AB9IL.net: The Crash of Voepass Flight 2283 (Sao Paulo)

A few hours ago, Voepass flight 2283 crashed in Vinhedo, Brazil, a town to the north of Sao Paulo. The near vertical impact took the lives of all passengers and crew who were aboard. The aircraft type was an ATR-72-500, a large regional turboprop. Multiple videos captured the falling plane prior to impact; others chronicled the fiery aftermath. In a few paragraphs below, I will give some insight, based on my experience as a former regional turboprop and heavy jet pilot.

Before we go on, let me say this: don't believe all of the clickbait and bullshit being posted to the internet about this crash. There are dozens of so-called exclusives and breaking news about Voepass 2283, which feature video or images from other crashes. I actually recognized some images taken from an ATR crash which happened in South Asia some time ago, used solelely as clickbait and not at all related to flight 2283. They want your attention and offer you nothing related to the truth.

What Happened?

Voepass 2283 seemed to fall from the sky in a condition known as a "flat spin." A spin is a condition in which the one or both wings are aerodynamically stalled: meeting the oncoming air at too high an angle to produce proper lift, instead generating turbulent eddies over the wings. Also, for a spin, one wing is in a worse condition and the airplane yaws (rotates about its vertical axis) and rolls (rotates about its longitudinal axis) toward that worse side. To an observer, the rotating airplane moves in a spiraling motion.

You may have seen planes spin for a second or so, during acrobatic flying at airshows, as "snap rolls" are rapid and deliberate spins executed while an airplane is moving horizontally over the ground. Perhaps you have seen other instances, maybe crashes, where an aircraft enters a stall, has some yaw imparted to it, and spirals downward toward the ground. In a flat spin, the aircraft falls while oriented right side up, not rolling much, but rotating in yaw, about its vertical axis.

For multiengine transport planes, flat spins are incredibly dangerous because of the difficulty of recovery. There is so much mass in the wings and tail, far fron the center of gravity, that angular momentum rather strongly tends to maintain the spin. Unless the pilot flying acts to break the stall and recover immediately from the spin, the situation tends to get worse, until recovery becomes impossible before hitting terrain.

Hazardous Icing Conditions

At the time of the accident, a significant risk of severe icing was present in the area, prompting the issuance of SIGMET bulletins. Issuance of such bulletins triggers, an obligation for airlines to not intentionally dispatch flights into icing conditions within the affected volumes of airspace. Likewise, pilots are always expected to be prudent about icing, and extra careful about icing in the areas designated by the SIGMET. Be aware that airspace covered by a SIGMET is not closed to traffic. Rather, pilots are alerted that dangers lurk there, for icing, turbulence, or other hazards.

Among the problems caused by icing on an aircraft, the increased stall speed and less predictable aerodynamics are among the worst. There is one type of clear icing, involving large supercooled water droplets, in which droplets strike the wing, then spread and run back along the surfaces. The stalling speed can literally increase until a plane snaps out of control while at normal speeds and G loadings. ATR turboprops have already crashed due to the sort of icing warned about in SIGMETS. Unfortunately, pilots won't necessarily know they've encountered severe icing until they can see it on the windshield or feel its effects.

The best way to handle severe icing, especially if it consists of large supercooled water droplets, is to stay out of it. The second best action, if pilots discover they are in it, is to get out of it ASAP. Descend to warmer air, turn to exit the visible moisture, and / or increase speed (to increase the total air temperature (TAT)), if possible. Obviously, make smart use of the available de-icing / anti-icing equipment.

When Twin Engine Planes Lose One

Large twin engine turboprop planes are like small propeller-driven twins. If the only problem is one failed engine, the plane may be flown on the remaining engine until a safe landing can be made. The worst time to experience an engine failure is on takeoff, where the challenge for pilots is to either stop before running out of runway, or continuing the takeoff and climbing out without hitting any obstacles. In such a scenario, there is usually just enough power available from the other engine to climb safely to an altitude where the pilots can secure the dead engine, configure the plane for single engine flight, and land at the nearest suitable airport.

Large turbopropeller powered twins get a boost in single engine performance (or rather, an allowance to fly at heavier weights) with a safety feature which automatically streamlines the propeller on the failed engine. Called "autofeather" or "autocoarsen," it acts automatically to prevent a windmilling propeller from adding more drag and blocking airflow across the associated wing. If autofeather fails, then the pilots must manually feather the propeller to avoid a significant loss of performance.

A critical problem with the loss of one engine is the matter of asymmetric thrust. Instead of the combined thrust of two engines acting through the plane's center, the thrust acts from the lateral offset of the good engine, and acts to yaw the plane. The pilot must maintain directional control by applying rudder pressure toward that working side, to nullify the yaw. It is especially critical that the plane be flown at a high enough airspeed to have enough force available from the rudder to fly straight, or the pilot will lose directional control. If control is lost, the airplane will roll and yaw toward the dead engine. If conditions are such that the airplane also reaches a stall situation, there is a good chance of getting into a spin. That is a very serious case of going from bad to worse - really worse.

Risks of Single Engine Flight at Higher Altitudes

One wouldn't expect a failed engine to be as dangerous during the descent phase of flight, when thrust is reduced. Voepass 2283 was at an altitude of 17,000 feet when its problems became very serious, and we don't yet know exactly what happened. I have a terrible suspicion that the flight had one engine fail, and for whatever reason, allowed the airspeed to decrease until a stall happened. If the good engine was providing power, the pilots may have gotten too slow to keep directional control (in yaw), increasing the probability of not just a stall, but also a spin.

