Narrative:

On departure from den a fuel imbalance was noted between the main tanks. We departed with the main tanks full and 1;700 pounds in the center tank. When I noticed the imbalance; the tanks read: number one 11.8; number two 10.6 and 400 in the center. We considered the fact that it was a gauge problem; or possibly that fuel was being transferred from the center into the wing tank. I turned off the center tank pumps and began cross-feeding to get within limits. During this time we analyzed the fuel system schematic and considered that the center tank scavenge jet pump might have malfunctioned and was allowing fuel to transfer from the center tank to the number one tank. Once the fuel was balanced; I turned on the center tank pumps to burn the last 400 pounds and again and saw the number one tank fuel quantity start to increase. During this time there was roughly a 500 pound imbalance after the center tank was depleted. I centered that imbalance one last time through cross feeding. We went into the QRH for guidance and additional information. The fuel burns were close; the pumps were configured correctly; cross-feed valve was operating correctly; and the fuel audit revealed us arriving with more fuel than planned into ZZZ; thus discrediting the possibility of a fuel leak. We did decide to time for 30 minutes per the QRH to verify the imbalance was less than 500 pounds. After 30 minutes; the imbalance was negligible. We continued on to ZZZ with the fuel gauges matched and no abnormal indications. We sent a mx request regarding the issue; and asked them to meet us at the gate. Prior to TOD it was determined that we would be overweight slightly for landing. We configured early and requested delay vectors from ATC to burn off the extra fuel. After a period of time in which the fuel on board (fob) indication should have come down enough for landing; we noted the number one main tank again increasing in quantity. At this point; once again questioning the accuracy of the fuel gauge; we decided to expeditiously land and deal with the fuel issue at the gate. We requested and were cleared for the visual approach.after an uneventful landing; the number one fuel quantity indicator began to fluctuate between 1;200 and 4;000 pounds and the number two gauge was at 4;000 pounds. Parked at the gate; the gauges read: number one 1;000; and number two 3;900. We coordinated with ZZZ ops and had the fuelers standby until maintenance had a chance to troubleshoot the problem. Maintenance showed up and verified that both the flight deck gauge and the wing gauge matched. Sticking the wing tanks confirmed that this was in fact the current fuel quantity. Maintenance then MEL'd the fuel gauge in the logbook and we observed the fueling process sticking the tanks; as per the MEL; with an info only write-up for the (M) procedure in the logbook.I truly believe that expectation bias was a major factor leading to this issue. For years I have cross-fed fuel using the same procedure; with the same outcome. This was no different. As I cross-fed; the gauges reacted as expected. The problem was the rate that I was transferring was greater than that indicated due to the malfunctioning gauge. After I thought we had 'fixed' the problem; the gauges seemed to be responding correctly and were matched; my thoughts drifted from the original issue. I also feel that my 'outside the box' analysis of a possible malfunctioning center tank jet scavenge pump should have been avoided. The fact that the number one tank increased by the amount of fuel in the center tank each time the center pumps were turned on was pure coincidence. Adhering to strict QRH checklist discipline would have prevented this. Hindsight is 20/20 and it's much easier to evaluate my actions now. The chain should have been broken early with the detection of a fuel quantity greater than the tank capacity. This thought did cross my mind; but the responseto the action that I used (cross feeding to balance) gave me the result that I expected; expectation bias.the main action that I will institute after this event is a thorough analysis before I crossfeed fuel. The givens are fuel flow in pounds per hour versus how many pounds to match the tanks. With this information you can obviously predetermine a time. There is no doubt in my mind that I will time each crossfeed operation from now on. I will also involve dispatch and maintenance control early on in a complex event such as this. My plan was to contact them after the 30 minute timing (per the QRH) if there was a fuel imbalance. There was not. Contacting them early on would possibly have changed my mindset toward the issue and given justification for allowing the presumed imbalance. The last item is checklist discipline. Even though I had a presumed action that would rectify the problem based on experience; I should have referred to the QRH initially (even for something as routine as balancing fuel). This could have been the break in the chain that would have determined a different outcome.

Related reports:

Google
 

Original NASA ASRS Text

Title: B737 flight crew reported difficulty accurately trouble shooting a malfunctioning fuel gauge.

