The 29-year-old pilot of a Cirrus SR22 was taken to a hospital with smoke inhalation after the aircraft descended on fire under its Cirrus Airframe Parachute System (CAPS) near Serra Azul in southeastern Brazil on Tuesday. The video shows the aircraft dropping gently (about 1700 fpm) under canopy as smoke billows from the engine compartment. The aircraft was completely consumed after touching down in a field, but the young pilot was able to escape. According to social media the plane was on its way from Pampulha airport, in Belo Horizonte, to Bom Despacho when something went wrong.
WATCH: Cirrus SR22 catches on fire with its parachute deployed near Serra Azul in Brazil.
— Breaking Aviation News & Videos (@aviationbrk) August 25, 2026
According to the rescue team, the 29-year-old pilot was taken to the João XXIII Hospital, in the capital of Minas Gerais, with smoke inhalation injuries.
The aircraft left Pampulha… pic.twitter.com/01iOfUZFq4
Early reports suggest the pilot activated the chute after some kind of failure resulted in an explosion and fire about 14 minutes into the flight. A cursory look at the SR22 POH for Brazil doesn’t yield any mention of fire in the scenarios presented for pulling the chute (page 85), but it does include this general disclaimer (page 80) that puts it all on the pilot. “The Cirrus Airframe Parachute System should be activated in the event of a life-threatening emergency where CAPS deployment is determined to be safer than continued flight and landing.” It will be up to Brazilian authorities to determine whether the pilot’s decision to pull the handle saved his life or almost killed him. Because it’s an American-made aircraft, the NTSB and FAA will also be involved.
1,700 fpm is what caught my eye. That had to be one heck of a not so soft landing.
1,700 fpm translates into a 20 mph dead stop crash. The dead stop part is what’s really going to hurt.
Shades of “It’s not the fall that gets ya, it’s the sudden stop?!” That being said, both the seat mounting frame and landing gear are designed to collapse progressively. It’s part of a system.
Yes, the insurance writes it off as totaled every time. Both wheel pants firmly implant themselves in the bottom sides of the aerodynamic lift devices.
No they don’t. There are a number of post-CAPS aircraft flying today. I think it’s something like 30%; maybe a little less.
Is the equivalent of being dropped 13ft while in a roll cage on landing gear and a seat design to act as a crumple zone.
There is additional deceleration to occupants due to the crumple zone effect of the landing gear and seat design. People walk away from chute deployments.
Why would parachute deployment have almost killed him? Descending at 1700 feet per minute would likely got him on the ground faster than than trying to reach an airport. And I imagine trying to land off airport with a cockpit full of smoke would have had a less than ideal outcome. Plus if there’s a fire in the cabin, you can position yourself out of the way of the flames rather than trying to rudder the plane with your feet on fire.
Marvin, may I softly and gently remind you that an emergency descent is not a smooth maneuver designed to make the passengers feel comfortable with your pilot skills, even though 1700FPM would normally be a high rate of descent for a Cirrus. 4000FPM can seem like forever in a jet if you really need to get down. Remember, Just One whiff of burning insulation from electrical wires can kill a pilot after rendering him unconscious in a matter of seconds or less. It’s not the same as standing on the downwind side of a campfire with smoke in your face and having to move your lawn chair to the other side in order to roast hot dogs and marshmallows. It is highly toxic beyond belief. And that is not the only toxic material that could produce smoke in an airplane.
What this means is that you need to bring that airplane out of the sky extremely aggressively and put it on the ground at the first possible landing opportunity.
He deployed at about 3000 ft AGL, so about 2 minutes to get to the ground. Are you really saying he could have safely flown it to the ground in less than 2 minutes?
Marvin, I appreciate the question, and you may have just saved a whole bunch of lives by asking it.
