Narrative:

Approaching zzzzz intersection at about XA18 (1:29 after takeoff) the autopilot disconnected; along with its aural warning and failure of a variety of cockpit displays and lights. I asked the first officer to manually fly the aircraft and maintain FL350. The left generator control off; left bus off; and left bus tie isln annunciators were illuminated; so we agreed that the appropriate checklist to run was (left and/or right) AC bus off. I asked the cabin crew to inspect the cabin; galleys; and lavatories for evidence of smoke or fire; and ran the checklist as we passed zzzzz. We were unable to restore power to the left AC bus from either the left generator or the APU; which left us with some concern that we had experienced an electrical short with the possibility of fire. We were almost 400 miles from our etp so we diverted towards ZZZ1. As we were turning off track I announced; 'mayday; mayday; mayday;' along with our position and intentions on 121.5; and was immediately answered by a company flight; who then coordinated between us and commercial radio on HF. Company flight's assistance was invaluable in lowering cockpit workload. We turned right; established a 15 mile offset from the track; descended to FL345 and found that maintaining that track was equivalent to a direct course to ZZZ1. The cabin crew reported no visible smoke or unusual smell; and inoperative systems consistent with the utility buses being unpowered. One flight attendant reported smelling 'exhaust' in the aft cabin well before the incident; but the odor was not present near the time of the incident and was not experienced by any of the other crewmembers. We received an ATC clearance to proceed direct to ZZZ1 at FL340. After establishing direct communication with dispatch via phone patch; we discussed designating ZZZ as a more suitable diversion airport. We agreed that benefits of better weather; emergency equipment; runways; and passenger and maintenance infrastructure outweighed the shorter distance to ZZZ1. We requested an ATC clearance direct to ZZZ and at XB17Z received clearance direct to ZZZ at FL340. At XB19Z we gained radar contact. We had a B737 check airman from another airline jumpseating with a seat in the cabin; and we asked him to join us in the cockpit for the descent and landing to possibly lower our workload and help identify errors. We followed fom procedures and suggested PA for a precautionary landing; and landed at ZZZ at XC30Z. Our ACARS was inoperative; and we received no reply to any of our in-range calls to ZZZ operations on any of the three frequencies shown in the route manual. ZZZ tower advised us of our gate. Electrical system failure; with task saturation due in part to QRH procedure and lack of information. From a regulatory/procedural standpoint; one area of concern was rvsm legality. I didn't have time to look up details to verify; but believed that the loss of all autopilots made us unable to legally remain in rvsm airspace. Having reviewed the fom after the fact; I should have asked commercial radio to relay to ATC that we were 'unable rvsm due equipment; but request to remain at FL340 under emergency conditions.' I believe it was made quite clear to commercial radio; however; that we did not have an autopilot and had declared an emergency. In that respect; I exercised my captain's emergency authority to remain in rvsm airspace under the circumstances. I discussed this with dispatch; who concurred