TBM900 out of controll roll right

[Sleepy81]

So, was on approach to LOWI. 13000 feet, turning from TULSI to RTT on AP with VPATH selected to hit RTT at 9500. Shortly after the right turn started, the aircraft abruptly and with no warning continued to turn right at full lock. Nosedived and hit the ground. I tried to take manual controll, ensuring AP was off, but I could not move any flightcontrols. I fly with pilot yoke hidden, but saw copilot yoke fully turned to the right, and it did not move at all when I moved my physical joystick. 

Wx was CAVOK, no icing (Pitot heat sw both on regardless). Else the flight was normal, with only a few sec when I hit above 100% max torque (should  not cause this problem regardless). Also, aircraft maintenance was showing all components brand new prior start from ETAD, so should not be any catastrophic failures, even though it feelt like the right wing broke off based on the violence of the maneuver.

Any ideas as to what happened? Attached X-plane log.txt and the TBM900 flightlog if that helps.

Else: fly in VR (Pimax 8KX), 5800X, 3080TI, 32GB DDR4, Win10.

log_220113_101634_ETAD.csv Log TBM900 out of controll.txt

[Sleepy81]

OK, so I played NTSB on this flight and it looks like the incident happened about 12:06:45. In the .csv the bank angle increases above 30 deg at that time, and continues to increase to inverted and beyond. I am a bit confused regarding the different roll coloums. One says Roll and is in line with that I saw. The other says RollC, and could be Roll Commanded by autopilot maybe? That pegs at 30 deg and stays that way until impact. I noticed that the FD was somewhat "crocked" during the flight. For about 80% of the flight it was tilted about 5 deg to the left, but the aircraft AP seamed to ignore that and followed the route fine with wings straight and level. The "tilt" didn not correlate with any crosswind corrections, and the AP flew straight and level with the correct crab to compensate anyways.

Also se that the AP suddenly changed from NAV/ALTS to ROL/NONE and then ROL/ALTS. That does not match what I saw in the cockpit. The AC was on AP NAV/ALTS when it happened, and the AP disconnected (light went out on the AP panels AP button) as far as I remember. So something weird was going on with the AP here...

 

Attached XLSX file if that helps.

TBM900 crash.xlsx

[Goran_M]

Have you checked the maintenance manager for damage to the airframe.  It looks as though you have broken control linkage.  I know you say everything checked out prior tot he flight, but it's a good idea to check post flight also.  A broken linkage can happen with faulty terrain, with the aircraft "dropping"

[Sleepy81]

I tried to check it after the crash, but all it said for all the components (including airframe) was “aircraft too damaged for repair” or something to that effect. Is/was there a way for me to check what was broken  after the crash, when the entire airplane suffered like 250Gs and everything is broken?

I do run Ortho4XP there, but then again I was about 10.000 feet above the terrain (at least the visible terrain (it was CAVOK at the time), so unless there are some calculations of terrain that is wrong (and not visible) it shouldn’t be that. 
 

Maybe will not figure this one out. But I’ll report back if it happens again. 

[Goran_M]

20 hours ago, Sleepy81 said:

but all it said for all the components (including airframe) was “aircraft too damaged for repair”

That's it.  You need to create a new airframe.  Without seeing the damage in the maintenance manager, it's a sure bet the control linkages were broken, and very likely, the pressure vessel was broken.  

[Sleepy81]

Ok, guess I had a catastrophic failure then, even though the aircraft was brand new and had only flown about 4 hrs. Seams kind of strange, but if you guys simulate random catastrophic failures, even the ones with extremely large MTBFs, then I guess I could have drawn a (un)lucky number. It for sure was not overstressed by maneuvering (on AP entire flight) or overspeeding. 
 

Still does not explain why AP commanded 30deg right bank the entire time, and why the copilots yoke was fully deflected right, and did not move when I moved my joystick. To me it looks more like there was a AP issue, and the AP commanded full right aileron, and somehow locked up when it reached 30 deg AOB. 
 

But if this happens again I still don’t understand how to check what went wrong. So assume it happens again: AC crashes, I go to the maintenance manager and all component’s say “too damaged for repair” (ie.  no way of knowing what broke first.), do I then create a new airframe and then go into the maintenance manager and it will show which components where broken prior the crash?

Edited January 15, 2022 by Sleepy81

[Goran_M]

The catastrophic failure can happen from incorrect terrain data.  This is not something we can control, and I strongly suspect this is what happened in your case.  Perhaps flag the spot as one to avoid, and load up at another gate.

The AP/control issue is the Autopilot trying to correct it's heading with a broken linkage.  

Another thing to check for is "accelerated wear".  Make sure this is turned off.

[dbw11]

On 1/15/2022 at 6:22 AM, Goran_M said:

That's it.  You need to create a new airframe.  Without seeing the damage in the maintenance manager, it's a sure bet the control linkages were broken, and very likely, the pressure vessel was broken.  

I have autopilot issues too (noted above) and I have done the new airframe creation route and the problem remains.

[Goran_M]

Does it happen at every airport?  Does it happen every time you load the aircraft?  

Try to load it at SCIP and see if the problem remains.