Narrative:

Delayed 4 hours for arrival into sfo due to high winds. Conditions at field were dusk or nearly dark; winds 010 @30 gust 45. Assigned and briefed LOC28L circle to land RW01R with break-out at san mateo bridge. Just prior to the bridge they switched us to circle land RW01L; we continued. As we begun the circle; we programmed the box for RW01L; 5 mile centerline; 3deg gp (there are no approaches for landing north at sfo.) we closed up the discontinuities and because the course did not auto-populate; we chose direct to first fix. The magenta line curved inside the 5 mile fix; it looked reasonable in relation to our position with the airport. We began circle at 1;900 feet configured to gear and flaps 15. We elected to stay in LNAV/VNAV and ap for safety. Turning base to final we received a GPWS alert and executed a go around. The go around went well; we climbed to 5;000 feet and were vectored for same approach again; but this time landing RW01R. We ran the checklists; and discussed what went wrong. We quickly identified our errors in using a 5 mile extension; and using too much automation for a visual/unusual maneuver. We decided to draw a 3 mile extension only as a reference; and flew without ap. For some reason the pdi did not work on this approach; so we put RW01R in the fix page; and used 3 to 1 to make sure were on gp. Lining up with the runway went well; but there are no approach lights/PAPI; lots of ground clutter; a very challenging runway to pick up at night in such a high work-load environment

Related reports:

Google
 

Original NASA ASRS Text

Title: A B737 flight crew went around after a GPWS alert sounded while they were circling to land at SFO on RWY 01L. On the next approach they shortened the final segment from five miles to three miles and were able to fly the approach without the GPWS sounding.

Narrative: Delayed 4 hours for arrival into SFO due to high winds. Conditions at field were dusk or nearly dark; winds 010 @30 gust 45. Assigned and briefed LOC28L Circle to Land RW01R with break-out at San Mateo Bridge. Just prior to the bridge they switched us to Circle land RW01L; we continued. As we begun the circle; we programmed the box for RW01L; 5 mile centerline; 3deg GP (there are no approaches for landing north at SFO.) We closed up the discontinuities and because the course did not auto-populate; we chose direct to first fix. The magenta line curved inside the 5 mile fix; it looked reasonable in relation to our position with the airport. We began circle at 1;900 feet configured to gear and flaps 15. We elected to stay in LNAV/VNAV and AP for safety. Turning base to final we received a GPWS alert and executed a go around. The go around went well; we climbed to 5;000 feet and were vectored for same approach again; but this time landing RW01R. We ran the checklists; and discussed what went wrong. We quickly identified our errors in using a 5 mile extension; and using too much automation for a visual/unusual maneuver. We decided to draw a 3 mile extension only as a reference; and flew without AP. For some reason the PDI did not work on this approach; so we put RW01R in the fix page; and used 3 to 1 to make sure were on GP. Lining up with the runway went well; but there are no approach lights/PAPI; lots of ground clutter; a very challenging runway to pick up at night in such a high work-load environment

Data retrieved from NASA's ASRS site 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.