NTSB Releases Preliminary Report On Cirrus/Swearingen Midair Collision


The NTSB has released its five-page preliminary report on the May 12 midair collision between two aircraft approaching busy Denver-area Centennial Airport (APA). No one was injured, despite the Cirrus SR22 colliding with a Swearingen SA226TC cargo plane with serious damage to both aircraft (the Swearingen landed safely—the Cirrus pilot pulled the CAPS whole-airplane parachute). Both were receiving ATC instructions for landings on the airport’s north-south parallel Runways 17L and 17R.

To recap, the Cirrus departed APA on a local flight to the west at around 9:21 a.m. local time. The Swearingen took off from Colorado’s Salida Airport (ANK) about 9:56 for a half-hour positioning flight to APA.

The report shows that at 10:22:43, 70 seconds before the collision, the cargo plane was established on a 5.5-mile straight-in approach to Runway 17L at APA, and the Cirrus was northbound, flying a right downwind leg for Runway 17R. From there, the NTSB report states, the Cirrus “continued the right-hand traffic pattern through the base leg, and then began to turn toward the final approach course for the runway. The airplanes collided at 10:23:53 when they were about 3.2 nm from APA. [The Swearingen] was aligned with runway 17L while [the Cirrus] was turning from base to final and heading about 146 degrees when the collision occurred.”

The ADS-B track illustrated in the report shows the Cirrus’s base leg took it east, through the extended centerline of its intended runway and also through the extended centerline of Runway 17L to which the Swearingen was lined up for landing. Citing a 146-degree heading at the time of the collision, the NTSB indicates that the Cirrus had yet to complete its turn from base to final for Runway 17R.

The report does not cite the altitude at which the collision occurred nor prevailing winds near the site of the incident. It also does not include information on ATC communications regarding either pilot reporting visual contact with the other aircraft.

Mark Phelps
Mark Phelps is a senior editor at AVweb. He is an instrument rated private pilot and former owner of a Grumman American AA1B and a V-tail Bonanza.

Other AVwebflash Articles


    • Well .. some people were adamant the Metroliner pilot should have seen the Cirrus.

      My question in response is ‘Where was the Cirrus when the Metroliner was stable on approach and 3.2NM from the runway?

      Given that the Cirrus was fast wonder if the Metroliner pilot could see the Cirrus at all when it was turning final. (Perhaps could see it on downwind leg.)

      Tip: Very difficult to see behind most airplanes. (Exceptions include observation aircraft and fighter aircraft.)

        • That comment was intended for the post that asked what the Cirrus pilot was doing flying fast etc.

      • Oh, return to base Keith. That the Cirrus hit the Metroliner might well mean it was visible from the Metroliner later than I blathered.

        Need position information, the preliminary ‘report’ only provides path information, does in effect show point of impact (where paths crossed), and does say “About 1022:43, N280KL was about 5.5 nm from APA and had completed a right turn to align with the
        final approach course to runway 17L. At this same time, N416DJ was on the downwind leg of the righthand traffic pattern for runway 17R just before he commenced a right turn to the base leg of the traffic pattern.”

  1. We didn’t learn much from that report that we didn’t already know. The only open question for me was what was the cirrus pilot doing flying so fast in the pattern and overshooting the runway centerline so badly?

    • Poor driving.

      It’s OK to drive poorly if you have a parachute, you are safe.


  2. These simultaneous parallel runway situations have always made me a bit apprehensive so I habitually make it a point to end up a little off to the “safe” side of centerline and then ease it in at a shallow angle rather than going for that crisp base-onto-perfect-centerline turn.

    • One renowned pilot used to fly a bit off centreline of airways once accurate GPS was in use.

      In your scenario you stay away from the other runway.

    • KSNA (Orange country – John Wayne) has this very requirement, as part of the VFR procedures to avoid this collision risk. I’ve always been under the impression that all airports with parallel runways close to each other would have this requirement. It’s just common sense.

      Several years ago, a friend watching a video recording of my landing into KSNA commented that I was not properly aligned with Runway 20L for the landing – subtly insinuating poor airmanship 😉 – I calmly said: Open up the chart supplement and look under “Remarks” where it says: “VFR Acft: To avoid overflights of Rwy 20R: Rwy 20L arr fly final at 15 deg angle to rwy….”

  3. The “very closely spaced” parallel runway configuration at APA (700 feet apart – the minimum for VFR parallel runways) is called out in this FAA “Hazards and Hot Spots” safety video. The video cautions pilots about overshooting the turn from base to final and conflicting with traffic approaching the parallel runway on close-in patterns, let alone well over 3 miles out.

    • Which means the Cirrus pilot didn’t check the available airport information prior to his flight.
      And failed to fly his aircraft at appropriate approach speeds for his aircraft. I think the Cirrus people tend to fly a little faster for their approaches after the crash Huston when the girl stalled in the pattern.

