Narrative:

Autopilot a was engaged; and to my best recollection; auto throttles were disengaged. Being vectored for the ILS 23 at buf; we were on a 060 outbound heading paralleling the localizer course on the north side. We were given a clearance to descend to 2;500 ft. Upon reaching we were given a right turn to 210 to intercept; maintain 2;500 until established; and cleared for the ILS 23. Captain selected approach mode on the MCP and turned the heading selector to 210 degrees. We were 6 miles from the runway and slowing. Captain called for and I selected flaps 5. At this point I saw we had not intercepted the localizer and that the glideslope indication was erratic. At about the same time; 3 things happened simultaneously; we intercepted the localizer; the aircraft pitch up to about 10 degrees attitude and we climbed. I said; 'watch your pitch attitude.' the captain did not disengage the autopilot and the aircraft pitched slightly down and then up abruptly to about 18 degrees approaching the yellow pitch limit indicator. I said; 'what are you doing?' he disengaged the autopilot but we had already climbed to 3;100 ft. All this happened rather quickly and we were just inside the outer marker and high. But we were able to become stabilized and on glidepath by 1;000 ft AGL by selecting flaps 40 and we landed normally. This was a confluence of miscues and lack of communication between the captain and me. The ATIS indicated high scattered clouds; 3 miles visibility; rain; and mist so he briefed the full ILS 23 approach. What we both failed to brief or even notice was the special notice to pilot's page immediately after the ILS 23 in the commercial charts. This page says not to engage approach mode until established on the localizer because the glideslope is unusable beyond 5 degrees north of the localizer course. The ATIS mentioned the part about the glideslope being unusable north of course as well. When we were being vectored; the visibility had improved to greater than 10 miles below 5;000 ft so we could see the airport the entire time. As we made the turn inbound I failed to notice that the captain had engaged approach mode instead of localizer mode until intercepting the localizer. We were vectored very close to the outer marker and it seems we were inside of 5 degrees of the localizer at the time the aircraft started pitching up abruptly. But I should have pointed out that he had approach mode selected with the glideslope indicating erratically. The other miscommunication I had with the captain was not suggesting he immediately disengage the autopilot at the first sign of unexpected pitch oscillations. He did so after the second pitch up almost to the pli; but I was ready to do so myself if he hadn't. In the final analysis we should have been more vigilant and taken heed of the special procedures pages in the approach charts and the ATIS. Especially after a long day that was approaching 11 hours.

Google
 

Original NASA ASRS Text

Title: A B737 Captain engaged the autopilot Approach mode while on a southeast heading to capture the BUF Runway 23 ILS localizer. The aircraft pitched up as the glide slope fluctuated and climbed 600' before the autopilot was disconnected.

Narrative: Autopilot A was engaged; and to my best recollection; auto throttles were disengaged. Being vectored for the ILS 23 at BUF; we were on a 060 outbound heading paralleling the localizer course on the north side. We were given a clearance to descend to 2;500 FT. Upon reaching we were given a right turn to 210 to intercept; maintain 2;500 until established; and cleared for the ILS 23. Captain selected approach mode on the MCP and turned the heading selector to 210 degrees. We were 6 miles from the runway and slowing. Captain called for and I selected flaps 5. At this point I saw we had not intercepted the localizer and that the glideslope indication was erratic. At about the same time; 3 things happened simultaneously; we intercepted the localizer; the aircraft pitch up to about 10 degrees attitude and we climbed. I said; 'Watch your pitch attitude.' The Captain did not disengage the autopilot and the aircraft pitched slightly down and then up abruptly to about 18 degrees approaching the yellow pitch limit indicator. I said; 'What are you doing?' He disengaged the autopilot but we had already climbed to 3;100 FT. All this happened rather quickly and we were just inside the outer marker and high. But we were able to become stabilized and on glidepath by 1;000 FT AGL by selecting flaps 40 and we landed normally. This was a confluence of miscues and lack of communication between the Captain and me. The ATIS indicated high scattered clouds; 3 miles visibility; rain; and mist so he briefed the full ILS 23 approach. What we both failed to brief or even notice was the special notice to Pilot's page immediately after the ILS 23 in the commercial charts. This page says not to engage approach mode until established on the localizer because the glideslope is unusable beyond 5 degrees north of the localizer course. The ATIS mentioned the part about the glideslope being unusable north of course as well. When we were being vectored; the visibility had improved to greater than 10 miles below 5;000 FT so we could see the airport the entire time. As we made the turn inbound I failed to notice that the Captain had engaged approach mode instead of localizer mode until intercepting the localizer. We were vectored very close to the outer marker and it seems we were inside of 5 degrees of the localizer at the time the aircraft started pitching up abruptly. But I should have pointed out that he had approach mode selected with the glideslope indicating erratically. The other miscommunication I had with the Captain was not suggesting he immediately disengage the autopilot at the first sign of unexpected pitch oscillations. He did so after the second pitch up almost to the PLI; but I was ready to do so myself if he hadn't. In the final analysis we should have been more vigilant and taken heed of the special procedures pages in the approach charts and the ATIS. Especially after a long day that was approaching 11 hours.

Data retrieved from NASA's ASRS site as of April 2012 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.