Narrative:

During descent into klga on the nanci arrival, we were given the standard clearance to cross somto intersection at 11000 ft. We were level at FL260 having made a VNAV path descent to cross the kerno intersection at FL260. Upon receipt of the clearance, we set 11000 ft in the MCP altitude window and verified that we had indeed programmed somto at 11000 ft. We also noted that our touchdown point from our intermediate altitude of FL260 was 25 NM. Our pitch at that point was VNAV. ATC then requested that we start our descent immediately to cross somto at 11000 ft. After selecting the descent page we observed our capture prompt to be missing, (indicating at active capture descent in progress). We then selected a vertical speed descent of 1000 FPM and after descending approximately 500- 600 reselected VNAV and observed an immediate return to the capture mode. We then entered into a brief discussion concerning the relative merits of a speed descent versus a path descent and did not notice the FMC's apparent failure to compute a path based on somto at 11000 ft, and its failure to capture a path that would have crossed somto at 11000 ft. That discussion was interrupted by ATC's query of 'are you going to be able to make somto at 11000 ft?' we both quickly referred to: a) vertical deviation on the descent page and B) distance from somto on our route legs page. It was then that we noticed the failure of the FMC to compute and capture the proper path. When I saw how close we were to somto (4 NM) I expected to see a +10000 ft or so on my vertical deviation, but in fact it was showing a - 400 ft from a downtrack waypoint constraint, wezul at 3000 ft. We advised ATC that we would not be able to meet the restriction and he asked us to expedite through 20000 ft, which we did. No further comment was made to ZDC concerning the somto clearance and we subsequently made a normal frequency change to new york approach control. Analysis: during the capture descent that we had been so careful to verify, our path somehow changed (or was deleted). It was obviously not deleted by either pilot, but it inexplicably disappeared from our descent page. (This is where I wish I had a video tape of the descent to more closely scrutinize the actual chain of events.) we do know this, when ATC questioned our ability to make the somto restriction and we quickly analyzed our relationship to the displayed path, the VNAV 'light' was not on. By definition, that means the mode cannot be disengaged. More accurately in this case, I suspect it also meant that the mode was not engaged. The first officer was quick to select a different pitch mode, level change, deployed full speed brakes, and an IAS command of 340 KIAS to expedite our descent. We had the opportunity to fly the nanci arrival 2 more times during this 3-DAY trip pairing and each time we tried to duplicate the chain of events with no success. (We made the somto restriction each time with normal indications.) conclusion: even though we properly programmed the descent path, we failed to monitor that descent and its subsequent capture of the proper path in order to meet the somto restriction. Thankfully, there was no traffic conflict. Recommendation: due to complaints from other pilots on this particular route, concerning their inability to program certain altitude constraints, perhaps there is a peculiarity about this particular arrival that may cause problems within the FMC regarding path computations. I recommend that industry experts examine this particular crossing restriction to insure that some sort of anomaly does not exist that might cause another crew to make the same mistake.

Google
 

Original NASA ASRS Text

Title: ALTDEV ALT XING RESTRICTION NOT MET.

