Narrative:

A pilot was plting a government aircraft on a training scenario. The scenario called for the pilot to land at X35 as a 'target aircraft' while being pursued. The pilot approached the airport from the southwest. He advised dunnellon unicom that he was entering a right base for runway 9. Approximately 3 mi from the airport the pilot, using an approved checklist, lowered the flaps to the first notch (15 degrees). As the aircraft slowed to gear extension speed, he lowered the landing gear, set the propellers at 2400 RPM, increased the mixture to full rich, turned the boost pumps on and proceeded to change tanks to the inboard tanks. As the pilot looked up at the fuel boost pump switches, he noticed the red 'right fuel boost inoperative' light illuminate. The pilot immediate turned both emergency fuel boost pumps on and switched both fuel tanks to the inboard tanks. At this time the right engine started surging, with rapid RPM indications up and down. The pilot then reset the fuel boost pump switch on the right engine, with the same surging results. The aircraft was turning to the right to set up the final approach. On final approach the pilot noticed that the left engine was now surging and both engines were not surging in unison, causing the aircraft to yaw 30-40 degrees nose left and right. The aircraft continued in a descent as the pilot was unable to maintain level flight. The pilot then decided to try the outboard tanks due to the rough engines on the inboard tanks. The surging of the engines continued and the pilot went back to the inboard tanks. The altitude of the aircraft was approximately 100 ft AGL at this time and continuing to descend. The pilot decided that there was not enough airspeed or altitude at this point to make the landing on the runway. To regain control of the aircraft, the pilot pulled the power levers to idle and selected an open field to conduct an off-airport landing. The actual selected landing site is located approximately 1/2 mi wnw from the end of runway 9 at dunnellon. The aircraft touched down at approximately 85 KIAS, with a nose high attitude. During the landing rollout, the aircraft struck a barbed-wire fence. As the aircraft rolled to a stop, the pilot turned off all switches and power and evacuate/evacuationed the aircraft through the left pilot's door. It is unclr at this time as to why the engines both quit, so a recommendation would be difficult. It appeared to the pilot that both engines were vapor-locking. Maybe the pilot should have changed fuel tanks earlier than he did at a high altitude and with a constant power setting. Callback conversation with reporter revealed the following information: reporter states that there was a small amount of contamination in the fuel but not enough to cause this type of problem. The normal procedure is to switch tanks prior to descent checklist use. In this case, reporter was by design so low that there was no actual descent. Piper uses a 5 mi 2000 ft frame for tank switching so one has altitude to recover if there is a problem. This operation did not allow for that altitude cushion to be able to troubleshoot. Some mechanics think it might have been a vapor lock which, because of low altitude, did not have time to clear. The actual decision time was 15-20 seconds. Damage to the aircraft was minimal. The fence was old and rickety with a fence post slightly denting the wing and the barbed wire creasing the nose of the aircraft.

Related reports:

Google
 

Original NASA ASRS Text

Title: PLT OF PA31 ON A TRAINING FLT HAS ENG SURGING ON BOTH ENGS BUT IN UNISON CREATING A 30-40 DEG YAW. AN OFF ARPT LNDG WAS MADE TO LAND WITH SOME ACFT CTL.