  4. Considering the heading of the Cirrus at impact, 146deg or just past a 45deg angle to the runway heading, and the excessive speed of the Cirrus indicates that the Cirrus was approaching from behind where the metro pilot would even have a chance to have visual…ie…in a blind spot. I’ll expect a few “I plead the 5th” on this one!

    • Agree.

      Plus the ML pilot’s job was to safely land his plane, which he did. Short final I’m focused on my landing. No one expects to be run over in that phase of flight.

      The ML had the mid top of his ship torn off. That could only occur if hit from the side.

      The pilot was looking forward, which is what we would expect on short final.

  5. Seeing several comments that the Cirrus was flying fast, I looked in the preliminary, and I don’t see that. That’s an unwarranted assumption. The Cirrus pretty clearly failed to turn in soon enough to line up with the centerline of 17R, but that doesn’t mean he was flying faster than he should have been.

    For myself, when I’ve landed at either APA or BJC, I’m very wary of the traffic on the other parallel. They’re close together. 700’ isn’t much, so that it doesn’t take much drifting off centerline to get too close. I can recall landing at BJC one time, wind out of the west as usual, already cleared to land on 30R, when the Bonanza aligning for 30L was asked if he saw me—tower could see that he was drifting toward me, as could I. He said he could not, yet he was right beside me. I just slowed down and let him get ahead of me as he drifted almost to the centerline of 30R before correcting.

    I wouldn’t call that Bonanza pilot incompetent or stupid or irresponsible, just that he wasn’t flying precisely. Nor would I call the Cirrus pilot incompetent, stupid, or irresponsible. I’ve made too many mistakes in my almost 5 decades of flying to use prejorative adjectives about others. I’ve instructed good students and poor students, and I’ve flown with good pilots and poor pilots, with skill levels and judgment ranging from excellent to nearly non-existent. Even the best make mistakes.

  6. Given the fact the Cirrus was in a right bank, maybe even a hard right bank, he would not be able to see the Swearington off to his left. Also, for a VFR traffic pattern, I think a base turn 3.5 NM out makes it harder to tell if you’re on the runway centerline. Keeping a tight traffic pattern would have made a drift to the left much easier to see.

  7. This AvWeb article has very little new information. The NTSB also has very little new information, but in the NTSB way, it says a bit more. It lays out and defines the final 10 seconds of the flights before impact. That is important.

    10 seconds is enough time to avoid an accident.

    The Cirrus obviously overshot the centerline of both runways and may have been on autopilot as others have pointed out.
    A key point that hasn’t been mentioned yet, is what the controllers were doing in these 10 seconds to maintain separation. We heard their voices, but don’t know much else of their participation. Separation is ultimately the responsibility of the pilots of course, but ATC is also there to maintain it.
    I think the final report will also go into a lot more detail on the controllers roles along with the pilots roles. I recommend reading this prelim report closely. (again, it doesn’t say much) The only names mentioned are those of the investigative trio.

    The testimony’s of those on board the aircrafts and of the controllers will be the key to the findings of the final report.

    Again, 10 seconds is enough time to avoid an accident. It didn’t in this case. Distractions can kill. They didn’t this time, but they might on any next flight! Departing and approaching airports, especially busy ones, are probably the most critical phase of all flights. We should all know this, but in this case one or more of the individuals involved didn’t practice it.

    I think that all of us will be awaiting the final report on this one!

  8. Honest to goodness, why do we do this? Why do we try to defend stupid.

    * The Cargo plane was *established and cleared* to land.
    * The Cargo pilot was sitting in the left seat so ANY right traffic would be hard to see and oh…

    He was straight in so he’s mainly scanning ahead,

    *The Cirrus pilot was fast in the pattern.
    *The Cirrus pilot overshot his own runway
    *The Cirrus pilot overshot the cargo plane’s runway, because

    He ran into the plane..

    * The Cirrus pilot f’ed up because he failed the most basic rule, situational awareness, fly the plane.

    Why defend it as pilots. Yes, we make mistakes and boy 99% they don’t involve slamming into other planes (or the ground) but when pilots *do* F up, let’s just acknowledge it, learn from it and stop trying to posit why an established airplane on final cleared to land has to F’ing worry about a plane about to slam into his side.

    The NTSB and the FAA will pass judgement, the insurance company will too, but since this pilot lived and if he gets to keep his ticket let us hope he learned some basic lessons:

    * Situational awareness will save your life
    * Proper traffic pattern speeds will save your life
    * Communications will save your life
    * and the day you don’t have a parachute and you ignore the top three, you’re life may not be saved.

    • All true.
      I also watched Paul’s great video on control towers on June 6th.
      Regardless of final responsibility (the pilot’s) a controller might have been able to prevent the accident in those final 10 seconds, even though the Cirrus obviously F’ed up on speed, position, and separation control and had final responsibility for separation. I believe this will be mentioned in the NTSB final.

      I agree with you. Just saying.