Pilatus PC-12 crash after take-off from Milan

8 Oct 21 19 Comments

On the 3rd of October 2021, a Pilatus PC-12 registration YR-PDV crashed shortly after taking off from Milan Linate Airport.

The Pilatus PC-12 is a single-engine turboprop designed and manufactured by Pilatus Aircraft in Switzerland. It is a popular aircraft for corporate transport (bizprop?) and short regional flights. In this instance, it was a private aircraft part-owned by Dan Petrescu, a Romanian and German national who has been nicknamed “the shadow billionnaire” by Romanian press, who write that the man was worth around three billion euros and that his plane was his passion.

Most media reports state that the billionaire purchased the Pilatus PC-12 in 2015 for four million euros along with a second party, although there seems to be some confusion as to which company the aircraft is registered under. He was the pilot of the aircraft which arrived in Milan from Bucharest a few days earlier, on the 30th of September. On the day of the crash, he reported his destination as Oblia Airport on the Italian island of Sardinia, where he owned a villa. There were seven passengers on board, family and friends of the pilot, including his young grandson. The pilot’s 98-year-old mother was already at the villa, waiting for their arrival.

The fully-laden Pilatus PC-12 departed from runway 36 at Linate airport at 13:04 local time.

The weather was drizzly with low clouds. There’s a reference to the flight being VFR (visual) but it is certainly clear that the aircraft disappeared into the clouds shortly after take off.

After entering the clouds, the aircraft turned for the southbound route to Sardinia but then turned again, heading towards the suburb of San Donato, southwest of the airport.

According to Romanian newspaper Adevarul, the pilot told ATC that he was diverting and asked to return to the airport. This request to return is noticeably lacking in other reports. The newspaper continues to explained an “expert theory” that the engine might have failed as the pilot attempted to turn back, explaining the sudden increase in speed, so I’m not sure we’re dealing with a reliable source.

The aircraft descended about a hundred feet and the speed increased. A few seconds later, the aircraft’s altitude dropped by another 150 feet. Then almost every report agrees that a Linate controller contacted the flight. “Why did you deviate? To avoid turbulence?”

“No,” came the response from the pilot. That was the last transmission. The aircraft entered a steep descent and disappeared from the radar screen. Shortly before impact, the aircraft was estimated at travelling 170 knots (320 km/h).

Stills from the Dashcam Video with the aircraft visible in the distant haze.

The PC-12 crashed into the roof of a two-story office building which was empty as the building was undergoing renovations. After the impact, the aircraft caught fire. There seems little evidence of an engine fire, although a number of eyewitnesses were quoted as seeing flames from the descending aircraft.

The following dashcam footage has been released by Colleriere della Sera:

The Italian National Agency for the Safety of Flight (ANSV) is investigating the crash, with the investigative bodies of the three involved countries: Romania (registration of the aircraft), Switzerland (manufacturer of the aircraft, Pilatus) and Canada (manufacture of the engine, Pratt & Whitney Canada). As of today, ANSV has only officially confirmed that they have recovered the data recorder from the wreckage. According to local newspaper Colleriere della Sera the investigation’s first task of the investigation is to understand why the aircraft deviated.

I note that on the aviation forums, pilot incapacitation is being put forward as a likely explanation. I’m wondering, though, if in fact the pilot suffered from spatial disorientation. Specifically, a rapid descent and crash after take off without visual references is extremely typical of somatogravic illusion.

Also known as the “Pitch-Up illusion”, somatogravic illusion was first identified in 1946 from research attempting to explain the high number of aircraft which crashed directly after take off during blackouts, which was originally believed to be caused by an unidentified fault in high-performance aircraft. The repeated fast descent into the ground was originally believed to be a specific issue with high performance aircraft. Research proved that the loss of control in dark conditions after take off is caused by the central nervous system misinterpreting sustained acceleration as pitching up. The data from the otolith organs is very close to the same and we disambiguate primarily through visual information. In the dark or in cloud, we lose the visual information that distinguishes acceleration from pitch-up events. Because somatogravic illusion most often happens shortly after take-off, there is very little chance to recover, thus the illusion is known to be especially fatal.

