Pilots are taught to “stay ahead of the airplane,” but there’s relatively little guidance on exactly what that means and how to do it. Years ago, a high-time CFII asked me if I knew the most important thing in instrument flying. I don’t recall my answer, but his was brilliant.
He said the most important thing was the “next two things.”
Often when I ask pilots flying technically advanced aircraft (TAA) to tell me the next two things they need to do, they almost never mention the need to push any buttons. Instead, I get generalizations about descending and turning, and so on.
If you’re not already a subscriber, what are you waiting for? Subscribe today to get the issue as soon as it is released in either Print or Digital formats.
But to make those things happen, one has to push the right buttons. Figuring out which buttons to push in advance can simplify a high-workload situation such as a missed
approach. Part of the problem is prioritization.
We’re often faced with multiple tasks that need our attention, and figuring out which ones are most important can be challenging. I teach pilots that generally the three highest priorities are rolling out on headings, leveling off at altitudes, and intercepting final approach courses. When these things are happening, you need to focus on them, and other tasks can usually wait.
- READ MORE: Exploring the Corner Case Approach for a Pilot
- READ MORE: What You Need to Know to Create an IFR Approach
The highest workload usually occurs toward the end of a flight, when flying an instrument approach or approaching a traffic pattern. A good practice is to try to move some of that workload earlier in the flight by doing setup tasks as soon as possible. But working ahead can sometimes lead to errors and automation surprises. Here are some that I’ve recently encountered.
A lot of pilots are confused about when to “activate an approach.” If you’re flying to an initial approach fix (IAF), there’s nothing you need to activate. Once you’re direct to the IAF, autosequencing of waypoints should occur without any further action from a pilot. However, if you’ve loaded an approach with vectors, typically you’ll later need to “activate vectors to final.”
Activating vectors to final draws a 30 nm line that extends out from the final approach fix (FAF). Once that’s been done, an autopilot can couple to the final approach course. However, you don’t want to activate vectors to final until you’re finished using your en route flight plan. As soon as a controller starts issuing headings to vector you to final, you’re done using the en route flight plan, and it’s a good time to activate vectors to final.
Last year, I was flying with a client in his Cirrus Vision Jet on an arrival to a Southern California airport, and we had been cleared to “descend via.” He had used the autopilot’s VNV key to select the vertical navigation mode. The autopilot was automatically flying the stepdowns on the arrival, and life was grand.
During the descent, the pilot began working ahead and loaded the ILS, which made sense. However, I heard him quietly say, “and now I need to activate the approach.” Before I could stop him, he had activated vectors to final, and life suddenly got complicated. Activating vectors to final caused the HSI’s course pointer needle to change from magenta (for GPS) to green (for the localizer), and the autopilot reverted to ROL mode (for wings level), and we were no longer following the arrival. Worse, vertical navigation mode was no longer active, as the autopilot had reverted to pitch mode.
![Avoid common automation surprises by focusing on critical flight priorities during complex instrument approaches. [Credit: Max Trescott]](https://www.flyingmag.com/wp-content/uploads/2026/09/Automation-Surprise.jpg?w=1024)
We were now descending at greater than 1,000 fpm over the San Gabriel Mountains, which rise as high as 8,000 feet. We no longer knew which waypoints we were between on the arrival and whether we were at a safe altitude.
After thinking about this for a moment, I pushed the autopilot’s ALT key, which caused us to immediately level off somewhere around 9,000 feet. I then called SoCal approach and explained that we accidentally “blew out our programming” and asked whether we were safe to fly at the present heading and altitude. The controller replied that we were safe, and then the hard work began.
Having good knowledge of your avionics and how to accomplish a task in more than one way is a valuable skill. That’s why I recommend that pilots use a simulator and spend hours trying out different ways to accomplish a task. More importantly, they need to know how to undo a task if they’ve started down the wrong rabbit hole.
In this case, I tried three different ways to get the arrival back before I succeeded. I don’t remember the exact sequence I tried, or which one finally won the day, but I believe they included:
» Activating the leg of the arrival we thought we were on.
» Reloading the arrival procedure using the PROC key.
» Reloading the entire flight plan.
Ultimately, we got back on the arrival within a minute or two. But it was a bit of a scramble to fix things.
More recently, I was working with a Vision Jet pilot who had just taken delivery of his airplane but hadn’t finished his type rating. We had previously talked about ways to reduce the workload when flying an approach with a series of stepdowns. Using the VNV key for vertical navigation is usually the easiest way to do that, but sometimes that mode won’t do the last one or two stepdowns.
I had pointed out that flying stepdowns manually is a lot of work using the autopilot, or hand-flying with the flight director, as each stepdown requires five separate actions. They are:
» Setting the ALT selector to a lower altitude.
» Pushing the VS (vertical speed) key.
» Setting the descent rate.
» Reducing the thrust lever or throttle.
» Increasing the thrust lever or throttle as the aircraft levels off.
I had pointed out that one way to work ahead is to preset the ALT selector to the next altitude, but that could only be done after the autopilot had leveled off at the prior altitude and was in ALT hold mode.
While approaching the pilot’s home airport, the approach controller gave us a series of stepdowns. One of them was from 6,000 feet to 5,000 feet. I noted that the pilot properly set the autopilot to level at 5,000 feet, and then I went back to reviewing a chart for the approach, as I hadn’t flown that one in years.
What I didn’t notice was that as we were passing through about 5,200 feet the pilot changed the ALT selector to 4,300 feet, which would be the next altitude after we joined the approach. But my mistake was bigger than his. I failed to monitor the altitude as we approached 5,000 feet and verify that the aircraft leveled at that altitude. Instead, I relied on having seen the autopilot set correctly. I didn’t anticipate that it would be changed later.
As I said earlier, leveling off at altitude is one of the top three priorities of an instrument pilot, and it takes precedence over reviewing a chart.
We subsequently flew below 5,000 feet and were called by ATC. We then climbed back up to 5,000 feet. We were never in danger, as we were in VMC, but a similar mistake in IMC could be catastrophic.
You may be wondering how I know now that the ALT selector was changed to 4,300 feet as we were passing through about 5,200 feet, if I didn’t notice that change when it occurred. It turns out that in some aircraft, ADS-B Out reports the altitude at which the ALT selector is set. After reviewing the ADS-B data for that flight, I could see when the ALT selector was set to 5,000 feet, and when it later changed to 4,300 feet.
Working ahead on the “next two things” and doing them as early as possible is a good strategy. But it doesn’t change the more important priorities of verifying that the autopilot rolls out on the desired heading, levels off at the selected altitude, and intercepts and joins the final approach course.
This column first appeared in the August Issue 973 of the FLYING print edition.
