Narrative:

While en route ATC gave us numerous vectors for spacing purposes. Finally we were cleared direct to the EMI VOR. Because we were only 9 mi from the fix and heading 090 degrees away from the VOR, the FMC commanded the autoplt into a steep right turn towards the fix. Just about the time we were established direct to the fix, we passed over the VOR. At this point we both noticed our CDI needles deflect 3/4 to the left and the FMC commanded the autoplt to turn left. This type of large CDI deflection while in LNAV is not normal, so we knew there was some kind of problem with the FMC. We checked the flight plan for the proper outbound heading from EMI, and sure enough, there was a problem. We then selected heading mode on our autoplt and began a turn to the right, using raw VOR data to head towards the outbound radial for J48. Just at this point ATC asked what heading we were on and we told them of our problem and we were correcting right. ATC then asked us to correct another 20 degrees right for the intercept. Luckily there was no conflict with other traffic and ATC told us there was no problem. After reviewing our situation it was clear to us that the FMC became confused while trying to sort out the direct EMI inputs, while being so close, and so extreme an angle to the fix. Fortunately, we noticed the problem in time to take action before getting way off course. It's important to note that while FMC automation is a great tool for us pilots, it is not perfect. We always have to be ready to go to manual mode when things go wrong. Supplemental information from acn 536574: ZDC was giving us sharp radar vectors 'for ZTL spacing.' after several turns we were cleared direct EMI when only 10 mi from the VOR at an adverse angle. I believe the FMC became confused due to our close proximity to the VOR and radical ground track called for by the ATC clearance. I believe ATC should be more cautious when clearing FMC equipped aircraft to a close proximity fix involving radical turns to re-establish a flight plan route. A better solution would have been to issue a radar vector to intercept the airway downstream from the subject fix.

Related reports:

Google
 

Original NASA ASRS Text

Title: A B737-300 FLC CORRECTS THEIR ACFT'S SUDDEN TURN OFF COURSE AFTER HAVING PROGRAMMED A SHORT RANGED 90 DEG INTERCEPT TO EMI, MD.

Narrative: WHILE ENRTE ATC GAVE US NUMEROUS VECTORS FOR SPACING PURPOSES. FINALLY WE WERE CLRED DIRECT TO THE EMI VOR. BECAUSE WE WERE ONLY 9 MI FROM THE FIX AND HEADING 090 DEGS AWAY FROM THE VOR, THE FMC COMMANDED THE AUTOPLT INTO A STEEP R TURN TOWARDS THE FIX. JUST ABOUT THE TIME WE WERE ESTABLISHED DIRECT TO THE FIX, WE PASSED OVER THE VOR. AT THIS POINT WE BOTH NOTICED OUR CDI NEEDLES DEFLECT 3/4 TO THE L AND THE FMC COMMANDED THE AUTOPLT TO TURN L. THIS TYPE OF LARGE CDI DEFLECTION WHILE IN LNAV IS NOT NORMAL, SO WE KNEW THERE WAS SOME KIND OF PROB WITH THE FMC. WE CHKED THE FLT PLAN FOR THE PROPER OUTBOUND HEADING FROM EMI, AND SURE ENOUGH, THERE WAS A PROB. WE THEN SELECTED HEADING MODE ON OUR AUTOPLT AND BEGAN A TURN TO THE R, USING RAW VOR DATA TO HEAD TOWARDS THE OUTBOUND RADIAL FOR J48. JUST AT THIS POINT ATC ASKED WHAT HEADING WE WERE ON AND WE TOLD THEM OF OUR PROB AND WE WERE CORRECTING R. ATC THEN ASKED US TO CORRECT ANOTHER 20 DEGS R FOR THE INTERCEPT. LUCKILY THERE WAS NO CONFLICT WITH OTHER TFC AND ATC TOLD US THERE WAS NO PROB. AFTER REVIEWING OUR SIT IT WAS CLR TO US THAT THE FMC BECAME CONFUSED WHILE TRYING TO SORT OUT THE DIRECT EMI INPUTS, WHILE BEING SO CLOSE, AND SO EXTREME AN ANGLE TO THE FIX. FORTUNATELY, WE NOTICED THE PROB IN TIME TO TAKE ACTION BEFORE GETTING WAY OFF COURSE. IT'S IMPORTANT TO NOTE THAT WHILE FMC AUTOMATION IS A GREAT TOOL FOR US PLTS, IT IS NOT PERFECT. WE ALWAYS HAVE TO BE READY TO GO TO MANUAL MODE WHEN THINGS GO WRONG. SUPPLEMENTAL INFO FROM ACN 536574: ZDC WAS GIVING US SHARP RADAR VECTORS 'FOR ZTL SPACING.' AFTER SEVERAL TURNS WE WERE CLRED DIRECT EMI WHEN ONLY 10 MI FROM THE VOR AT AN ADVERSE ANGLE. I BELIEVE THE FMC BECAME CONFUSED DUE TO OUR CLOSE PROX TO THE VOR AND RADICAL GND TRACK CALLED FOR BY THE ATC CLRNC. I BELIEVE ATC SHOULD BE MORE CAUTIOUS WHEN CLRING FMC EQUIPPED ACFT TO A CLOSE PROX FIX INVOLVING RADICAL TURNS TO RE-ESTABLISH A FLT PLAN RTE. A BETTER SOLUTION WOULD HAVE BEEN TO ISSUE A RADAR VECTOR TO INTERCEPT THE AIRWAY DOWNSTREAM FROM THE SUBJECT FIX.

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.