Narrative:

This was my first flight as a captain on an oceanic crossing although I've had numerous xings as a first officer. It was my responsibility to chkout 2 first officers on this flight as to the procedures involved in oceanic xings. On all my previous xings, I had sole responsibility for the loading, monitoring, and plotting of the navigation system. Now that I was captain, I felt that this responsibility should be shared among the crew members to reduce load and serve as a means of double-checking each other. So, I implemented the following procedures: one pilot loads all waypoints and a second pilot checks the coordinates for errors, checks distances between each set of waypoints and compares them to the flight plan, and finally if all is correct he circles the waypoint number on the flight plan. One pilot maintains a navigation log which compares actual position to intended position at each waypoint. It also monitors and logs various outputs from the navigation units. One pilot maintains a log on the master flight plan of various conditions (winds, temperature, speeds, fuel time, etc). Finally one pilot maintains a plotting chart and plots a present position point 10 mins after crossing each fix. The eastbound flight went smoothly, both first officers quickly learned and used these procedures west/O error. Our problem arose on the westbound return flight. We had one shoot flight from eham to lfpb to pick-up more passengers and fuel before starting our oceanic crossing. While waiting for the passengers in eham I loaded the oceanic crossing fltplan into the global's memory. I thought this might save some time on the ground in lfpb. While loading the coordinates, I made the classic 1 degree error loading 50N 15W instead of 51N 15W for one of the coordinates, I had planned to have the first officers double-check these coordinates on the ground in lfpb when they loaded and double-checked the INS coordinates. After landing in lfpb, I had the first officers reload both INS's with the new fltplan coordinates using all the previous procedures. At this point, I had to leave the aircraft to take care of some paperwork, so I told them to load and double-check everything before my return. Little did I know that when they found the fltplan in the global's memory they assumed that I had already double-checked it, so my 1 degree error was never detected. All crew members now assumed that the navigation units were free of errors. After takeoff from lfpb the autoplt was engaged with the global as its navigation source. (Note: our global displays its loaded fltplan course on our radar screen). 10 mins after passing 50N 08W we made our first plot. I showed us just slightly south of the intended course (it was also just slightly north of our actual course) so this did not trigger any alarm bells. Since the global was the only unit loaded incorrectly and it displays its fltplan course on the radar, everything looked like we were right on track. I estimate that approximately midway to 50N 15W is where I noticed the discrepancy. The INS's global did not match. Knowing I was south of course I made an immediate right turn toward the correct course until we could figure out what was wrong. We found the error in the global and corrected it. When we passed 51N 15W we were less than 3 mins off our initial crossing estimate, so using this as a guide I estimate we were approximately 15 to 20 mi off course. (420 KTS = 7 mi/min, so 2-3 min = 14-21 mi). I believe this navigation error occurred because of a breakdown in communications in the cockpit. Maybe in retrospect, I should have double-checked everything the first officers did but I didn't want to instill any mistrust in them. This error seemed to be a difficult one to catch even if we would have checked the distances between waypoints, the difference in going to 50N 15W is only 3.8 mi with a mere 12 degree heading change. This may not have been enough for them to notice. One procedure which could have allowed us to detect this error much sooner would have been to have one of the INS's as the navigation source for the autoplt instead of the global. In this confign the error would have been detected by an actual deviation from the global's plotted course on the radar screen. This would provide another means of double-checking what the autoplt is actually flying (INS) against the fltplan loaded in the global. On all future oceanic flts this will be a standard operating procedure for our company.

Google
 

Original NASA ASRS Text

Title: POSITION DEVIATION ON THE START OF A TRANS ATLANTIC CROSSING.