Narrative: DURING DSCNT INTO KLGA ON THE NANCI ARR, WE WERE GIVEN THE STANDARD CLRNC TO CROSS SOMTO INTXN AT 11000 FT. WE WERE LEVEL AT FL260 HAVING MADE A VNAV PATH DSCNT TO CROSS THE KERNO INTXN AT FL260. UPON RECEIPT OF THE CLRNC, WE SET 11000 FT IN THE MCP ALT WINDOW AND VERIFIED THAT WE HAD INDEED PROGRAMMED SOMTO AT 11000 FT. WE ALSO NOTED THAT OUR TOUCHDOWN POINT FROM OUR INTERMEDIATE ALT OF FL260 WAS 25 NM. OUR PITCH AT THAT POINT WAS VNAV. ATC THEN REQUESTED THAT WE START OUR DSCNT IMMEDIATELY TO CROSS SOMTO AT 11000 FT. AFTER SELECTING THE DSCNT PAGE WE OBSERVED OUR CAPTURE PROMPT TO BE MISSING, (INDICATING AT ACTIVE CAPTURE DSCNT IN PROGRESS). WE THEN SELECTED A VERT SPD DSCNT OF 1000 FPM AND AFTER DSNDING APPROX 500- 600 RESELECTED VNAV AND OBSERVED AN IMMEDIATE RETURN TO THE CAPTURE MODE. WE THEN ENTERED INTO A BRIEF DISCUSSION CONCERNING THE RELATIVE MERITS OF A SPD DSCNT VERSUS A PATH DSCNT AND DID NOT NOTICE THE FMC'S APPARENT FAILURE TO COMPUTE A PATH BASED ON SOMTO AT 11000 FT, AND ITS FAILURE TO CAPTURE A PATH THAT WOULD HAVE CROSSED SOMTO AT 11000 FT. THAT DISCUSSION WAS INTERRUPTED BY ATC'S QUERY OF 'ARE YOU GOING TO BE ABLE TO MAKE SOMTO AT 11000 FT?' WE BOTH QUICKLY REFERRED TO: A) VERT DEV ON THE DSCNT PAGE AND B) DISTANCE FROM SOMTO ON OUR RTE LEGS PAGE. IT WAS THEN THAT WE NOTICED THE FAILURE OF THE FMC TO COMPUTE AND CAPTURE THE PROPER PATH. WHEN I SAW HOW CLOSE WE WERE TO SOMTO (4 NM) I EXPECTED TO SEE A +10000 FT OR SO ON MY VERT DEV, BUT IN FACT IT WAS SHOWING A - 400 FT FROM A DOWNTRACK WAYPOINT CONSTRAINT, WEZUL AT 3000 FT. WE ADVISED ATC THAT WE WOULD NOT BE ABLE TO MEET THE RESTRICTION AND HE ASKED US TO EXPEDITE THROUGH 20000 FT, WHICH WE DID. NO FURTHER COMMENT WAS MADE TO ZDC CONCERNING THE SOMTO CLRNC AND WE SUBSEQUENTLY MADE A NORMAL FREQ CHANGE TO NEW YORK APCH CTL. ANALYSIS: DURING THE CAPTURE DSCNT THAT WE HAD BEEN SO CAREFUL TO VERIFY, OUR PATH SOMEHOW CHANGED (OR WAS DELETED). IT WAS OBVIOUSLY NOT DELETED BY EITHER PLT, BUT IT INEXPLICABLY DISAPPEARED FROM OUR DSCNT PAGE. (THIS IS WHERE I WISH I HAD A VIDEO TAPE OF THE DSCNT TO MORE CLOSELY SCRUTINIZE THE ACTUAL CHAIN OF EVENTS.) WE DO KNOW THIS, WHEN ATC QUESTIONED OUR ABILITY TO MAKE THE SOMTO RESTRICTION AND WE QUICKLY ANALYZED OUR RELATIONSHIP TO THE DISPLAYED PATH, THE VNAV 'LIGHT' WAS NOT ON. BY DEFINITION, THAT MEANS THE MODE CANNOT BE DISENGAGED. MORE ACCURATELY IN THIS CASE, I SUSPECT IT ALSO MEANT THAT THE MODE WAS NOT ENGAGED. THE FO WAS QUICK TO SELECT A DIFFERENT PITCH MODE, LEVEL CHANGE, DEPLOYED FULL SPD BRAKES, AND AN IAS COMMAND OF 340 KIAS TO EXPEDITE OUR DSCNT. WE HAD THE OPPORTUNITY TO FLY THE NANCI ARR 2 MORE TIMES DURING THIS 3-DAY TRIP PAIRING AND EACH TIME WE TRIED TO DUPLICATE THE CHAIN OF EVENTS WITH NO SUCCESS. (WE MADE THE SOMTO RESTRICTION EACH TIME WITH NORMAL INDICATIONS.) CONCLUSION: EVEN THOUGH WE PROPERLY PROGRAMMED THE DSCNT PATH, WE FAILED TO MONITOR THAT DSCNT AND ITS SUBSEQUENT CAPTURE OF THE PROPER PATH IN ORDER TO MEET THE SOMTO RESTRICTION. THANKFULLY, THERE WAS NO TFC CONFLICT. RECOMMENDATION: DUE TO COMPLAINTS FROM OTHER PLTS ON THIS PARTICULAR RTE, CONCERNING THEIR INABILITY TO PROGRAM CERTAIN ALT CONSTRAINTS, PERHAPS THERE IS A PECULIARITY ABOUT THIS PARTICULAR ARR THAT MAY CAUSE PROBLEMS WITHIN THE FMC REGARDING PATH COMPUTATIONS. I RECOMMEND THAT INDUSTRY EXPERTS EXAMINE THIS PARTICULAR XING RESTRICTION TO INSURE THAT SOME SORT OF ANOMALY DOES NOT EXIST THAT MIGHT CAUSE ANOTHER CREW TO MAKE THE SAME MISTAKE.

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.