Narrative:

Approximately 2 hours into the cruise phase of this flight (continuously smooth conditions), the #2 autoplt began an uncommanded 1000 FPM climb. My first officer and I both noticed it almost immediately. (Additionally, 'autoplt trim' was annunciated on both FMA's during this event.) my first officer (PF) reactively reached for the vertical speed wheel on the DFGS panel. He quickly and very smoothly input a vertical speed descent of about -500 FPM. The aircraft reversed its climb into a descent. Then, suddenly, the vertical speed changed (without input) to -2500 FPM. At this time I stated 'just disconnect the autoplt' and return to FL310. During this yahoo, the altitude went almost to FL312, then to FL296-FL297 (I saw FL296.5). Almost simultaneously, center (ZFW, I believe) called us on the altitude deviation, and exactly as I saw it. Well, my first officer hand flew for 15-20 seconds and retrimmed the aircraft, then engaged the #2 autoplt. The autoplt held perfectly for approximately 2-3 mins. Then, once again, suddenly began an immediate 1000 FPM vertical climb. At this time, my first officer disconnected the #2 autoplt, stabilized the aircraft, and then engaged the #1 autoplt. The #1 autoplt worked flawlessly for the remainder of the flight. I asked center if this had caused any conflict and he stated no. Upon gate arrival at dtw, a maintenance entry was made and maintenance was consulted. The mechanic reviewed the fault history and found a history of previous faults in the area of DFGS altdevs. Callback conversation with reporter revealed the following information: reporter stated that he had performed some research on the aircraft after the incident occurred. He stated that when he and the first officer had preflted the aircraft, that they went back 3 days into the logbook for any unresolved aircraft problems. This is in accord with company policy. Neither one of them saw anything related to an autoplt problem. They spoke with a mechanic later who stated that he had done a bite check on the aircraft's computer and found that within the past 5 days there had been a history of 5 altdevs. A week after the incident the reporter noted that the aircraft went back to the main maintenance base, perhaps on a ferry flight. The only information he could find related to that maintenance procedure was a reference to an MEL 08-1295CC and a remark regarding the DFGS being mentioned with an 'option one' remark and 'uncommanded roll' for autoplt #1. He supposed that the aircraft was returned to service after that.

Related reports:

Google
 

Original NASA ASRS Text

Title: MD80. SEVERAL ALTDEVS OCCUR WHEN THE #2 AUTOPLT MALFUNCTIONS AND CAUSES CLIMB AND DESCEND OSCILLATIONS.

Narrative: APPROX 2 HRS INTO THE CRUISE PHASE OF THIS FLT (CONTINUOUSLY SMOOTH CONDITIONS), THE #2 AUTOPLT BEGAN AN UNCOMMANDED 1000 FPM CLB. MY FO AND I BOTH NOTICED IT ALMOST IMMEDIATELY. (ADDITIONALLY, 'AUTOPLT TRIM' WAS ANNUNCIATED ON BOTH FMA'S DURING THIS EVENT.) MY FO (PF) REACTIVELY REACHED FOR THE VERT SPD WHEEL ON THE DFGS PANEL. HE QUICKLY AND VERY SMOOTHLY INPUT A VERT SPD DSCNT OF ABOUT -500 FPM. THE ACFT REVERSED ITS CLB INTO A DSCNT. THEN, SUDDENLY, THE VERT SPD CHANGED (WITHOUT INPUT) TO -2500 FPM. AT THIS TIME I STATED 'JUST DISCONNECT THE AUTOPLT' AND RETURN TO FL310. DURING THIS YAHOO, THE ALT WENT ALMOST TO FL312, THEN TO FL296-FL297 (I SAW FL296.5). ALMOST SIMULTANEOUSLY, CTR (ZFW, I BELIEVE) CALLED US ON THE ALTDEV, AND EXACTLY AS I SAW IT. WELL, MY FO HAND FLEW FOR 15-20 SECONDS AND RETRIMMED THE ACFT, THEN ENGAGED THE #2 AUTOPLT. THE AUTOPLT HELD PERFECTLY FOR APPROX 2-3 MINS. THEN, ONCE AGAIN, SUDDENLY BEGAN AN IMMEDIATE 1000 FPM VERT CLB. AT THIS TIME, MY FO DISCONNECTED THE #2 AUTOPLT, STABILIZED THE ACFT, AND THEN ENGAGED THE #1 AUTOPLT. THE #1 AUTOPLT WORKED FLAWLESSLY FOR THE REMAINDER OF THE FLT. I ASKED CTR IF THIS HAD CAUSED ANY CONFLICT AND HE STATED NO. UPON GATE ARR AT DTW, A MAINT ENTRY WAS MADE AND MAINT WAS CONSULTED. THE MECH REVIEWED THE FAULT HISTORY AND FOUND A HISTORY OF PREVIOUS FAULTS IN THE AREA OF DFGS ALTDEVS. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: RPTR STATED THAT HE HAD PERFORMED SOME RESEARCH ON THE ACFT AFTER THE INCIDENT OCCURRED. HE STATED THAT WHEN HE AND THE FO HAD PREFLTED THE ACFT, THAT THEY WENT BACK 3 DAYS INTO THE LOGBOOK FOR ANY UNRESOLVED ACFT PROBS. THIS IS IN ACCORD WITH COMPANY POLICY. NEITHER ONE OF THEM SAW ANYTHING RELATED TO AN AUTOPLT PROB. THEY SPOKE WITH A MECH LATER WHO STATED THAT HE HAD DONE A BITE CHK ON THE ACFT'S COMPUTER AND FOUND THAT WITHIN THE PAST 5 DAYS THERE HAD BEEN A HISTORY OF 5 ALTDEVS. A WK AFTER THE INCIDENT THE RPTR NOTED THAT THE ACFT WENT BACK TO THE MAIN MAINT BASE, PERHAPS ON A FERRY FLT. THE ONLY INFO HE COULD FIND RELATED TO THAT MAINT PROC WAS A REF TO AN MEL 08-1295CC AND A REMARK REGARDING THE DFGS BEING MENTIONED WITH AN 'OPTION ONE' REMARK AND 'UNCOMMANDED ROLL' FOR AUTOPLT #1. HE SUPPOSED THAT THE ACFT WAS RETURNED TO SVC AFTER THAT.

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.