Narrative:

During descent into btm, we were visually dodging numerous thunderstorms and rain showers. There was a heavy storm just a couple of mi west of the airport and a light shower immediately east of the field. The wind was 070 degrees at 22 KTS gusts to 28 KTS. We flew crosswind for runway 11 in a break in the rain and turned left downwind. A low fuel pressure caution for the left engine occurred and the captain turned the electric pumps on. The caution extinguished. I continued a visual pattern and on final briefed possible windshear given the WX. We experienced an increase in performance with airspeed +10 -0 KTS and vertical speed varying +/-300 FPM. I announced the performance increase and continued because it was below the SOP thresholds. At about 300 ft AGL I determined that the indications called for a go around and initiated a visual missed approach by calling it and adding power. We yawed strongly to the left and I saw the left engine at 0% torque. I called for maximum power, flaps 15 degrees (go around setting) and announced the engine failure. The captain verified the failure and feathered the propeller. (Autofeather does not arm in this scenario.) because we were surrounded by high terrain on all sides within 2 mi I flew at V2 with flaps 15 degrees and circled back to the airport. We landed without further incident with no injuries or damage. After landing we discovered the left fuel tank empty and the right tank barely below the level it was before takeoff. The xfeed must have been open in-flight. Although it is possible that we had inadvertently opened it in-flight, neither the captain nor I noticed any of the 4 annunciators which would have been lit up in that case. Maintenance has not yet inspected the valve to see if it is stuck open even though selected closed. Contributing factors include: crew inattention to the fuel gauges. We should have noticed the left tank was getting low on fuel. The QRH (checklist for abnormal and emergency procedures) was insufficient. The only action for low fuel pressure is turn on the electrical pumps. There should be a check of fuel level. Another factor was the WX and terrain, which effectively had us committed to land once in the traffic pattern, so we were not inclined to deeply troubleshoot the original low fuel pressure caution. I should add that the most significant factor in our successful outcome was the outstanding simulator training I had just completed only 2 weeks ago. I am convinced that the extreme scenarios that we see in the simulator gave me the skills and confidence to succeed in real life. Airplane training would not do that, as it is too hazardous to practice this type of problem in an airplane. Callback conversation with reporter revealed the following information: reporter states that this seems to be an isolated problem and may have been the result of human error or the fact that maintenance was performed the night before. The right fuel tank had been opened at that time and it is possible the filter may have been clogged or some xfeed selector malfunction occurred. The company has not yet determined the problem. The captain unfortunately did not follow company procedure, he immediately electrically feathered the propeller without following the normal checklist procedures.

Related reports:

Google
 

Original NASA ASRS Text

Title: FLC OF EMB120 EXECUTES A GAR ON APCH DUE TO WINDSHEAR ACTIVITY. ON PWR UP THE L ENG INDICATES 0 PERCENT TORQUE. ON LNDG THE L TANK WAS EMPTY AND THE R TANK BARELY BELOW THE LEVEL AT DEP.

