Narrative:

This incident occurred while flying an older pa23r160 on a multi engine cross country training flight from hnl to lih to lny to hnl on dec/xa/00. Prior to leaving hnl, I checked both main fuel tanks, and I saw fuel in both cells slightly below top-off level. The aircraft's main cells each hold 37 gallons, of which 36 gallons are usable, and that gives between 4.5 and 5 hours of engine time at 7-8 gph per engine. Before we left hnl, I spoke to the mei who flew the aircraft on the preceding flight, and he said he had topped off the 2 main tanks after his first instruction flight of the day, and he had put only .2 on the engines since then. No one else had flown the aircraft that day. He did not take the second instructional flight because of an out of limits magneto drop on the left engine. The aircraft had been downed, and 2 plugs replaced on the left engine. That work was just being completed when my student and I arrived for our flight at about XA20. I estimated the cross country would take about 3.4 hours, and with only .2 on the engines after the main fuel cells were topped off, we had more than enough fuel for the proposed cross country and the required reserve. As we were preparing to leave hnl, the FBO fuel truck came by, and as a precaution, I asked the driver to top off both main fuel cells, and he said yes. I took the caps off of both tanks and left them on the wings next to the filler tubes with the access doors open. The driver got out of his truck, and started refueling procedures. We did not see him put fuel in the cells. My student and I walked around the corner of the hangar to talk to another instructor about a different aircraft, and when we returned just a few mins later, the fill caps were in place and the access doors were closed. We did not check the fuel levels again before leaving hnl. However, I was confident that even if the cells had not been topped off, we had sufficient fuel for the cross country. My student and I got in the aircraft, received our IFR clearance to lih, executed a normal run-up (left engine magnetos were good), and started the flight a little before XB00. The 1.1 hour flight to lih was uneventful, and on landing we taxied to the downwind end of the runway, picked up our IFR clearance to lny, and then took off. There was no AVGAS at lih that evening because the generator on the fuel pump was inoperative. The engine mixtures were properly leaned through the cruise portion of the flight to lih. The 1.6 hour flight to lny was also uneventful, and after we landed, we shut down the aircraft and stretched our legs. We had 2.7 hours on the hobbs meter when we shut down on lny, and that would put us back in hnl with about 3.4 hours on the engines, as I expected. Again, the engines were properly leaned for the cruise portion of the flight to lny. While at lny we preflted the aircraft, checked oil levels, and looked in the fuel cells. We did not see any fuel in the cells, but that is common with the old apaches, because the slope of the cells is steep, and the last 1/2 - 1/3 of the fuel is not visible through the filler cap. We drained fuel samples from both engine fuel sumps. We let lny VFR a little before XC00 and called ZHN for flight following. We were climbing to 4500 ft MSL, but leveled off at 3500 ft MSL to avoid a small line of clouds on our course. About 10 mins after lifting off lny, the right engine shut down. I was flying the aircraft when the right engine shut down, and after going through the engine out procedures and checklist, I decided to feather the right engine and land as soon as possible. I called ZHN, declared an emergency, and turned toward the mkk airport, which was about 15 NM northeast of our position. We were about 30 mi from lny and about 35 mi from hnl. We proceeded to the mkk airport on the left engine with the right engine feathered. I know the terrain around the mkk airport very well, so I maintained 3500 ft MSL until we were within 4 mi of the airport, and I then made a straight-in approach and landing on runway 5. The tower at mkk airport was unmanned, because it was after closing, so I activated the runway lights with my transmitter on the tower frequency 125.7. We landed the aircraft without further incident and taxied to the transient parking area. We secured the aircraft and decided to wait until morning to inspect the aircraft closely. I notified the aircraft owner by voice mail message on his business telephone and on his cell phone. The next morning we preflight the aircraft, which included draining fuel samples from the sumps on both engines. I turned on the master switch and the engine fuel boost pumps and checked the fuel pressure on both engines. Pressure on the left engine was within limitations, but the pressure on the right engine was zero. Also, I could hear from the high speed of the right boost pump that it was not pumping gas. A friend of mine flew in from hnl, and I told him I thought the right engine shutdown may have occurred from fuel starvation. We xferred about 1 1/2 gallons of fuel from one of his fuel cells to the right fuel cell on the apache and started the engine. The propeller unfeathered properly, and the engine ran fine, with all engine instruments reading within limitations. Even though it didn't make sense to me, I continued to think that the right engine had shut down because of fuel starvation. I started the left engine, but it ran rough. I checked the magnetos, but they seemed to be functioning properly. I secured the engines and tied down the aircraft. My student and I returned to hnl with my friend, and I again notified the owner of the aircraft by voice mail at the 2 numbers. The following day, I returned to mkk airport and xferred 20 gallons of fuel from the cells on the aircraft I had flown to mkk to the apache, putting 10 gallons in each of the apache's main fuel cells. I started the left engine, and it continued to run rough, so I shut it down. I secured the apache and returned to hnl. The next day, I finally made contact with the owner of the apache, and he told me the right fuel cell drains faster than the left fuel cell. We discussed the fuel usage of the aircraft, and he said the engines may use as much as 9 gph. If the right cell was topped off with 36 gallons of fuel, and if all of the fuel was burned by the engine in 3 hours, the fuel flow would have averaged 12 gph. That rate of fuel flow would have over enriched the engine, fouled the plugs, and probably caused an engine failure. The problem arose from fuel starvation in the right engine. Contributing factors were not knowing that the right fuel cell would lose or use fuel at a rate close to 12 gph. The problem was discovered when the right engine shut down, and my corrective action was feather the right engine and land as soon as possible. There were no injuries or property damage. The rough running left engine was due to a fouled plug, which did not occur until the morning after the incident. Human performance considerations included planning and calculations for fuel usage based on 7-8 gph per engine, which was incorrect. As for action or inaction, we had known the condition of the right fuel cell, we would have planned a shorter cross country or returned to hnl. The only explanation that makes sense is: 1) the right fuel cell did not have 36 gallons in it when we started the cross country, or 2) the right fuel cell had 36 gallons but leaked about 9 gallons, or 3) the capacity of the right fuel cell is not 36 usable gallons of fuel.