that since it was an emergency situation and we had received an ATC clearance to remain at FL340; rvsm was not an issue requiring a descent. To lower workload if this situation happens again; I suggest a note be added to the QRH procedure such as: 'advise ATC unable rvsm due equipment.' if required; obtain clearance to lower altitude; or to remain at altitude under emergency conditions.' procedurally; we had an issue with the QRH checklist; which directs resetting the right bus tie switch. The checklist is not clear as to whether only the switchassociated with the failed side should be selected off; or if both switches should be cycled in all three situations covered by this checklist (left; right; or left and right AC bus off). We felt that since we could not reestablish power to the left bus; the electrical system was compromised by more than a simple generator failure. (I note here that after our arrival in ZZZ; a team of technicians had still not found the cause of our problem after troubleshooting for three days.) we were not confident that if we selected the right bus tie switch off we could get the right bus back. This would leave us with both AC buses off. At the very least; we thought that if we performed this step we would temporarily lose the only navigation information we had; and then have to reinitialize the FMC route; etc. We felt that it was more prudent to keep the navigation information we had running; and decided not to perform this step of the checklist. I suggest that the QRH be revised to clarify this issue. Given the trouble we had with the checklist; and the difficulty in properly choosing alternatives presented; I suggest separating this into two different checklists: 'left or right AC bus off;' and 'left and right AC bus off.' the way it is now; it is a long; confusing checklist with multiple opportunities for mistakes or misunderstanding. Another issue with the checklist was that it was incomplete in its description of inoperative systems. The only inoperative systems shown in either the left bus off or bus isolated checklists are: all autopilots; left and C flight directors; and flap indicator. In our situation; the left AC transfer bus was also apparently unpowered. It would have helped us if there had been a note to this effect in the QRH. There is a list of components on this bus (operative components with heading operating); but I did not see this until after the fact. With only one IRU and one DME; FMC position updating was not adequate when we reached the ZZZ terminal area. The first officer was using heading select and flight director to intercept the ILS; and the flight director and navigation display were still showing that we had not reached localizer alive when raw data showed that we were centered on the localizer. Additionally; fuel pumps powered by the left AC bus were not mentioned in the QRH. While this may seem obvious; I'll point out that under the stress of being almost two hours from land at night with a possible electrical fire; not much is obvious that isn't staring you in the face. We did notice that half of our fuel pumps were not running; which didn't concern us. What was not initially obvious was that the left engine was no longer burning fuel from the center tank; and fuel imbalance was occurring at a rate of about 4000 pounds per hour. Using normal crossfeed procedures to correct the imbalance would have left us with only one electrical pump supplying