Every pilot has to make the decision what he will do in any given situation, and not knowing what his options were, I won’t speculate on that. But we should ask, Was there an airport nearby? How many airports did he pass before making the decision? How long had he waited to make the decision after he noticed the smoke? Etc…etc…If he pulled the chute at 3000 feet, how many minutes before that and what altitude was he at a few minutes before pulling the chute? Did he get the faint smell of smoke in time to make a decision, or did thick, toxic smoke just engulf the cockpit instantly? We don’t have those answers yet.
But let me answer your question with a question, with the assumption that unbreathable, toxic smoke engulfed the cockpit instantly at 3000 feet. Can you survive breathing highly toxic smoke for 2 minutes? What happens if you open the door on a cirrus in order to get one breath of fresh air?
You can certainly open a Cirrus door inflight without consequence. Accidental door openings happen and as long as the pilot in distracted by it, it is a non-issue.
He was 14 minutes away from his departure airport. Even if it was over an airport, and maybe even on the downwind, he would be hard-pressed to have the airplane on the ground and stopped in under two minutes.
He was over farmland and residential terrain. Is he would have tried to land in some farm field, he liked would have end up upside and on fire.
Yes, you’re correct. You certainly can open a Cirrus door in flight, unauthorized. And sometimes they just pop open on takeoff partially disturbing airflow to the tail like an Aztec. So the next question is, “when you open that door in flight because of smoke, how does that affect the airflow to the fire in the cockpit that is about about to burn your soon to be unconscious body, which is struggling for breath in toxic smoke?”
No. The question isn’t about smoke. The question is how do you propose to get on the ground faster than 2 minutes?
Second question is, are you a bot designed to prompt engagement.
Furthermore, opening the doors is part of the smoke in cockpit procedure, so your unauthorized comment is as nonsensical as the test.
typo (rest not text)
Yes, the question is about smoke. And I sense that a whole generation of presumptuous young pilots have not been trained, much less “properly trained “ to understand and deal with one of the most deadly enemies in aviation, and that is smoke.
Ok bot.
The fastest route to the ground in a Cirrus is an emergency descent to 1000′ AGL followed by a CAPS deployment. In the case of an engine fire, the emergency descent creates high airflow through the cowling which helps suppress the fire, and then the descent under canopy takes significantly less than a minute, and you can egress immediately upon touchdown, unlike a forced landing where you first have to stop from ~70mph.
Have you seen any reports of whether it was an engine fire or an electrical fire? The script above says fire was seen coming from the engine compartment, but I’m not sure how qualified the source was.
I’m not sure if it’s been confirmed, but that’s a *very* big fire to be electrical. Much more likely to be something fuel-fed.
There was an on-fire Cirrus descending under parachute at Boulder, CO several years back. The pilot and/or any passengers burned to death long before the plane reached the ground.
I’d like to think that, were I in a similar position, I’d try to get the plane a bit lower before opening the ‘chute.
Based on ADSB he deployed about 3000 ft AGL, Cirrus recommends 2000 ft at the minimum CAPS decision altitude.
Most emergencies allow time to “wind the clock” meaning there’s time to think through options and consult a checklist. Not so with a fire. Seconds count. A lesson backed by multiple accident reports. Hopefully Cirrus documents and training teach this.
Swiss Air 111 comes to mind. If I remember, there were an ungodly number of checklist items and steps the crew had to perform/sort thru before turning their attention to landing the aircraft.
As pilots, we want to believe there is a “best option” to handle a given situation, and if we study, and train, we’ll be able to make the right call. But this incident demonstrates that sometimes there is no best option – just a coin toss, and that frustrates us.
“activated in the event of a life-threatening emergency”. I believe he made the right decision. He was PIC, his call. After all, how large do the flames have to be to be called “life-threatening”? I’m glad to see he is still among us. A smart call.
Agreed. Considering he was able to get out on his own, he clearly made the right decision.
I’m surprised (although this is the internet and shouldn’t be) about criticism to pull the chute.