At an altitude of 17,000 feet, a heavily loaded ATR-72 may need a lot of power to fly level, or to level off from a descent. As the air is almost half as dense as air at sea level, single engine flight, and especially single engine flight if the dead engine's propeller is unfeathered, may be quite precarious. I am speculating, as I don't have the flight data and performance charts, but it seems that Voepass 2283 was in an extremely risky situation. The safe way out would be to pitch downward for a gravity-assisted speed boost, reduce engine thrust for a reduction in power asymmetry, and get better control of the plane. Of course, in crowded skies, the pilot not flying would need to declare the emergency and work with ATC to get a better altitude and safe spacing from other traffic.

Was the Autopilot Involved?

There is yet another issue to consider: the autopilot. Autopilots are nice to have, as they can be configured to fly the plane for you, reducing fatigue, and allowing the pilot flying to broaden his awareness to other aspects of the flight's situation.Autopilots can be troublesome too, as they often don't handle failed engines well. They'll fight the situation in simpleminded autopilot ways, until they lose control of the plane. I recall there was a crash several years ago near Indonesia, in which the pilot flying had engaged the autopilot, and then didn't notice the failure of an engine. When the autoflight system could neither maintain speed nor directional control, the plane stalled, spun, and crashed into the ocean below. The crew did try to recover, but things had gotten too bad too quickly, and they hit the surface before they could level off.

Other Possibilities

While I see similarities with the way flight 2283 came down, there may have been altogether different circumstances involved. Perhaps they had a structural failure or jammed flight controls. Perhaps they had problems with icing. Perhaps something even more unforseen happened. Aviation is full of surprises, mostly of the dangerous and deadly kind. I have certainly had my own share of experiences where, if not for dumb luck or the grace of God, I would have been killed long ago.

May the passengers and crew of Voepass flight 2283 rest in peace. If and when the Last Trumpet sounds, and we all are restored to an afterlife, whole and uncorrupted, may the friends and loved ones be reunited in joy.

My Comments after the Preliminary Report

It is saddening not only that this crash happened, but also that all of the key factors could have been handled in such a way that the crash could have been prevented. I want to warn everyone, pilots and non-pilots alike, that we live in an imperfect world. People are imperfect and so are the machines humans create. Our knowledge is imperfect and so are our safety procedures. If you see a bunch of ill-informed trolls putting down the pilots as "incompetent" or "hired for their connections and not skill," just reject the trash talk and move on to a better forum.

Here is a short list of things which seemed to matter in this accident:

1. A chain of events can kill any of us on our next flights. Be prudent, never complacent or cocky.
2. Deicing boots are not effective and not certified for severe icing; don't even trust them in moderate icing. I have witnessed them doing a crappy job, but I am still here to say this.
3. It is better to be aggressive and get out of bad icing conditions than to be patient and cooperative. A "PAN PAN PAN" call may be appropriate. The Voepass crew did not seem to recognize their increasingly dangerous situation.
4. Autopilots offload the work of flying, so you can broaden your attention to see the overall situation. However, autopilots are poker players: they don't reveal stress or difficulty. They abruptly relinquish control back to the pilots, who are probably going to be surprised and not ready to grab the wheel and fly a plane snapping into a violent maneuver. More than a few jetliners have gone way the hell out of control for this.
5. Considering the icing on the ATR-72's wings, the normal stall warning system would have been ineffective and likely to not detect an impending stall. The wings would no longer have their precisely engineered curves, but some kind of rough, random, and uneven surface. The plane stalled when it rolled into a right turn toward a navigational waypoint.
6. The last thing the crew could have done in order to recover would have been to set power to idle and pitch the plane down to break the stall. Firewalling the throttles would have made the spin worse, flattening the spin. Worse still, was the fact that firewall power would produce uneven thrust - and resultant yaw. Remember, one air conditioning pack was inoperative, enabling that engine to have more air and produce 6% to 7% more thrust.

One unfavorable thing after another added up to this crash happening. For starters, there were flight at relatively high altitudes, without much excess power available for fast flight, along with extra care needed for even thrust (one pack inoperative), the presence of at least moderate, if not severe icing, and flight on the autopilot, which did not give hints as to how badly the plane may have been handling.

Flight control positions were not available, so we do not know what inputs the captain was making, as the flying pilot. He may have been applying full power and trying to pitch up and climb back to their assigned altitude (which would not work in a stall). He may have been applying foreward pressure, intending to pitch down and into the relative wind, breaking the stall. If so, there would have been the issue of the horizontal stabilizer and elevator being "shadowed" by the wings. In other words, a deeply stalled airplane, especially "T" tailed ones, will have a problem with airflow being blocked by the stalled wings.

The remainder of the information given by the crash investigators just gives color and vividness to the crash of Voepass flight 2283. The bottom line is how how multiple things, not lethal in isolation, acted together to result in the crash. No reasonable crew whould ever see their situation as perfectly safe; various hazards are always present and must be considered and somehow mitigated. As pilots, we are stalked by the grim reaper, who tries to back us into a corner and kill us. Stay awake, watch for those compounding problems, the loss of safety margin, and GTFO more sooner than later.

I have seen it enough times, that acting early to keep small problems from becoming big ones can save lives.

© 2005 - 2024 AB9IL.net, All Rights Reserved.
Written and curated by Philip Collier / AB9IL.
About Philip Collier / AB9IL, Commentaries and Op-Eds, Contact, Privacy Policy and Affiliate Disclosure, XML Sitemap.
This website is reader-supported. As an Amazon affiliate, I earn from qualifying purchases.