TBM 900 down, 2 fatalities, Truckee, 3/30

Kenny Phillips

Kenny Phillips

Final Approach
Joined
Jul 29, 2018
Messages
5,634
Display Name
Display name:
Kenny Phillips
Near the airport whilst attempting to land, didn't appear to hit any houses.
 
KTRK 310047Z 00000KT 1/2SM SN OVC009 M01/M01 A2964

KTRK 310155Z AUTO 00000KT 1/2SM SN OVC009 M01/M01 A2966 RMK AO2 P0003

TAF KTRK 301755Z 3018/3118 VRB05KT 3SM BR VCSH BKN010
FM302000 03007KT 2SM -SHSN BKN020
FM310600 31005KT 1/2SM SN BKN010
FM311600 VRB05KT P6SM VCSH BKN020
 
He was on the missed and looks like he was doing 145 kts when they lost him my guess is ice , FIKI doesn’t save you when it gets really bad and the auto pilot should have been flying the plane sad story though , it’s hard for us to accept these kind of accidents sometimes, totally capable aircraft and pilot but they happen , fuel should not have been a issue unless he down loaded fuel for his passenger load which seems to happen at times
 
Autopilot should have been flying the plane? If it was carrying ice? No thanks. One word comes to mind on that. Roselawn.
 
Last edited:
I have flown 1000 s of miles in icing conditions with the auto pilot on in the winter it happens every day
 
The thing that jumps out at me is the turn. Very little climb, should have been on the way to 12,000 in the turn. Instead, it was fairly level with a quick turn rate in the first 90° and an accelerating groundspeed. The turn relaxed quite a bit, still fairly level, then reversed (while still accelerating) just before control (or orientation) was lost.
 
The turn looks ok to me and he was climbing 1000 fmp at 145 kts , the more I look at it maybe he did have a fuel problem but again sad story
 
tbroadhurst said:
I have flown 1000 s of miles in icing conditions with the auto pilot on in the winter it happens every day
Click to expand...
I wasn’t making a general statement about using the autopilot when flying in icing conditions. I have plenty of experience flying in ice and if I were on a missed approach with an immediate climbing turn carrying ice in a small turboprop with boots, I’d prefer to hand fly the airplane.
 
tbroadhurst said:
The turn looks ok to me and he was climbing 1000 fmp at 145 kts , the more I look at it maybe he did have a fuel problem but again sad story
Click to expand...
It vacillated between 6900 and 7100 while accelerating throughout the ~40 seconds after initiating the missed. I wouldn’t call it much of a climb at all.
 
I'll assume that has an inertial separator, and it was being used, but we'll likely know before long.
 
I fly the RNAV 20 frequently with IR students. The missed is a full 180 degree right turn with terrain all around. Sounds like a classic got-slow, stall-spin. Perhaps the airplane was never cleaned up, and had ice, but the visibility is almost certainly a factor. It was snowing last night with poor visibility there but much better in Reno.

News is reporting “moderate snow falling in the Truckee area at the time of the crash with visibility of a half-mile, though winds were light."

That sounds about right. LP minimums are one mile visibility for Category B.

So frustrating that this continues to happen.
 
DesertNomad said:
I fly the RNAV 20 frequently with IR students. The missed is a full 180 degree right turn with terrain all around. Sounds like a classic got-slow, stall-spin. Perhaps the airplane was never cleaned up, and had ice, but the visibility is almost certainly a factor. It was snowing last night with poor visibility there but much better in Reno.

News is reporting “moderate snow falling in the Truckee area at the time of the crash with visibility of a half-mile, though winds were light."

That sounds about right. LP minimums are one mile visibility for Category B.

So frustrating that this continues to happen.
Click to expand...
Agree with all, but if the data are correct it looks like the entry to the loss of control was a little more like a spiral than a pure stall/spin. Of course, that could have been exacerbated by ice and/or wing loading into an accelerated stall. But the groundspeed of 140-150 knots around that time, where the left turn begins, stands out.

Consider the changes in acceleration: an initial climb, then a tight, level right turn, followed by a gentle, accelerating, level right turn, followed by fairly straight and level, still accelerating. Then a gentle left turn and sharp left with a sudden dive.

Could be a lot of other things, but sadly, the pattern drawn by the data (again, assuming it’s correct) is unfortunately a little too familiar.

