LOC Recoveries

The put your head down and close your eyes method ain't working. It's time to implement formal, realistic training and testing to prevent loss of control.


Unusual attitude recovery is a standard part of the FAA checkride for nearly every rating, plus flight reviews and instrument proficiency checks. Pilots dutifully don a hood, put their heads down as an instructor or examiner puts the airplane through a few gyrations and then says, “You’ve got it.” The pilot looks up to see that the airplane is either in a nose-low or nose-high bank and usually makes a power change to help control speed, levels the wings and returns the nose to the horizon before something breaks. A check is placed in the unusual attitudes box and the pilot and instructor move on to other tasks.

Unfortunately, every year a few of the pilots who went through that drill within the preceding 24 months lose control of their airplanes in flight. They either tear the wings off in a diving spiral, hit the ground in a diving spiral before the wings come off, stall and fail to return to controlled flight before hitting the ground. Otherwise, they demonstrate they cannot return the airplane to controlled flight prior to ground impact following an autopilot malfunction, flight control system malfunction or upset due to environmental factors. With loss of control one of the biggest single causes of accidents, it seems to me that it’s time to beef up unusual attitude training into something that might help pilots help themselves when the airplane is doing something weird and the overwhelming thought in the pilot’s head is all “OhmigawdI’mgonnadieandit’sgonnahurt.” I think it’s time for unusual attitude recovery training to morph into loss-of-control (LOC) recovery training at all levels of initial and recurrent training.

Just A Different Name?

Definitions are important—loss-of-control recovery training is more than the unusual attitude recovery training I just described for several reasons. While it encompasses unusual attitude recovery, it also includes ground training in recognizing the in-flight circumstances that put a pilot at high risk of loss of control.

These circumstances include: low altitude and low energy; distraction/fixation when in IMC; mistakes in programming automation; environmental factors such as ice, turbulence and wind shear; and malfunction/failure of automation or control/trim systems. Real LOC recovery training teaches that because loss-of-control events come from a diverse set of causes, they cannot be solved by a single set of procedures. Solutions to true LOC require pilot ingenuity and skill when facing rare or unique situations. It also teaches that when the real thing happens, there is a period in which the pilot is “startled” (I think that’s much too tame a word for what the pilot feels when things start going south rapidly) and how to work through the startle. The concepts of “measured response” and “alternate control strategy” need to be taught.

Finally, a LOC event may be accompanied by a change in the flying characteristics of the airplane. Unusual attitude recovery scenario training almost invariably stops when the pilot gets the airplane back to level flight—LOC recovery training goes further because strategies may have to be developed to allow the pilot to keep the airplane in level flight and then maneuver it to a landing.

New Concepts

What makes for good LOC recovery training that can be conducted in a reasonable amount of time and reviewed during recurrent training without cutting out other important material? On the ground, LOC discussions can be worked into sections on aerodynamics and weight and balance—to enhance a pilot’s understanding of the why of LOC and increased risk at higher altitudes, higher weights and aft c.g. locations.

It’s also a good place to introduce the concept that full control deflection may be required to recover from a LOC event, and why. It’s a good time to discuss the concept that to recover from a stall, the angle of attack has to be reduced and that it’s okay to lose some altitude to do so—trying to power out of a stall is not always successful, especially at higher altitudes where available power is reduced.

Risk factors for LOC and how to be alert for increased risk should be covered on the ground. Any time the airplane is in a state of low energy at low altitude—shortly after takeoff or in the pattern—a pilot should be aware that a system malfunction, distraction, focus on one item to the detriment of others, wind shear, turbulence or immediate traffic conflict can result in an inadvertent stall in a remarkably short period of time. Techniques for reducing the risk—such as a sterile cockpit rule, use of an angle of attack indicator, trimming for appropriate speeds and how to correctly program automation—should be reviewed.

