On the Way Down

VNAV and the stabilized approach

May 1, 2006

We've all heard the rumor that a stabilized approach can lead to a good landing. Great, we can always use more good landings. But what's a stabilized approach and how do I get there?

A great way to get to a stabilized approach is to start with vertical navigation (VNAV) planning. The basic premise of VNAV is pretty simple: Don't hit the ground unless you intend to. And when you do hit the ground, do it in such a way that you and your passengers can walk away unassisted and the airplane can be used again with just minor repairs.

VNAV is basically descent planning. Whether it's planning to arrive at traffic pattern altitude on the downwind, planning to arrive at the glideslope intercept on the ILS on speed and glidepath, or planning to make sure you hit all of the 296 altitude and speed restrictions on the BLAAH 4 arrival into Hooterville International — you're VNAVing.

On the other hand, any VNAV plan that includes the words "I think," "Maybe," "Let's give it a try," or "Watch this" is generally not a good plan and is more like wishful thinking. It's kind of like when they say the lottery drawing this week is worth $100 million, with the short note at the bottom that says you should not play the lottery for investment purposes.

What's a stabilized approach?

So, good VNAV planning leads to a stabilized approach. Great! But what's a stabilized approach?

  1. Don't know, never done one.
  2. My heart rate is 190, but at least it's stabilized there during the approach.
  3. A reasonably safe and stable approach from which a normal landing can be made.

No matter what our experience level is, there is a tremendous temptation to try to save an unstable approach. At times, it can take a lot of discipline to go around. As pilots we generally hate doing that, as we often think we've failed. We hate failing. Plus, in this era of $100-per-gallon fuel, it costs our wallet as well as our pride.

I was flying an approach into San Francisco a while back and got turned in on the ILS extremely high and fast. Hey, I'm a team player, so I did my best to recover it. The only things stable during the approach were the buried glideslope and vertical speed indicators as we approached the outer marker.

I was tempted to continue and save the approach — I knew I could save it. Unfortunately, there are a lot of pilots and airplanes sticking out of hillsides like lawn darts — they knew they could save it too.

Did we continue? We didn't meet any stabilized approach criteria, so we were forced by company policy and just plain good judgment to hit the eject button. Stabilized approach criteria for us meant being no more than a dot high on the glideslope at the outer marker and within 10 knots of final approach speed descending at less than 1,000 feet per minute at either 1,000 feet agl in instrument meteorological conditions or 500 feet agl in visual meteorological conditions.

Although it was air traffic control's fault for turning us in so high, it was our fault, too, for not having a VNAV descent plan that told us right away that we were not going to make it. Unfortunately, our plan consisted of "I think I can save it" rather than "I know I can save it" with a speed and descent rate planned to back it up.

We're not always as great as we think we are

I have a friend who when asked how he is, generally answers, "Practically perfect in every way." Now, after knowing Kelly since junior high school, I realize that he is not really grounded in reality on this particular subject. As pilots, we're like that sometimes when we evaluate our own flying.

Two friends were taking a quick flight south in a twin for some warm-weather beach time. Things were relatively low-key except for the fact that the weather at their destination was not very good and was deteriorating quickly, because of growing afternoon thunderstorms.

As sometimes happens, ATC did not provide a timely descent clearance. So, by the time they started down they were impressively above any reasonable glidepath and their VNAV plan was hopelessly out the window. However, our friends were "can do" guys and the weather was closing in fast. With a hotel room on the beach waiting for them, they were really motivated to get in and land. Taking time for a gentle descent just didn't seem like an option.

So, they "banzai'd" it down and around the weather — a left here, a right there, increase the descent to make it under that cloud, shallow it out to make it over that one. You get the picture. They started the VOR approach really high, but that was OK because they were really fast, too. When they finally got on the glidepath at 500 feet, it was quite a surprise after spending the majority of the last 20 minutes chasing it. Something about a blind squirrel finding an acorn comes to mind. Of course, the landing was a greaser, so it was all good.

They were pretty excited about the saved approach and great landing and expected a standing ovation from their passengers. It didn't occur to them that their passengers might have a different view of the flight. Instead of hearing "great landing" and "Chuck Yeager ain't got nothin' on you," our friends heard comments over dinner about the uncomfortable descent and difficulties in moving around the nearly vertical cabin.