I’d like to know why the climb stopped, why the tight initial turn rate on the missed. Makes me consider the autopilot modes in use (or not) and what was in the nav system (or not) at the time.
 
blancolirio channel report on this accident
 
That sure sounds like he stalled it , those planes have such wonderful avionics that could fly the plane a lot better than they did , I wonder why they don’t use it , all he had to do was hit the go around button and work the lever let the AP and flight director fly the plane , sad story and it can be a very unforgiving experience, a lot of us have been real close and we were lucky some aren’t
 
tbroadhurst said:
That sure sounds like he stalled it , those planes have such wonderful avionics that could fly the plane a lot better than they did , I wonder why they don’t use it , all he had to do was hit the go around button and work the lever let the AP and flight director fly the plane , sad story and it can be a very unforgiving experience, a lot of us have been real close and we were lucky some aren’t
Click to expand...
That plane should have had "envelope protection" to prevent stalls, but it does allow one to fly the plane into the ground.
 
tbroadhurst said:
That sure sounds like he stalled it , those planes have such wonderful avionics that could fly the plane a lot better than they did , I wonder why they don’t use it , all he had to do was hit the go around button and work the lever let the AP and flight director fly the plane , sad story and it can be a very unforgiving experience, a lot of us have been real close and we were lucky some aren’t
Click to expand...
It's not as easy as that. A go around in IMC is one of the hardest maneuvers in flying. The transition from getting ready to land and climbing into
the soup is a big one. In my view, it needs to be a well-choregraphed sequence of events that should be practiced regularly. The first thing
(after smoothly applying takeoff power and TOGA) is to go from landing flaps to takeoff flaps. Then you bring the gear up (after positive
rate) and accelerate while making sure that you fly into that flight director - you must believe that the ~10 degrees pitch up or so needs to be
complied with - you can't let your mind fool you because you feel things differently. You must trust your instruments because spatial disorientation
in that phase can easily happen due to the rapid transition. After 10 seconds or so, you take out the approach flaps and go to flaps zero. All the
while you're either hand flying the plane (while making that 180 degree turn in Truckee) or you put your autopilot on at ~ 500 feet AGL - depending
on the type of autopilot. You also must remember that TOGA (at least in the Garmin G3000) will automatically put the plane in NAV/FMS mode - so
the published missed is automatically followed by the flight director. I believe it you hit the NAV button, you'll probably undo that / turn that off.
And that'll make it even harder while you're trying to figure out what went wrong. And you might not be able to figure anything out if you're
close to being overwhelmed with what you're currently doing. That's why practicing this is so important.

This is not an easy endeavor - as I said above, it must be practiced. And that's what I do - I try to 'build' that sequence into my muscle memory so
I hopefully don't wind up like this TBM.

I hate to say it, but if you only do this twice a year (the 6/6/6 rule and recurrent), you're asking for trouble in the real world. That's because
I do not think that's enough in a high-performance aircraft.

And yes, it could have been ice too but I haven't watched blancolirio yet - that would just make it harder / worse.

PS: I'm not just making any of this up. I was flying with someone who I knew very well (I was in the right seat) who lost
control of the airplane on in a missed approach at Nantucket in a Baron. I took over (because I was flying a lot
more than him back then) and diverted to Providence. On the ground in Providence, he admitted that I saved
his life - and my own too. I promised myself after that that I would build these procedures into
my muscle memory - because that could easily have been me if I slacked off and flew in IMC without
being ready for the missed approach - or the myriad of other things that come alone with instrument flying in
real weather. The problem with Nantucket was the the weather was fine elsewhere - it's just that fog rolled
in as it often does at "fog island." That's why we must always be ready for these scenarios - even if we think /
the forecast says the weather will be OK at the destination.
 
Last edited:
The 960 is highly automated with auto throttles. I'm wondering that if pushing the toga button with the AP engaged will initiate a climb and add power automatically?

I'm thinking a tail stall will be the ultimate culprit.
 
PaulS said:
The 960 is highly automated with auto throttles. I'm wondering that if pushing the toga button with the AP engaged will initiate a climb and add power automatically?

I'm thinking a tail stall will be the ultimate culprit.
Click to expand...
It does.
 
If you push the GA button, go to takeoff thrust, and let the airplane fly the missed with no further input, what airspeed will it fly? I see the missed approach for the RNAV 20 has a max speed of 200 KIAS until after the initial turn.
 
