Narrative:

I was planning on a VFR flight from sgh to dlz (about 30 mi, 12 or so mins, away). My twin commander has a set of regular fuel gauges and also a digital fuel meter, both of which had always been in agreement as best I could tell. I have had the aircraft for just over 1 yr and was told by the previous owners that the tanks had been drained and the quantities verified the yr prior to my purchasing the aircraft. They had rptedly found that the main fuel tank gauge read about 20-30 gals low, an assumption that was supported by repetitive matching to the digital counter over many fill-ups. On preflight, the main tank was showing just above 30 gals and the wing system had between 5 and 10 gals showing in each (40-50 gals total). The digital counter showed 65 gals remaining. Based on these numbers I determined that I had enough fuel to go to dlz with 30+ mins reserve to spare. The engines together burn about 80 gph (1.3 gals per min) on takeoff and 50 gph (0.83 gals per min) at cruise. Total fuel burn should have been about 15-20 gals. About 5 mins after takeoff, the main tank gauge was suddenly showing more like 10 gals (there is some fluctuation in motion). I was uncomfortable with this degree of change and decided to go back to sgh and get gas. As I turned back, the right engine quit. Both engines were running off main tank at the time and the gauge showed 5-10 gals remaining when it quit. I turned the boost pumps on and selected the right wing tank. The engine restarted. The left motor started to quit so I also switched that one to wing tank and it regained power as well. I immediately headed for the nearest field which was madison county, about 9 mi away. As I was preparing to land on a 1.5 mi final, the right engine once gain quit as the wing ran dry. I switched it back to main tank hoping for any overflow. The left quit on 3/4 or so mi final. Further attempts at engine starts or tank switches were deemed futile by me and I concentrated on a smooth forced landing in a field short of the madison county airport. This was accomplished. There was no damage to the aircraft or the field and no injuries. When all the investigations were over, I put 20 gals of AVGAS in the main tank. The motors started up immediately and I taxied the airplane across the field to the fuel pumps. The wing tanks took the appropriate amount of fuel. The main tanks took about 35 gals less than expected (taking the 20 gals added in the field into account). In analyzing this situation, the glaring concern is that I was assuming a certain amount of fuel present in the aircraft and that assumption was incorrect. I had previously ascertained where 'actual' empty was on the wing tanks, but had never done so on the main -- it just didn't seem such a good idea. Indications over a yr of flying the aircraft seemed to support the notion that the main tank gauge read a bit low, when in fact it appears that it reads slightly high. The discrepancy between the amount of fuel that the aircraft took versus the amount it should have taken needs to be resolved. Either the digital fuel meter badly underestimates the amount of fuel burned, or the tanks aren't really holding the amount that they should to begin with. Based on the post accident fill-up, the latter may be the case. In the end, I was lulled into not believing the manual gauges and not adding enough margin of error for their degree of inherent inaccuracy. If it turns out that I have a collapsed fuel bladder (as one mechanic suspects) this may have helped lead me down this path. In the end, the aircraft could have some trapped fuel, though I think this is less likely. A short trip on a VFR day may have added some complacency from a human factors point of view. My plan of corrective action is to drain the fuel system and verify the actual quantities. If a discrepancy is confirmed, visual check of the 9 fuel bladders will be indicated to ensure that one of the bladders hasn't collapsed or that there isn't any trapped fuel. I plan to have the fuel indicators re-calibrated as part of this process. No matter what the findings, I willobviously allow for a larger margin of error related to the fuel gauges going forward. Any further actions would be based upon findings going forward. Callback conversation with reporter revealed the following information: the aircraft has been flown since with no problem. Further flts are planned near the airport in order to burn off most of the fuel so it will not have to be drained by maintenance when they begin their investigation.