both engines; which we were reluctant to do under the circumstances. We opened the crossfeed valve and left all operating pumps running; allowing both engines to feed from the center tank. I suggest including in the QRH a note such as; 'note: fuel will not feed from the center tank to the engine on the failed AC bus side unless the crossfeed valve is open.' we also found that we could not control cockpit temperature; which became quite cold in flight. On touchdown; the cockpit immediately became very hot; and the cockpit windows started to rapidly fog over until we turned the left pack off. This could be mentioned as a consideration at the end of the checklist.callback conversation with reporter revealed the following information: the reporter spoke with two very knowledgeable maintenance people and got two opinions about the cause of this anomaly. The most probable is the left CDU failure with a circuit breaker that did not open with a massive internal CDU failure. Testing the CDU in the right position popped the right CDU circuit breaker. When the left CDU circuit breaker was pulled and the CDU put back in the left position the bus was powerednormally until the circuit breaker was pushed in at which time the left bus and left transfer bus went down. The odd thing about this is that the CDU is only protected by a 2.5 amp circuit breaker but the circuit breaker failed. This brings up a question of the reliability of the circuit breakers and if some cycle or inspect method should be developed for dc monitoring. The second thought about the bus and gen failure is an aft galley coffee maker that had a high resistance.

Related reports:

Google
 

Original NASA ASRS Text

Title: A B757'S L GEN FAILED RESULTING IN THE LOSS OF L AC; L AC TRANSFER BUSES AND OPENING THE BUS TIE ISOLATION BREAKER. AN EMER WAS DECLARED.

Narrative: APPROACHING ZZZZZ INTERSECTION AT ABOUT XA18 (1:29 AFTER TAKEOFF) THE AUTOPILOT DISCONNECTED; ALONG WITH ITS AURAL WARNING AND FAILURE OF A VARIETY OF COCKPIT DISPLAYS AND LIGHTS. I ASKED THE FO TO MANUALLY FLY THE AIRCRAFT AND MAINTAIN FL350. THE LEFT GENERATOR CONTROL OFF; LEFT BUS OFF; AND LEFT BUS TIE ISLN ANNUNCIATORS WERE ILLUMINATED; SO WE AGREED THAT THE APPROPRIATE CHECKLIST TO RUN WAS (L AND/OR R) AC BUS OFF. I ASKED THE CABIN CREW TO INSPECT THE CABIN; GALLEYS; AND LAVATORIES FOR EVIDENCE OF SMOKE OR FIRE; AND RAN THE CHECKLIST AS WE PASSED ZZZZZ. WE WERE UNABLE TO RESTORE POWER TO THE LEFT AC BUS FROM EITHER THE LEFT GENERATOR OR THE APU; WHICH LEFT US WITH SOME CONCERN THAT WE HAD EXPERIENCED AN ELECTRICAL SHORT WITH THE POSSIBILITY OF FIRE. WE WERE ALMOST 400 MILES FROM OUR ETP SO WE DIVERTED TOWARDS ZZZ1. AS WE WERE TURNING OFF TRACK I ANNOUNCED; 'MAYDAY; MAYDAY; MAYDAY;' ALONG WITH OUR POSITION AND INTENTIONS ON 121.5; AND WAS IMMEDIATELY ANSWERED BY A COMPANY FLIGHT; WHO THEN COORDINATED BETWEEN US AND COMMERCIAL RADIO ON HF. COMPANY FLIGHT'S ASSISTANCE WAS INVALUABLE IN LOWERING COCKPIT WORKLOAD. WE TURNED RIGHT; ESTABLISHED A 15 MILE OFFSET FROM THE TRACK; DESCENDED TO FL345 AND FOUND THAT MAINTAINING THAT TRACK WAS EQUIVALENT TO A DIRECT COURSE TO ZZZ1. THE CABIN CREW REPORTED NO VISIBLE SMOKE OR UNUSUAL SMELL; AND INOPERATIVE SYSTEMS CONSISTENT WITH THE UTILITY BUSES BEING UNPOWERED. ONE FLIGHT ATTENDANT REPORTED SMELLING 'EXHAUST' IN THE AFT CABIN WELL BEFORE THE INCIDENT; BUT THE ODOR WAS NOT PRESENT NEAR THE TIME OF THE INCIDENT AND WAS NOT EXPERIENCED BY ANY OF THE OTHER