Narrative:

While on a radar vector to runway 1L at ams, approach control asked us to verify our altitude. First officer responded with '3000 ft assigned.' approach control advised us that our mode C was reading 2400 ft and advised us to check our altimeter setting to 993 hectopascals. We did, and responded 993 hectopascals. The 993 hectopascals was in green numbers and looked normal. Both altimeters were the same and were reading 3000 ft. We were in VFR conditions at the time with good visibility and we had the airport in sight. We were asked at least one more time to check our altimeter setting. Again we did, and responded to the correct setting. We then noticed that our standby altimeter was reading 2400 ft. About this time, we were intercepting the electronic GS. As we were descending through about 1000 ft, I cycled my altimeter setting selector push button (I pushed once to standard, and a second time back to local). No change occurred on the altimeter setting display. It remained 993 hectopascals in green numbers, however, the altimeter did wind down rapidly approximately 500 ft. We landed without incident. At this point, we did not make any logbook entry. We did not know if the ATC equipment had malfunctioned, if it was our error, or if an aircraft malfunction had occurred. After arriving at our overnight hotel, and having time to analyze and discuss the situation with each other, we decided to locate the crew that was flying this aircraft back to dtw. We advised the crew of the situation and advised them to be aware of a possible altimeter switching problem. We also asked them to try to further analyze and write up any resulting discrepancy, if any was discovered. No terrain clearance, or aircraft separation problem occurred as a result of this problem. This has been reported to our air safety and maintenance departments. Supplemental information from acn 457517: standby altimeter was set by the captain to equivalent inches of mercury value of 29.30. The transition level for this airport is not a fixed/set value and for this time period was set by ATC at 3000 ft. Due to forecast WX and winds, the crew planned an approach/landing for runway 27, however, ATIS and ACARS landing information indicated runway 19R was active for our arrival (first change). During the time the flight was vectored from sugol (IAF 30 NM out) to final approach localizer intercept, the runway advised as active changed multiple times. ATIS advised runway 19R. The first amsterdam controller advised runway 27 and the final approach controller advised runway 1L. All of these changes occurred during the busiest and last 15 mins of the flight. Each approach change was reprogrammed into the FMC with the associated approach briefing conducted. The controller advised us to rechk our altimeter settings to '993 and 2993.' each time the first officer responded to the final controller with the above data the final controller responded that the mode C readout still showed 2400 ft MSL. This multiple exchange (approximately 4 times) of rechking the altimetry data resulted in the same scenario until the captain noticed and pointed out that the standby altimeter indicated 2500 ft MSL. In VMC, the captain rechked and manually pressed the button used to reset the altimeter setting from standby to the current altimeter setting (and vice versa). The first officer did the same and received the same result, all the while the altimeter setting remained exactly the same, ie, 993 hectopascals and green. Upon the last query by the final approach controller, the first officer reported a 'possible malfunction of the altimeter system' and would have it checked out. 2 possible technical problems in this situation: 1) the altimeter was reset by the flight crew by pushing the reset button as is usually done at transition level and the computer system subsequently reset the altitude readout to show an 'inches of mercury MSL altitude readout.' 2) the altimeter was not reset by the flight crew and the computer system made a change in the altitude readout to show an 'inches of mercury MSL altitude readout' when the settings showed hectopascals as being entered and active (green). The airline has removed the altimeter check from the approach check. The most valuable place for the altimeter to get verified is prior to commencing theapch. Amsterdam's approach information handling was laborious at best. ATIS, initial controller and final controller need to be synchronized so as to not add to the already busy workload of terminal arrival procedures which is only exacerbated by the fact that this was a heavy jet with a 2-MAN crew where there is typically 4 present during this phase. The fact that this flight flew all through the night over the north atlantic region under rvsm limitations only adds to the stressful demands of this particular flight. The multiple runway changes occurred during the flight's busiest time in a foreign land, where controllers using a heavy accent is exacerbated also by typically very rapid speech patterns. The fact that the flight level transition is not fixed, therefore not published and therefore not known until the flight contacts approach control, contributes to the confusion factor when crews on the B747-400 typically fly to areas of the world where such a flight level transition altitude is consistently set at a fixed altitude and published as such. Most of the B747-400 crews fly to amsterdam on an occasional basis at best. This particular crew consisted of a captain who flies to amsterdam about once per yr and the first officer, although experienced on the B747-400 for many yrs, had been to amsterdam only twice before, the last being over 3 yrs ago. Conclusion/recommendations: investigate the software engineering of the FMC/CADC interface to determine what caused the malfunctioning display anomaly. Reinstall a 'hard' altimeter setting challenge and response on the approach check (12000 ft). Interface with amsterdam controllers/control facility to eliminate multiple runway changes during/after the initial approach phase has begun.

