High & Hot

We're high. Really high. I'm not sure just how far above a 3 degree glidepath we are, because Runway 35 at Philadelphia is not served by an ILS or a VASI, and the GPS approach we loaded in the FMS is not displaying a glidepath because we intercepted the approach inside the final approach fix. If any of the above were available, though, I'm pretty sure they would be pegged. This runway is somewhat short and wide, so if we look high, we're really high.

"Flaps 5, speed 140,"

"Flaps 5, speed 140."

We are cooking along over the ground with a 20 knot tailwind. The Delaware River is quickly approaching, and Runway 35's 6500 feet of pavement loom beyond it. My FO, who is flying the airplane, pitches up slightly to get below the maximum speed for Flaps Full. I glance over at him; his facial expression mirrors my own thoughts: this sucks!

"Flaps Full, V-approach, landing check."

"Flaps Full, V-approach, landing check. Gear?"

"Down for 35"

"Down for 35. Flaps Full set, flight attendants notified, cleared to land 35, landing check complete. 1000 feet, instruments, uh...normal?"

We're fully configured now and the airplane is coming down good. I have serious doubts about whether it'll be enough to salvage this screwed up approach. There's plenty of temptation to do so. Philly is busy today, really busy. There's a reason they asked us whether we could accept a visual approach to 35. There was about an 8 knot tailwind component so I had looked up the performance; we were well under the maximum weight for landing on 35 with a 10 knot tailwind. Of course, when we told approach we could do it we had no idea they'd be clearing us for the visual from a 4 mile base at 1500 feet with a speed restriction of 190 knots until turning final and a 20 knot tailwind aloft. Mind you, Piedmont Dash 8's have been making successful visual approaches in those conditions all afternoon. But this is not a turboprop, its a slippery jet whose very low-drag efficiency is proving its Achilles' heel in this moment. It should have been apparent to me from the moment they cleared us for the approach that this was unworkable, and I should have requested a vector back around. But I didn't realize it then, and once committed to a certain course of action, there's a strong mission-completion bias in the heart of most pilots, a natural tendency to play the cards one's been dealt and make it work. There is, however, a point at which the laws of physics win out over the steeliest determination.

We're approaching 500 feet above the ground, the point at which our company requires that we have a visual approach stabilized - that is, configured, on glideslope, and on speed. Time to evaluate how things are going. We're configured, and coming down quickly. That's good. However, we're nearly 30 knots above Vref, and the sight picture is showing that we're still pretty high. Suddenly I'm not flying a NewCo JungleBus into Philly, I'm on a Southwest 737 approaching Runway 8 at Burbank. There are only two ways this can end: in a go-around, or in a gas station across Hollywood Way. The choice is clear.

"Screw this, man, go around."

"Go around, Flaps 4."

"Flaps 4. Philly tower, NewCo 1808 going around.

"Roger, NewCo 1808, fly runway heading, maintain 2000, contact approach 124.5"

"Runway heading two thousand twenty four point five NewCo eighteen oh eight."

"Heading."

"Heading. Here's 2000 set for the missed."

"One thousand for two thousand. Flight level change, speed 210."

"Flight level change...uh, speed 180 for Flaps 4."

"Ah, right, 180. Flaps 3."

"Flaps 3."

"Flaps 2, speed 210."

"Flaps 2, speed 210. Philly approach, NewCo 1808 on the missed from Philly 350 heading and two thousand."

"NewCo 1808 roger."

"Flaps 1."

"Flaps 1."

"Flaps zero...speed FMS."

"Flaps zero, speed FMS."

Well, that wasn't so bad. A whole lot happened in those two or three minutes, and although a go-around is theoretically a routine maneuver, this is only the second one I've done in the JungleBus. You do them often in the sim, however, and that training comes back to you quickly. The passengers, on the other hand, probably aren't quite so familiar with what just happened. Once Philly Approach turns us onto the downwind for an ILS to 27R, I make a PA to the main cabin.

"Ladies and Gentlemen, what you just experienced is known as a go-around. Long story short, air traffic control brought us a little too close in to the airport while we were high and fast, and we weren't able to descend steeply enough. I decided the safest course of action was to climb away for another approach. We're being brought around for another runway now, and I expect that we'll be landing in the next five to ten minutes. Thank you."

When things don't go as planned, it's easy to get thrown off your game. A big part of having a professional pilot mentality is being able to put setbacks behind you and concentrating on the task at hand. I assure myself that the go-around was the right decision, check our fuel state, and set the FMS up for the next approach. There's still a little voice nagging me in the back of my head: you accepted the approach clearance when you shouldn't have. I silence the voice as best as I can for the landing, but it's apparent that I'm still a little distracted: I forget the "Flaps Up, After Landing Check" call upon clearing the runway until my First Officer reminds me.

As the passengers deplane, many of them thank us for getting them there safely, something I've noticed more of in the weeks after Cactus 1549 and Colgan 3407. Our lead flight attendant notes that many of them seemed anxious during the maneuver but relaxed once I made the PA. I realize that while accepting what turned out to be an unflyable approach clearance was a mistake, it wasn't an egregious one like a decision to continue the approach would've been. Risking a monumental blunder in hopes of covering up a minor misjudgement would have been unforgivable. Now that I've made the mistake of accepting a visual approach from 4 miles out at 1500 feet and 190 knots with a tailwind, I won't do it again. Because many others have made the mistake of continuing grossly unstabilized approaches, that's thankfully a lesson I won't have to learn through my own errors.