Risk Factors

Programming automation is a time of remarkably increased risk in all phases of flight because of distraction and the potential that improper programming may cause the airplane to do something unexpected, which can lead to an upset or spatial disorientation and loss of control once the pilot shuts off the automation. I’ve looked at accidents that involved loss of control within moments after a pilot was given a vector for an approach and turned the wrong way because he hadn’t loaded the approach correctly into the aircraft’s automation, the airplane didn’t do what the pilot expected and he disconnected the automation while trying to sort things out. A corollary to this sort of accident involves pilots who are over-reliant on automation and get into trouble when it is not available.

I firmly believe that ground training should include a frank discussion of the startle factor. A certain percentage of pilots suffer from the macho feeling that they can handle anything and will never be surprised in flight. LOC training needs to bring out that the startle factor, particularly where the airplane does something unusual rapidly—and there may be things flying around the cockpit and the pilot may smack his or her head against the ceiling or side wall—means that no pilot is going to be cool and calm enough to evaluate accurately what is happening and why.

Measured responses have to be taught to avoid the often counterproductive tendency to react before analyzing. The hoary aviation axiom that when facing a critical situation, “the first thing to do is wind your watch,” has basis in truth. There’s a strong tendency of pilots to fixate on one indication, thereby fail to comprehend and correctly diagnose the underlying problem. The solution is to gain control of the airplane and then determine and eliminate the cause of the problem.

As pilots age, they tend to take longer to take in what they see during an instrument scan—their scan time takes longer, it takes longer to carry out tasks such as changing frequencies and they may be more likely to fixate—all of which increases the risk of the airplane entering a bank or wandering off altitude when hand flying. Awareness of the increased risk and the need to keep a scan going will help aging pilots avoid loss of control events.

The fact that full deflection of one or more of the flight controls may be necessary should be emphasized. Few pilots have put the ailerons or rudders to the stops in flight and may only have run out of elevator authority when making a full-stall landing. As an aerobatic instructor, I’ve observed that most pilots are hesitant to put the controls to the stops even when they know it is necessary, such as when learning to do an aileron roll. In addition, it should be pointed out that it may be necessary to continue an unusual control deflection once the airplane is back to level flight because of a jammed control, runaway trim that cannot be reversed or structural damage to the airplane resulting from the LOC event.

Factors Resulting In Upsets

The list of things that can interfere with recovering from an inflight upset and require a training strategy is long, but here are some highlights:


George may not be your friend. Incorrect programming may have gotten you into an LOC situation in the first place. Understanding how to overpower or disable an autopilot or trim system can be critical.


Thanks to the element of surprise, our initial reaction may be delayed or inappropriate, making the situation worse. Understanding the situation and correctly applying correctives is critical. You don’t want to make it worse.

Control Inputs

It may take substantial effort to manipulate the controls effectively to respond appropriately, and the natural first response may be incorrect. Practice (and understanding) improves the likelihood of getting it right.

Which Airplane?

In order to maximize the probability that a pilot will successfully recover from a LOC event, she or he should train for it in an airplane that duplicates the handling and control response of the airplanes he or she regularly flies. If the pilot flies an airplane with a yoke, training should be in an airplane with a yoke. Most general aviation airplanes have relatively slow roll rates and require putting the ailerons to the stops to get whatever roll rate is built in. That means training in an airplane that has

a slow roll rate so that the pilot can get used to what may be hard work in recovering from a roll past the vertical with the nose dropping.

I do not advocate LOC recovery training in high-performance aerobatic aircraft for the important reason that they are way too easy to fly. Recovery from inverted flight takes no effort in them, whereas the same recovery in a different airplane takes both hands on the yoke, cranking it as far as it will go and holding it there.

In my opinion, the ideal airplanes for LOC recovery training are the Cessna 150/152 Aerobat, aerobatic Beech Musketeer and aerobatic E33C/F33C Beech Bonanza. All have control responses that are similar to the general aviation airplanes most pilots fly, including modest roll rates, and it takes work to recover from the attitudes that can be generated in an LOC upset.

