Narrative:

I was the captain of an air carrier medium large transport en route from the island of maui to honolulu. The reason for this report was a deviation of 1200-1700 ft from the assigned altitude of 16000 ft. The cause of the deviation was an emergency situation in which the autoplt went into a strong pitch-up rather than a mild pitch-down when it should have been doing its automatic altitude capture maneuver. The aircraft was climbing at 320 KIAS per honolulu center's instruction. At approximately 14200 ft, a slow decrease in climb rate was initiated so that the automatic altitude capture would occur at a lower climb rate and yield a smooth level off at 16000 ft. At approximately 15600 ft (as the automatic capture was being anticipated), there was a sharp jolt which felt like an exaggerated version of the jolt which occurs when the autoplt is disengaged while the elevator servo is applying elevator input. The jolt was accompanied by an immediate pitching up of the aircraft and the resultant increase in climb rate. I thought that we had experienced an autoplt disconnect instead of capture, and proceeded to push forward on the yoke to level off. As I applied forward pressure, I was surprised at the high amount of force required, and instead of trimming immediately with the thumb switch, I checked the autoplt elevator servo indicator, and noticed that it was full scale up and that the elevator out of trim light was on. At that point I realized that the autoplt had captured, and not disconnected. I was unable to safely disconnect it manually at that point because of the flight attendants and serving carts in the aisle. I thought that I had succeeded in overpwring the autoplt, and had leveled at 16200-16300, but the force increased, and I immediately instructed the first officer to 'call center and tell them we're busting altitudes'. I made a PA announcement telling the passenger to make sure that their seat belts were securely fastened because we would probably be getting a much bigger jolt than the first one very shortly due to an autoplt malfunction. The flight attendant were instructed to secure the cabin and be seated. I then went back to a 2 handed push on the yoke and told the first officer to make another reduction. As we climbed in cavu conditions with very low power, the speed bled off rapidly, and the autoplt was successfully disconnected at approximately 200 KIAS. The elevator servo indicator was still showing full scale nose up, but because of the low speed, the jolt was smaller than the capture jolt, and the pitch-down was slight. Due to the 'smooth' disconnect, there was no aircraft damage, and no injuries. The first officer and I both thought that we had 'topped out' between 17200-17300 ft, but a phone conversation with honolulu center revealed that they reported 17700 ft.

Related reports:

Google
 

Original NASA ASRS Text

Title: ACR MLG EXPERIENCES PITCHUP DUE TO AUTOPLT MALFUNCTION.

Narrative: I WAS THE CAPT OF AN ACR MLG ENRTE FROM THE ISLAND OF MAUI TO HONOLULU. THE REASON FOR THIS RPT WAS A DEV OF 1200-1700 FT FROM THE ASSIGNED ALT OF 16000 FT. THE CAUSE OF THE DEV WAS AN EMER SITUATION IN WHICH THE AUTOPLT WENT INTO A STRONG PITCH-UP RATHER THAN A MILD PITCH-DOWN WHEN IT SHOULD HAVE BEEN DOING ITS AUTOMATIC ALT CAPTURE MANEUVER. THE ACFT WAS CLBING AT 320 KIAS PER HONOLULU CTR'S INSTRUCTION. AT APPROX 14200 FT, A SLOW DECREASE IN CLB RATE WAS INITIATED SO THAT THE AUTOMATIC ALT CAPTURE WOULD OCCUR AT A LOWER CLB RATE AND YIELD A SMOOTH LEVEL OFF AT 16000 FT. AT APPROX 15600 FT (AS THE AUTOMATIC CAPTURE WAS BEING ANTICIPATED), THERE WAS A SHARP JOLT WHICH FELT LIKE AN EXAGGERATED VERSION OF THE JOLT WHICH OCCURS WHEN THE AUTOPLT IS DISENGAGED WHILE THE ELEVATOR SERVO IS APPLYING ELEVATOR INPUT. THE JOLT WAS ACCOMPANIED BY AN IMMEDIATE PITCHING UP OF THE ACFT AND THE RESULTANT INCREASE IN CLB RATE. I THOUGHT THAT WE HAD EXPERIENCED AN AUTOPLT DISCONNECT INSTEAD OF CAPTURE, AND PROCEEDED TO PUSH FORWARD ON THE YOKE TO LEVEL OFF. AS I APPLIED FORWARD PRESSURE, I WAS SURPRISED AT THE HIGH AMOUNT OF FORCE REQUIRED, AND INSTEAD OF TRIMMING IMMEDIATELY WITH THE THUMB SWITCH, I CHKED THE AUTOPLT ELEVATOR SERVO INDICATOR, AND NOTICED THAT IT WAS FULL SCALE UP AND THAT THE ELEVATOR OUT OF TRIM LIGHT WAS ON. AT THAT POINT I REALIZED THAT THE AUTOPLT HAD CAPTURED, AND NOT DISCONNECTED. I WAS UNABLE TO SAFELY DISCONNECT IT MANUALLY AT THAT POINT BECAUSE OF THE FLT ATTENDANTS AND SERVING CARTS IN THE AISLE. I THOUGHT THAT I HAD SUCCEEDED IN OVERPWRING THE AUTOPLT, AND HAD LEVELED AT 16200-16300, BUT THE FORCE INCREASED, AND I IMMEDIATELY INSTRUCTED THE FO TO 'CALL CTR AND TELL THEM WE'RE BUSTING ALTS'. I MADE A PA ANNOUNCEMENT TELLING THE PAX TO MAKE SURE THAT THEIR SEAT BELTS WERE SECURELY FASTENED BECAUSE WE WOULD PROBABLY BE GETTING A MUCH BIGGER JOLT THAN THE FIRST ONE VERY SHORTLY DUE TO AN AUTOPLT MALFUNCTION. THE FA WERE INSTRUCTED TO SECURE THE CABIN AND BE SEATED. I THEN WENT BACK TO A 2 HANDED PUSH ON THE YOKE AND TOLD THE FO TO MAKE ANOTHER REDUCTION. AS WE CLBED IN CAVU CONDITIONS WITH VERY LOW PWR, THE SPD BLED OFF RAPIDLY, AND THE AUTOPLT WAS SUCCESSFULLY DISCONNECTED AT APPROX 200 KIAS. THE ELEVATOR SERVO INDICATOR WAS STILL SHOWING FULL SCALE NOSE UP, BUT BECAUSE OF THE LOW SPD, THE JOLT WAS SMALLER THAN THE CAPTURE JOLT, AND THE PITCH-DOWN WAS SLIGHT. DUE TO THE 'SMOOTH' DISCONNECT, THERE WAS NO ACFT DAMAGE, AND NO INJURIES. THE FO AND I BOTH THOUGHT THAT WE HAD 'TOPPED OUT' BTWN 17200-17300 FT, BUT A PHONE CONVERSATION WITH HONOLULU CTR REVEALED THAT THEY RPTED 17700 FT.

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.