approached the end of the runway. Shortly thereafter, the engine began to hesitate, and then lose all power. With minimal altitude, the pilot elected to maintain the airplane's heading and land on a dirt road ahead. He confirmed the fuel boost pump was on, and applied full rich fuel mixture control with no change in power. Just prior to landing he changed course for a larger road, and during the landing flare the right main landing gear caught the edge of the road and collapsed. The airplane struck a pit and spun around, sustaining substantial damage to the left wing and aft fuselage. The pilot stated that the fuel servo had recently been overhauled due to a low fuel pressure problem. He reinstalled the unit himself following overhaul, and this was the first flight since. An airframe and powerplant mechanic examined the engine following the accident under the direction of the NTSB investigator-in-charge. Examination of the fuel inlet line, which connected the fuel servo to the fuel flow divider, revealed that it was approximately 1 1/3 turns loose at the 'B' nut fitting at the fuel flow divider. The mechanic did not observe any other anomalies that would have precluded normal operation.