Google
 

Original NASA ASRS Text

Title: A TWIN AERO COMMANDER 685 HAS EACH ENG QUIT DUE TO APPARENT FUEL STARVATION. RESTART OF BOTH ENGS IS ACCOMPLISHED WHILE HDG FOR THE NEAREST ARPT AFTER SWITCHING FUEL TANKS. SUBSEQUENT DUAL ENG FAILURE IS FOLLOWED BY SUCCESSFUL LNDG JUST SHORT OF THE PLANNED LNDG RWY.

Narrative: I WAS PLANNING ON A VFR FLT FROM SGH TO DLZ (ABOUT 30 MI, 12 OR SO MINS, AWAY). MY TWIN COMMANDER HAS A SET OF REGULAR FUEL GAUGES AND ALSO A DIGITAL FUEL METER, BOTH OF WHICH HAD ALWAYS BEEN IN AGREEMENT AS BEST I COULD TELL. I HAVE HAD THE ACFT FOR JUST OVER 1 YR AND WAS TOLD BY THE PREVIOUS OWNERS THAT THE TANKS HAD BEEN DRAINED AND THE QUANTITIES VERIFIED THE YR PRIOR TO MY PURCHASING THE ACFT. THEY HAD RPTEDLY FOUND THAT THE MAIN FUEL TANK GAUGE READ ABOUT 20-30 GALS LOW, AN ASSUMPTION THAT WAS SUPPORTED BY REPETITIVE MATCHING TO THE DIGITAL COUNTER OVER MANY FILL-UPS. ON PREFLT, THE MAIN TANK WAS SHOWING JUST ABOVE 30 GALS AND THE WING SYS HAD BTWN 5 AND 10 GALS SHOWING IN EACH (40-50 GALS TOTAL). THE DIGITAL COUNTER SHOWED 65 GALS REMAINING. BASED ON THESE NUMBERS I DETERMINED THAT I HAD ENOUGH FUEL TO GO TO DLZ WITH 30+ MINS RESERVE TO SPARE. THE ENGS TOGETHER BURN ABOUT 80 GPH (1.3 GALS PER MIN) ON TKOF AND 50 GPH (0.83 GALS PER MIN) AT CRUISE. TOTAL FUEL BURN SHOULD HAVE BEEN ABOUT 15-20 GALS. ABOUT 5 MINS AFTER TKOF, THE MAIN TANK GAUGE WAS SUDDENLY SHOWING MORE LIKE 10 GALS (THERE IS SOME FLUCTUATION IN MOTION). I WAS UNCOMFORTABLE WITH THIS DEG OF CHANGE AND DECIDED TO GO BACK TO SGH AND GET GAS. AS I TURNED BACK, THE R ENG QUIT. BOTH ENGS WERE RUNNING OFF MAIN TANK AT THE TIME AND THE GAUGE SHOWED 5-10 GALS REMAINING WHEN IT QUIT. I TURNED THE BOOST PUMPS ON AND SELECTED THE R WING TANK. THE ENG RESTARTED. THE L MOTOR STARTED TO QUIT SO I ALSO SWITCHED THAT ONE TO WING TANK AND IT REGAINED PWR AS WELL. I IMMEDIATELY HEADED FOR THE NEAREST FIELD WHICH WAS MADISON COUNTY, ABOUT 9 MI AWAY. AS I WAS PREPARING TO LAND ON A 1.5 MI FINAL, THE R ENG ONCE GAIN QUIT AS THE WING RAN DRY. I SWITCHED IT BACK TO MAIN TANK HOPING FOR ANY OVERFLOW. THE L QUIT ON 3/4 OR SO MI FINAL. FURTHER ATTEMPTS AT ENG STARTS OR TANK SWITCHES WERE DEEMED FUTILE BY ME AND I CONCENTRATED ON A SMOOTH FORCED LNDG IN A FIELD SHORT OF THE MADISON COUNTY ARPT. THIS WAS ACCOMPLISHED. THERE WAS NO DAMAGE TO THE ACFT OR THE FIELD AND NO INJURIES. WHEN ALL THE