Narrative:

Shortly after level off, just north of orl, I noticed a small indicated lateral fuel unbalance of less than 500 pounds. The left main tank was lower than the right main and total fuel used was slightly greater than forecast, but consistent with the vectoring we had experienced departing pbi. The center tank indicator was inoperative and on dmi. The autoplt, which had functioned normally on the previous flight before the ron in pbi, was unusable due to severe porpoising. The fuel used was within 100 pounds for the respective engines and no lateral unbalance was noted while hand flying the aircraft. At no time during the flight did the difference in fuel used exceed 100 pounds nor was any lateral unbalance experienced (aileron trim .5 units left toward the lesser indicated fuel). Due to these indications, I was reasonably convinced that we had an indicator error. I continued to monitor the fuel quantity and the indicated unbalance appeared to be increasing. The quick reference checklist was reviewed for any applicable procedure and none was found. Although a fuel leak should be indicated by decreasing quantity accompanied by a fuel inlet pressure low light, it occurred to me that there was a possibility of a leak of insufficient volume to create a low inlet pressure. I requested the first officer to make a visual inspection of the left wing and the #1 engine nacelle. No abnormalities were noted as a result of this inspection. This information further encouraged me to believe that the problem was in the left main tank indicator. However, as a precaution, I idented a minimum of 3 suitable alternate airfields that could be reached via an idle descent from any point in the remaining flight planned route, should there be an actual fuel loss. North of jax, approaching dbn, the indicated lateral fuel unbalance was approaching the maximum of 1500 pounds. There still was no lateral trim problem nor any difference in fuel used. I discussed our options with the first officer, including the possibility of shutting down #1 engine, in the event that the fuel loss was real. Should an emergency landing be made, it was agreed that we would shut down #1 engine immediately upon touchdown, use only idle reverse on #2 engine to prevent blowing any fuel present forward over the aircraft, and land as close as possible to the approach side of the touchdown zone due to virtually no reverse and manual spoilers (automatic spoilers on dmi). I instructed the first officer to declare an emergency at the first indication of a fuel inlet low pressure light and to request an immediate approach to atl (visible ahead some 60 NM). If atl was not immediately available, he was to request mcn (just behind our left wing). Other suitable alternates were idented as columbus, fulton county, athens, and augusta. Shortly thereafter the indicated fuel quantity in the left main tank reached zero and the fuel inlet pressure low light flickered. The fuel crossfeed was immediately opened and an emergency declared according to plan. ATC offered a straight in approach (against the current flow) to runway 27L. An idle descent to the OM was accomplished and we were required to configure the aircraft (including landing gear) in order to place ourselves in the proper position to accomplish a safe landing. The before landing/final checklist was accomplished. The approach, landing, and minimum fuel go around had been briefed during the descent/approach checklist and a routine landing was accomplished on runway 27L at atl with equipment requested and standing by. I personally briefed the passenger and tried to reassure them of the precautionary nature of our actions. The center tank had been stuck at zero quantity prior to departing pbi. Maintenance postflt revealed that 5600 pounds had migrated to the center tank. We were unable to discern this migration due to the inoperative center tank indicator. It is interesting to note that neither the autoplt nor fuel problems were in evidence prior to the 'B' check accomplished during our ron by contract maintenance in pbi. Callback conversation with reporter revealed the following information: the reporter is being investigated by FAA. The FAA contends the reporter should have diverted on the first indication of fuel imbalance on the gauges. Therptr contends that aircraft flew as if the gauge indication was in error and felt there would have been poor judgement to divert on an unsupported fuel gauge indication. The reporter was told that some poppet valves failed allowing the fuel to get into the fueling plumbing and somehow the fuel was able to move via the fueling manifold to the center tank. Being the center tank gauge was placarded inoperative and reading 0 he would have no center tank gauge indication. He had personally observed the 'drip stick' for the center tank prior to takeoff to read '0.' the total fuel quantity gauge read there was fuel on the aircraft to support the reporter that there in fact was fuel in the left tank. However, the total fuel quantity gauge was indicating the fuel in the center tank. The fuel did not migrate through the crossfeed manifold, but rather, through the fueling manifold. The analyst has had the dc-9 migrate fuel into the center tank which engineers say should not happen, but, no problem existed because the center tank gauge was showing what was happening and turning on the 4/center tank boost pumps evacuate/evacuationed the center tank of fuel and solved the problem.