CREWMEMBERS. WE RECEIVED AN ATC CLEARANCE TO PROCEED DIRECT TO ZZZ1 AT FL340. AFTER ESTABLISHING DIRECT COMMUNICATION WITH DISPATCH VIA PHONE PATCH; WE DISCUSSED DESIGNATING ZZZ AS A MORE SUITABLE DIVERSION AIRPORT. WE AGREED THAT BENEFITS OF BETTER WEATHER; EMERGENCY EQUIPMENT; RUNWAYS; AND PASSENGER AND MAINTENANCE INFRASTRUCTURE OUTWEIGHED THE SHORTER DISTANCE TO ZZZ1. WE REQUESTED AN ATC CLEARANCE DIRECT TO ZZZ AND AT XB17Z RECEIVED CLEARANCE DIRECT TO ZZZ AT FL340. AT XB19Z WE GAINED RADAR CONTACT. WE HAD A B737 CHECK AIRMAN FROM ANOTHER AIRLINE JUMPSEATING WITH A SEAT IN THE CABIN; AND WE ASKED HIM TO JOIN US IN THE COCKPIT FOR THE DESCENT AND LANDING TO POSSIBLY LOWER OUR WORKLOAD AND HELP IDENTIFY ERRORS. WE FOLLOWED FOM PROCEDURES AND SUGGESTED PA FOR A PRECAUTIONARY LANDING; AND LANDED AT ZZZ AT XC30Z. OUR ACARS WAS INOPERATIVE; AND WE RECEIVED NO REPLY TO ANY OF OUR IN-RANGE CALLS TO ZZZ OPERATIONS ON ANY OF THE THREE FREQUENCIES SHOWN IN THE ROUTE MANUAL. ZZZ TOWER ADVISED US OF OUR GATE. ELECTRICAL SYSTEM FAILURE; WITH TASK SATURATION DUE IN PART TO QRH PROCEDURE AND LACK OF INFORMATION. FROM A REGULATORY/PROCEDURAL STANDPOINT; ONE AREA OF CONCERN WAS RVSM LEGALITY. I DIDN'T HAVE TIME TO LOOK UP DETAILS TO VERIFY; BUT BELIEVED THAT THE LOSS OF ALL AUTOPILOTS MADE US UNABLE TO LEGALLY REMAIN IN RVSM AIRSPACE. HAVING REVIEWED THE FOM AFTER THE FACT; I SHOULD HAVE ASKED COMMERCIAL RADIO TO RELAY TO ATC THAT WE WERE 'UNABLE RVSM DUE EQUIPMENT; BUT REQUEST TO REMAIN AT FL340 UNDER EMERGENCY CONDITIONS.' I BELIEVE IT WAS MADE QUITE CLEAR TO COMMERCIAL RADIO; HOWEVER; THAT WE DID NOT HAVE AN AUTOPILOT AND HAD DECLARED AN EMERGENCY. IN THAT RESPECT; I EXERCISED MY CAPTAIN'S EMERGENCY AUTHORITY TO REMAIN IN RVSM AIRSPACE UNDER THE CIRCUMSTANCES. I DISCUSSED THIS WITH DISPATCH; WHO CONCURRED THAT SINCE IT WAS AN EMERGENCY SITUATION AND WE HAD RECEIVED AN ATC CLEARANCE TO REMAIN AT FL340; RVSM WAS NOT AN ISSUE REQUIRING A DESCENT. TO LOWER WORKLOAD IF THIS SITUATION HAPPENS AGAIN; I SUGGEST A NOTE BE ADDED TO THE QRH PROCEDURE SUCH AS: 'ADVISE ATC UNABLE RVSM DUE EQUIPMENT.' IF REQUIRED; OBTAIN CLEARANCE TO LOWER ALTITUDE; OR TO REMAIN AT ALTITUDE UNDER EMERGENCY CONDITIONS.' PROCEDURALLY; WE HAD AN ISSUE WITH THE QRH CHECKLIST; WHICH DIRECTS RESETTING THE RIGHT BUS TIE SWITCH. THE CHECKLIST IS NOT CLEAR AS TO WHETHER ONLY THE SWITCHASSOCIATED WITH THE FAILED SIDE SHOULD BE SELECTED OFF; OR IF BOTH SWITCHES SHOULD BE CYCLED IN ALL THREE SITUATIONS COVERED BY THIS CHECKLIST (L; R; OR L AND R AC BUS OFF). WE FELT THAT SINCE WE COULD NOT REESTABLISH POWER TO THE LEFT BUS; THE ELECTRICAL SYSTEM WAS COMPROMISED BY MORE THAN A SIMPLE GENERATOR FAILURE. (I NOTE HERE THAT AFTER OUR ARRIVAL IN ZZZ; A TEAM OF TECHNICIANS HAD STILL NOT FOUND THE CAUSE OF OUR PROBLEM AFTER TROUBLESHOOTING FOR THREE DAYS.) WE WERE NOT CONFIDENT THAT IF WE SELECTED THE RIGHT BUS TIE SWITCH OFF WE COULD GET THE RIGHT BUS BACK. THIS WOULD LEAVE US WITH BOTH AC BUSES OFF. AT THE VERY LEAST; WE THOUGHT THAT IF WE PERFORMED THIS STEP WE WOULD TEMPORARILY LOSE THE ONLY NAVIGATION INFORMATION WE HAD; AND THEN HAVE TO REINITIALIZE THE FMC ROUTE; ETC. WE FELT THAT IT WAS MORE PRUDENT TO KEEP THE NAVIGATION INFORMATION WE HAD RUNNING; AND DECIDED NOT TO PERFORM THIS STEP OF THE CHECKLIST. I SUGGEST THAT THE QRH BE REVISED TO CLARIFY THIS ISSUE. GIVEN THE TROUBLE WE HAD WITH THE CHECKLIST; AND THE DIFFICULTY IN PROPERLY CHOOSING ALTERNATIVES PRESENTED; I