Narrative: On departure from DEN a fuel imbalance was noted between the main tanks. We departed with the main tanks full and 1;700 pounds in the center tank. When I noticed the imbalance; the tanks read: number one 11.8; number two 10.6 and 400 in the center. We considered the fact that it was a gauge problem; or possibly that fuel was being transferred from the center into the wing tank. I turned off the center tank pumps and began cross-feeding to get within limits. During this time we analyzed the fuel system schematic and considered that the center tank scavenge jet pump might have malfunctioned and was allowing fuel to transfer from the center tank to the number one tank. Once the fuel was balanced; I turned on the center tank pumps to burn the last 400 pounds and again and saw the number one tank fuel quantity start to increase. During this time there was roughly a 500 pound imbalance after the center tank was depleted. I centered that imbalance one last time through cross feeding. We went into the QRH for guidance and additional information. The fuel burns were close; the pumps were configured correctly; cross-feed valve was operating correctly; and the fuel audit revealed us arriving with more fuel than planned into ZZZ; thus discrediting the possibility of a fuel leak. We did decide to time for 30 minutes per the QRH to verify the imbalance was less than 500 pounds. After 30 minutes; the imbalance was negligible. We continued on to ZZZ with the fuel gauges matched and no abnormal indications. We sent a MX request regarding the issue; and asked them to meet us at the gate. Prior to TOD it was determined that we would be overweight slightly for landing. We configured early and requested delay vectors from ATC to burn off the extra fuel. After a period of time in which the Fuel On Board (FOB) indication should have come down enough for landing; we noted the number one main tank again increasing in quantity. At this point; once again questioning the accuracy of the fuel gauge; we decided to expeditiously land and deal with the fuel issue at the gate. We requested and were cleared for the visual approach.After an uneventful landing; the number one fuel quantity indicator began to fluctuate between 1;200 and 4;000 pounds and the number two gauge was at 4;000 pounds. Parked at the gate; the gauges read: number one 1;000; and number two 3;900. We coordinated with ZZZ Ops and had the Fuelers standby until Maintenance had a chance to troubleshoot the problem. Maintenance showed up and verified that both the flight deck gauge and the wing gauge matched. Sticking the wing tanks confirmed that this was in fact the current fuel quantity. Maintenance then MEL'd the fuel gauge in the logbook and we observed the fueling process sticking the tanks; as per the MEL; with an Info Only write-up for the (M) procedure in the logbook.I truly believe that expectation bias was a major factor leading to this issue. For years I have cross-fed fuel using the same procedure; with the same outcome. This was no different. As I cross-fed; the gauges reacted as expected. The problem was the rate that I was transferring was greater than that indicated due to the malfunctioning gauge. After I thought we had 'fixed' the problem; the gauges seemed to be responding correctly and were matched; my thoughts drifted from the original issue. I also feel that my 'outside the box' analysis of a possible malfunctioning center tank jet scavenge pump should have been avoided. The fact that the number one tank increased by the amount of fuel in the center tank each time the center pumps were turned on was pure coincidence. Adhering to strict QRH checklist discipline would have prevented this. Hindsight is 20/20 and it's much easier to evaluate my actions now. The chain should have been broken early with the detection of a fuel quantity greater than the tank capacity. This thought did cross my mind; but the responseto the action that I used (cross feeding to balance) gave me the result that I expected; expectation bias.The main action that I will institute after this event is a thorough analysis before I crossfeed fuel. The givens are fuel flow in pounds per hour versus how many pounds to match the tanks. With this information you can obviously predetermine a time. There is no doubt in my mind that I will time each crossfeed operation from now on. I will also involve Dispatch and Maintenance Control early on in a complex event such as this. My plan was to contact them after the 30 minute timing (per the QRH) if there was a fuel imbalance. There was not. Contacting them early on would possibly have changed my mindset toward the issue and given justification for allowing the presumed imbalance. The last item is checklist discipline. Even though I had a presumed action that would rectify the problem based on experience; I should have referred to the QRH initially (even for something as routine as balancing fuel). This could have been the break in the chain that would have determined a different outcome.

Data retrieved from NASA's ASRS site 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.