Narrative:

On jan/xx/95 I was PF, flight backup, flight X from jfk to bru. We were being vectored for the ILS runway 25L approach. WX conditions at the time were VMC. Bru approach gave us a heading to intercept the localizer then cleared us for the approach. The center autoplt was on and approach mode armed. Localizer and GS captured while in the turn. Airplane began a descent on autoplt. However, raw data showed us right of course and below GS. First officer called out altitude at about the same time bru approach queried our altitude. Captain directed me to climb. At that point the flight director was unreliable, commanding a turn away from the localizer and a descent away from GS, I disconnected the autoplt and autothrottle then corrected onto the ILS runway 25L using raw data. Remainder of flight was uneventful. Supplemental information from acn 295045: at approximately XA20Z, flight backup was flying aircraft with autoplt on, in ground contact, at 2000 ft MSL, in right downwind to bru runway 25L. 5 NM short of bun, approach instructed us to turn right to 220 degrees, and cleared us for the ILS runway 25L. Flight backup set heading select 220 and armed approach. We were right of localizer and below GS. Localizer capture occurred in the turn and autoplt continued turn toward the 253 degree runway heading. GS captured during the turn, and autoplt began a descent below 2000 ft. Raw data indicated we were right of localizer. The PNF first officer called 'altitude' and I directed the flight backup to climb. At this point, the runway was in sight. Flight backup disconnected the autoplt, autothrottles, and began a climb toward the GS and a slight left turn toward the localizer. Approach control questioned our altitude and then handed us over to tower. Both flight directors were commanding a right turn away from the localizer, and a descent away from the GS. I cycled both flight directors off then on, and armed the approach mode. The remainder of the flight was uneventful with normal flight director indications. Callback conversation with reporter revealed the following information: the captain states he was in the copilot seat of the B-767-300, the relief pilot doing the flying in the captain's seat. The localizer had not left the full displacement when the aircraft started to turn. The heading bug automatically aligned with the ILS and the autoplt captured a false localizer and commenced descent on a false GS. (On apr/xx/95 a caution was sent to the company pilots regarding numerous false captures at brussels and toronto.) the captain switched off the autoplt, autothrottle, and flight director and rearmed the flight director which then indicated normally. The airplane, however, was hand flown to the landing with the flight director indicating normally. During the false capture the raw data was indicating correctly as observed visually with the airport in sight while the flight director and autoplt were chasing a false capture to the right of and below course. The 3 IRU's alignments were accurate throughout the approach and during the false capture. Industry procedure is to arm the approach at the cue 'cleared for the approach' as was done in this case. This captain arms the approach when he is flying when he sees the raw data localizer move across the case finessing with heading select to a moderate intercept angle. He suggests this procedure may have precluded this false capture and might be considered a better procedure industry wide. He finds this especially useful in south america where facility surveys are less accurate.

Google
 

Original NASA ASRS Text

Title: B-767-300 CAPTURES FALSE LOC AND GS.