SUGGEST SEPARATING THIS INTO TWO DIFFERENT CHECKLISTS: 'L OR R AC BUS OFF;' AND 'L AND R AC BUS OFF.' THE WAY IT IS NOW; IT IS A LONG; CONFUSING CHECKLIST WITH MULTIPLE OPPORTUNITIES FOR MISTAKES OR MISUNDERSTANDING. ANOTHER ISSUE WITH THE CHECKLIST WAS THAT IT WAS INCOMPLETE IN ITS DESCRIPTION OF INOPERATIVE SYSTEMS. THE ONLY INOPERATIVE SYSTEMS SHOWN IN EITHER THE L BUS OFF OR BUS ISOLATED CHECKLISTS ARE: ALL AUTOPILOTS; L AND C FLIGHT DIRECTORS; AND FLAP INDICATOR. IN OUR SITUATION; THE LEFT AC TRANSFER BUS WAS ALSO APPARENTLY UNPOWERED. IT WOULD HAVE HELPED US IF THERE HAD BEEN A NOTE TO THIS EFFECT IN THE QRH. THERE IS A LIST OF COMPONENTS ON THIS BUS (OPERATIVE COMPONENTS WITH HDG OPERATING); BUT I DID NOT SEE THIS UNTIL AFTER THE FACT. WITH ONLY ONE IRU AND ONE DME; FMC POSITION UPDATING WAS NOT ADEQUATE WHEN WE REACHED THE ZZZ TERMINAL AREA. THE FO WAS USING HEADING SELECT AND FLIGHT DIRECTOR TO INTERCEPT THE ILS; AND THE FLIGHT DIRECTOR AND NAV DISPLAY WERE STILL SHOWING THAT WE HAD NOT REACHED LOCALIZER ALIVE WHEN RAW DATA SHOWED THAT WE WERE CENTERED ON THE LOCALIZER. ADDITIONALLY; FUEL PUMPS POWERED BY THE LEFT AC BUS WERE NOT MENTIONED IN THE QRH. WHILE THIS MAY SEEM OBVIOUS; I'LL POINT OUT THAT UNDER THE STRESS OF BEING ALMOST TWO HOURS FROM LAND AT NIGHT WITH A POSSIBLE ELECTRICAL FIRE; NOT MUCH IS OBVIOUS THAT ISN'T STARING YOU IN THE FACE. WE DID NOTICE THAT HALF OF OUR FUEL PUMPS WERE NOT RUNNING; WHICH DIDN'T CONCERN US. WHAT WAS NOT INITIALLY OBVIOUS WAS THAT THE LEFT ENGINE WAS NO LONGER BURNING FUEL FROM THE CENTER TANK; AND FUEL IMBALANCE WAS OCCURRING AT A RATE OF ABOUT 4000 LBS PER HOUR. USING NORMAL CROSSFEED PROCEDURES TO CORRECT THE IMBALANCE WOULD HAVE LEFT US WITH ONLY ONE ELECTRICAL PUMP SUPPLYING BOTH ENGINES; WHICH WE WERE RELUCTANT TO DO UNDER THE CIRCUMSTANCES. WE OPENED THE CROSSFEED VALVE AND LEFT ALL OPERATING PUMPS RUNNING; ALLOWING BOTH ENGINES TO FEED FROM THE CENTER TANK. I SUGGEST INCLUDING IN THE QRH A NOTE SUCH AS; 'NOTE: FUEL WILL NOT FEED FROM THE CENTER TANK TO THE ENGINE ON THE FAILED AC BUS SIDE UNLESS THE CROSSFEED VALVE IS OPEN.' WE ALSO FOUND THAT WE COULD NOT CONTROL COCKPIT TEMPERATURE; WHICH BECAME QUITE COLD IN FLIGHT. ON TOUCHDOWN; THE COCKPIT IMMEDIATELY BECAME VERY HOT; AND THE COCKPIT WINDOWS STARTED TO RAPIDLY FOG OVER UNTIL WE TURNED THE LEFT PACK OFF. THIS COULD BE MENTIONED AS A CONSIDERATION AT THE END OF THE CHECKLIST.CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: THE REPORTER SPOKE WITH TWO VERY KNOWLEDGEABLE MAINTENANCE PEOPLE AND GOT TWO OPINIONS ABOUT THE CAUSE OF THIS ANOMALY. THE MOST PROBABLE IS THE L CDU FAILURE WITH A CIRCUIT BREAKER THAT DID NOT OPEN WITH A MASSIVE INTERNAL CDU FAILURE. TESTING THE CDU IN THE R POSITION POPPED THE R CDU CIRCUIT BREAKER. WHEN THE L CDU CIRCUIT BREAKER WAS PULLED AND THE CDU PUT BACK IN THE L POSITION THE BUS WAS POWEREDNORMALLY UNTIL THE CIRCUIT BREAKER WAS PUSHED IN AT WHICH TIME THE L BUS AND L TRANSFER BUS WENT DOWN. THE ODD THING ABOUT THIS IS THAT THE CDU IS ONLY PROTECTED BY A 2.5 AMP CIRCUIT BREAKER BUT THE CIRCUIT BREAKER FAILED. THIS BRINGS UP A QUESTION OF THE RELIABILITY OF THE CIRCUIT BREAKERS AND IF SOME CYCLE OR INSPECT METHOD SHOULD BE DEVELOPED FOR DC MONITORING. THE SECOND THOUGHT ABOUT THE BUS AND GEN FAILURE IS AN AFT GALLEY COFFEE MAKER THAT HAD A HIGH RESISTANCE.

Data retrieved from NASA's ASRS site as of May 2009 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.