Related reports:

Google
 

Original NASA ASRS Text

Title: PA23 INSTRUCTOR AND TRAINEE RAN OUT OF GAS AND HAD #2 ENG FAIL IN ZHN CLASS E AIRSPACE.

Narrative: THIS INCIDENT OCCURRED WHILE FLYING AN OLDER PA23R160 ON A MULTI ENG XCOUNTRY TRAINING FLT FROM HNL TO LIH TO LNY TO HNL ON DEC/XA/00. PRIOR TO LEAVING HNL, I CHKED BOTH MAIN FUEL TANKS, AND I SAW FUEL IN BOTH CELLS SLIGHTLY BELOW TOP-OFF LEVEL. THE ACFT'S MAIN CELLS EACH HOLD 37 GALLONS, OF WHICH 36 GALLONS ARE USABLE, AND THAT GIVES BTWN 4.5 AND 5 HRS OF ENG TIME AT 7-8 GPH PER ENG. BEFORE WE LEFT HNL, I SPOKE TO THE MEI WHO FLEW THE ACFT ON THE PRECEDING FLT, AND HE SAID HE HAD TOPPED OFF THE 2 MAIN TANKS AFTER HIS FIRST INSTRUCTION FLT OF THE DAY, AND HE HAD PUT ONLY .2 ON THE ENGS SINCE THEN. NO ONE ELSE HAD FLOWN THE ACFT THAT DAY. HE DID NOT TAKE THE SECOND INSTRUCTIONAL FLT BECAUSE OF AN OUT OF LIMITS MAGNETO DROP ON THE L ENG. THE ACFT HAD BEEN DOWNED, AND 2 PLUGS REPLACED ON THE L ENG. THAT WORK WAS JUST BEING COMPLETED WHEN MY STUDENT AND I ARRIVED FOR OUR FLT AT ABOUT XA20. I ESTIMATED THE XCOUNTRY WOULD TAKE ABOUT 3.4 HRS, AND WITH ONLY .2 ON THE ENGS AFTER THE MAIN FUEL CELLS WERE TOPPED OFF, WE HAD MORE THAN ENOUGH FUEL FOR THE PROPOSED XCOUNTRY AND THE REQUIRED RESERVE. AS WE WERE PREPARING TO LEAVE HNL, THE FBO FUEL TRUCK CAME BY, AND AS A PRECAUTION, I ASKED THE DRIVER TO TOP OFF BOTH MAIN FUEL CELLS, AND HE SAID YES. I TOOK THE CAPS OFF OF BOTH TANKS AND LEFT THEM ON THE WINGS NEXT TO THE FILLER TUBES WITH THE ACCESS DOORS OPEN. THE DRIVER GOT OUT OF HIS TRUCK, AND STARTED REFUELING PROCS. WE DID NOT SEE HIM PUT FUEL IN THE CELLS. MY STUDENT AND I WALKED AROUND THE CORNER OF THE HANGAR TO TALK TO ANOTHER INSTRUCTOR ABOUT A DIFFERENT ACFT, AND WHEN WE RETURNED JUST A FEW MINS LATER, THE FILL CAPS WERE IN PLACE AND THE ACCESS DOORS WERE CLOSED. WE DID NOT CHK THE FUEL LEVELS AGAIN BEFORE LEAVING HNL. HOWEVER, I WAS CONFIDENT THAT EVEN IF THE CELLS HAD NOT BEEN TOPPED OFF, WE HAD SUFFICIENT FUEL FOR THE XCOUNTRY. MY STUDENT AND I GOT IN THE ACFT, RECEIVED OUR IFR CLRNC TO LIH, EXECUTED A NORMAL RUN-UP (L ENG MAGNETOS WERE GOOD), AND STARTED THE FLT A LITTLE BEFORE XB00. THE 1.1 HR FLT TO LIH WAS UNEVENTFUL, AND ON LNDG WE TAXIED TO THE DOWNWIND END OF THE RWY, PICKED UP OUR IFR CLRNC TO LNY, AND THEN TOOK OFF. THERE WAS NO AVGAS AT LIH THAT EVENING BECAUSE THE GENERATOR ON THE FUEL PUMP WAS