Narrative:

This problem occurred at the end of a flight in my 1984 BE58 baron from nashville, tn, to atlanta, GA, on oct/xx/96. I was given an intercept vector and cleared by atlanta approach for the VOR/DME or GPS runway 27 approach circle to land runway 2R at peachtree-dekalb airport, atlanta, GA. I had elected to fly the GPS approach as loaded in the database on my kln 90B GPS with the autoplt armed for the approach. The autoplt coupled the approach and I began a descent from 3000 ft MSL to 2500 ft MSL as specified by the approach procedure. Before reaching 2500 ft, I noted the moving map display on the GPS showed the airplane south of the approach course which was confirmed by cross reference to the second VOR indicator. At this point, I was contacted by atlanta approach who stated he showed me south of the approach course and canceled my approach clearance. As I began the missed approach procedure and disconnected the autoplt, I asked for another approach and received vectors. I also informed the controller I had flown a GPS procedure. He suggested making the second approach a VOR approach and I agreed. At this point, I also asked him to repeat the previous vector. When he repeated it, I realized I had turned past it to the left and began an immediate turn back to the vector. The controller said it appeared I was turning to 360 degrees and I informed him I had turned back to 120 degrees as requested. The controller cleared me onto the approach course again. I intercepted the appropriate VOR radial, flew the approach by hand according to the published approach, broke out well above the published minimum, and landed on runway 2R without incident. I am still not sure what caused the problem with the original GPS approach. I have practiced many similar approachs with a flight instructor without problems since having the GPS installed in may 1996. The database was also current (the GPS will not allow the approach to activate if the database is out of date). I am currently reviewing the GPS manual and intend to fly GPS approachs with my flight instructor this week to try to duplicate this situation and to make sure I understand the procedures necessary to properly couple such an approach to the autoplt. It is possible that I should have had the GPS on obs mode instead of leg mode at this particular segment of the approach, I will test this possibility during the approachs with my flight instructor. I should have made sure I was firmly established on the new missed approach vector before discussing the new approach procedure with the controller. I allowed this distraction to cause me to fly through the vector before I verified the vector and corrected my heading. I realize the judicious use of the autoplt using the heading mode would have been very helpful at this point in managing the increased workload of setting up the new approach on VOR 1 and the HSI. In my opinion, nothing occurred that compromised the safety of the flight since the deviation from the approach course was idented by both the pilot and the controller before a descent to the final MDA was initiated. However, the occurrence certainly demonstrates the importance of using all available resources in monitoring an instrument approach, particularly one that can be flown using either VOR or GPS. Callback conversation with reporter revealed the following information: the next weekend the reporter went out with an instructor pilot and they practiced approachs, using the GPS. Reporter found that his problem had been that he had failed to switch from the 'leg' phase on the GPS to the 'obs' position. While in the 'leg' position the GPS will want to take the aircraft to the next fix within the database if coupled to the autoplt. If the flight director is also on it can be very confusing to the pilot. If receiving vectors, one must select 'obs' to override this database default and then the aircraft will take the vectors needed for the approach. This is only when off the final approach course. Once on the final approach course, the 'leg' position is once again selected.

Google
 

Original NASA ASRS Text

Title: PVT PLT USING GPS FOR IAP PROC VOR DME APCH RWY 27 AT PDK SUFFERS A HEADING TRACK POS DEV. MAKES MISSED APCH UPON HAVING ATL APCH CTLR CANCEL HIS APCH CLRNC.

