Narrative:

Aircraft automation malfunction induced by 'on approach logic' associated with false glide slope (GS) capture.flight was uneventful; weather not a factor. Autopilot; autothrottles were on. Approach gave us a vector to the west of the initial approach fix (IAF) to provide separation between flights; minimum practical speed assigned by ATC; 170 knots. We were given a vectors; then cleared direct [the] IAF intersection and cleared the ILS DME runway 11 no.1 from (IAF). ATC requested final approach speed for traffic separation; 150 knots (approximate; as I can't remember exactly). Aircraft was in approach mode (extended centerline from IAF and flaps set at 15). Pilot flying (PF); after approach clearance; selected app mode on the MCP. Approximately 3 NM west south west of IAF; the aircraft autopilot 'captured' what appears now to be a false GS signal (not apparent at the time; as a green GS illuminated on the pfd window); and the aircraft began a gentle descent. A crew member commented that the GS capture seemed a bit early as 5000 feet is the initial approach altitude at IAF. Before the crew could intervene; and at approximately 4750 feet MSL; the aircraft; on its own; began a climb; with additional thrust (thrust levers moved forward) back to 5000 feet MSL. A red bar momentarily popped up over the GS window of the pfd (both displays; PF; and pilot monitoring); then; after maybe 5 seconds it went away. The crew commented about the anomaly and continued the approach. Inside IAF with the field in sight; localizer captured and annunciated; the aircraft again captured the GS. All appeared normal except for the fact the captain's raw GS date on his display seemed erratic; moving up and down about 1/8th of a dot rapidly. Automation appeared normal.event. Sometime after GS capture; and before final gear and flaps (approximately 3500 feet MSL); the pilot monitoring noticed the airspeed; with auto throttles still connected; was showing about 10 knots below flaps 15 bug (again; I think 150 knots selected in the speed window); that being 140 knots!!!(approximately 5 knots above the yellow 'foot') at that time; automation was disconnected and power was manually applied. Approach and landing was uneventful.I'm not a flight guidance guru; but in 10 years flying the 777; I've never witnessed a 10 knots slow auto throttle commanded actual speed. 3 knots maybe; not 10. In my opinion; it appears the aircraft's approach logic algorithm malfunctioned due to the first red bar GS annunciation and didn't fully rectify that input; and the auto throttle command logic became corrupted; or stopped working all together. Stall monitoring; it appears; must have still been operating correctly because the aircraft speed never entered the 'foot' and stayed 5 knots above that speed. Also; in retrospect; I don't recall seeing a green trend arrow as the aircraft returned to the assigned window speed; or during the approach and landing (I was hyper aware at this point and kept thinking something was wrong with the display; I just couldn't put my finger on it till now!).the aircraft was written up with two separate entries for to the false GS and auto throttles. Complacency and finger memory expectations. You open the speed window and always get that speed; well until now. You expect to get a trend arrow as a crutch to monitor airspeed. You may not. After 10 hours of flying all night; this type of malfunction is very insidious. Stay in the loop.

Related reports:

Google
 

Original NASA ASRS Text

Title: A B777-200 flight crew reported the aircraft autothrottles were malfunctioning; allowing a speed 10 knots below the selected speed.

Narrative: Aircraft automation malfunction induced by 'on approach logic' associated with false glide slope (GS) capture.Flight was uneventful; weather not a factor. Autopilot; autothrottles were ON. Approach gave us a vector to the west of the initial approach fix (IAF) to provide separation between flights; minimum practical speed assigned by ATC; 170 knots. We were given a vectors; then cleared direct [the] IAF intersection and cleared the ILS DME Runway 11 No.1 from (IAF). ATC requested final approach speed for traffic separation; 150 knots (approximate; as I can't remember exactly). Aircraft was in approach mode (extended centerline from IAF and flaps set at 15). Pilot Flying (PF); after approach clearance; selected APP mode on the MCP. Approximately 3 NM west south west of IAF; the aircraft autopilot 'captured' what appears now to be a false GS signal (NOT apparent at the time; as a green GS illuminated on the PFD window); and the aircraft began a gentle descent. A crew member commented that the GS capture seemed a bit early as 5000 feet is the initial approach altitude at IAF. Before the crew could intervene; and at approximately 4750 feet MSL; the aircraft; on its own; began a climb; with additional thrust (thrust levers moved forward) back to 5000 feet MSL. A red bar momentarily popped up over the GS window of the PFD (both displays; PF; and Pilot Monitoring); then; after maybe 5 seconds it went away. The crew commented about the anomaly and continued the approach. Inside IAF with the field in sight; LOC captured and annunciated; the aircraft again captured the GS. All appeared normal except for the fact the Captain's raw GS date on his display seemed erratic; moving up and down about 1/8th of a dot rapidly. Automation appeared normal.EVENT. Sometime after GS capture; and before final gear and flaps (approximately 3500 feet MSL); the pilot monitoring noticed the airspeed; with auto throttles still connected; was showing about 10 knots below flaps 15 bug (again; I think 150 knots selected in the speed window); that being 140 knots!!!(approximately 5 knots above the yellow 'foot') At that time; automation was disconnected and power was manually applied. Approach and landing was uneventful.I'm not a flight guidance guru; but in 10 years flying the 777; I've never witnessed a 10 knots slow auto throttle commanded actual speed. 3 knots maybe; not 10. In my opinion; it appears the aircraft's approach logic algorithm malfunctioned due to the first red bar GS annunciation and didn't fully rectify that input; and the auto throttle command logic became corrupted; or stopped working all together. Stall monitoring; it appears; must have still been operating correctly because the aircraft speed never entered the 'foot' and stayed 5 knots above that speed. Also; in retrospect; I don't recall seeing a green trend arrow as the aircraft returned to the assigned window speed; or during the approach and landing (I was hyper aware at this point and kept thinking something was wrong with the display; I just couldn't put my finger on it till now!).The aircraft was written up with two separate entries for to the false GS and auto throttles. Complacency and finger memory expectations. You open the speed window and always get that speed; well until now. You expect to get a trend arrow as a crutch to monitor airspeed. You may not. After 10 hours of flying all night; this type of malfunction is VERY INSIDIOUS. Stay in the loop.

Data retrieved from NASA's ASRS site 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.