Narrative: DURING DSCNT INTO BTM, WE WERE VISUALLY DODGING NUMEROUS TSTMS AND RAIN SHOWERS. THERE WAS A HVY STORM JUST A COUPLE OF MI W OF THE ARPT AND A LIGHT SHOWER IMMEDIATELY E OF THE FIELD. THE WIND WAS 070 DEGS AT 22 KTS GUSTS TO 28 KTS. WE FLEW XWIND FOR RWY 11 IN A BREAK IN THE RAIN AND TURNED L DOWNWIND. A LOW FUEL PRESSURE CAUTION FOR THE L ENG OCCURRED AND THE CAPT TURNED THE ELECTRIC PUMPS ON. THE CAUTION EXTINGUISHED. I CONTINUED A VISUAL PATTERN AND ON FINAL BRIEFED POSSIBLE WINDSHEAR GIVEN THE WX. WE EXPERIENCED AN INCREASE IN PERFORMANCE WITH AIRSPD +10 -0 KTS AND VERT SPD VARYING +/-300 FPM. I ANNOUNCED THE PERFORMANCE INCREASE AND CONTINUED BECAUSE IT WAS BELOW THE SOP THRESHOLDS. AT ABOUT 300 FT AGL I DETERMINED THAT THE INDICATIONS CALLED FOR A GAR AND INITIATED A VISUAL MISSED APCH BY CALLING IT AND ADDING PWR. WE YAWED STRONGLY TO THE L AND I SAW THE L ENG AT 0% TORQUE. I CALLED FOR MAX PWR, FLAPS 15 DEGS (GAR SETTING) AND ANNOUNCED THE ENG FAILURE. THE CAPT VERIFIED THE FAILURE AND FEATHERED THE PROP. (AUTOFEATHER DOES NOT ARM IN THIS SCENARIO.) BECAUSE WE WERE SURROUNDED BY HIGH TERRAIN ON ALL SIDES WITHIN 2 MI I FLEW AT V2 WITH FLAPS 15 DEGS AND CIRCLED BACK TO THE ARPT. WE LANDED WITHOUT FURTHER INCIDENT WITH NO INJURIES OR DAMAGE. AFTER LNDG WE DISCOVERED THE L FUEL TANK EMPTY AND THE R TANK BARELY BELOW THE LEVEL IT WAS BEFORE TKOF. THE XFEED MUST HAVE BEEN OPEN INFLT. ALTHOUGH IT IS POSSIBLE THAT WE HAD INADVERTENTLY OPENED IT INFLT, NEITHER THE CAPT NOR I NOTICED ANY OF THE 4 ANNUNCIATORS WHICH WOULD HAVE BEEN LIT UP IN THAT CASE. MAINT HAS NOT YET INSPECTED THE VALVE TO SEE IF IT IS STUCK OPEN EVEN THOUGH SELECTED CLOSED. CONTRIBUTING FACTORS INCLUDE: CREW INATTENTION TO THE FUEL GAUGES. WE SHOULD HAVE NOTICED THE L TANK WAS GETTING LOW ON FUEL. THE QRH (CHKLIST FOR ABNORMAL AND EMER PROCS) WAS INSUFFICIENT. THE ONLY ACTION FOR LOW FUEL PRESSURE IS TURN ON THE ELECTRICAL PUMPS. THERE SHOULD BE A CHK OF FUEL LEVEL. ANOTHER FACTOR WAS THE WX AND TERRAIN, WHICH EFFECTIVELY HAD US COMMITTED TO LAND ONCE IN THE TFC PATTERN, SO WE WERE NOT INCLINED TO DEEPLY TROUBLESHOOT THE ORIGINAL LOW FUEL PRESSURE CAUTION. I SHOULD ADD THAT THE MOST SIGNIFICANT FACTOR IN OUR SUCCESSFUL OUTCOME WAS THE OUTSTANDING SIMULATOR TRAINING I HAD JUST COMPLETED ONLY 2 WKS AGO. I AM CONVINCED THAT THE EXTREME SCENARIOS THAT WE SEE IN THE SIMULATOR GAVE ME THE SKILLS AND CONFIDENCE TO SUCCEED IN REAL LIFE. AIRPLANE TRAINING WOULD NOT DO THAT, AS IT IS TOO HAZARDOUS TO PRACTICE THIS TYPE OF PROB IN AN AIRPLANE. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: RPTR STATES THAT THIS SEEMS TO BE AN ISOLATED PROB AND MAY HAVE BEEN THE RESULT OF HUMAN ERROR OR THE FACT THAT MAINT WAS PERFORMED THE NIGHT BEFORE. THE R FUEL TANK HAD BEEN OPENED AT THAT TIME AND IT IS POSSIBLE THE FILTER MAY HAVE BEEN CLOGGED OR SOME XFEED SELECTOR MALFUNCTION OCCURRED. THE COMPANY HAS NOT YET DETERMINED THE PROB. THE CAPT UNFORTUNATELY DID NOT FOLLOW COMPANY PROC, HE IMMEDIATELY ELECTRICALLY FEATHERED THE PROP WITHOUT FOLLOWING THE NORMAL CHKLIST PROCS.

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.