AINsight: A Visual Approach Gone Bad

 - August 26, 2022, 11:57 AM

A visual approach is the first type of approach taught during primary flight training. A well-flown VFR traffic pattern in a light aircraft is benign. At each point, downwind, abeam the numbers, base leg, and turning final, student pilots are taught to fly the appropriate power setting, airspeed, and configuration, all leading to a nice stable uneventful approach and landing. 

In a large turbine-powered aircraft, a visual approach is far less common or predictable. A past Flight Safety Foundation (FSF) study found that 41 percent of all approach and landing accidents began with a visual approach.

For a pilot, the primary goals of a visual approach are to be safe, legal, and try not to embarrass yourself. On occasion, the latter happens. If not properly planned and executed, a visual approach has the potential to lead to complete chaos on the flight deck. 

Case in point, I will share a personal “experience” not only to show my vulnerabilities but to demonstrate that a pilot with nearly 40 years of flying experience will, on occasion, embarrass themself.

A visual approach gone bad will become deeply rooted in your memory. I may not remember what I had for dinner last night, but can recall every exacting detail of this botched visual approach as if it happened yesterday, even though this event occurred almost four years ago.

Here’s the set-up: it was the fourth and final leg of the day from a small Midwest airport to a little larger Midwest airport. The scheduled flight time for this repositioning flight was only 45 minutes, and the weather was CAVU across the entire region. 

I was the pilot flying and had a highly experienced first officer—although he was new to the company—who would be the pilot monitoring (PM). At this point, the sun was rising from the East—the same direction that we were flying. Before top-of-descent, the PM had copied down ATIS, computed the landing data for a reduced flap landing, and had sent the in-range message with the most important canned note: “call hotel van.”

Next, I would thoroughly brief the approach—a “visual backed up by the ILS”—and we discussed the nuances of a reduced flap landing. At lighter weights, a reduced flap landing will provide a Vapp that is more appropriate for a heavy jet (plus it's quieter, more fuel efficient, etc.). Covered were the required items such as selecting the appropriate landing data, corrected Vapp in the FMC, GPWS flap override switch to “ON,” and a discussion about a reduction in drag, which would require us to slow down earlier than normal.

The descent checklist was completed and, as we neared top-of-descent, we could see the airport. We were both familiar with this airport since it’s a regional hub for our operation.

Passing through 16,000 feet, the en route center controller handed us off to approach control. The approach controller then cleared us direct to the final approach fix (FAF) and gave us a further descent clearance to 3,000 feet (2,500 feet was the FAF altitude). 

At this point, the aircraft was in the clean configuration at 250 knots, and we were on a dog leg to the final approach course. We seemed to be squared away.

With the runway in sight and sensing a higher workload for the controller (he was also working a combined tower and ground frequency) and a desire to continue the descent to 2,500 feet, we both agreed that we could fly the visual approach. The PM requested it and the controller obliged and cleared us for the visual approach. 

This is where it started to go bad. Although we were established “on path” to cross the FAF at 2,500 feet, I was fast—way fast. As I extended the speed brakes, my inner “John Wayne” kicked in and without further evaluation, I blurted out “autopilot off, autothrottles off” and flipped the flight director switch to flight path vector. Safety “experts” say there is an “an overreliance on automation” and “not enough hand flying”—that’s what was going on in the back of my head.

In retrospect, this was a horrible idea. Not only did I increase my workload, but I loaded up the PM by now having him manipulate the speeds on the mode control panel, in addition to active monitoring duties, configuring the aircraft, setting the missed approach altitude, and running the landing checklist. 

Through some miracle, a flurry of commands and arms flailing about the cockpit, we were able to configure the aircraft and slow a bit. Approaching the FAF at about 1,500 feet above field elevation, the aircraft was on path and the speed was slowing through Vapp+40.

Our company’s stabilized approach criteria called for no more than Vapp+10 at 1,000 feet above field elevation. At this point, I realized that the culprit was not only poor situational awareness on my part, but a strong southwesterly tailwind aloft. 

Not at all pleased with my work, I made the decision to discontinue the approach, which with our aircraft is a nice, controlled procedure that is essentially a very soft go-around. 

After my FO confessed my sins to the tower controller, he then cleared us to enter left traffic for a VFR pattern to the same runway. At this point, still hand-flying, I reverted to my training of the proper configuration and speeds on downwind, abeam the numbers, base, and final. The result was a decent landing and an overworked first officer. (I did apologize—a lot.) 

During the ride to the hotel, the first officer and I unpacked the botched approach. In addition to the tailwind aloft and poor situational awareness, a direct clearance to the FAF reduced our track mileage to the runway—this hastened the deceleration portion of the descent. These are all common gotchas when attempting to fly a visual approach.

In addition, we discussed whether this visual approach was safe and legal. While there were several errors made during the descent and approach, the decision to discontinue the approach (as backed by a no-fault go-around policy) trapped most of the errors and prevented them from becoming an undesired aircraft state. So, yes, we were safe since we chose not to continue an unstabilized approach below 1,000 feet above field elevation.

According to the definition of a visual approach, we were completely legal. The approach was authorized by ATC, the airport was VFR, and we always kept the airport in sight during the approach and remained clear of clouds. 

In hindsight, we were completely safe and legal, but I was completely embarrassed. If I had to do it over again, I would have considered—after a long day—getting vectors to an ILS approach or squaring the turn to final versus a slight dogleg.

Most importantly, I should have relied more on automation by keeping the autopilot and autothrottles engaged, contrary to what the experts say. By disconnecting the automation, I increased the workload on the PM not only during the approach but also during the discontinued approach and subsequent VFR traffic pattern.

FSF has produced a series of informative approach and landing accident reduction briefings notes, including one on visual approaches. This is a great resource and should be helpful in guiding any future discussions on visual approaches.

The opinions expressed in this column are those of the author and not necessarily endorsed by AIN Media Group.