Narrative:

Flight sequence of 3 legs began nov/tue/95 at XX25 EST from sdf to phl then rdu and return to sdf. On leg #1 to phl, I noticed FMC programmed dscnts were slightly abnormal, at a slower rate with approximately 1/2 throttle instead of power off. Checks of FMC and IRS data revealed no abnormalities. On leg #2 to rdu dscnts were again started by the FMC a little earlier than I thought necessary. On the ground at rdu I again tried to find something concrete to write up for maintenance without success. Our 3 IRU's showed error rates of .5, .5, .6, (NM/hour I think) on the ground. Our operations manual (pilot) shows no data and I considered this acceptable. On leg #3 rdu to sdf, I again tried to find the problem. This time FMC controlled dscnts were perfect power off. I assumed the 'anomaly' (boeing's favorite word for B757 glitches) had fixed itself. When outside the FAF at sdf I discovered a 5 mi 'map shift.' the map showed us 5 mi west of our actual position, even though both vors were in 'automatic' for IRS updating. On the ground at sdf with brake set we showed 10 KTS ground speed and IRU error rates of 4.6, 4.6, 4.5. Even though we showed no map shift on final to phl or rdu, we wonder if it had in fact occurred en route. Had we missed any crossing restrs? Were we on course? This EFIS 'magic' is our primary navigation method (approved) until on approach. I will never trust it again. RMI needles with a DME readout have never let me down. (At this time maintenance has no answers, still investigating.) callback conversation with reporter revealed the following information: the captain was flying a B757-200 freighter equipped with rolls-royce engines. The aircraft's data base showed that the localizer and the DME at sdf were not co-located when, in fact, they are and this part of the problem has been corrected. The reporter is pursuing the descent problem still. He said that one area of suspicion stems from the fact that the earlier aircraft had pratt and whitney engines. The fmcs show the correct labels, but perhaps the incorrect data base is installed. This analyst believes that another cause could be a large wind change as the aircraft descended the first 2 times. A great increase in the effective headwind would cause the computer to adjust by shallowing its descent rate and adding power.

Related reports:

Google
 

Original NASA ASRS Text

Title: ACFT EQUIP PROB. ACR HAS AN APPARENT MAP SHIFT, BUT THE PROB WAS A DATA BASE ERROR.

Narrative: FLT SEQUENCE OF 3 LEGS BEGAN NOV/TUE/95 AT XX25 EST FROM SDF TO PHL THEN RDU AND RETURN TO SDF. ON LEG #1 TO PHL, I NOTICED FMC PROGRAMMED DSCNTS WERE SLIGHTLY ABNORMAL, AT A SLOWER RATE WITH APPROX 1/2 THROTTLE INSTEAD OF PWR OFF. CHKS OF FMC AND IRS DATA REVEALED NO ABNORMALITIES. ON LEG #2 TO RDU DSCNTS WERE AGAIN STARTED BY THE FMC A LITTLE EARLIER THAN I THOUGHT NECESSARY. ON THE GND AT RDU I AGAIN TRIED TO FIND SOMETHING CONCRETE TO WRITE UP FOR MAINT WITHOUT SUCCESS. OUR 3 IRU'S SHOWED ERROR RATES OF .5, .5, .6, (NM/HR I THINK) ON THE GND. OUR OPS MANUAL (PLT) SHOWS NO DATA AND I CONSIDERED THIS ACCEPTABLE. ON LEG #3 RDU TO SDF, I AGAIN TRIED TO FIND THE PROB. THIS TIME FMC CONTROLLED DSCNTS WERE PERFECT PWR OFF. I ASSUMED THE 'ANOMALY' (BOEING'S FAVORITE WORD FOR B757 GLITCHES) HAD FIXED ITSELF. WHEN OUTSIDE THE FAF AT SDF I DISCOVERED A 5 MI 'MAP SHIFT.' THE MAP SHOWED US 5 MI W OF OUR ACTUAL POS, EVEN THOUGH BOTH VORS WERE IN 'AUTO' FOR IRS UPDATING. ON THE GND AT SDF WITH BRAKE SET WE SHOWED 10 KTS GND SPD AND IRU ERROR RATES OF 4.6, 4.6, 4.5. EVEN THOUGH WE SHOWED NO MAP SHIFT ON FINAL TO PHL OR RDU, WE WONDER IF IT HAD IN FACT OCCURRED ENRTE. HAD WE MISSED ANY XING RESTRS? WERE WE ON COURSE? THIS EFIS 'MAGIC' IS OUR PRIMARY NAV METHOD (APPROVED) UNTIL ON APCH. I WILL NEVER TRUST IT AGAIN. RMI NEEDLES WITH A DME READOUT HAVE NEVER LET ME DOWN. (AT THIS TIME MAINT HAS NO ANSWERS, STILL INVESTIGATING.) CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: THE CAPT WAS FLYING A B757-200 FREIGHTER EQUIPPED WITH ROLLS-ROYCE ENGS. THE ACFT'S DATA BASE SHOWED THAT THE LOC AND THE DME AT SDF WERE NOT CO-LOCATED WHEN, IN FACT, THEY ARE AND THIS PART OF THE PROB HAS BEEN CORRECTED. THE RPTR IS PURSUING THE DSCNT PROB STILL. HE SAID THAT ONE AREA OF SUSPICION STEMS FROM THE FACT THAT THE EARLIER ACFT HAD PRATT AND WHITNEY ENGS. THE FMCS SHOW THE CORRECT LABELS, BUT PERHAPS THE INCORRECT DATA BASE IS INSTALLED. THIS ANALYST BELIEVES THAT ANOTHER CAUSE COULD BE A LARGE WIND CHANGE AS THE ACFT DSNDED THE FIRST 2 TIMES. A GREAT INCREASE IN THE EFFECTIVE HEADWIND WOULD CAUSE THE COMPUTER TO ADJUST BY SHALLOWING ITS DSCNT RATE AND ADDING PWR.

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.