It initially climbed several hundred feet while at a pretty slow groundspeed. It turned really tightly initially (faster then the autopilot should allow), then more gently, then for 10-15 seconds it was almost straight and level while accelerating to 150kts before it suddenly turned left and departed controlled flight. If we’re considering stalls, the ability to make that initial climb at the much slower groundspeed (reminder that the surface wind was nil), followed by a tight but successful turn at a moderately slow airspeed, doesn’t align with the airspeed being much higher a minute later with the climb and turn being nearly nonexistent, then a stall occurring.

However, a tail stall could make sense if it was concurrent with raising the flaps at that moment. That’s the only aerodynamic explanation I could see for it seemingly having plenty of performance then suddenly not.

But to me there are a lot of things that potentially point to the autopilot not being activated in the missed, or it (or the nav system) not being in a mode that enabled it to fly the published missed, with a high potential for spatial disorientation in the moments that followed.
 
Flying Keys said:
However, a tail stall could make sense if it was concurrent with raising the flaps at that moment. That’s the only aerodynamic explanation I could see for it seemingly having plenty of performance then suddenly not.
Click to expand...
If it happened with a configuration change, a tail stall would happen when adding flaps, not retracting them.
 
MauleSkinner said:
If it happened with a configuration change, a tail stall would happen when adding flaps, not retracting them.
Click to expand...
Ah, right. It makes it even less likely that it was the case. I’m just trying to make the mental leap that a stall led to the departure and not the other way around. I missed a step.
 
MauleSkinner said:
If you push the GA button, go to takeoff thrust, and let the airplane fly the missed with no further input, what airspeed will it fly? I see the missed approach for the RNAV 20 has a max speed of 200 KIAS until after the initial turn.
Click to expand...
Good question - I forgot to add FLC (flight level change) into what I said above. You can either go VNAV/FLC in what I fly.
Or just go FLC and dial in 200 knots for what I do. But the dial is still cumbersome when you're really busy with the other
stuff that I mentioned. If you go VNAV and then FLC, the airspeed will automatically go to 200 because that's in the program.
But, like anything, there's no panacea - VNAV/FLC can be dangerous if there's an engine failure, because, if that happens, I'm
supposed to fly out at approx 114 knots (Vapp) and that MUST be dialed into the FLC / VSI roll 'wheel.'

Even with the TBM 960's autothrottles, I imagine that you still need to disengage the autopilot and then turn it back on at
about 450 to 500 feet AGL. At least that's what I need to do without autothrottles.

The bottom line is that there's a lot going on. And, quite honestly, I think that autothrottles could mess up someone like
me whose so used to doing it the other way for years.

Muscle memory and habit are a funny thing. Sometimes, when I'm calling for a clarence at my airport in Florida, the
first thing I say is "Westchester clearance...." My brain automatically says that sometimes (when I'm just going through
the motions) because that's where I learned to fly....and that's where I've done a lot of my flying. I think that they
laugh at that a bit...but it's definitely habit / 'brain muscle memory.'

PS: I even had a really good golf instructor tell me that, under pressure, your 'new golf swing' (after taking a few
lessons) will completely fall apart unless it's built into muscle memory. That's why there must be consistent
practice of the proper repetitions - to build a complex sequence into our muscle memory.
The problem with flying is that it's not golf - the stakes are much higher.
 
CaptainDoug said:
Good question - I forgot to add FLC (flight level change) into what I said above. You can either go VNAV/FLC in what I fly.
Or just go FLC and dial in 200 knots for what I do. But the dial is still cumbersome when you're really busy with the other
stuff that I mentioned. If you go VNAV and then FLC, the airspeed will automatically go to 200 because that's in the program.
But, like anything, there's no panacea - VNAV/FLC can be dangerous if there's an engine failure, because, if that happens, I'm
supposed to fly out at approx 114 knots (Vapp) and that MUST be dialed into the FLC / VSI roll 'wheel.'

Even with the TBM 960's autothrottles, I imagine that you still need to disengage the autopilot and then turn it back on at
about 450 to 500 feet AGL. At least that's what I need to do without autothrottles.

The bottom line is that there's a lot going on. And, quite honestly, I think that autothrottles could mess up someone like
me whose so used to doing it the other way for years.

Muscle memory and habit are a funny thing. Sometimes, when I'm calling for a clarence at my airport in Florida, the
first thing I say is "Westchester clearance...." My brain automatically says that sometimes (when I'm just going through
the motions) because that's where I learned to fly....and that's where I've done a lot of my flying. I think that they
laugh at that a bit...but it's definitely habit / 'brain muscle memory.'

