Narrative:

We were on an extended final approach to runway 26 at phx, and had been cleared to 'maintain 4000 ft until 9 DME on the localizer, cleared for the visual approach to runway 26.' I was manually flying the aircraft, as all 3 autoplts were inoperative (right and C on dpi/MEL, left failed en route). After leveling at 4000 ft we were advised of traffic,'...regional jet off your left is going to runway 25L.' we saw a crj on a left base leg at our 10 O'clock position at approximately our altitude. I remarked to the first officer that it looked like it would work as long as the crj didn't overshoot his turn to final. As we approached 9 DME, the crj had started his turn to final, but had not yet rolled out of his turn. I realized that the crj had overshot his turn and I started climbing at the same time that the TCASII issued the aural RA, 'climb, climb, climb.' I climbed more aggressively and watched the crj pass underneath us in his turn, and then out of view. TCASII then issued the aural RA, 'monitor vertical speed, monitor vertical speed,' and the ivsi showed a red arc prohibiting any descent. I leveled off at about 4300 ft, but couldn't see the crj so I started an s-turn to the right. The TCASII annunciated, 'clear of conflict,' and shortly we saw the rj well ahead of and below us, intercepting the final for runway 25L. I then commenced a descent and turned back towards the runway, landing without further incident on runway 26. The first officer told the controller we had climbed and were s-turning due to an RA caused by the rj overshooting his turn to final. Had we not climbed at all, a collision may have occurred. I don't know if the crj ever saw us. The TCASII was certainly helpful. Although I had just started climbing based on my visual assessment, the closure rate of the crj was rapid and the TCASII aural warning reinforced my decision and added more urgency. Had the TCASII not given the 'monitor vertical speed' aural and visual, immediately followed by the 'clear of conflict,' I would have followed my initial climb with a go around, since we lost visual contact with the crj when it passed beneath us. The controller's first advisory about the crj was also helpful, as we hadn't seen it before he mentioned it. I'm not sure if the autoplt being inoperative was a factor. My workload was higher, and it's possible that if the autoplt had been engaged that I could have been watching the crj for a higher percentage of time and made an earlier assessment of his trajectory. As it was, I remember seeing him in his turn, not being concerned, checking the instrument panel, and then looking back at the crj and becoming immediately concerned. That momentary panel scan may have cost a second or two in reaction time, but may have been offset by not losing any reaction time in the process of disengaging/overpowering the autoplt had it been engaged. The crj was never on the same approach or tower frequency as we were. We were not told that he was maintaining visual separation with us, nor were we instructed to maintain visual separation with him. I would speculate that he was told about us, and cleared for a visual approach from his base leg. I believe that the sun would have been directly in the crj pilot's eyes during their turn to final, significantly impairing their ability to see the runway they were turning towards, as well as us. I believe that when parallel approachs are being conducted, aircraft should be vectored to a heading that allows some distance at a 30-45 degree intercept angle to final, rather than a heading that requires or allows a 90 degree turn to final, as is common at phx. I can attest from personal experience that the visual cues available during a 90 degree turn to final are not always adequate to prevent an overshoot.

Related reports:

Google
 

Original NASA ASRS Text

Title: NMAC BTWN A B757 ON FINAL FOR VISUAL TO RWY 26 AND A CRJ IN A TURN FOR A VISUAL APCH TO RWY 25L, 9 NM E OF PHX, AZ.

