Narrative:

Kennedy 1 departure; canarsie climb off of runway 31L. The FMS was loaded correctly. It showed direct cri; then D176B which is along the cri 176 degree radial outbound per the departure procedure. This fix is only 2 miles past cri. What happens in the FMS/FGC autopilot will fly direct cri; then initially make the turn on the 176 degree radial; but then almost instantly 'smart turn' to the left (off course) toward coate intersection which is the next fix on the route. At this point in time; without delay; we hand flew the aircraft; switched to conventional navigation and flew the aircraft to the correct route. All at the same time; the ATC controller noticed we were trying to fix this. He had us turn to a heading for radar vectors. You must understand that this all happened within 3 seconds of crossing cri with the FMS. Also contributing to the FMS's quick reactions; in my opinion; would be we had a 40 KT tailwind crossing cri turning outbound. A couple things would (may) fix this. 1) leaving in a 'no link' may help and prevent the FMS from 'smart turning' to coate intersection. 2) having the FMS database fixed to have an extended fix outbound the cri 176 degree radial. Like many other departures; the 2 miles 176 degree radial fix is there to show compliance with the crossing altitude. However; there needs to be another fix outbound until ATC vectors the flight off the cri climb. 3) only fly the departure with conventional navigation; even though the procedure is in the database. I know this procedure is a 'hot topic.' for that reason we had the conventional navigation all set in case the FMS did something we did not like. We are both experienced and have flown out of jfk many times (I was based there). I know this is important. Besides filing this report in case ATC files a deviation; I truly believe this is more important to be fixed in the database; or to fly this without the FMS guidance. Please call with questions. I fully believe the database needs to be updated also. Callback conversation with reporter revealed the following information: reporter advised the aircraft is equipped with dual FMS systems that must be programmed independently. He believes the captain (the pilot flying) most likely closed the route discontinuity ('no link') between the crossing fix and coate and the error went undetected when the route check was performed. Doing so would result in exactly the track described in the event. Reporter further stated he had flown the SID on numerous occasions and this was the only time this happened. Reporter believes company training puts too much emphasis on eliminating route discontinuities (no links) and not enough on recognizing when they are in the original coding to comply with a procedure which includes a 'conditional' aspect. In this case; the cri 176R is to be flown until a vector to the enroute fix is provided -- ergo; the time/distance spent on the track is conditional on when the vector is provided by ATC.

Google
 

Original NASA ASRS Text

Title: E145 FMS on the KENNEDY ONE SID; CANARSIE climb attempted to proceed from the CRI 176/2 crossing restriction fix directly to COATE intersection vice flying the 176 radial for vectors per the SID.

Narrative: Kennedy 1 Departure; CANARSIE climb off of Runway 31L. The FMS was loaded correctly. It showed direct CRI; then D176B which is along the CRI 176 degree radial outbound per the departure procedure. This fix is only 2 miles past CRI. What happens in the FMS/FGC autopilot will fly direct CRI; then initially make the turn on the 176 degree radial; but then almost instantly 'smart turn' to the left (off course) toward COATE Intersection which is the next fix on the route. At this point in time; without delay; we hand flew the aircraft; switched to conventional navigation and flew the aircraft to the correct route. All at the same time; the ATC Controller noticed we were trying to fix this. He had us turn to a heading for radar vectors. You must understand that this all happened within 3 seconds of crossing CRI with the FMS. Also contributing to the FMS's quick reactions; in my opinion; would be we had a 40 KT tailwind crossing CRI turning outbound. A couple things would (may) fix this. 1) Leaving in a 'no link' may help and prevent the FMS from 'smart turning' to COATE Intersection. 2) Having the FMS database fixed to have an extended fix outbound the CRI 176 degree radial. Like many other departures; the 2 miles 176 degree radial fix is there to show compliance with the crossing altitude. However; there needs to be another fix outbound until ATC vectors the flight off the CRI climb. 3) Only fly the departure with conventional navigation; even though the procedure is in the database. I know this procedure is a 'hot topic.' For that reason we had the conventional navigation all set in case the FMS did something we did not like. We are both experienced and have flown out of JFK many times (I was based there). I know this is important. Besides filing this report in case ATC files a deviation; I truly believe this is more important to be fixed in the database; or to fly this without the FMS guidance. Please call with questions. I fully believe the database needs to be updated also. Callback conversation with Reporter revealed the following information: Reporter advised the aircraft is equipped with dual FMS systems that must be programmed independently. He believes the Captain (the pilot flying) most likely closed the route discontinuity ('no link') between the crossing fix and COATE and the error went undetected when the route check was performed. Doing so would result in exactly the track described in the event. Reporter further stated he had flown the SID on numerous occasions and this was the only time this happened. Reporter believes company training puts too much emphasis on eliminating route discontinuities (no links) and not enough on recognizing when they are in the original coding to comply with a procedure which includes a 'conditional' aspect. In this case; the CRI 176R is to be flown until a vector to the enroute fix is provided -- ergo; the time/distance spent on the track is conditional on when the vector is provided by ATC.

Data retrieved from NASA's ASRS site as of April 2012 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.