Surprise needs to be built into LOC recovery training—which makes tossing something sudden at a pilot during a flight review or IPC part of the training. The instructor can have a little fun coming up with ways to startle a pilot and perhaps induce a LOC such as shouting or slapping an open palm on an interior surface while the pilot is flying the airplane a knot or two above stall. Rolling in excessive nose-up or -down trim while establishing an unusual attitude and then forbidding the pilot from resetting it once level flight is established demonstrates how hard it can be to keep an impaired airplane under control following recovery.

Avoiding rolling pullups, which put intense strain on airframes, should be emphasized. During a nose-low, steeply banked event, the pilot needs to learn to get the wings level prior to raising the nose and understand why forward pressure on the yoke may be needed as the airplane naturally seeks to slow to its trim speed.

Finally, there seems to be a shortcoming in the way recovery from a nose-high impending LOC event is taught in the piston-airplane world. Most procedures call for going to full power to delay stall onset and immediately working to force the nose down with the yoke to get back to level flight. In the jet world, pilots are taught to also enter a bank of about 45 degrees, because it may not be possible to get the nose down with the yoke (e.g., trim runaway, center of gravity shift). Entering a bank will lower the nose and help avoid a stall. Should the elevator or trim tab be jammed or cargo has shifted aft, establishing a steep turn may allow the pilot to maintain altitude at a speed above stall while fixing the problem. I believe that too often general aviation pilots focus on trying to force the nose down, wings level, in an impending nose-high LOC event and are unable to do so before stalling and losing control.


I’m a strong proponent of simulators for LOC avoidance and recovery training. They allow numerous sequences to be set up and run in short order. In addition, some simulators can simulate control hard overs—something that is difficult to do in an airplane.

The down side is that simulators cannot simulate the G loads that a pilot may be subjected to during the upset and recovery. In order to lower the nose to avoid stalling the airplane in a nose-high LOC event, the pilot may have to go to partial or negative G, which the simulator cannot duplicate. Having loose items in the cabin float around and/or hit the pilot on the back of the head may be a reality of a successful LOC recovery, which means that it must be practiced in an airplane.


Teaching plain vanilla unusual attitude recovery for more than 80 years hasn’t been successful in keeping the loss of control accident rate under control. Moving to LOC avoidance and recovery training that emphasizes the why of LOC, the risk factors leading to it, avoiding it and realistic in-flight simulations and recovery is the logical next step. Besides, the training is a lot of fun.

This article originally appeared in the August 2017 issue of Aviation Safety magazine.

For more great content like this, subscribe to Aviation Safety!

Other AVwebflash Articles


  1. Many good points raised in this article, especially the startle response and how that can make a bad situation worse. I teach Aerobatics and also have an upset recovery and prevention syllabus but there are no aerobatic aircraft at any local flight school so I only teach on students who bring their own aerobatic airplane, so it is not accessible to most local GA pilots.

    This problem has been recognized by AOPA and they are now pushing “Expanded Envelope Training”. This is all the elements of UPRT and LOC training that can be done inside the flight envelope of a normal non aerobatic airplane, which is actually quite a lot. Most pilots fly their airplanes firmly in the middle of the flight maneuvering envelope. EET explores the edges of that envelope in order to give them the skills and confidence to recover if they ever get into a bad and getting worse place.

  2. If you grew up in the snow belt your Dad probably took you out on a snowy parking lot to figure out driving in the snow…no matter how many times you’re told “steer into the skid” it doesn’t click until you do. Unfortunately, GA training seems to more emphasize moving to Florida over real world exposure.

    Regardless of aircraft chosen for training, a windscreen full of nothing but ground or sky is the biggest hurdle for some…if you can get pilots accustomed to this view, there’s hope for rational application of procedures instead of a frozen processor between the ears.