Google
 

Original NASA ASRS Text

Title: FUEL MIGRATES FROM L MAIN TANK TO CTR TANK PROMPTING FLC TO DECLARE AN EMER AND LAND.

Narrative: SHORTLY AFTER LEVEL OFF, JUST N OF ORL, I NOTICED A SMALL INDICATED LATERAL FUEL UNBALANCE OF LESS THAN 500 LBS. THE L MAIN TANK WAS LOWER THAN THE R MAIN AND TOTAL FUEL USED WAS SLIGHTLY GREATER THAN FORECAST, BUT CONSISTENT WITH THE VECTORING WE HAD EXPERIENCED DEPARTING PBI. THE CTR TANK INDICATOR WAS INOP AND ON DMI. THE AUTOPLT, WHICH HAD FUNCTIONED NORMALLY ON THE PREVIOUS FLT BEFORE THE RON IN PBI, WAS UNUSABLE DUE TO SEVERE PORPOISING. THE FUEL USED WAS WITHIN 100 LBS FOR THE RESPECTIVE ENGS AND NO LATERAL UNBALANCE WAS NOTED WHILE HAND FLYING THE ACFT. AT NO TIME DURING THE FLT DID THE DIFFERENCE IN FUEL USED EXCEED 100 LBS NOR WAS ANY LATERAL UNBALANCE EXPERIENCED (AILERON TRIM .5 UNITS LEFT TOWARD THE LESSER INDICATED FUEL). DUE TO THESE INDICATIONS, I WAS REASONABLY CONVINCED THAT WE HAD AN INDICATOR ERROR. I CONTINUED TO MONITOR THE FUEL QUANTITY AND THE INDICATED UNBALANCE APPEARED TO BE INCREASING. THE QUICK REF CHKLIST WAS REVIEWED FOR ANY APPLICABLE PROC AND NONE WAS FOUND. ALTHOUGH A FUEL LEAK SHOULD BE INDICATED BY DECREASING QUANTITY ACCOMPANIED BY A FUEL INLET PRESSURE LOW LIGHT, IT OCCURRED TO ME THAT THERE WAS A POSSIBILITY OF A LEAK OF INSUFFICIENT VOLUME TO CREATE A LOW INLET PRESSURE. I REQUESTED THE FO TO MAKE A VISUAL INSPECTION OF THE L WING AND THE #1 ENG NACELLE. NO ABNORMALITIES WERE NOTED AS A RESULT OF THIS INSPECTION. THIS INFO FURTHER ENCOURAGED ME TO BELIEVE THAT THE PROB WAS IN THE L MAIN TANK INDICATOR. HOWEVER, AS A PRECAUTION, I IDENTED A MINIMUM OF 3 SUITABLE ALTERNATE AIRFIELDS THAT COULD BE REACHED VIA AN IDLE DSCNT FROM ANY POINT IN THE REMAINING FLT PLANNED RTE, SHOULD THERE BE AN ACTUAL FUEL LOSS. N OF JAX, APCHING DBN, THE INDICATED LATERAL FUEL UNBALANCE WAS APCHING THE MAX OF 1500 LBS. THERE STILL WAS NO LATERAL TRIM PROB NOR ANY DIFFERENCE IN FUEL USED. I DISCUSSED OUR OPTIONS WITH THE FO, INCLUDING THE POSSIBILITY OF SHUTTING DOWN #1 ENG, IN THE EVENT THAT THE FUEL LOSS WAS REAL. SHOULD AN EMER LNDG BE MADE, IT WAS AGREED THAT WE WOULD SHUT DOWN #1 ENG IMMEDIATELY UPON TOUCHDOWN, USE ONLY IDLE REVERSE ON #2 ENG TO PREVENT BLOWING ANY FUEL PRESENT FORWARD OVER THE ACFT, AND LAND AS CLOSE AS POSSIBLE TO THE APCH SIDE OF THE TOUCHDOWN ZONE DUE TO VIRTUALLY NO REVERSE AND MANUAL SPOILERS (AUTO SPOILERS ON DMI). I INSTRUCTED THE FO TO DECLARE AN EMER AT THE FIRST INDICATION OF A FUEL INLET LOW PRESSURE LIGHT AND TO REQUEST AN IMMEDIATE APCH TO ATL (VISIBLE AHEAD SOME 60 NM). IF ATL WAS NOT IMMEDIATELY AVAILABLE, HE WAS