Narrative: ON JAN/XX/95 I WAS PF, FLT BACKUP, FLT X FROM JFK TO BRU. WE WERE BEING VECTORED FOR THE ILS RWY 25L APCH. WX CONDITIONS AT THE TIME WERE VMC. BRU APCH GAVE US A HDG TO INTERCEPT THE LOC THEN CLRED US FOR THE APCH. THE CTR AUTOPLT WAS ON AND APCH MODE ARMED. LOC AND GS CAPTURED WHILE IN THE TURN. AIRPLANE BEGAN A DSCNT ON AUTOPLT. HOWEVER, RAW DATA SHOWED US RIGHT OF COURSE AND BELOW GS. FO CALLED OUT ALT AT ABOUT THE SAME TIME BRU APCH QUERIED OUR ALT. CAPT DIRECTED ME TO CLB. AT THAT POINT THE FLT DIRECTOR WAS UNRELIABLE, COMMANDING A TURN AWAY FROM THE LOC AND A DSCNT AWAY FROM GS, I DISCONNECTED THE AUTOPLT AND AUTOTHROTTLE THEN CORRECTED ONTO THE ILS RWY 25L USING RAW DATA. REMAINDER OF FLT WAS UNEVENTFUL. SUPPLEMENTAL INFO FROM ACN 295045: AT APPROX XA20Z, FLT BACKUP WAS FLYING ACFT WITH AUTOPLT ON, IN GND CONTACT, AT 2000 FT MSL, IN R DOWNWIND TO BRU RWY 25L. 5 NM SHORT OF BUN, APCH INSTRUCTED US TO TURN R TO 220 DEGS, AND CLRED US FOR THE ILS RWY 25L. FLT BACKUP SET HDG SELECT 220 AND ARMED APCH. WE WERE R OF LOC AND BELOW GS. LOC CAPTURE OCCURRED IN THE TURN AND AUTOPLT CONTINUED TURN TOWARD THE 253 DEG RWY HDG. GS CAPTURED DURING THE TURN, AND AUTOPLT BEGAN A DSCNT BELOW 2000 FT. RAW DATA INDICATED WE WERE R OF LOC. THE PNF FO CALLED 'ALT' AND I DIRECTED THE FLT BACKUP TO CLB. AT THIS POINT, THE RWY WAS IN SIGHT. FLT BACKUP DISCONNECTED THE AUTOPLT, AUTOTHROTTLES, AND BEGAN A CLB TOWARD THE GS AND A SLIGHT L TURN TOWARD THE LOC. APCH CTL QUESTIONED OUR ALT AND THEN HANDED US OVER TO TWR. BOTH FLT DIRECTORS WERE COMMANDING A R TURN AWAY FROM THE LOC, AND A DSCNT AWAY FROM THE GS. I CYCLED BOTH FLT DIRECTORS OFF THEN ON, AND ARMED THE APCH MODE. THE REMAINDER OF THE FLT WAS UNEVENTFUL WITH NORMAL FLT DIRECTOR INDICATIONS. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: THE CAPT STATES HE WAS IN THE COPLT SEAT OF THE B-767-300, THE RELIEF PLT DOING THE FLYING IN THE CAPT'S SEAT. THE LOC HAD NOT LEFT THE FULL DISPLACEMENT WHEN THE ACFT STARTED TO TURN. THE HDG BUG AUTOMATICALLY ALIGNED WITH THE ILS AND THE AUTOPLT CAPTURED A FALSE LOC AND COMMENCED DSCNT ON A FALSE GS. (ON APR/XX/95 A CAUTION WAS SENT TO THE COMPANY PLTS REGARDING NUMEROUS FALSE CAPTURES AT BRUSSELS AND TORONTO.) THE CAPT SWITCHED OFF THE AUTOPLT, AUTOTHROTTLE, AND FLT DIRECTOR AND REARMED THE FLT DIRECTOR WHICH THEN INDICATED NORMALLY. THE AIRPLANE, HOWEVER, WAS HAND FLOWN TO THE LNDG WITH THE FLT DIRECTOR INDICATING NORMALLY. DURING THE FALSE CAPTURE THE RAW DATA WAS INDICATING CORRECTLY AS OBSERVED VISUALLY WITH THE ARPT IN SIGHT WHILE THE FLT DIRECTOR AND AUTOPLT WERE CHASING A FALSE CAPTURE TO THE R OF AND BELOW COURSE. THE 3 IRU'S ALIGNMENTS WERE ACCURATE THROUGHOUT THE APCH AND DURING THE FALSE CAPTURE. INDUSTRY PROC IS TO ARM THE APCH AT THE CUE 'CLRED FOR THE APCH' AS WAS DONE IN THIS CASE. THIS CAPT ARMS THE APCH WHEN HE IS FLYING WHEN HE SEES THE RAW DATA LOC MOVE ACROSS THE CASE FINESSING WITH HDG SELECT TO A MODERATE INTERCEPT ANGLE. HE SUGGESTS THIS PROC MAY HAVE PRECLUDED THIS FALSE CAPTURE AND MIGHT BE CONSIDERED A BETTER PROC INDUSTRY WIDE. HE FINDS THIS ESPECIALLY USEFUL IN SOUTH AMERICA WHERE FACILITY SURVEYS ARE LESS ACCURATE.

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.