Having said that, there was no apparent attempt to recover or avoid the building, even though the spatial illusion would have passed the moment the aircraft was free of the cloud.

The Pilatus PC-12 carries a Lightweight Data Recorder (LDR) which carries the flight data parameters and audio recordings from the cockpit area as well as possibly, as an optional extra, video from a camera installed into the cockpit. Once the data is recovered and released, we will no doubt know more.


  • Somatogravic illusion may be a credible cause, but there is little or no concrete information here to draw any (hypothetical) conclusions.
    Maybe some of the other readers of this blog may have an opinion?

  • If the idea of the flight being VFR pans out, we can probably assume the pilot didn’t have an instrument rating? In that case, flying into cloud is often fatal.
    It turns out maybe they should have hired a better pilot.

  • even though the spatial illusion would have passed the moment the aircraft was free of the cloud.
    Today’s response comes from Rocky Horror: ~”I’ll remove the cause, but not the symptoms.” How long does it take someone’s brain to reboot from that level of bad information — especially when they are already overloaded (altitude and direction changes, answering air-traffic control)? It will be interesting to hear details from the flight recorder (the crash made news in the US, but with not even this much information), but I doubt it will have the key info of just when the plane was in clouds or (relatively) clear air. Possibly we need a new acronym to go with CFIT, e.g., Controlled Flight Into Clouds — which for a pilot without instrument experience is begging for trouble. I assume he had some hood time in order to get his basic license, but a hood shuts out all queues instead of giving false ones, and he may not have done even hood time since getting his license.

  • I just looked it up. In the US, you can fly a Turbo-Prop single without IFR or Type ratings as long as you stay under the 12,500 pounds. †he PC-12 comes in under the limit.

    In the US one can also fly in IFR conditions without filing a flight plan in uncontrolled airspace.

    Stupid but true.

    • Not exactly stupid; there’s a lot more open space in the US than in Europe. And the rules (at least when I learned them half a century ago) were that in airspace below 1200′ above ground level and not near an airfield with an instrument approach, you could fly with 1 mile visibility rather than the 3 miles for standard VFR. (This may not be a precise statement — as I said, it’s been a long time — but I think it’s close.) Anything higher, or closer to a landing zone, requires full VFR weather or an IFR plan. (Charts showed the exact definition of “closer”; it usually involved a few miles around the runway and an extension for the instrument approach.) Between the Mississippi and the Rockies there is a lot of flat land without much on it — maybe not as empty as Australia (where I saw a full-size passenger jet flying on ADF), but there’s a reason the coasts call it “flyover country”. I have commented previously on the libertarian delusions popular in the US, but there are some areas so open that the restrictions for denser areas aren’t critical and can reduce connection.

  • Maybe some rules have changed since I retired from flying, but certainly in my days there was no requirement for “hood time” for a basic licence, meaning: PPL, VFR, no type- or twin rating.
    In some jurisdictions the weight limits differ. The US is fairly liberal in that respect. For instance, I flew a Turbo Commander for a delivery flight from Weston (EIWT), Ireland, to the USA after a relatively short introduction from the owner. It was US registered and I held a restricted FAA CPL, ME and IR. OK, my main licence was an Irish ATPL with lots of time on different types of aircraft, including jets and turboprops, but if I had been a PPL with only a basic ME rating the same would have applied.
    In many (European) countries, to operate this type would have required a type rating. For instance, in the Netherlands a basic twin rating would have restricted the pilot to aircraft with a MTOW of 2000 kg. The same would apply to the PC12, an SE licence would not suffice.
    In the USA these aircraft do not require a rating, although operators or owners can avail of a course including simulator time. Much depends on whether or not the aircraft was bought new, because in this case this training is usually offered for free as part of the purchase.
    There is often a tendency to “skip” a professional training course because it costs money – although the owner was a billionaire, so that was probably not a factor. What could have been off-putting is the extremely tedious process of getting TSA clearance to be allowed to undergo such training if it takes place in the USA. So there may have been a temptation to fly the aircraft without prior conversion training. Which brings me to the suspicion that the owner may not have been very experienced, did not have an instrument rating and the lack of professional training might have caused an overload, including sensory disorientation, when he found himself in IMC.
    It would be interesting to find if the jurisdiction in which the aircraft was registered, Romania, adhered to FAA regulations, or if not, whether the PC12 required a type rating there. The pilot, presumably the owner, had bought the aircraft about 6 years prior to the crash, but how many hours of flying did he accumulate on it? What was his overall skills level? Did he fly it regularly or only occasionally?
    Please not that this entry is following my own train of thought, it is not intended to discredit the pilot. It is not a conclusion, only a hypothesis.