INOP. THE ENG MIXTURES WERE PROPERLY LEANED THROUGH THE CRUISE PORTION OF THE FLT TO LIH. THE 1.6 HR FLT TO LNY WAS ALSO UNEVENTFUL, AND AFTER WE LANDED, WE SHUT DOWN THE ACFT AND STRETCHED OUR LEGS. WE HAD 2.7 HRS ON THE HOBBS METER WHEN WE SHUT DOWN ON LNY, AND THAT WOULD PUT US BACK IN HNL WITH ABOUT 3.4 HRS ON THE ENGS, AS I EXPECTED. AGAIN, THE ENGS WERE PROPERLY LEANED FOR THE CRUISE PORTION OF THE FLT TO LNY. WHILE AT LNY WE PREFLTED THE ACFT, CHKED OIL LEVELS, AND LOOKED IN THE FUEL CELLS. WE DID NOT SEE ANY FUEL IN THE CELLS, BUT THAT IS COMMON WITH THE OLD APACHES, BECAUSE THE SLOPE OF THE CELLS IS STEEP, AND THE LAST 1/2 - 1/3 OF THE FUEL IS NOT VISIBLE THROUGH THE FILLER CAP. WE DRAINED FUEL SAMPLES FROM BOTH ENG FUEL SUMPS. WE LET LNY VFR A LITTLE BEFORE XC00 AND CALLED ZHN FOR FLT FOLLOWING. WE WERE CLBING TO 4500 FT MSL, BUT LEVELED OFF AT 3500 FT MSL TO AVOID A SMALL LINE OF CLOUDS ON OUR COURSE. ABOUT 10 MINS AFTER LIFTING OFF LNY, THE R ENG SHUT DOWN. I WAS FLYING THE ACFT WHEN THE R ENG SHUT DOWN, AND AFTER GOING THROUGH THE ENG OUT PROCS AND CHKLIST, I DECIDED TO FEATHER THE R ENG AND LAND ASAP. I CALLED ZHN, DECLARED AN EMER, AND TURNED TOWARD THE MKK ARPT, WHICH WAS ABOUT 15 NM NE OF OUR POS. WE WERE ABOUT 30 MI FROM LNY AND ABOUT 35 MI FROM HNL. WE PROCEEDED TO THE MKK ARPT ON THE L ENG WITH THE R ENG FEATHERED. I KNOW THE TERRAIN AROUND THE MKK ARPT VERY WELL, SO I MAINTAINED 3500 FT MSL UNTIL WE WERE WITHIN 4 MI OF THE ARPT, AND I THEN MADE A STRAIGHT-IN APCH AND LNDG ON RWY 5. THE TWR AT MKK ARPT WAS UNMANNED, BECAUSE IT WAS AFTER CLOSING, SO I ACTIVATED THE RWY LIGHTS WITH MY XMITTER ON THE TWR FREQ 125.7. WE LANDED THE ACFT WITHOUT FURTHER INCIDENT AND TAXIED TO THE TRANSIENT PARKING AREA. WE SECURED THE ACFT AND DECIDED TO WAIT UNTIL MORNING TO INSPECT THE ACFT CLOSELY. I NOTIFIED THE ACFT OWNER BY VOICE MAIL MESSAGE ON HIS BUSINESS TELEPHONE AND ON HIS CELL PHONE. THE NEXT MORNING WE PREFLT THE ACFT, WHICH INCLUDED DRAINING FUEL SAMPLES FROM THE SUMPS ON BOTH ENGS. I TURNED ON THE MASTER SWITCH AND THE ENG FUEL BOOST PUMPS AND CHKED THE FUEL PRESSURE ON BOTH ENGS. PRESSURE ON THE L ENG WAS WITHIN LIMITATIONS, BUT THE PRESSURE ON THE R ENG WAS ZERO. ALSO, I COULD HEAR FROM THE HIGH SPD OF THE R BOOST PUMP THAT IT WAS NOT PUMPING GAS. A FRIEND OF MINE FLEW IN FROM HNL, AND I TOLD HIM I THOUGHT THE R ENG SHUTDOWN MAY HAVE OCCURRED FROM FUEL