INVESTIGATIONS WERE OVER, I PUT 20 GALS OF AVGAS IN THE MAIN TANK. THE MOTORS STARTED UP IMMEDIATELY AND I TAXIED THE AIRPLANE ACROSS THE FIELD TO THE FUEL PUMPS. THE WING TANKS TOOK THE APPROPRIATE AMOUNT OF FUEL. THE MAIN TANKS TOOK ABOUT 35 GALS LESS THAN EXPECTED (TAKING THE 20 GALS ADDED IN THE FIELD INTO ACCOUNT). IN ANALYZING THIS SIT, THE GLARING CONCERN IS THAT I WAS ASSUMING A CERTAIN AMOUNT OF FUEL PRESENT IN THE ACFT AND THAT ASSUMPTION WAS INCORRECT. I HAD PREVIOUSLY ASCERTAINED WHERE 'ACTUAL' EMPTY WAS ON THE WING TANKS, BUT HAD NEVER DONE SO ON THE MAIN -- IT JUST DIDN'T SEEM SUCH A GOOD IDEA. INDICATIONS OVER A YR OF FLYING THE ACFT SEEMED TO SUPPORT THE NOTION THAT THE MAIN TANK GAUGE READ A BIT LOW, WHEN IN FACT IT APPEARS THAT IT READS SLIGHTLY HIGH. THE DISCREPANCY BTWN THE AMOUNT OF FUEL THAT THE ACFT TOOK VERSUS THE AMOUNT IT SHOULD HAVE TAKEN NEEDS TO BE RESOLVED. EITHER THE DIGITAL FUEL METER BADLY UNDERESTIMATES THE AMOUNT OF FUEL BURNED, OR THE TANKS AREN'T REALLY HOLDING THE AMOUNT THAT THEY SHOULD TO BEGIN WITH. BASED ON THE POST ACCIDENT FILL-UP, THE LATTER MAY BE THE CASE. IN THE END, I WAS LULLED INTO NOT BELIEVING THE MANUAL GAUGES AND NOT ADDING ENOUGH MARGIN OF ERROR FOR THEIR DEG OF INHERENT INACCURACY. IF IT TURNS OUT THAT I HAVE A COLLAPSED FUEL BLADDER (AS ONE MECH SUSPECTS) THIS MAY HAVE HELPED LEAD ME DOWN THIS PATH. IN THE END, THE ACFT COULD HAVE SOME TRAPPED FUEL, THOUGH I THINK THIS IS LESS LIKELY. A SHORT TRIP ON A VFR DAY MAY HAVE ADDED SOME COMPLACENCY FROM A HUMAN FACTORS POINT OF VIEW. MY PLAN OF CORRECTIVE ACTION IS TO DRAIN THE FUEL SYS AND VERIFY THE ACTUAL QUANTITIES. IF A DISCREPANCY IS CONFIRMED, VISUAL CHK OF THE 9 FUEL BLADDERS WILL BE INDICATED TO ENSURE THAT ONE OF THE BLADDERS HASN'T COLLAPSED OR THAT THERE ISN'T ANY TRAPPED FUEL. I PLAN TO HAVE THE FUEL INDICATORS RE-CALIBRATED AS PART OF THIS PROCESS. NO MATTER WHAT THE FINDINGS, I WILLOBVIOUSLY ALLOW FOR A LARGER MARGIN OF ERROR RELATED TO THE FUEL GAUGES GOING FORWARD. ANY FURTHER ACTIONS WOULD BE BASED UPON FINDINGS GOING FORWARD. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: THE ACFT HAS BEEN FLOWN SINCE WITH NO PROB. FURTHER FLTS ARE PLANNED NEAR THE ARPT IN ORDER TO BURN OFF MOST OF THE FUEL SO IT WILL NOT HAVE TO BE DRAINED BY MAINT WHEN THEY BEGIN THEIR INVESTIGATION.

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.