Unfortunately, I've been in either the backseat or the front seat on more than a few of these wild rides. While it could have been a Cessna 310 going to Florida, this was in a Boeing, with just shy of 200 passengers going into Cancun, Mexico. And although passengers are generally pretty nice, the flight attendants will tell you exactly what they think over dinner, and occasionally even draw very graphic pictures for you. Needless to say, dinner and the evening were on the pilots.

So, amigo, how do we make VNAV work for us?

First, it's about knowing what your airplane will do. Some airplanes are pretty slick, while some are the aerodynamic equivalent of a brick wall. Sometimes there are large variations within a single type. With 737s, the -400s come down like a rock, while the -800s with winglets are much slicker (if you can even say that about a 737 with a straight face) and have a much harder time going down and slowing down at the same time.

Are there speed brakes or other drag devices (like landing gear or flaps) to assist in going down or slowing down? What are the gear (assuming it's not welded in place) and flap speed limitations, and how do we slow down to them? How many feet per mile can we descend? Are there engine or structural limitations to worry about? If we're flying a finicky turbocharged piston airplane, we may be limited to a one-inch reduction in manifold pressure per minute.

Second, absent outside guidance (such as our CFI, the FBO we rent from, or the company we fly for) it's about setting and knowing our own personal minimums; i.e., what we are comfortable with as pilot in command. There are several things to consider....

If we are starting our descent above 18,000 feet, we'll have an altitude change as we transition from 29.92 to the local altimeter setting — 10 feet for every one-one-hundredth of an inch on the altimeter. If it's 30.12 on the ground, then we'll have a change of 200 feet going through 18,000 as we reset the altimeters.

What's the wind like? If it's a big tailwind, we may need to start down earlier. The rate we need is a function of our groundspeed throughout the descent, not the indicated or true airspeed. If we've got a huge headwind, we can delay our descent, although we may want to descend earlier if the winds below are more favorable.

Do we have to slow down somewhere during the descent? We know that our groundspeed generally will decrease as we get lower. How quickly can we slow to enter the traffic pattern, and are there other speed limits along the way? How many knots per mile will our airplane slow down — for the jet I usually fly, it's about 10 knots per mile without drag.

Finally, what about our passengers? Some can tolerate several thousand feet a minute; some have trouble with any descent. Generally, when I'm flying a nonpressurized airplane, I plan 500 fpm. Then if I want to go down a little faster, and my passengers are OK with it, great. If not, I'm set. A great flight can turn really unpleasant if your passengers are uncomfortable with the descent. And what about small children? We've all heard them on the airlines, screaming during the descent — and that's with a pressurized cabin descending in the 500-fpm range, regardless of what the airplane's doing.

How about your needs? Me, I got really sick a few years ago and instead of staying in bed, I went flying, skiing, and flying again. Really dumb. Because of that, my ears have gone from 2,000-fpm ears to 500-fpm ears. Bummer.

And we can't assume we're home free if we happen to reside in the high-rent district and fly a pressurized airplane — we can still get ahead of the pressurization system at times. Generally, the risk of that is highest during a short flight with a large airport-elevation change. Like going from Big Bear, California, at more than 6,500 feet msl to Ontario, California, at just more than 900 feet msl in just a few minutes. You could find yourself on the ground near sea level while the cabin is well above you struggling to catch up. And it will catch up — it'll usually dump as soon as it thinks it's on the ground — and that will not be pleasant. I guarantee you'll only do that once.

It all comes down to math

As in so many things aviation, it all comes down to math. Can I afford to fly today? Can I afford to buy that airplane? Can I afford the divorce that may follow the airplane purchase?

The math starts with how many feet you have to lose compared with how many feet you can lose over a given distance. If we need to lose 10,000 feet, going 120 knots over the ground, and want a descent rate around 500 fpm, when do we start down? Five hundred feet per minute into 10,000 feet is 20 minutes; multiply that by my speed of two miles a minute for a result of 40 miles from the airport. We can adjust for a straight in, a downwind arrival at pattern altitude, or a landing or approach from the other direction.

Most ILS glidepaths are set at 3 degrees, which seems to be the norm for descents in aviation. That's about 318 feet per nautical mile. If we're doing 120 knots, that's roughly two miles a minute with a descent rate of 636 feet per minute. And just like that, we have a plan.