STARVATION. WE XFERRED ABOUT 1 1/2 GALLONS OF FUEL FROM ONE OF HIS FUEL CELLS TO THE R FUEL CELL ON THE APACHE AND STARTED THE ENG. THE PROP UNFEATHERED PROPERLY, AND THE ENG RAN FINE, WITH ALL ENG INSTS READING WITHIN LIMITATIONS. EVEN THOUGH IT DIDN'T MAKE SENSE TO ME, I CONTINUED TO THINK THAT THE R ENG HAD SHUT DOWN BECAUSE OF FUEL STARVATION. I STARTED THE L ENG, BUT IT RAN ROUGH. I CHKED THE MAGNETOS, BUT THEY SEEMED TO BE FUNCTIONING PROPERLY. I SECURED THE ENGS AND TIED DOWN THE ACFT. MY STUDENT AND I RETURNED TO HNL WITH MY FRIEND, AND I AGAIN NOTIFIED THE OWNER OF THE ACFT BY VOICE MAIL AT THE 2 NUMBERS. THE FOLLOWING DAY, I RETURNED TO MKK ARPT AND XFERRED 20 GALLONS OF FUEL FROM THE CELLS ON THE ACFT I HAD FLOWN TO MKK TO THE APACHE, PUTTING 10 GALLONS IN EACH OF THE APACHE'S MAIN FUEL CELLS. I STARTED THE L ENG, AND IT CONTINUED TO RUN ROUGH, SO I SHUT IT DOWN. I SECURED THE APACHE AND RETURNED TO HNL. THE NEXT DAY, I FINALLY MADE CONTACT WITH THE OWNER OF THE APACHE, AND HE TOLD ME THE R FUEL CELL DRAINS FASTER THAN THE L FUEL CELL. WE DISCUSSED THE FUEL USAGE OF THE ACFT, AND HE SAID THE ENGS MAY USE AS MUCH AS 9 GPH. IF THE R CELL WAS TOPPED OFF WITH 36 GALLONS OF FUEL, AND IF ALL OF THE FUEL WAS BURNED BY THE ENG IN 3 HRS, THE FUEL FLOW WOULD HAVE AVERAGED 12 GPH. THAT RATE OF FUEL FLOW WOULD HAVE OVER ENRICHED THE ENG, FOULED THE PLUGS, AND PROBABLY CAUSED AN ENG FAILURE. THE PROB AROSE FROM FUEL STARVATION IN THE R ENG. CONTRIBUTING FACTORS WERE NOT KNOWING THAT THE R FUEL CELL WOULD LOSE OR USE FUEL AT A RATE CLOSE TO 12 GPH. THE PROB WAS DISCOVERED WHEN THE R ENG SHUT DOWN, AND MY CORRECTIVE ACTION WAS FEATHER THE R ENG AND LAND ASAP. THERE WERE NO INJURIES OR PROPERTY DAMAGE. THE ROUGH RUNNING L ENG WAS DUE TO A FOULED PLUG, WHICH DID NOT OCCUR UNTIL THE MORNING AFTER THE INCIDENT. HUMAN PERFORMANCE CONSIDERATIONS INCLUDED PLANNING AND CALCULATIONS FOR FUEL USAGE BASED ON 7-8 GPH PER ENG, WHICH WAS INCORRECT. AS FOR ACTION OR INACTION, WE HAD KNOWN THE CONDITION OF THE R FUEL CELL, WE WOULD HAVE PLANNED A SHORTER XCOUNTRY OR RETURNED TO HNL. THE ONLY EXPLANATION THAT MAKES SENSE IS: 1) THE R FUEL CELL DID NOT HAVE 36 GALLONS IN IT WHEN WE STARTED THE XCOUNTRY, OR 2) THE R FUEL CELL HAD 36 GALLONS BUT LEAKED ABOUT 9 GALLONS, OR 3) THE CAPACITY OF THE R FUEL CELL IS NOT 36 USABLE GALLONS OF FUEL.

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.