Narrative: WE WERE ON AN EXTENDED FINAL APCH TO RWY 26 AT PHX, AND HAD BEEN CLRED TO 'MAINTAIN 4000 FT UNTIL 9 DME ON THE LOC, CLRED FOR THE VISUAL APCH TO RWY 26.' I WAS MANUALLY FLYING THE ACFT, AS ALL 3 AUTOPLTS WERE INOP (R AND C ON DPI/MEL, L FAILED ENRTE). AFTER LEVELING AT 4000 FT WE WERE ADVISED OF TFC,'...REGIONAL JET OFF YOUR L IS GOING TO RWY 25L.' WE SAW A CRJ ON A L BASE LEG AT OUR 10 O'CLOCK POS AT APPROX OUR ALT. I REMARKED TO THE FO THAT IT LOOKED LIKE IT WOULD WORK AS LONG AS THE CRJ DIDN'T OVERSHOOT HIS TURN TO FINAL. AS WE APCHED 9 DME, THE CRJ HAD STARTED HIS TURN TO FINAL, BUT HAD NOT YET ROLLED OUT OF HIS TURN. I REALIZED THAT THE CRJ HAD OVERSHOT HIS TURN AND I STARTED CLBING AT THE SAME TIME THAT THE TCASII ISSUED THE AURAL RA, 'CLB, CLB, CLB.' I CLBED MORE AGGRESSIVELY AND WATCHED THE CRJ PASS UNDERNEATH US IN HIS TURN, AND THEN OUT OF VIEW. TCASII THEN ISSUED THE AURAL RA, 'MONITOR VERT SPD, MONITOR VERT SPD,' AND THE IVSI SHOWED A RED ARC PROHIBITING ANY DSCNT. I LEVELED OFF AT ABOUT 4300 FT, BUT COULDN'T SEE THE CRJ SO I STARTED AN S-TURN TO THE R. THE TCASII ANNUNCIATED, 'CLR OF CONFLICT,' AND SHORTLY WE SAW THE RJ WELL AHEAD OF AND BELOW US, INTERCEPTING THE FINAL FOR RWY 25L. I THEN COMMENCED A DSCNT AND TURNED BACK TOWARDS THE RWY, LNDG WITHOUT FURTHER INCIDENT ON RWY 26. THE FO TOLD THE CTLR WE HAD CLBED AND WERE S-TURNING DUE TO AN RA CAUSED BY THE RJ OVERSHOOTING HIS TURN TO FINAL. HAD WE NOT CLBED AT ALL, A COLLISION MAY HAVE OCCURRED. I DON'T KNOW IF THE CRJ EVER SAW US. THE TCASII WAS CERTAINLY HELPFUL. ALTHOUGH I HAD JUST STARTED CLBING BASED ON MY VISUAL ASSESSMENT, THE CLOSURE RATE OF THE CRJ WAS RAPID AND THE TCASII AURAL WARNING REINFORCED MY DECISION AND ADDED MORE URGENCY. HAD THE TCASII NOT GIVEN THE 'MONITOR VERT SPD' AURAL AND VISUAL, IMMEDIATELY FOLLOWED BY THE 'CLR OF CONFLICT,' I WOULD HAVE FOLLOWED MY INITIAL CLB WITH A GAR, SINCE WE LOST VISUAL CONTACT WITH THE CRJ WHEN IT PASSED BENEATH US. THE CTLR'S FIRST ADVISORY ABOUT THE CRJ WAS ALSO HELPFUL, AS WE HADN'T SEEN IT BEFORE HE MENTIONED IT. I'M NOT SURE IF THE AUTOPLT BEING INOP WAS A FACTOR. MY WORKLOAD WAS HIGHER, AND IT'S POSSIBLE THAT IF THE AUTOPLT HAD BEEN ENGAGED THAT I COULD HAVE BEEN WATCHING THE CRJ FOR A HIGHER PERCENTAGE OF TIME AND MADE AN EARLIER ASSESSMENT OF HIS TRAJECTORY. AS IT WAS, I REMEMBER SEEING HIM IN HIS TURN, NOT BEING CONCERNED, CHKING THE INST PANEL, AND THEN LOOKING BACK AT THE CRJ AND BECOMING IMMEDIATELY CONCERNED. THAT MOMENTARY PANEL SCAN MAY HAVE COST A SECOND OR TWO IN REACTION TIME, BUT MAY HAVE BEEN OFFSET BY NOT LOSING ANY REACTION TIME IN THE PROCESS OF DISENGAGING/OVERPOWERING THE AUTOPLT HAD IT BEEN ENGAGED. THE CRJ WAS NEVER ON THE SAME APCH OR TWR FREQ AS WE WERE. WE WERE NOT TOLD THAT HE WAS MAINTAINING VISUAL SEPARATION WITH US, NOR WERE WE INSTRUCTED TO MAINTAIN VISUAL SEPARATION WITH HIM. I WOULD SPECULATE THAT HE WAS TOLD ABOUT US, AND CLRED FOR A VISUAL APCH FROM HIS BASE LEG. I BELIEVE THAT THE SUN WOULD HAVE BEEN DIRECTLY IN THE CRJ PLT'S EYES DURING THEIR TURN TO FINAL, SIGNIFICANTLY IMPAIRING THEIR ABILITY TO SEE THE RWY THEY WERE TURNING TOWARDS, AS WELL AS US. I BELIEVE THAT WHEN PARALLEL APCHS ARE BEING CONDUCTED, ACFT SHOULD BE VECTORED TO A HEADING THAT ALLOWS SOME DISTANCE AT A 30-45 DEG INTERCEPT ANGLE TO FINAL, RATHER THAN A HEADING THAT REQUIRES OR ALLOWS A 90 DEG TURN TO FINAL, AS IS COMMON AT PHX. I CAN ATTEST FROM PERSONAL EXPERIENCE THAT THE VISUAL CUES AVAILABLE DURING A 90 DEG TURN TO FINAL ARE NOT ALWAYS ADEQUATE TO PREVENT AN OVERSHOOT.

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.