After dispersing both fire bottles and shutting down the engine in the Cirrus SF50 simulator for my first emergency, I pulled the chute at 7000 feet and it took many minutes to get down to earth. My instructor suggested next time gliding for a bit in my intact airplane then pulling the chute a few thousand feet over the conveniently located airport under where the fire (instructor) occurred.
DAMN those Cfi’s! ;-D
Jetjock60, The first time I flew a real Vision Jet after getting the rating in the simulator, the CFI asked “Does it fly like the sim?” I replied “Not at all” and he looked shocked. I said “Nothing broke.”
Ha! Yeah David. My first type was in the 737 when I got hired on in the mid 90’s. The instructor said,”Prepare yourself for the unluckiest two weeks in the most unreliable airplane you ever had!”
If you’re going to position yourself over a conveniently located airport to pull the chute, wouldn’t you just land the plane?
In general, you get yelled at by Cirrus instructors if you don’t use the Chute. (“The insurance company owns the plane.”) I this case it was heavy IFR down to the ground.
I have flown about 5 different Cirrus SR-22’s over the years for various customers and one of the customers, who was also a junior pilot said, “My wife won’t let me fly in anything but a Cirrus because it has a parachute.”😀😀😀 Hey, I don’t know, maybe it’s safer to die in a burning airplane with a parachute deployed than to die in a crash before it has a chance to burn. (Tic). Maybe Cirrus should put a label on the parachute that says “in case of fire, remember that it takes a lot longer to get it on the ground from altitude with parachute deployed. Smoke from fire may cause difficulty breathing” or something like that. You know, like a bottle of poison says “keep away from children.”
Just an observation, but I have seen way too many of these Cirrus accidents with the 550 Continental spewing oil up the windshield, engine failing in flight and other issues. I don’t remember so many Cessna, Piper and Beech airplanes making the news for the same reasons. Cirrus is a fun little airplane, but I get the distinct impression that a lot new pilots believe that they can’t die in it because if anything goes wrong, they can just blow the parachute. In all fairness, the statistics might be higher simply because it is such a popular airplane. But my observation remains unchanged.
So you are saying he could/should have safely flown it to the ground in a smoked filled cabin in less than the 2 minutes it took for the chute to get him to the ground?
I guess you just have to decide whether you can breathe thick toxic smoke that is rapidly increasing for 2 minutes or not. Firefighters and paramedics say it will kill you.
So how would you propose to get down faster?
Don’t wait until the smoke is boiling in the cockpit. First whiff that touches your nostrils, get it on the ground. If you are above minimum parachute altitude dive to CAPS altitude ASAP and then pull it if there is no airport or smooth field. If there is an airport right below you, land immediately like any other airplane. If the choice is trees, mountains, water, night time or buildings, 35kts with the parachute is better than 65 knots without it. Then ask yourself if you can really swim a quarter mile to shore in the dark after bumping your head in a CAPS water landing. Also ask yourself if you really want to deploy a parachute in icing conditions. You can’t cut away the chute if it ices up like the airframe did. In icing conditions, it will.
I tell all my copilots on the pre takeoff brief after all the V1 discussions and return to airport emergencies, “If things get bad enough, let’s just get it on the ground somewhere.” Most aviation emergencies can be dealt with in a timely fashion. But fire/smoke, airframe structural failure and loss of directional control close to the ground, are all killers with little opportunity to recover.
The question I have is: Why do these engines suffer so many catastrophic failures? Such failures were almost completely eliminated in the automotive world with better engineering and sophisticated ECUs.
I am not advocating for the use of automotive engines but aircraft engine reliability and diagnostics improve drastically (along with power and efficiency) when we adopt modern engine controls and other technologies.
I got a funny feeling that if you told your wife, girlfriend that you’re going to hit the ground at 1,700 fpm or 20 mph she’s going to re-think the chute. I don’t care what kind of compression construction is built into the Cirrus. The first thought that comes to mind is “ooohhhhh… my back.”