    • Exactly my thought — somebody rich might like flying in a fancy machine and do it as much as possible, or might think he can hop in the cockpit whenever he needs to be somewhere quickly; cf JFK Jr.’s final flight, or Thurman Munson’s death in a Citation he bought after less than two years of flying.

      • Petrescu’s licence was IFR. He flew before Cirrus. Also with p12 he had a lot of experience, not only carrying a type rating. He flew more than once to Canada from Europe. So, let’s wait for the black box.

  • Vifor,
    Yes, agreed. I already made it quite clear (I hope) that I was just reflecting on potential causes, not attempting to come to a conclusion. That would be premature and put unfair blame on Petrescu.

    Chip made an earlier comment about “a lot of airspace in the USA”, referring to a more relaxed attitude towards flying in limited visibility or clouds without an instrument rating. The UK used to have an “IMC rating” which was some sort of in-between rating. Many accidents took place when pilots inadvertently got into IMC and not all that seldom took place near airports. The philosophy of allowing lower minima to prevail because of “a lot more open space” seems to me to be irrelevant.
    Having said that, the minimum visibility for VFR flights outside controlled airspace in the Netherlands used to be one nautical mile with a minimum cloud base of 500 ft.
    In my early days of commercial aviation I was a banner-towing pilot and because we got paid by the flying hour we translated that into 1.5 km.
    Sometimes the weather cleared to just about that, we took off and returned hours later. In quite a few cases the improvement was only temporary and the vis may have dropped to 700 m. At least, we cruised at 40 kts and literally knew the location of just about every high tree, every chimney, radio mast, TV aerial. In some cases we would have to pull up into the overcast to clear high tension wires. Yet, very few pilots lost their way. In those days GPS would have been something from a Sci-Fi book. We had our liquid compass and a VFR chart, that was all!
    I suppose that the banner had a stabilising effect on the aircraft, nobody seems to have been affected by vertigo.
    To be honest, we broke the rules many times in a major way. The oversight from the authorities was not very strict. We did not carry any electronics, except for a VHF-com. So we did not show up on the radar either, or maybe as a flock of birds with corresponding speed. Or too low anyway. Yes, we got away with it. Professional? No, I must admit.

    The examiner who gave me my first rating on the Citation happened to be the one who had passed Munson.
    Munson’s crash was not due to spatial disorientation. He was doing single pilot circuits in his Citation. Some press reports mention the presence of an “instructor”. This person was actually an instructor on light SE aircraft and did not have an instrument rating, nor a ME rating and was certainly not qualified on the Citation.
    Munson had been on final approach when a light aircraft cut in ahead of him and forced him to go around. The go-around was performed properly, but it seems that Munson’s lack of general experience put him under a lot of pressure. He forgot to extend flaps on the next approach, but flew at the normal Vref as if he had extended the flaps.
    The Citation 500 had no aural stall warning, instead it had strips on the leading edges to cause a “rumble”, a vibration of the control column at the approach to a stall. The Citation had more than enough power to recover if the pilot reacted. All that would have been required was to shove the power levers forward to the take-off setting and extend the first notch of flaps. It would not even have been necessary to push the controls forward, it would have flown out of the stall smoothly without any loss of altitude.
    Unfortunately, Munson must have become overloaded. He did not react, continued the approach and the aircraft stalled, killing both on board.
    My examiner was under investigation when I did my test and he gave me a very tough time. He told me afterwards that he was not going to pass anyone unless he had made double-sure that he or she would be up to it. He was very bitter about the investigation. He had been blamed for passing Munson with very little total experience, but there was no statute to limit a pilot from becoming type rated due to low total flying experience. I met the same pilot years later in Tucson where I did my Learjet rating. He was a former examiner then, the FAA had indeed pulled his examiner’s qualification.