Related reports:

Google
 

Original NASA ASRS Text

Title: A B747-400 2 PLT CREW EXPERIENCES AN ALTDEV DURING THEIR APCH TO SHIPOL ARPT. THEIR DUAL FMC-CADC COMPUTER SET ALTIMETERS DISPLAYED AN ALT 500 FT HIGHER THAN ACTUAL, ACCORDING TO THE CTLR AT EHAM, FO.

Narrative: WHILE ON A RADAR VECTOR TO RWY 1L AT AMS, APCH CTL ASKED US TO VERIFY OUR ALT. FO RESPONDED WITH '3000 FT ASSIGNED.' APCH CTL ADVISED US THAT OUR MODE C WAS READING 2400 FT AND ADVISED US TO CHK OUR ALTIMETER SETTING TO 993 HECTOPASCALS. WE DID, AND RESPONDED 993 HECTOPASCALS. THE 993 HECTOPASCALS WAS IN GREEN NUMBERS AND LOOKED NORMAL. BOTH ALTIMETERS WERE THE SAME AND WERE READING 3000 FT. WE WERE IN VFR CONDITIONS AT THE TIME WITH GOOD VISIBILITY AND WE HAD THE ARPT IN SIGHT. WE WERE ASKED AT LEAST ONE MORE TIME TO CHK OUR ALTIMETER SETTING. AGAIN WE DID, AND RESPONDED TO THE CORRECT SETTING. WE THEN NOTICED THAT OUR STANDBY ALTIMETER WAS READING 2400 FT. ABOUT THIS TIME, WE WERE INTERCEPTING THE ELECTRONIC GS. AS WE WERE DSNDING THROUGH ABOUT 1000 FT, I CYCLED MY ALTIMETER SETTING SELECTOR PUSH BUTTON (I PUSHED ONCE TO STANDARD, AND A SECOND TIME BACK TO LCL). NO CHANGE OCCURRED ON THE ALTIMETER SETTING DISPLAY. IT REMAINED 993 HECTOPASCALS IN GREEN NUMBERS, HOWEVER, THE ALTIMETER DID WIND DOWN RAPIDLY APPROX 500 FT. WE LANDED WITHOUT INCIDENT. AT THIS POINT, WE DID NOT MAKE ANY LOGBOOK ENTRY. WE DID NOT KNOW IF THE ATC EQUIP HAD MALFUNCTIONED, IF IT WAS OUR ERROR, OR IF AN ACFT MALFUNCTION HAD OCCURRED. AFTER ARRIVING AT OUR OVERNIGHT HOTEL, AND HAVING TIME TO ANALYZE AND DISCUSS THE SIT WITH EACH OTHER, WE DECIDED TO LOCATE THE CREW THAT WAS FLYING THIS ACFT BACK TO DTW. WE ADVISED THE CREW OF THE SIT AND ADVISED THEM TO BE AWARE OF A POSSIBLE ALTIMETER SWITCHING PROB. WE ALSO ASKED THEM TO TRY TO FURTHER ANALYZE AND WRITE UP ANY RESULTING DISCREPANCY, IF ANY WAS DISCOVERED. NO TERRAIN CLRNC, OR ACFT SEPARATION PROB OCCURRED AS A RESULT OF THIS PROB. THIS HAS BEEN RPTED TO OUR AIR SAFETY AND MAINT DEPTS. SUPPLEMENTAL INFO FROM ACN 457517: STANDBY ALTIMETER WAS SET BY THE CAPT TO EQUIVALENT INCHES OF MERCURY VALUE OF 29.30. THE TRANSITION LEVEL FOR THIS ARPT IS NOT A FIXED/SET VALUE AND FOR THIS TIME PERIOD WAS SET BY ATC AT 3000 FT. DUE TO FORECAST WX AND WINDS, THE CREW PLANNED AN APCH/LNDG FOR RWY 27, HOWEVER, ATIS AND ACARS LNDG INFO INDICATED RWY 19R WAS ACTIVE FOR OUR ARR (FIRST CHANGE). DURING THE TIME THE FLT WAS VECTORED FROM SUGOL (IAF 30 NM OUT) TO FINAL APCH LOC INTERCEPT, THE RWY ADVISED AS ACTIVE CHANGED MULTIPLE TIMES. ATIS ADVISED RWY 19R. THE FIRST AMSTERDAM CTLR ADVISED RWY 27 AND THE FINAL APCH CTLR ADVISED RWY 1L. ALL OF THESE CHANGES OCCURRED DURING THE BUSIEST AND LAST 15 MINS OF THE FLT. EACH APCH CHANGE WAS REPROGRAMMED INTO THE FMC WITH THE ASSOCIATED APCH BRIEFING CONDUCTED. THE CTLR ADVISED US TO RECHK OUR ALTIMETER SETTINGS TO '993 AND 2993.' EACH TIME THE FO RESPONDED TO THE FINAL CTLR WITH THE ABOVE DATA THE FINAL CTLR RESPONDED