TO REQUEST MCN (JUST BEHIND OUR L WING). OTHER SUITABLE ALTERNATES WERE IDENTED AS COLUMBUS, FULTON COUNTY, ATHENS, AND AUGUSTA. SHORTLY THEREAFTER THE INDICATED FUEL QUANTITY IN THE L MAIN TANK REACHED ZERO AND THE FUEL INLET PRESSURE LOW LIGHT FLICKERED. THE FUEL CROSSFEED WAS IMMEDIATELY OPENED AND AN EMER DECLARED ACCORDING TO PLAN. ATC OFFERED A STRAIGHT IN APCH (AGAINST THE CURRENT FLOW) TO RWY 27L. AN IDLE DSCNT TO THE OM WAS ACCOMPLISHED AND WE WERE REQUIRED TO CONFIGURE THE ACFT (INCLUDING LNDG GEAR) IN ORDER TO PLACE OURSELVES IN THE PROPER POS TO ACCOMPLISH A SAFE LNDG. THE BEFORE LNDG/FINAL CHKLIST WAS ACCOMPLISHED. THE APCH, LNDG, AND MINIMUM FUEL GAR HAD BEEN BRIEFED DURING THE DSCNT/APCH CHKLIST AND A ROUTINE LNDG WAS ACCOMPLISHED ON RWY 27L AT ATL WITH EQUIP REQUESTED AND STANDING BY. I PERSONALLY BRIEFED THE PAX AND TRIED TO REASSURE THEM OF THE PRECAUTIONARY NATURE OF OUR ACTIONS. THE CTR TANK HAD BEEN STUCK AT ZERO QUANTITY PRIOR TO DEPARTING PBI. MAINT POSTFLT REVEALED THAT 5600 LBS HAD MIGRATED TO THE CTR TANK. WE WERE UNABLE TO DISCERN THIS MIGRATION DUE TO THE INOP CTR TANK INDICATOR. IT IS INTERESTING TO NOTE THAT NEITHER THE AUTOPLT NOR FUEL PROBS WERE IN EVIDENCE PRIOR TO THE 'B' CHK ACCOMPLISHED DURING OUR RON BY CONTRACT MAINT IN PBI. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: THE RPTR IS BEING INVESTIGATED BY FAA. THE FAA CONTENDS THE RPTR SHOULD HAVE DIVERTED ON THE FIRST INDICATION OF FUEL IMBALANCE ON THE GAUGES. THERPTR CONTENDS THAT ACFT FLEW AS IF THE GAUGE INDICATION WAS IN ERROR AND FELT THERE WOULD HAVE BEEN POOR JUDGEMENT TO DIVERT ON AN UNSUPPORTED FUEL GAUGE INDICATION. THE RPTR WAS TOLD THAT SOME POPPET VALVES FAILED ALLOWING THE FUEL TO GET INTO THE FUELING PLUMBING AND SOMEHOW THE FUEL WAS ABLE TO MOVE VIA THE FUELING MANIFOLD TO THE CTR TANK. BEING THE CTR TANK GAUGE WAS PLACARDED INOP AND READING 0 HE WOULD HAVE NO CTR TANK GAUGE INDICATION. HE HAD PERSONALLY OBSERVED THE 'DRIP STICK' FOR THE CTR TANK PRIOR TO TKOF TO READ '0.' THE TOTAL FUEL QUANTITY GAUGE READ THERE WAS FUEL ON THE ACFT TO SUPPORT THE RPTR THAT THERE IN FACT WAS FUEL IN THE L TANK. HOWEVER, THE TOTAL FUEL QUANTITY GAUGE WAS INDICATING THE FUEL IN THE CTR TANK. THE FUEL DID NOT MIGRATE THROUGH THE CROSSFEED MANIFOLD, BUT RATHER, THROUGH THE FUELING MANIFOLD. THE ANALYST HAS HAD THE DC-9 MIGRATE FUEL INTO THE CTR TANK WHICH ENGINEERS SAY SHOULD NOT HAPPEN, BUT, NO PROB EXISTED BECAUSE THE CTR TANK GAUGE WAS SHOWING WHAT WAS HAPPENING AND TURNING ON THE 4/CTR TANK BOOST PUMPS EVACED THE CTR TANK OF FUEL AND SOLVED THE PROB.

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.