Narrative:

On an intercept heading at 6000 ft MSL, after being cleared for a visual approach to runway 26R in atl at night, the first officer selected the approach mode of the autoplt. The use of the autoplt approach mode was intended to reduce pilot workload, but in fact had the opposite effect. The GS captured when the aircraft was approximately 1/2 DOT below the GS indication. The aircraft began a slow climb and while the first officer attempted to deselect the approach mode twice, the aircraft climbed approximately 500 ft. The captain instructed the first officer to disconnect the autoplt and return to 6000 ft. She made inputs to the MCP that would give appropriate flight director commands to the first officer. Approach mode was re-engaged, GS and localizer then captured and the visual approach was continued. The flight crew was not aware of any traffic conflicts that occurred as a result of the altitude deviation. The first officer stated during the debrief that he had never seen this 'pitch-up to the GS' anomaly of the autoplt on the line or in the simulator. A contributing factor in the response time of the first officer to recognize and correct the poor performance of the autoplt may have been fatigue. This was the last leg of a 4-DAY trip in which he had flown with 4 different capts. Supplemental information from acn 534072: level at 5000 ft MSL, 180 degree heading, we were assigned a 230 degree heading and cleared visual approach to runway 26R to follow a heavy B767. I disconnected the autoplt and flew the plane back to 5000 ft. There was no inquiry into the deviation. The conditions were night, VMC, visibility 10 mi.

Related reports:

Google
 

Original NASA ASRS Text

Title: A B737-800 FLT SUFFERS AN ALTDEV WHEN THE APCH COUPLER ATTEMPTS TO CAPTURE THE GS FROM BELOW WHILE OFF THE LOC ON AN INTERCEPT HDG 13 MI E OF ATL, GA.

Narrative: ON AN INTERCEPT HDG AT 6000 FT MSL, AFTER BEING CLRED FOR A VISUAL APCH TO RWY 26R IN ATL AT NIGHT, THE FO SELECTED THE APCH MODE OF THE AUTOPLT. THE USE OF THE AUTOPLT APCH MODE WAS INTENDED TO REDUCE PLT WORKLOAD, BUT IN FACT HAD THE OPPOSITE EFFECT. THE GS CAPTURED WHEN THE ACFT WAS APPROX 1/2 DOT BELOW THE GS INDICATION. THE ACFT BEGAN A SLOW CLB AND WHILE THE FO ATTEMPTED TO DESELECT THE APCH MODE TWICE, THE ACFT CLBED APPROX 500 FT. THE CAPT INSTRUCTED THE FO TO DISCONNECT THE AUTOPLT AND RETURN TO 6000 FT. SHE MADE INPUTS TO THE MCP THAT WOULD GIVE APPROPRIATE FLT DIRECTOR COMMANDS TO THE FO. APCH MODE WAS RE-ENGAGED, GS AND LOC THEN CAPTURED AND THE VISUAL APCH WAS CONTINUED. THE FLC WAS NOT AWARE OF ANY TFC CONFLICTS THAT OCCURRED AS A RESULT OF THE ALTDEV. THE FO STATED DURING THE DEBRIEF THAT HE HAD NEVER SEEN THIS 'PITCH-UP TO THE GS' ANOMALY OF THE AUTOPLT ON THE LINE OR IN THE SIMULATOR. A CONTRIBUTING FACTOR IN THE RESPONSE TIME OF THE FO TO RECOGNIZE AND CORRECT THE POOR PERFORMANCE OF THE AUTOPLT MAY HAVE BEEN FATIGUE. THIS WAS THE LAST LEG OF A 4-DAY TRIP IN WHICH HE HAD FLOWN WITH 4 DIFFERENT CAPTS. SUPPLEMENTAL INFO FROM ACN 534072: LEVEL AT 5000 FT MSL, 180 DEG HDG, WE WERE ASSIGNED A 230 DEG HDG AND CLRED VISUAL APCH TO RWY 26R TO FOLLOW A HVY B767. I DISCONNECTED THE AUTOPLT AND FLEW THE PLANE BACK TO 5000 FT. THERE WAS NO INQUIRY INTO THE DEV. THE CONDITIONS WERE NIGHT, VMC, VISIBILITY 10 MI.

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.