THAT THE MODE C READOUT STILL SHOWED 2400 FT MSL. THIS MULTIPLE EXCHANGE (APPROX 4 TIMES) OF RECHKING THE ALTIMETRY DATA RESULTED IN THE SAME SCENARIO UNTIL THE CAPT NOTICED AND POINTED OUT THAT THE STANDBY ALTIMETER INDICATED 2500 FT MSL. IN VMC, THE CAPT RECHKED AND MANUALLY PRESSED THE BUTTON USED TO RESET THE ALTIMETER SETTING FROM STANDBY TO THE CURRENT ALTIMETER SETTING (AND VICE VERSA). THE FO DID THE SAME AND RECEIVED THE SAME RESULT, ALL THE WHILE THE ALTIMETER SETTING REMAINED EXACTLY THE SAME, IE, 993 HECTOPASCALS AND GREEN. UPON THE LAST QUERY BY THE FINAL APCH CTLR, THE FO RPTED A 'POSSIBLE MALFUNCTION OF THE ALTIMETER SYS' AND WOULD HAVE IT CHKED OUT. 2 POSSIBLE TECHNICAL PROBS IN THIS SIT: 1) THE ALTIMETER WAS RESET BY THE FLC BY PUSHING THE RESET BUTTON AS IS USUALLY DONE AT TRANSITION LEVEL AND THE COMPUTER SYS SUBSEQUENTLY RESET THE ALT READOUT TO SHOW AN 'INCHES OF MERCURY MSL ALT READOUT.' 2) THE ALTIMETER WAS NOT RESET BY THE FLC AND THE COMPUTER SYS MADE A CHANGE IN THE ALT READOUT TO SHOW AN 'INCHES OF MERCURY MSL ALT READOUT' WHEN THE SETTINGS SHOWED HECTOPASCALS AS BEING ENTERED AND ACTIVE (GREEN). THE AIRLINE HAS REMOVED THE ALTIMETER CHK FROM THE APCH CHK. THE MOST VALUABLE PLACE FOR THE ALTIMETER TO GET VERIFIED IS PRIOR TO COMMENCING THEAPCH. AMSTERDAM'S APCH INFO HANDLING WAS LABORIOUS AT BEST. ATIS, INITIAL CTLR AND FINAL CTLR NEED TO BE SYNCHRONIZED SO AS TO NOT ADD TO THE ALREADY BUSY WORKLOAD OF TERMINAL ARR PROCS WHICH IS ONLY EXACERBATED BY THE FACT THAT THIS WAS A HVY JET WITH A 2-MAN CREW WHERE THERE IS TYPICALLY 4 PRESENT DURING THIS PHASE. THE FACT THAT THIS FLT FLEW ALL THROUGH THE NIGHT OVER THE NORTH ATLANTIC REGION UNDER RVSM LIMITATIONS ONLY ADDS TO THE STRESSFUL DEMANDS OF THIS PARTICULAR FLT. THE MULTIPLE RWY CHANGES OCCURRED DURING THE FLT'S BUSIEST TIME IN A FOREIGN LAND, WHERE CTLRS USING A HVY ACCENT IS EXACERBATED ALSO BY TYPICALLY VERY RAPID SPEECH PATTERNS. THE FACT THAT THE FLT LEVEL TRANSITION IS NOT FIXED, THEREFORE NOT PUBLISHED AND THEREFORE NOT KNOWN UNTIL THE FLT CONTACTS APCH CTL, CONTRIBUTES TO THE CONFUSION FACTOR WHEN CREWS ON THE B747-400 TYPICALLY FLY TO AREAS OF THE WORLD WHERE SUCH A FLT LEVEL TRANSITION ALT IS CONSISTENTLY SET AT A FIXED ALT AND PUBLISHED AS SUCH. MOST OF THE B747-400 CREWS FLY TO AMSTERDAM ON AN OCCASIONAL BASIS AT BEST. THIS PARTICULAR CREW CONSISTED OF A CAPT WHO FLIES TO AMSTERDAM ABOUT ONCE PER YR AND THE FO, ALTHOUGH EXPERIENCED ON THE B747-400 FOR MANY YRS, HAD BEEN TO AMSTERDAM ONLY TWICE BEFORE, THE LAST BEING OVER 3 YRS AGO. CONCLUSION/RECOMMENDATIONS: INVESTIGATE THE SOFTWARE ENGINEERING OF THE FMC/CADC INTERFACE TO DETERMINE WHAT CAUSED THE MALFUNCTIONING DISPLAY ANOMALY. REINSTALL A 'HARD' ALTIMETER SETTING CHALLENGE AND RESPONSE ON THE APCH CHK (12000 FT). INTERFACE WITH AMSTERDAM CTLRS/CTL FACILITY TO ELIMINATE MULTIPLE RWY CHANGES DURING/AFTER THE INITIAL APCH PHASE HAS BEGUN.

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.