Narrative: A PLT WAS PLTING A GOV ACFT ON A TRAINING SCENARIO. THE SCENARIO CALLED FOR THE PLT TO LAND AT X35 AS A 'TARGET ACFT' WHILE BEING PURSUED. THE PLT APCHED THE ARPT FROM THE SW. HE ADVISED DUNNELLON UNICOM THAT HE WAS ENTERING A R BASE FOR RWY 9. APPROX 3 MI FROM THE ARPT THE PLT, USING AN APPROVED CHKLIST, LOWERED THE FLAPS TO THE FIRST NOTCH (15 DEGS). AS THE ACFT SLOWED TO GEAR EXTENSION SPD, HE LOWERED THE LNDG GEAR, SET THE PROPS AT 2400 RPM, INCREASED THE MIXTURE TO FULL RICH, TURNED THE BOOST PUMPS ON AND PROCEEDED TO CHANGE TANKS TO THE INBOARD TANKS. AS THE PLT LOOKED UP AT THE FUEL BOOST PUMP SWITCHES, HE NOTICED THE RED 'R FUEL BOOST INOP' LIGHT ILLUMINATE. THE PLT IMMEDIATE TURNED BOTH EMER FUEL BOOST PUMPS ON AND SWITCHED BOTH FUEL TANKS TO THE INBOARD TANKS. AT THIS TIME THE R ENG STARTED SURGING, WITH RAPID RPM INDICATIONS UP AND DOWN. THE PLT THEN RESET THE FUEL BOOST PUMP SWITCH ON THE R ENG, WITH THE SAME SURGING RESULTS. THE ACFT WAS TURNING TO THE R TO SET UP THE FINAL APCH. ON FINAL APCH THE PLT NOTICED THAT THE L ENG WAS NOW SURGING AND BOTH ENGS WERE NOT SURGING IN UNISON, CAUSING THE ACFT TO YAW 30-40 DEGS NOSE L AND R. THE ACFT CONTINUED IN A DSCNT AS THE PLT WAS UNABLE TO MAINTAIN LEVEL FLT. THE PLT THEN DECIDED TO TRY THE OUTBOARD TANKS DUE TO THE ROUGH ENGS ON THE INBOARD TANKS. THE SURGING OF THE ENGS CONTINUED AND THE PLT WENT BACK TO THE INBOARD TANKS. THE ALT OF THE ACFT WAS APPROX 100 FT AGL AT THIS TIME AND CONTINUING TO DSND. THE PLT DECIDED THAT THERE WAS NOT ENOUGH AIRSPD OR ALT AT THIS POINT TO MAKE THE LNDG ON THE RWY. TO REGAIN CTL OF THE ACFT, THE PLT PULLED THE PWR LEVERS TO IDLE AND SELECTED AN OPEN FIELD TO CONDUCT AN OFF-ARPT LNDG. THE ACTUAL SELECTED LNDG SITE IS LOCATED APPROX 1/2 MI WNW FROM THE END OF RWY 9 AT DUNNELLON. THE ACFT TOUCHED DOWN AT APPROX 85 KIAS, WITH A NOSE HIGH ATTITUDE. DURING THE LNDG ROLLOUT, THE ACFT STRUCK A BARBED-WIRE FENCE. AS THE ACFT ROLLED TO A STOP, THE PLT TURNED OFF ALL SWITCHES AND PWR AND EVACED THE ACFT THROUGH THE L PLT'S DOOR. IT IS UNCLR AT THIS TIME AS TO WHY THE ENGS BOTH QUIT, SO A RECOMMENDATION WOULD BE DIFFICULT. IT APPEARED TO THE PLT THAT BOTH ENGS WERE VAPOR-LOCKING. MAYBE THE PLT SHOULD HAVE CHANGED FUEL TANKS EARLIER THAN HE DID AT A HIGH ALT AND WITH A CONSTANT PWR SETTING. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: RPTR STATES THAT THERE WAS A SMALL AMOUNT OF CONTAMINATION IN THE FUEL BUT NOT ENOUGH TO CAUSE THIS TYPE OF PROB. THE NORMAL PROC IS TO SWITCH TANKS PRIOR TO DSCNT CHKLIST USE. IN THIS CASE, RPTR WAS BY DESIGN SO LOW THAT THERE WAS NO ACTUAL DSCNT. PIPER USES A 5 MI 2000 FT FRAME FOR TANK SWITCHING SO ONE HAS ALT TO RECOVER IF THERE IS A PROB. THIS OP DID NOT ALLOW FOR THAT ALT CUSHION TO BE ABLE TO TROUBLESHOOT. SOME MECHS THINK IT MIGHT HAVE BEEN A VAPOR LOCK WHICH, BECAUSE OF LOW ALT, DID NOT HAVE TIME TO CLR. THE ACTUAL DECISION TIME WAS 15-20 SECONDS. DAMAGE TO THE ACFT WAS MINIMAL. THE FENCE WAS OLD AND RICKETY WITH A FENCE POST SLIGHTLY DENTING THE WING AND THE BARBED WIRE CREASING THE NOSE OF THE ACFT.

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.