Narrative:

Returning on VFR flight; aircraft appeared to have very slow fuel leak from right tank. Mechanic had request[ed] I burn off excess fuel in the right wing bladder in order to repair/replace bladder. I was cross-feeding left engine from right tank and right engine was also drawing fuel from right tank to empty tank. I believed; based upon shadin and jpi fuel flow gauges; that there was approximately 20 gallons remaining in the right bladders. I had intended to switch both engines to feed off their respective tanks on final approach. I was on a practice GPS approach over the gulf; when both engines stopped at 2;000 MSL and several miles offshore. The entire trip I had reminded myself I was cross-feeding; but when I attempted to stop cross-feeding; only the left engine restarted. I did not think about the right engine [which] was drawing off an empty side of the plane; because I believed there was still 20 gallons in the right tank. To add to this problem the left engine was not putting out full power due to an undetermined reason (the mechanic was going to go through the left engine at the same time the right wing bladder leak was being fixed the following week).I called approach; whom I was only monitoring but not communicating with; and informed them of the problem. [I] let them know I was having engine problems and [may need a] potential rescue. For [someone to] know to look for me in the event of a bad outcome. Approach was helpful in partial vectoring.the mistakes I made:1. [I] believed [the] fuel flow gauges (usually highly accurate) were correct with a slow leak in a bladder or fuel line.2. [I] hesitated in engine out procedures due to [loosing] both engines. I knew] that left tank was nearly full and mistakenly believed the right tank had 20 gallons.3. I did not feather the right engine due to the 1;000 foot altitude when I recovered operation of the fuel starved [left] engine (I forgot I was cross-feeding and did not correctly identify the right engine as windmilling). I was concerned about loss of the left engine due to pre-existing mechanical concerns. [This] diverted my attention; and lost altitude rather than analyze the whole situation. I was concerned I was at 1;000 feet and had 10 miles to go to land. I was distracted with the poor engine function on the left engine and diverted my attention from diagnosing the right engine and possible ditching sites in the bay. The left engine had been in the shop several times over the last year over poor power output; so I was distracted with the functioning engine operating rather than diagnosing the right engine which I could have restarted had I check[ed] fuel management. I limped into the airport slowing losing altitude. The plane was lightly loaded which helped. I do not think these events would have happened had the failure occurred at 5;000 ft instead of close to sea level and over rough and cold ocean water at low altitude.the dual failure was with plane at least 40% full of fuel; beginning approach and diverting attention from a fuel failed engine; and over open ocean. Entry into the practice IFR GPS approach was partially diverting my attention also. This is about concurrent problems in a near emergency; which does happen in emergencies. If all these events [happened] separately [there] would not have been an issue. I have learned to always monitor ATC even if not communicating.

Related reports:

Google
 

Original NASA ASRS Text

Title: Light twin pilot reports attempting to run the right wing tank dry by feeding both engines from that tank; in order to facilitate planned maintenance; but is not prepared for the consequences when his efforts prove successful.

Narrative: Returning on VFR flight; aircraft appeared to have very slow fuel leak from right tank. Mechanic had request[ed] I burn off excess fuel in the right wing bladder in order to repair/replace bladder. I was cross-feeding left engine from right tank and right engine was also drawing fuel from right tank to empty tank. I believed; based upon Shadin and JPI fuel flow gauges; that there was approximately 20 gallons remaining in the right bladders. I had intended to switch both engines to feed off their respective tanks on final approach. I was on a practice GPS approach over the Gulf; when both engines stopped at 2;000 MSL and several miles offshore. The entire trip I had reminded myself I was cross-feeding; but when I attempted to stop cross-feeding; only the left engine restarted. I did not think about the right engine [which] was drawing off an empty side of the plane; because I believed there was still 20 gallons in the right tank. To add to this problem the left engine was not putting out full power due to an undetermined reason (the mechanic was going to go through the left engine at the same time the right wing bladder leak was being fixed the following week).I called Approach; whom I was only monitoring but not communicating with; and informed them of the problem. [I] let them know I was having engine problems and [may need a] potential rescue. For [someone to] know to look for me in the event of a bad outcome. Approach was helpful in partial vectoring.The mistakes I made:1. [I] believed [the] fuel flow gauges (usually highly accurate) were correct with a slow leak in a bladder or fuel line.2. [I] hesitated in engine out procedures due to [loosing] both engines. I knew] that left tank was nearly full and mistakenly believed the right tank had 20 gallons.3. I did not feather the right engine due to the 1;000 foot altitude when I recovered operation of the fuel starved [left] engine (I forgot I was cross-feeding and did not correctly identify the right engine as windmilling). I was concerned about loss of the left engine due to pre-existing mechanical concerns. [This] diverted my attention; and lost altitude rather than analyze the whole situation. I was concerned I was at 1;000 feet and had 10 miles to go to land. I was distracted with the poor engine function on the left engine and diverted my attention from diagnosing the right engine and possible ditching sites in the bay. The left engine had been in the shop several times over the last year over poor power output; so I was distracted with the functioning engine operating rather than diagnosing the right engine which I could have restarted had I check[ed] fuel management. I limped into the airport slowing losing altitude. The plane was lightly loaded which helped. I do not think these events would have happened had the failure occurred at 5;000 FT instead of close to sea level and over rough and cold ocean water at low altitude.The dual failure was with plane at least 40% full of fuel; beginning approach and diverting attention from a fuel failed engine; and over open ocean. Entry into the practice IFR GPS approach was partially diverting my attention also. This is about concurrent problems in a near emergency; which does happen in emergencies. If all these events [happened] separately [there] would not have been an issue. I have learned to always monitor ATC even if not communicating.

Data retrieved from NASA's ASRS site as of July 2013 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.