Narrative:

On jul/xx/89, I was working local control at kennedy (jfk) tower. At XX38:22 local time, I cleared small aircraft X for takeoff off of runway 22R. The kennedy two SID calls for 22R departures to 'fly runway heading for vectors...' and to 'climb and maintain 5000'.' as small aircraft X cleared the end of the runway, I cleared several heavy jet arrivals from 22L across runway 22R and to their ramps. I then instructed small aircraft X to contact departure. At some time prior to XX41:10 UTC, I observed the runway to be clear and the previous departure, small aircraft X to be in a left turn (via the BRITE scope) clear of the departure course. I would estimate that the course flown by small aircraft X diverged approximately 45 degree to the left of runway heading. It is not uncommon for departure control to vector slow moving aircraft left or right of course to accommodate faster moving departures behind. At XX41:10 UTC I cleared air carrier Y for takeoff, exactly 2 minutes and 48 seconds after small aircraft X. While air carrier Y was rolling, small aircraft X called me. I responded to his call twice but he did not answer to my return calls. When air carrier Y was clear of the runway I instructed him to contact departure control. Small aircraft X was not being vectored by departure as I thought. The pilot had not even contacted departure control yet! The pilot had, for reasons unknown, made a left turn-out on his own and then made a right turn to an approximately heading of 310 degree, across the departure path of air carrier Y. At the time that I instructed air carrier Y to contact departure, the radar indicated that small aircraft X was still southbound. At approximately XX43:30 UTC, air carrier Y saw small aircraft X at his 10:30 position, 500' horizontal, same altitude, on a converging course, and took evasive action. The radar printout indicates that small aircraft X also took evasive action. 15 mins afterwards, I was told by departure control that 'small aircraft X cut in front of air carrier Y and air carrier Y was not too pleased'. I did not know that a loss of separation had occurred, much less than that it occurred in my air traffic area. I was informed of the near midair collision report 2 days later. Contributing factors: 1) pilot of small aircraft X did not have low altitude en route charts for our area. We gave him the tower copies just prior to his takeoff! 2) pilot of small aircraft X did not turn on his transponder until approximately 3 minutes after departure. His first transponder return on radar indicates he was at 1700' MSL (and AGL). By comparison, air carrier Y first return showed 400'. 3) pilot of small aircraft X did not contact departure control until approximately 6 minutes after his was instructed to do so. 4) pilot of small aircraft X turned left after departure when SID required runway heading. 5) pilot of small aircraft X leveled off at 1800 MSL when SID required climb to 5000' MSL. 6) pilot of small aircraft X called me at XX42:03 UTC, and I answered at XX42:13 UTC, and again a few seconds later, but got no response. I believe he intended to 'fess up' to being lost or disoriented but thought twice and changed his mind. I believe that this was most likely his biggest mistake, one that almost cost him his life. 7) conflict alert system did not alarm. This could well have been caused by the fact that the computer had only had 5 'radar hits' on small aircraft X prior to his crossing paths with air carrier Y. It usually takes 3 to 4 hits for the computer just to identify and tag a target. Elimination of any one of factors 2 thru 6 would have prevented this incident.

Related reports:

Google
 

Original NASA ASRS Text

Title: A DEPARTING ACR CAME IN CLOSE PROX TO A PREVIOUSLY RELEASED SMA.

