Narrative:

I was flying a piper supercub with the side window open. From postflt inspection, it appears that 1/2 the sliding side window blew off the airplane and struck the left fuel tank drain while departing the aircraft. I did not notice since the window was already open. I remember the window being present at takeoff. The left drain started leaking. Beforehand I flew 10 mins off the left tank to verify fuel flow then flew 3 hours on the right tank until it indicated 1/8. The fuel indicator on the supercub is a ball in a tube which can be impossible to see when completely full or empty. Since I could see no ball on the left indicator and I had verified the tank full on the ground, I believed it to be full and switched to the left tank. About 1 min later the engine quit but restarted upon going to the right tank. I still believed the left tank to be full and pulled slight positive and negative G's to see if either would reveal the ball in the left tank indicator. The only effect was that the ball for the right tank 'unstuck' and went to empty. I turned straight for the airport and climbed to 4500 ft from 3500 ft. The engine quit again before reaching the airport and I landed in a field about 1 mi short of the runway. I believe the combination of leaking left tank, misleading left fuel gauge, and stuck right fuel gauge were beyond my control. The off-field landing itself was uneventful.

Google
 

Original NASA ASRS Text

Title: PLT OF A PIPER PA18 SUPERCUB MADE AN OFF ARPT EMER LNDG AFTER THE ENG QUIT DUE TO FUEL STARVATION. ONE OF THE CABIN SIDE WINDOWS CAME OFF INFLT EARLIER STRIKING THE L FUEL TANK DRAIN VALVE CAUSING A FUEL LEAK NOT NOTICED BY THE RPTR. THERE WAS NO DAMAGE TO THE ACFT.

Narrative: I WAS FLYING A PIPER SUPERCUB WITH THE SIDE WINDOW OPEN. FROM POSTFLT INSPECTION, IT APPEARS THAT 1/2 THE SLIDING SIDE WINDOW BLEW OFF THE AIRPLANE AND STRUCK THE L FUEL TANK DRAIN WHILE DEPARTING THE ACFT. I DID NOT NOTICE SINCE THE WINDOW WAS ALREADY OPEN. I REMEMBER THE WINDOW BEING PRESENT AT TKOF. THE L DRAIN STARTED LEAKING. BEFOREHAND I FLEW 10 MINS OFF THE L TANK TO VERIFY FUEL FLOW THEN FLEW 3 HRS ON THE R TANK UNTIL IT INDICATED 1/8. THE FUEL INDICATOR ON THE SUPERCUB IS A BALL IN A TUBE WHICH CAN BE IMPOSSIBLE TO SEE WHEN COMPLETELY FULL OR EMPTY. SINCE I COULD SEE NO BALL ON THE L INDICATOR AND I HAD VERIFIED THE TANK FULL ON THE GND, I BELIEVED IT TO BE FULL AND SWITCHED TO THE L TANK. ABOUT 1 MIN LATER THE ENG QUIT BUT RESTARTED UPON GOING TO THE R TANK. I STILL BELIEVED THE L TANK TO BE FULL AND PULLED SLIGHT POSITIVE AND NEGATIVE G'S TO SEE IF EITHER WOULD REVEAL THE BALL IN THE L TANK INDICATOR. THE ONLY EFFECT WAS THAT THE BALL FOR THE R TANK 'UNSTUCK' AND WENT TO EMPTY. I TURNED STRAIGHT FOR THE ARPT AND CLBED TO 4500 FT FROM 3500 FT. THE ENG QUIT AGAIN BEFORE REACHING THE ARPT AND I LANDED IN A FIELD ABOUT 1 MI SHORT OF THE RWY. I BELIEVE THE COMBINATION OF LEAKING L TANK, MISLEADING L FUEL GAUGE, AND STUCK R FUEL GAUGE WERE BEYOND MY CTL. THE OFF-FIELD LNDG ITSELF WAS UNEVENTFUL.

Data retrieved from NASA's ASRS site as of July 2007 and automatically converted to unabbreviated mixed upper/lowercase text. This report is for informational purposes with no guarantee of accuracy. See NASA's ASRS site for official report.