You just have to remind them that the speed of a Cirrus at touchdown in a full-stall landing is over 65 miles per hour.
Also, with 151 CAPS “saves” to date, there have been a handful of back injuries, usually from people who are not in the recommended position at impact (which is just to have your back firmly against the seat back).
The issue isn’t the speed at touchdown, it how much distance is available to decelerate afterwards which will determine how much “G” you’ll feel. Plus remember, the Cirrus has the airbag in the seat belt to further protect.
Yes, and in a forced landing, the space available to decelerate is unpredictable and potentially very short. Meanwhile the CAPS deceleration is known, engineered, tested, and proven over 150 times.
The airbags, by the way, have almost never deployed in an accident. I’m happy to have them, but I think they’re a marginal benefit.
I would argue that the reason they have nearly never deployed is that they are designed for longitudinal acceleration, not vertical. Since most of the Cirrus “news” events we hear about involve the ‘chute, it’s of little surprise we don’t hear of airbag deployment. I don’t disagree that the ‘chute is of benefit, just that there are circumstances were we need a bit more granularity and forethought as to whether a “conventional” forced landing approach to a situation might be best. I believe in this instance as others have stated, the pilot “got lucky”.
As a Cirrus owner, I follow the accident reports closely on the COPA forums, including ones that don’t make the broader news. Even in non-CAPS events, the airbags rarely deploy.
There *have* been some deployments, and I think NTSB even cited them as having likely prevented injury in an accident report. That’s why I say I’m happy to have them. If the very specific circumstances arise, they can be useful.
Anyway, as to parachute vs forced landing, in this circumstance, the correct answer was to pull CAPS. Might have been better to pull it just a little lower (1000′ AGL ideally). Emergency descent to 1000′ AGL followed by CAPS deployment is the fastest route to the ground and egress. This has been studied in simulators.
I have some limited experience in an SR22 from years ago when a gentleman asked me to provide him instruction on the G1000 Perspective system in a G3. As I recall, the min altitude on takeoff for CAPS deployment has gradually increased over time from 400 to 500 and now 600 feet. However I noticed you are saying that the company recommends deployment at 2K? Is this new? Or strictly from an enroute situation. What am I missing here?
The 400-600 numbers are minimum, while 2000 is considered ideal. For example 400 might be survivable, while 1000-2000 would assure maximum deceleration. I think Cirrus recommends 2000 to allow an additional safety margin.
Cool. Thanks!
The original altitude was 500 feet, I believe. When the G5 was certified with an increased gross weight (3400 -> 3600 pounds), the parachute had to be enlarged, and that larger parachute takes a little bit longer to fully deploy, so the altitude for the G5 and later aircraft is 600 feet. These are considered the minimum altitudes for deployment from level flight to achieve full canopy inflation and to give the line cutters a chance to fire and get the airplane into a flat attitude for touchdown. These numbers are most relevant for something like an engine failure at takeoff.
But pulling higher gives everything the best chance to be fully stable and gives the occupants a little more time to secure the cabin and themselves. So Cirrus recommends that if you’re in a situation where, say, you lose the engine up high and you’re trying to glide to a runway, you make the decision at 2000′.
Makes sense. I suppose it’s to minimize the swinging that would occur before reaching the ground? The 400 ft. number stuck in my head for some reason. Maybe from the original SR20.
They do say in the POH that deployments as low as 400′ have been successful, so perhaps that’s what you’re remembering. (In fact, there have been even lower deployments, although you’re really on the hairy edge there and some have resulted in some injury.)
In reviewing the POH just now, I found that right before it mentions that 400′ number, it says that there is “no minimum altitude”. However, what’s taught is 500′ as the “CAPS available” height AGL for the G3 and before, and 600′ for the G5 and later. (And yes, there was no G4…)
Did I miss the part where it says he pulled the handle after the fire started? No idea if he “should have” or “shouldn’t have”, but I’m pretty sure he is glad he did.