  • IIRC (as it has been a very long while), the lower minima for uncontrolled airspace in the US are very similar to what you cite for the Netherlands. The relevance of “lots of open space” is that the number of public airFIELDS (not just airports) is massively outnumbered (~3:1 in my memory) by private strips; many of these are remote enough that a pilot would have to get seriously lost (or climb to what is known to be an illegal altitude for the conditions) to infringe on controlled airspace.

    Your discussion of Munson’s crash buttresses my case; he was a seriously inexperienced pilot who got overloaded. Disorientation is one contributor to overload, as is having been in disorienting conditions; somebody being careless in front of you is another.

  • Chip, yes that is now clear.
    And another “yes”: Munson was indeed, in your own words, “a seriously inexperienced pilot who got overloaded.”
    Insofar as I remember, the examiner (his name must be still somewhere in my old log books) was blamed for not putting some limitations on Munson. He told me that there was no regulation to base such a limit on. A candidate either failed, or passed. No “wriggle room” to restrict the successfully newly minted holder of the type rating.
    The stunt that he pulled on me was: At the end of the check ride we did a few circuits, including some touch-and-goes. When we were on the last one he asked if it would be OK if we would make another one, whereupon he would take control because he, as an examiner, did not get a lot of handling himself. I agreed (of course).
    We were on very, very short finals, full flaps, just about to set down. I had already closed the power levers, expecting the call “my controls”.
    But instead, he suddenly called “go around”. I reacted whereupon he pulled the power back on one engine.
    A triple whammy: I expected him to take over, instead he gave me a totally unexpected wave-off AND a simulated engine failure.
    This, and he admitted it in his post-flight briefing, was way beyond the requirements. But I handled it and he was satisfied that he could sign me out. Which led to his explanation about Munson.

  • Vifor debunked the lack of experience and spatial disorientation theory. The pilot (Petrescu) WAS experienced, an IF rating and had completed lots of international trips.
    So it seems that the only theory that is plausible is a sudden malfunction of the major instruments.
    Maybe it will be revealed?

  • Nobody mentioned the auto pilot.I fly a PC 12 NG and if you configure the plane properly for take off. Have your altitude pre selected, with your speed bug set its impossible to spin in if you hit the AP button. The wings level, you pitch up for altitude and your speed is set so you don’t stall. You control the direction with the heading bug. Failing this yes somthing happened, we will just have to wait and see

  • Full trim down runaway…
    Anyone flown a PC12 ?
    Did you notice it lack a trim wheel on the pedestal ?
    Are you aware of the trim runaway procedure on a PC12 ?
    It is a serious possibility …

  • Elevator linkage disconnect would cause the decent symptoms as described, uncontrolled descent with no ability to recover. Turns in an attempt to see if first rush or wind shear was causing the loss of altitude. This would have been psychologically overwhelming and numbing to the pilot, enough so to forgo radio communications once he discovered the had no control over the planes descent.
    I rule out instrument failure because he did not report it in his last communication when asked why he deviated. I don’t think he knew what the problem was yet at that point. If he was suffering physically I think he would have reported it in his last communication. Instead he responded short with “No” and no explanation. An indication that perhaps he was trying to diagnose what was going on at a critical pace. Human factor in the midst of the unknown can create mild shock resulting in comm block.

    So sorry to hear about he and his family. A tragic loss.

    • Something is wrong . He should have committed to his planned fligh as of the METAR he would have been free and clear of weather above 5000 feet . But just at that altitude he went unplanned and unprepared to turn back to the airport ? Perhaps someone was I’ll in the plane making him careless in flying the airplane? In any case his flight path was strange as he overshot his turn . Some serious stress and inability to have contacted the tower for vectors and assistance . Flying the AP in a PC-12 will not normally kill you .

Post a comment:

Your email address will not be published. Required fields are marked *


This site uses Akismet to reduce spam. Learn how your comment data is processed.