PS: I even had a really good golf instructor tell me that, under pressure, your 'new golf swing' (after taking a few
lessons) will completely fall apart unless it's built into muscle memory. That's why there must be consistent
practice of the proper repetitions - to build a complex sequence into our muscle memory.
The problem with flying is that it's not golf - the stakes are much higher.
Click to expand...
Sounds like a good chance that “letting the automation take care of it” could have been perceived as “too much work.”
 
MauleSkinner said:
Sounds like a good chance that “letting the automation take care of it” could have been perceived as “too much work.”
Click to expand...
In the M600 implementation of the G3000 at least, press of the GA button results in a specific climb angle, and TOGA power.
 
dans2992 said:
In the M600 implementation of the G3000 at least, press of the GA button results in a specific climb angle, and TOGA power.
Click to expand...
I presume by “climb angle” you mean “pitch attitude?” What airspeed does that give you after you clean up?
 
Ah, yes, sorry, pitch attitude.

I’ll have to check what speed that yields once clean, I move over to FLC pretty quick.
 
Maybe someone with TBM 960 experience can chime in and describe how the automation will work.

The initial step of the missed approach is simple. Climbing right turn to an altitude well above the minimums on a heading that is almost 180 degrees opposite of what you were flying. If a pilot is proficient in using the automation to its fullest, everything will work. Unless there is another issue working against it (say an autopilot disengaging on its own due to servo loads reaching limits with an out of trim airplane; due to ice for example) or one of the automation features failing. If a pilot is proficient in hand flying he could say "At the MAP I'll select G/A, kick off the autopilot, execute a climbing right turn and then reactivate the automation when I'm ready" and it should work. Unless there is another issue working against him (say spacial disorientation). Either way it's performed, proficiency matters. As mentioned earlier there is a lot going on, and you have to be ready.
 
MauleSkinner said:
I presume by “climb angle” you mean “pitch attitude?” What airspeed does that give you after you clean up?
Click to expand...

TrueCourse said:
Maybe someone with TBM 960 experience can chime in and describe how the automation will work.

The initial step of the missed approach is simple. Climbing right turn to an altitude well above the minimums on a heading that is almost 180 degrees opposite of what you were flying. If a pilot is proficient in using the automation to its fullest, everything will work. Unless there is another issue working against it (say an autopilot disengaging on its own due to servo loads reaching limits with an out of trim airplane; due to ice for example) or one of the automation features failing. If a pilot is proficient in hand flying he could say "At the MAP I'll select G/A, kick off the autopilot, execute a climbing right turn and then reactivate the automation when I'm ready" and it should work. Unless there is another issue working against him (say spacial disorientation). Either way it's performed, proficiency matters. As mentioned earlier there is a lot going on, and you have to be ready.
Click to expand...
According to the POH, TOGA with autopilot gives 10° pitch up and max torque until 500’ AGL, when envelope protection kicks in. Same as takeoff. I don’t know what speeds and climb rates that yields, as you said, someone with experience in type will need to chime in. But the AP should also limit the turn rate to 3°/sec and looks like it was beyond that initially.
 
Hengelo said:
How will they know?
Click to expand...

By studying the data from the Avionics. I imagine in a tail stall, the pitch angle of the aircraft will not match the expected pitch angle based on the elevator input.

I'm familiar with the G1000 Cirrus, which will pitch to a climb attitude and fly straight. There are no autothrottles, so you need to apply full power, clean the aircraft up, and engage either heading or nav mode depending on what your instructions are. I assume the 960 is similar, except with autothrottles.
 
Flying Keys said:
But the AP should also limit the turn rate to 3°/sec and looks like it was beyond that initially.
Click to expand...
Which is why I’m wondering where the speed would end up. I see quite a few pilots who haven’t got the automation management skills to ensure that the airspeed wouldn’t exceed the limits of this procedure. I also see two ways pilots deal with that…either automation off and hand-fly, which could result in what you noted, or leave the automation on and hope there’s enough margin built into the procedure that they don’t hit anything.
 
CaptainDoug said:
It's not as easy as that. A go around in IMC is one of the hardest maneuvers in flying. The transition from getting ready to land and climbing into
the soup is a big one.
Click to expand...
Agreed. During training we do a lot of them. Mostly simulated rather than because we flew to minimums and saw nothing. In real world GA, they are very rare. So this busy workload time is something we are not very good at.

Either way, the accident strikes me as the worst kind of missionitis, coming home to a very familiar airport in lousy conditions.
 
You must log in or register to reply here.
Back
Top