On May 16, 1998, at 1925 eastern daylight time, a Piper PA-28-151, N56969, was substantially damaged during a forced landing near the Princeton Airport (39N), Princeton, New Jersey. One passenger received minor injuries. The certificated commercial pilot and two passengers were not injured. Visual meteorological conditions prevailed for the personal flight that originated at 39N approximately 1920. No flight plan was filed for the flight conducted under 14 CFR Part 91. The pilot reported to the Montgomery Township Police that he departed 39N with 3 passengers and "...at least 40 gallons of fuel in the tanks." The pilot stated the engine lost power during climbout and he attempted to return to the airport. He determined he could not reach the airport and performed a forced landing to a field. In a written statement, the pilot said approximately 400 feet on climbout the engine began to run "rough", backfire, and the airplane would not climb. He said there were no forced landing areas available and he turned for the airport. The pilot said he "lost" the engine in the middle of the turn. An airport employee who witnessed the departure of N56969 was interviewed by telephone. The employee held a commercial pilot certificate and stated that he noticed the airplane taxiing for takeoff from "...the wrong runway." He said the winds and the windsock favored a departure to the east. The witness said he contacted the airport manager by two-way radio in an effort to warn the pilot. He said the airport manager attempted to contact the pilot of N56969 over the UNICOM frequency several times with no response. The witness said N56969 took off to the west from runway 28 "...with a tailwind." When questioned about the sound of the engine during the takeoff roll, he said: "The engine sounded fine. It didn't sound rough. It sounded like a normal take off to me." Winds reported at Trenton, New Jersey, 13 miles southwest of 39N, were from 110 degrees at 9 knots. The engine was examined and a test run was performed under the supervision of a Federal Aviation Administration (FAA) Aviation Safety Inspector at the Textron Lycoming factory, Williamsport, Pennsylvania, on October 6, 1998. Examination of the engine revealed the carburetor was separated from the engine during the accident sequence. The carburetor was reattached to the engine, and the engine was installed in the test cell. After two failed attempts at engine start, a carburetor float was freed and the engine started immediately on the third attempt. The engine ran to rated power. According to the engine test summary: "The engine test run results show that the engine was capable of running and demonstrating operational power. There was only one discrepancy or anomaly that would indicate that the engine was not capable of running and producing power prior to mishap, this was the sticking float in the carburetor. It also needs to be noted that the carburetor was shipped upside down in a separate box. The carburetor float could have stuck while up side down in the shipping box." After completion of the test run, the carburetor was removed from the engine and shipped to Precision Airmotive Corporation, Everett, Washington, for examination and flow test. The examination and flow test was completed under the supervision of an NTSB investigator on December 4, 1998. Examination of test results revealed the carburetor flowed within operational limits. The airplane's weight and balance data sheet listed a maximum takeoff weight of 2,325 pounds. The actual takeoff weight of the airplane at departure could not be verified due to an undetermined amount of baggage on board. However, calculations revealed that the weight of the fuel and the occupants only, resulted in an airplane weight of 2,366 pounds. The pilot reported 362 hours of total flight experience, 120 hours of which were in make and model.