Narrative: ON JUL/XX/89, I WAS WORKING LCL CTL AT KENNEDY (JFK) TWR. AT XX38:22 LCL TIME, I CLRED SMA X FOR TKOF OFF OF RWY 22R. THE KENNEDY TWO SID CALLS FOR 22R DEPS TO 'FLY RWY HDG FOR VECTORS...' AND TO 'CLIMB AND MAINTAIN 5000'.' AS SMA X CLRED THE END OF THE RWY, I CLRED SEVERAL HEAVY JET ARRIVALS FROM 22L ACROSS RWY 22R AND TO THEIR RAMPS. I THEN INSTRUCTED SMA X TO CONTACT DEP. AT SOME TIME PRIOR TO XX41:10 UTC, I OBSERVED THE RWY TO BE CLEAR AND THE PREVIOUS DEP, SMA X TO BE IN A LEFT TURN (VIA THE BRITE SCOPE) CLEAR OF THE DEP COURSE. I WOULD ESTIMATE THAT THE COURSE FLOWN BY SMA X DIVERGED APPROX 45 DEG TO THE LEFT OF RWY HDG. IT IS NOT UNCOMMON FOR DEP CTL TO VECTOR SLOW MOVING ACFT LEFT OR RIGHT OF COURSE TO ACCOMMODATE FASTER MOVING DEPS BEHIND. AT XX41:10 UTC I CLRED ACR Y FOR TKOF, EXACTLY 2 MINUTES AND 48 SECONDS AFTER SMA X. WHILE ACR Y WAS ROLLING, SMA X CALLED ME. I RESPONDED TO HIS CALL TWICE BUT HE DID NOT ANSWER TO MY RETURN CALLS. WHEN ACR Y WAS CLEAR OF THE RWY I INSTRUCTED HIM TO CONTACT DEP CTL. SMA X WAS NOT BEING VECTORED BY DEP AS I THOUGHT. THE PLT HAD NOT EVEN CONTACTED DEP CTL YET! THE PLT HAD, FOR REASONS UNKNOWN, MADE A LEFT TURN-OUT ON HIS OWN AND THEN MADE A RIGHT TURN TO AN APPROX HDG OF 310 DEG, ACROSS THE DEP PATH OF ACR Y. AT THE TIME THAT I INSTRUCTED ACR Y TO CONTACT DEP, THE RADAR INDICATED THAT SMA X WAS STILL SBND. AT APPROX XX43:30 UTC, ACR Y SAW SMA X AT HIS 10:30 POSITION, 500' HORIZONTAL, SAME ALT, ON A CONVERGING COURSE, AND TOOK EVASIVE ACTION. THE RADAR PRINTOUT INDICATES THAT SMA X ALSO TOOK EVASIVE ACTION. 15 MINS AFTERWARDS, I WAS TOLD BY DEP CTL THAT 'SMA X CUT IN FRONT OF ACR Y AND ACR Y WAS NOT TOO PLEASED'. I DID NOT KNOW THAT A LOSS OF SEPARATION HAD OCCURRED, MUCH LESS THAN THAT IT OCCURRED IN MY ATA. I WAS INFORMED OF THE NMAC REPORT 2 DAYS LATER. CONTRIBUTING FACTORS: 1) PLT OF SMA X DID NOT HAVE LOW ALT ENRTE CHARTS FOR OUR AREA. WE GAVE HIM THE TWR COPIES JUST PRIOR TO HIS TKOF! 2) PLT OF SMA X DID NOT TURN ON HIS XPONDER UNTIL APPROX 3 MINUTES AFTER DEP. HIS FIRST XPONDER RETURN ON RADAR INDICATES HE WAS AT 1700' MSL (AND AGL). BY COMPARISON, ACR Y FIRST RETURN SHOWED 400'. 3) PLT OF SMA X DID NOT CONTACT DEP CTL UNTIL APPROX 6 MINUTES AFTER HIS WAS INSTRUCTED TO DO SO. 4) PLT OF SMA X TURNED LEFT AFTER DEP WHEN SID REQUIRED RWY HDG. 5) PLT OF SMA X LEVELED OFF AT 1800 MSL WHEN SID REQUIRED CLIMB TO 5000' MSL. 6) PLT OF SMA X CALLED ME AT XX42:03 UTC, AND I ANSWERED AT XX42:13 UTC, AND AGAIN A FEW SECONDS LATER, BUT GOT NO RESPONSE. I BELIEVE HE INTENDED TO 'FESS UP' TO BEING LOST OR DISORIENTED BUT THOUGHT TWICE AND CHANGED HIS MIND. I BELIEVE THAT THIS WAS MOST LIKELY HIS BIGGEST MISTAKE, ONE THAT ALMOST COST HIM HIS LIFE. 7) CONFLICT ALERT SYSTEM DID NOT ALARM. THIS COULD WELL HAVE BEEN CAUSED BY THE FACT THAT THE COMPUTER HAD ONLY HAD 5 'RADAR HITS' ON SMA X PRIOR TO HIS XING PATHS WITH ACR Y. IT USUALLY TAKES 3 TO 4 HITS FOR THE COMPUTER JUST TO IDENTIFY AND TAG A TARGET. ELIMINATION OF ANY ONE OF FACTORS 2 THRU 6 WOULD HAVE PREVENTED THIS INCIDENT.

Data retrieved from NASA's ASRS site as of August 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.