Involve air traffic control in the plan. Give controllers a heads-up that you're looking for lower, and often they'll give you a pilot's discretion descent to a lower altitude, letting you descend when you want to. On the other hand, if they ask you to do something different and you're not cool with it, just tell them. They're generally pretty understanding, but don't be unnerved if they give you a vector for controller amusement to allow for your descent. Taking a few turns here and there is a very small price to pay for the comfort of your passengers and/or engine. Besides, we all love to fly, so why are we so often in such a hurry to get back on the ground?

Meep meep

Failure to fly a stabilized approach is a recognized cause of many accidents and incidents. As someone who learned to fly in Southern California, I have spent untold hours at the intersection of runways 8/26 and 15/33 at Burbank's Bob Hope Airport in everything from a Cessna 172 to a Citation X watching airplanes go racing by like the Road Runner trying to land on Runway 8. It seems airplanes always float and touch down late, and their pilots are always hitting the brakes hard to avoid that rather menacing-looking barrier fence at the end of the runway.

I wasn't surprised when I read about the Acme Airlines 737 that left its mark in the airport fence after a high, fast, and unstabilized approach followed by a high, fast, and long landing. That's a bad plan, with a bad result. Good VNAV planning and execution generally lead to a stabilized approach and landing. I say "generally" because as pilots, we can screw up nearly anything, but at least we're setting ourselves up with the best chance for success. And that's all we can ask for.


Marc K. Henegar, of Bend, Oregon, is a Boeing 737 captain.


Going Down in Style

Removing the mystery from vertical navigation planning

By Thomas B. Haines

Pilots flying with newer moving maps — whether handheld or panel mount — often can avoid the math of VNAV calculations by using the vertical navigation functions built into the equipment. Although air traffic control will usually determine when you'll start down while flying IFR, with a VNAV profile in the moving map, you can provide a gentle reminder if ATC gets busy. When flying VFR, VNAV can help you avoid an engine-cooling descent.

Garmin, maker of the most popular panel-mount and handheld GPSs, provides one of the easiest to use VNAV functions. On the Garmin GNS 530, for example, VNAV gets its own button on the front of the unit. Push it and you are presented with a page where you can enter the altitude you would like to descend to — either msl or agl. Another entry allows you to set the distance from a particular waypoint — either your next waypoint or any waypoint in your flight plan — where you would like to arrive at the selected altitude. The point can be a certain number of miles either before or after a waypoint, or by setting zero miles, you'll arrive at the altitude right at the waypoint. You can then choose your desired vertical speed for the descent.

With those entries in place, the system will flash a message — if you choose — to remind you to start the descent. Throughout the descent it will show the vertical speed required (VSR) to arrive at the descent point. Prior to reaching the top of the descent profile, the system shows the current VSR to meet the profile, which might be only a few feet per minute if you're many miles from the top of the descent. As you move closer to that point, the VSR number will increase. Just before it meets your target descent rate — perhaps, 500 fpm — the system will remind you to start down. The display also will show how many minutes it will be until you arrive at the top of the descent point. Match your descent rate to the VSR and you'll arrive right on schedule.

The function can be used not only for final descents, but also to meet ATC-assigned crossing restrictions. If the controller requests that you cross five miles east of XYZ VOR at 5,000 feet, you can program that into the box and know that if you follow the VSR readout you'll always get to the point at just the right altitude.

The VSR readout also can be extremely helpful in an emergency. In the case of a sick engine or even a complete engine failure, if you have an airport as your next waypoint you can check the VSR. If you are able to maintain the displayed VSR or less you know you'll make it to the airport. If you end up descending at a rate greater than that shown, you know you had better prepare for an off-airport landing. Remember that winds can change with altitude. The system will constantly recalculate based on changing conditions, but what looks like a slam dunk at one altitude may look less promising at a different one, so always allow yourself multiple options.

Most of the Garmin handheld models and units from several other manufacturers incorporate similar functionality.


E-mail the author at thomas.haines@aopa.org.


ASF Safety Seminar

The Last Five Miles is an AOPA Air Safety Foundation Safety Seminar, which focuses on techniques for making those last few miles before touchdown both safe and enjoyable. You will learn:

  • Three good reasons to descend early...and one time when it would be a big mistake.
  • When straight-in approaches can be both safe and legal.
  • How to safely handle traffic pattern conflicts.

Visit the Web site for more information.