Narrative: THIS WAS MY FIRST FLT AS A CAPT ON AN OCEANIC XING ALTHOUGH I'VE HAD NUMEROUS XINGS AS A F/O. IT WAS MY RESPONSIBILITY TO CHKOUT 2 F/OS ON THIS FLT AS TO THE PROCS INVOLVED IN OCEANIC XINGS. ON ALL MY PREVIOUS XINGS, I HAD SOLE RESPONSIBILITY FOR THE LOADING, MONITORING, AND PLOTTING OF THE NAV SYS. NOW THAT I WAS CAPT, I FELT THAT THIS RESPONSIBILITY SHOULD BE SHARED AMONG THE CREW MEMBERS TO REDUCE LOAD AND SERVE AS A MEANS OF DOUBLE-CHKING EACH OTHER. SO, I IMPLEMENTED THE FOLLOWING PROCS: ONE PLT LOADS ALL WAYPOINTS AND A SECOND PLT CHKS THE COORDINATES FOR ERRORS, CHKS DISTANCES BTWN EACH SET OF WAYPOINTS AND COMPARES THEM TO THE FLT PLAN, AND FINALLY IF ALL IS CORRECT HE CIRCLES THE WAYPOINT NUMBER ON THE FLT PLAN. ONE PLT MAINTAINS A NAV LOG WHICH COMPARES ACTUAL POS TO INTENDED POS AT EACH WAYPOINT. IT ALSO MONITORS AND LOGS VARIOUS OUTPUTS FROM THE NAV UNITS. ONE PLT MAINTAINS A LOG ON THE MASTER FLT PLAN OF VARIOUS CONDITIONS (WINDS, TEMP, SPDS, FUEL TIME, ETC). FINALLY ONE PLT MAINTAINS A PLOTTING CHART AND PLOTS A PRESENT POS POINT 10 MINS AFTER XING EACH FIX. THE EBND FLT WENT SMOOTHLY, BOTH F/OS QUICKLY LEARNED AND USED THESE PROCS W/O ERROR. OUR PROB AROSE ON THE WBND RETURN FLT. WE HAD ONE SHOOT FLT FROM EHAM TO LFPB TO PICK-UP MORE PAXS AND FUEL BEFORE STARTING OUR OCEANIC XING. WHILE WAITING FOR THE PAXS IN EHAM I LOADED THE OCEANIC XING FLTPLAN INTO THE GLOBAL'S MEMORY. I THOUGHT THIS MIGHT SAVE SOME TIME ON THE GND IN LFPB. WHILE LOADING THE COORDINATES, I MADE THE CLASSIC 1 DEG ERROR LOADING 50N 15W INSTEAD OF 51N 15W FOR ONE OF THE COORDINATES, I HAD PLANNED TO HAVE THE F/OS DOUBLE-CHK THESE COORDINATES ON THE GND IN LFPB WHEN THEY LOADED AND DOUBLE-CHKED THE INS COORDINATES. AFTER LNDG IN LFPB, I HAD THE F/OS RELOAD BOTH INS'S WITH THE NEW FLTPLAN COORDINATES USING ALL THE PREVIOUS PROCS. AT THIS POINT, I HAD TO LEAVE THE ACFT TO TAKE CARE OF SOME PAPERWORK, SO I TOLD THEM TO LOAD AND DOUBLE-CHK EVERYTHING BEFORE MY RETURN. LITTLE DID I KNOW THAT WHEN THEY FOUND THE FLTPLAN IN THE GLOBAL'S MEMORY THEY ASSUMED THAT I HAD ALREADY DOUBLE-CHKED IT, SO MY 1 DEG ERROR WAS NEVER DETECTED. ALL CREW MEMBERS NOW ASSUMED THAT THE NAV UNITS WERE FREE OF ERRORS. AFTER TKOF FROM LFPB THE AUTOPLT WAS ENGAGED WITH THE GLOBAL AS ITS NAV SOURCE. (NOTE: OUR GLOBAL DISPLAYS ITS LOADED FLTPLAN COURSE ON OUR RADAR SCREEN). 10 MINS AFTER PASSING 50N 08W WE MADE OUR FIRST PLOT. I SHOWED US JUST SLIGHTLY S OF THE INTENDED COURSE (IT WAS ALSO JUST SLIGHTLY N OF OUR ACTUAL COURSE) SO THIS DID NOT TRIGGER ANY ALARM BELLS. SINCE THE GLOBAL WAS THE ONLY UNIT LOADED INCORRECTLY AND IT DISPLAYS ITS FLTPLAN COURSE ON THE RADAR, EVERYTHING LOOKED LIKE WE WERE RIGHT ON TRACK. I ESTIMATE THAT APPROX MIDWAY TO 50N 15W IS WHERE I NOTICED THE DISCREPANCY. THE INS'S GLOBAL DID NOT MATCH. KNOWING I WAS S OF COURSE I MADE AN IMMEDIATE RIGHT TURN TOWARD THE CORRECT COURSE UNTIL WE COULD FIGURE OUT WHAT WAS WRONG. WE FOUND THE ERROR IN THE GLOBAL AND CORRECTED IT. WHEN WE PASSED 51N 15W WE WERE LESS THAN 3 MINS OFF OUR INITIAL XING ESTIMATE, SO USING THIS AS A GUIDE I ESTIMATE WE WERE APPROX 15 TO 20 MI OFF COURSE. (420 KTS = 7 MI/MIN, SO 2-3 MIN = 14-21 MI). I BELIEVE THIS NAV ERROR OCCURRED BECAUSE OF A BREAKDOWN IN COMS IN THE COCKPIT. MAYBE IN RETROSPECT, I SHOULD HAVE DOUBLE-CHKED EVERYTHING THE F/OS DID BUT I DIDN'T WANT TO INSTILL ANY MISTRUST IN THEM. THIS ERROR SEEMED TO BE A DIFFICULT ONE TO CATCH EVEN IF WE WOULD HAVE CHKED THE DISTANCES BTWN WAYPOINTS, THE DIFFERENCE IN GOING TO 50N 15W IS ONLY 3.8 MI WITH A MERE 12 DEG HDG CHANGE. THIS MAY NOT HAVE BEEN ENOUGH FOR THEM TO NOTICE. ONE PROC WHICH COULD HAVE ALLOWED US TO DETECT THIS ERROR MUCH SOONER WOULD HAVE BEEN TO HAVE ONE OF THE INS'S AS THE NAV SOURCE FOR THE AUTOPLT INSTEAD OF THE GLOBAL. IN THIS CONFIGN THE ERROR WOULD HAVE BEEN DETECTED BY AN ACTUAL DEV FROM THE GLOBAL'S PLOTTED COURSE ON THE RADAR SCREEN. THIS WOULD PROVIDE ANOTHER MEANS OF DOUBLE-CHKING WHAT THE AUTOPLT IS ACTUALLY FLYING (INS) AGAINST THE FLTPLAN LOADED IN THE GLOBAL. ON ALL FUTURE OCEANIC FLTS THIS WILL BE A STANDARD OPERATING PROC FOR OUR COMPANY.

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.