Narrative:

On approach; we were flying a visual. We were talking to approach and were on a vector 260 (downwind to final). We were then given a left hand turn to 210 and I dialed in the course. Autopilot was engaged; heading select was selected and the aircraft did not turn. The first officer commented that this appeared to be the same problem we had going into our departing city (see previous report). I checked the rudder trim and it was centered. I then went into cws and turned to the 210 heading. As we rolled out; I reengaged heading select and the plane started to drift to the right passing through a 220 heading. Approach gave us a 260 heading again and the aircraft rolled to the new heading with no problem. In the turn; however; I noticed that the fuel gauges were counting up and down rapidly (like when you're being fueled on the ground). The left gauge read 7099 and the right gauge read 9099. I asked the first officer when did we get a 2000# imbalance? We were then given a heading of 340 and asked if we had the field. On the turn to final; the plane over banked past 45 degrees and we got the bank angle warning. I disconnected the autopilot and noticed the gauges were counting up and down again and they had now settled at 6099 and 9999. Now hand flying; there was no doubt in my mind that this imbalance was real. We checked the fuel switches. The crossfeed valve was closed with no light and the center pump switches were off. We had transferred fuel from the center tank on the climb out of our departing city and the first officer had pointed out the low pressure light. I closed the crossfeed at that time and turned off the pumps. At that time we had 9000# a side and were about 30-35 minutes out. When we briefed our first approach into the field (visibility xz left) we had about 8500# a side. We continued a low approach to runway xx told tower that we would need to go back to approach for some delay vectors to look at a problem and we thought it would take 20-30 minutes. We opened up the crossfeed valve; and turned off the boost pumps to the low side (right side). We left the gear down and the flaps at 5. Approach climbed us to 9000' MSL and we kept the speed at 180 knots. Without checklist guidance; I thought this would be the best configuration to balance the fuel quickly and without too much noise to the passengers. I made an announcement to the passengers and we tried to find any guidance in the QRH and fom to help us balance fuel. We checked all fuel cbs and noticed that they were in. After about 25 minutes; the fuel was within 1200# and we requested a vector back to the field and landed on xyl with a 700# imbalance. First; we both thought that this was an autopilot malfunction to start; since we had seen a similar situation on the previous leg. So my attention was originally focused on the autopilot. There is only a mention in the limitations about the 1000# imbalance landing restriction. It might be important to ask boeing how much of a max imbalance can occur before the autopilot can handle the load and what is the max imbalance the plane can handle while being hand flown. I can assure you that it is not a lot of fun. I would like to see a fuel imbalance checklist for the -300; since we don't have one and a fuel annunciation light in the event of an imbalance like on the -700. Although in this case I'm not sure that it would have helped since I didn't know I could trust my fuel gauges. I would also like to see this on a pt if I'm in a -300 simulator and discuss with the pilot group what the best configuration is to fly so no one gets on the back side of the power curve.

Related reports:

Google
 

Original NASA ASRS Text

Title: A B737-300 developed a fuel imbalance with the crossfeed closed during descent. The crew delays landing until the fuel was balanced.

Narrative: On approach; we were flying a visual. We were talking to Approach and were on a vector 260 (downwind to final). We were then given a left hand turn to 210 and I dialed in the course. Autopilot was engaged; heading select was selected and the aircraft did not turn. The First Officer commented that this appeared to be the same problem we had going into our departing city (see previous report). I checked the rudder trim and it was centered. I then went into CWS and turned to the 210 heading. As we rolled out; I reengaged heading select and the plane started to drift to the right passing through a 220 heading. Approach gave us a 260 heading again and the aircraft rolled to the new heading with no problem. In the turn; however; I noticed that the fuel gauges were counting up and down rapidly (like when you're being fueled on the ground). The left gauge read 7099 and the right gauge read 9099. I asked the First Officer when did we get a 2000# imbalance? We were then given a heading of 340 and asked if we had the field. On the turn to final; the plane over banked past 45 degrees and we got the bank angle warning. I disconnected the autopilot and noticed the gauges were counting up and down again and they had now settled at 6099 and 9999. Now hand flying; there was no doubt in my mind that this imbalance was real. We checked the fuel switches. The crossfeed valve was closed with no light and the center pump switches were off. We had transferred fuel from the center tank on the climb out of our departing city and the First Officer had pointed out the Low Pressure light. I closed the crossfeed at that time and turned off the pumps. At that time we had 9000# a side and were about 30-35 minutes out. When we briefed our first approach into the field (VIS XZ L) we had about 8500# a side. We continued a low approach to Runway XX told Tower that we would need to go back to approach for some delay vectors to look at a problem and we thought it would take 20-30 minutes. We opened up the crossfeed valve; and turned off the boost pumps to the low side (right side). We left the gear down and the flaps at 5. Approach climbed us to 9000' MSL and we kept the speed at 180 knots. Without checklist guidance; I thought this would be the best configuration to balance the fuel quickly and without too much noise to the Passengers. I made an announcement to the Passengers and we tried to find any guidance in the QRH and FOM to help us balance fuel. We checked all fuel CBs and noticed that they were in. After about 25 minutes; the fuel was within 1200# and we requested a vector back to the field and landed on XYL with a 700# imbalance. First; we both thought that this was an autopilot malfunction to start; since we had seen a similar situation on the previous leg. So my attention was originally focused on the autopilot. There is only a mention in the Limitations about the 1000# imbalance landing restriction. It might be important to ask Boeing how much of a max imbalance can occur before the autopilot can handle the load and what is the max imbalance the plane can handle while being hand flown. I can assure you that it is not a lot of fun. I would like to see a fuel imbalance checklist for the -300; since we don't have one and a fuel annunciation light in the event of an imbalance like on the -700. Although in this case I'm not sure that it would have helped since I didn't know I could trust my fuel gauges. I would also like to see this on a PT if I'm in a -300 simulator and discuss with the Pilot group what the best configuration is to fly so no one gets on the back side of the power curve.

Data retrieved from NASA's ASRS site as of April 2012 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.