    Couple nits to pick for those learning this subject…”low altitude low energy” muddies the fact that altitude is your biggest energy store, not airspeed (especially for GA)…suggest just ref’ing low energy as the energy concept always seems poorly explained in GA training. Jet recovery is different as it lacks the luxury of propwash over the tail…there’s a reason jets don’t do hammerheads. The “unusual attitude” photo presents a confusing perspective which doesn’t help explain the concept unless one flies without seatbelts and performs 60 banks on their head…

    Overall a good conversation to have on a regular basis…IFR requires currency, “one and done” probably ain’t going to cut it for slow flight/unusual attitudes.

  3. The original article appeared in 2017. What training has been added from the article’s suggestions? I suspect very little. I have always disliked the “usual training” for recovery from unusual attitudes where the instructor puts the aircraft in an unusual attitude for the student or examinee to recover. I had a great learning experience when my instructor had me looking down and I continued to be on the controls with his instructions to to turn left, turn right, pull back, etc. so I was the one who got us in an unusual attitude and became nicely disoriented by my own hands and only then I had to recover from my own induced unusual attitude. This served two purposes – first I became disoriented by my own control inputs (great learning experience) and secondly then I had to recover from an unusual attitude that I got us into. Good pilots seek to always expand their skills and it starts with ground work and lots of “What if…” and then practicing them in the air where it really counts.

  4. In some 3000 hours and about 50 years of flying, I’ve had two notable LOCs, both of which were handled well because I’d had basic aerobatic training, about 10 hours worth. I commend aerobatics as a way to understand how to get back “within the envelope” without breaking anything. It’s been some 40+ years since I had the aerobatics course, but I remember many of its lessons well.

    The first, which occurred within months after I took the aerobatics course, was being caught in the wake turbulence of a B737 while flying in a TR182 at 7000’ MSL/1800’ AGL north of Denver. As the airplane rolled very suddenly well beyond wings vertical, I applied full opposite aileron and most importantly pushed hard on the yoke. The natural tendency of most pilots untrained in aerobatics is to pull on the yoke when inverted. Although in this case, the airplane did not go fully wings level inverted, it was enough that pushing was important, especially because at the time the altitude was only about 1800’ AGL. As I recall, there was very little loss of altitude in the recovery.

    The second was several years later, in a Mooney 231 at about 11,000’ MSL/7000’ AGL at night southwest of Newcastle, WY on an IFR flight plan. The autopilot was on, the air was clear and smooth, and I was frankly enjoying just sitting there, when suddenly the autopilot commanded a hard right turn. I think I hit both the yoke disconnect and panel disconnect at about the same time and returned the airplane to level flight, but not before the airplane had turned laterally about 90° and lost several hundred feet of altitude. The delay caused by the “startle effect” was significant. As I was returning onto course and altitude, ATC called and asked if everything was alright. Later examination of the autopilot by an avionics shop was unable to determine why the event had occurred, but it caused me to forever distrust autopilots.

    Learning about an airplane’s performance outside of the normal flight envelope and how to cope with such unusual events is a real safety measure. Whether that’s in a basic aerobatics course, or in a dedicated upset recovery course or LOC recovery course, every pilot owes it to him/herself and passengers to take the course and become proficient in it. It does wonders for confidence, too.

  5. With head down, eyes closed and arms crossed, my instructor would place the 172 in an “unusual” attitude. As he was doing this, I would mentally follow along and when he said “your airplane”, I’d open my eyes, grab the yoke and perform what was required to “recover”. No big deal I thought. It was easy.

    Instead, try this:
    On a whim, the instructor had me close my eyes with my hands on the yoke, I’d fly the airplane and he would give me verbal instructions: turn left, turn right, level off, climb, descend, turn left, level off…now, open your eyes and recover!

    I have never been so discombobulated! I was all cattywompus. I thought we were nose up, we’d be nose down. Nose down, we’d be in a right climbing turn. It would take me a good 2-3 seconds to get unscrewed, and then another second or two to recover. I have no idea what was the difference between my instructor placing the A/C in an “unusual” attitude and him having me place it there, but it was different and we inadvertently found a new training aid.

