JOWinTokyo wrote:
I'm not sure of the significance of that. Could it be an indication that maybe they were in a big hurry to get the plane on the ground? A fire on board for example?
fpetrutiu wrote:
holczakker wrote:
Video of the crash:
https://www.linkedin.com/embed/feed/upd ... 2267472898
No high vertical speed, no fire seen from the outside.
Looks like CFIT to me. Lined up with the runway, looks like stable approach under minimums to break out from the clouds, impacted the small hill before the runway as they did not have enough thrust to go around by the time they realized they were about to impact trees.
I'll answer both in the one reply. By the FAF, they were meant to be at 2,700ft AMSL, which they were. However, they were travelling at 230kts indicated then. The FAF is 6.2nm from the runway. Generally at that point, you want to be 160-180kts by then (subject to speed restrictions, of course). From the IAF and IF (17nm and 11nm from the runway), they were meant to be flying at or slower than 230kts and 210kts respectively. The aircraft exceeded those speed restrictions, and actually made the aircraft overshoot the centreline for the ILS. So from the beginning, they were already going too fast.
By the time they reach 1,000ft, operators state usually that the aircraft MUST be stable if flying in IMC, otherwise a go around must be initiated. It was IMC in Vilnius at the time. The aircraft was almost certainly not. To be considered 'stable', the aircraft MUST be at landing speed (there is a leeway of being below VREF no more than by 5 knots and no faster than 20 knots above VREF). The aircraft MUST be on profile, both laterally and vertically. Vertically for an ILS, you should not be more than 1 dot below or above the glide by then and anytime beyond then. The aircraft MUST be configured to land (so landing gear down, flaps in landing configuration, speedbrake stowed). Unless briefed, vertical speed CANNOT be greater than -1,000fpm. Pitch and bank must be appropriate too. They were travelling at 200kts indicated when at that 1,000ft gate, which is certainly too fast to be at a landing speed and too fast to have their flaps in a landing configuration. The aircraft was certainly not stable at 1,000ft.
By 1,000ft and thereafter, their groundspeed should have been somewhere between 120-150kts GS with a vertical speed somewhere between -600fpm to 750fpm.
When they joined final, their groundspeed was around 200kts. On a 3-degree glide, like runway 19, target vertical speed at that GS should be -1,000fpm. The aircraft was descending faster than that for most of the final approach segment. As GS decreased slightly, VS should have decreased slightly too to around -850fpm with a 170kts GS. The VS did decrease to that at times, but did not stay there was subsequently increased again. So by the sounds of it, the aircraft was already below glide for most of the final approach segment.
By 500ft, the crew should have been visual with the runway. I can't find a NOTAM that indicates that the PAPIs were out-of-service.
In the last 30 seconds of ADS-B data, groundspeed dropped by quite a bit to 145kts, with vertical speed increasing to as much as -2,000fpm, before reducing to -1,500fpm right before the end. The headwind component should have been around 20-25kts by then (if no windshear was present). 30 seconds beforehand, GS was at 159kts and VS at -1,000fpm.
As reported elsewhere, there were no issues reported by the crew to ATC. If a fire did occurred, a "MAYDAY" would have been declared with track-shortening given to the aircraft. Neither occurred. The crew was handed off to Tower, in which they acknowledged. However in the LiveATC audio feed, the crew can't be heard contacting the Tower. Both the Tower controller and the Approach controller tried twice to contact the aircraft and also separately issued landing clearances to the aircraft twice.
Footnote: I am not blaming the crew, nor the operator, or anyone really. At the end of the day, pilots are products of their training departments/institutions. Mechanical machines, contraptions, and technology are all prone to malfunctions and errors. I am just stating the data that is available and what the data should look like. I am interesting to see what will be found on the black boxes when they get readout and analysed.