Narrative: THIS PROB OCCURRED AT THE END OF A FLT IN MY 1984 BE58 BARON FROM NASHVILLE, TN, TO ATLANTA, GA, ON OCT/XX/96. I WAS GIVEN AN INTERCEPT VECTOR AND CLRED BY ATLANTA APCH FOR THE VOR/DME OR GPS RWY 27 APCH CIRCLE TO LAND RWY 2R AT PEACHTREE-DEKALB ARPT, ATLANTA, GA. I HAD ELECTED TO FLY THE GPS APCH AS LOADED IN THE DATABASE ON MY KLN 90B GPS WITH THE AUTOPLT ARMED FOR THE APCH. THE AUTOPLT COUPLED THE APCH AND I BEGAN A DSCNT FROM 3000 FT MSL TO 2500 FT MSL AS SPECIFIED BY THE APCH PROC. BEFORE REACHING 2500 FT, I NOTED THE MOVING MAP DISPLAY ON THE GPS SHOWED THE AIRPLANE S OF THE APCH COURSE WHICH WAS CONFIRMED BY CROSS REF TO THE SECOND VOR INDICATOR. AT THIS POINT, I WAS CONTACTED BY ATLANTA APCH WHO STATED HE SHOWED ME S OF THE APCH COURSE AND CANCELED MY APCH CLRNC. AS I BEGAN THE MISSED APCH PROC AND DISCONNECTED THE AUTOPLT, I ASKED FOR ANOTHER APCH AND RECEIVED VECTORS. I ALSO INFORMED THE CTLR I HAD FLOWN A GPS PROC. HE SUGGESTED MAKING THE SECOND APCH A VOR APCH AND I AGREED. AT THIS POINT, I ALSO ASKED HIM TO REPEAT THE PREVIOUS VECTOR. WHEN HE REPEATED IT, I REALIZED I HAD TURNED PAST IT TO THE L AND BEGAN AN IMMEDIATE TURN BACK TO THE VECTOR. THE CTLR SAID IT APPEARED I WAS TURNING TO 360 DEGS AND I INFORMED HIM I HAD TURNED BACK TO 120 DEGS AS REQUESTED. THE CTLR CLRED ME ONTO THE APCH COURSE AGAIN. I INTERCEPTED THE APPROPRIATE VOR RADIAL, FLEW THE APCH BY HAND ACCORDING TO THE PUBLISHED APCH, BROKE OUT WELL ABOVE THE PUBLISHED MINIMUM, AND LANDED ON RWY 2R WITHOUT INCIDENT. I AM STILL NOT SURE WHAT CAUSED THE PROB WITH THE ORIGINAL GPS APCH. I HAVE PRACTICED MANY SIMILAR APCHS WITH A FLT INSTRUCTOR WITHOUT PROBS SINCE HAVING THE GPS INSTALLED IN MAY 1996. THE DATABASE WAS ALSO CURRENT (THE GPS WILL NOT ALLOW THE APCH TO ACTIVATE IF THE DATABASE IS OUT OF DATE). I AM CURRENTLY REVIEWING THE GPS MANUAL AND INTEND TO FLY GPS APCHS WITH MY FLT INSTRUCTOR THIS WK TO TRY TO DUPLICATE THIS SIT AND TO MAKE SURE I UNDERSTAND THE PROCS NECESSARY TO PROPERLY COUPLE SUCH AN APCH TO THE AUTOPLT. IT IS POSSIBLE THAT I SHOULD HAVE HAD THE GPS ON OBS MODE INSTEAD OF LEG MODE AT THIS PARTICULAR SEGMENT OF THE APCH, I WILL TEST THIS POSSIBILITY DURING THE APCHS WITH MY FLT INSTRUCTOR. I SHOULD HAVE MADE SURE I WAS FIRMLY ESTABLISHED ON THE NEW MISSED APCH VECTOR BEFORE DISCUSSING THE NEW APCH PROC WITH THE CTLR. I ALLOWED THIS DISTR TO CAUSE ME TO FLY THROUGH THE VECTOR BEFORE I VERIFIED THE VECTOR AND CORRECTED MY HEADING. I REALIZE THE JUDICIOUS USE OF THE AUTOPLT USING THE HEADING MODE WOULD HAVE BEEN VERY HELPFUL AT THIS POINT IN MANAGING THE INCREASED WORKLOAD OF SETTING UP THE NEW APCH ON VOR 1 AND THE HSI. IN MY OPINION, NOTHING OCCURRED THAT COMPROMISED THE SAFETY OF THE FLT SINCE THE DEV FROM THE APCH COURSE WAS IDENTED BY BOTH THE PLT AND THE CTLR BEFORE A DSCNT TO THE FINAL MDA WAS INITIATED. HOWEVER, THE OCCURRENCE CERTAINLY DEMONSTRATES THE IMPORTANCE OF USING ALL AVAILABLE RESOURCES IN MONITORING AN INST APCH, PARTICULARLY ONE THAT CAN BE FLOWN USING EITHER VOR OR GPS. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: THE NEXT WEEKEND THE RPTR WENT OUT WITH AN INSTRUCTOR PLT AND THEY PRACTICED APCHS, USING THE GPS. RPTR FOUND THAT HIS PROB HAD BEEN THAT HE HAD FAILED TO SWITCH FROM THE 'LEG' PHASE ON THE GPS TO THE 'OBS' POS. WHILE IN THE 'LEG' POS THE GPS WILL WANT TO TAKE THE ACFT TO THE NEXT FIX WITHIN THE DATABASE IF COUPLED TO THE AUTOPLT. IF THE FLT DIRECTOR IS ALSO ON IT CAN BE VERY CONFUSING TO THE PLT. IF RECEIVING VECTORS, ONE MUST SELECT 'OBS' TO OVERRIDE THIS DATABASE DEFAULT AND THEN THE ACFT WILL TAKE THE VECTORS NEEDED FOR THE APCH. THIS IS ONLY WHEN OFF THE FINAL APCH COURSE. ONCE ON THE FINAL APCH COURSE, THE 'LEG' POS IS ONCE AGAIN SELECTED.

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.