    • Further, and I think this applies to recovery and other maneuvers such as stalls..when we train, the maneuver is “to do” or “not do” something.

      “Let’s set up for a power off stall”..Ok, we are going to stall the airplane…check.

      “Let’s set up for slow flight”..OK, we are NOT going to stall the airplane…check.

      What happens when we are in slow flight and inadvertently stall the airplane? In training, we do NOT inadvertently do things. If we do, we see it as a failure, not a real life teaching moment.

      After my “upset recovery” moment, my instructor and I started to do things a little differently. We would climb to altitude and instead of practicing stalls, he’d point out a farmhouse to circle and we’d get slower, and slower, angle of attack increasing, slower, pitching higher, focusing on the farmhouse and my perfect circle, until the stall was as much of a surprise as we could make it. Now recover.

      Once the “flying the plane part” is coming together, and the student can do the maneuvers, stop focusing on the maneuvers. Start focusing on real life scenarios. It’s not “turns around a point”, it’s making a 360 for spacing-it’s circling a grass field to check for activity before you set up to land.

      It’s not a “power on stall”, it’s this is what it looks like during a go-around. Here’s the point in which you could be surprised-don’t let it surprise you.

      Let the stall happen(at altitude of course) and you’ll get a two-fer. The stall-what not to do, and the recovery-what to do.

  6. Unfortunately this is another recycled Avweb article from 5+ years ago that could stand to be updated to reflect the current state of the industry. “I think it’s time for unusual attitude recovery training to morph into loss-of-control (LOC) recovery training at all levels of initial and recurrent training.” Many people agree, and in 2017 (when the article was written) the FAA and EASA started proposing requirements for formalized UPRT training, at least at the ATP level. By 2019 formalized UPRT training became a requirement for new CPL and ATP certificates by EASA. Although the FAA has yet to make it a requirement, they have published guidance in the form of ACs and the latest version of the airplane flying handbook that specifies basically the same thing. In addition, the author contends that “Real LOC recovery training teaches that because loss-of-control events come from a diverse set of causes, they cannot be solved by a single set of procedures. Solutions to true LOC require pilot ingenuity and skill when facing rare or unique situations.” This opinion is contrary with the EASA, FAA, and industry guidance. In fact there is one basic recovery strategy that can be used to recover from all upsets and unusual attitudes with the only exception being a fully developed spin. PUSH-ROLL-POWER-STABILIZE is now the single procedure endorsed by the FAA, EASA, and other industry stakeholders as the procedure to be used for unusual attitudes and upsets up to and including incipient (but not fully developed) spins. See chapter 5 in the airplane flying handbook. There is an alternate strategy which reverses the POWER and ROLL steps which can optionally be used in some nose high upsets which MAY work better for some aircraft. But in general if you only remember the basic strategy it will work. “pilot ingenuity and skill” are in very short supply during an upset encounter when the startle factor comes into play. This is why a simple approach that works for all situations is so important. I agree with the author’s conclusion that “Teaching plain vanilla unusual attitude recovery for more than 80 years hasn’t been successful in keeping the loss of control accident rate under control”. I just wish Avweb would publish new, updated content instead of finding old articles and reposting them. Times change…

  7. First rule like so many other aviation procedures is to not make it worse.

    My old acft had separate post stall recovery (essentially neutralize controls) and spin recovery procedures (pro-spin…meaning if you guessed wrong it would induce a spin if not already in a spin). What I passed along was if the acft wasn’t doing what you expected then apply post stall recovery procedures while assessing for a spin. If you weren’t in a spin, you were likely recovered by the time you figured it out…if you were in a spin you had not made it worse and gave yourself time to evaluate direction of spin to apply correct inputs.

    • Centralize, Analyze is what I teach my aerobic students to do when they don’t know what to do next

  8. Aerobatic is what I meant before the @#&$@& autocorrect “fixed it” although come to think about it Centralize, Analyze works for aerobics too 🤔