Taiwan twin engine crash. Did one engine flameout or two?

I don't know ATRs from nothing.....But the King Air I flew had "Rudder Boost." It applied itself automatically during an engine failure.

Something else was going on there.....poor training, more than one failure, overloaded, poor maintenance.....something.
 
bobmrg said:
Knowing no more that what I saw on TV, I thought it was a classic Vmc roll, the kind with which every nascent multiengine pilot must deal during training and on the checkride.
Click to expand...

Honest question Bob. How can you tell it is a VMC rollover vs the beginnings of a stall/spin?

With the initial rate of descent and nose-high attitude, I am not thinking he had anywhere near full power on the good engine when the left wing dropped.

I could be wrong, but it seems to me that he had reduced the power on the good engine in order to maintain control and then got too slow and stalled it.
 
kgruber said:
Something else was going on there.....poor training, more than one failure, overloaded, poor maintenance.....something.
Click to expand...

Could be all if the above if there is much truth to the rumors about company culture.
 
Jim_R said:
No, the left is definitely turning slower. Blurrier=faster. Right=blurrier. (We are looking at the belly of the plane.)
Click to expand...


Correct, I got my left and rights mixed up! I edited my post. Sometimes I wonder how I am allowed to fly airplanes.
 
Fearless Tower said:
Honest question Bob. How can you tell it is a VMC rollover vs the beginnings of a stall/spin?

With the initial rate of descent and nose-high attitude, I am not thinking he had anywhere near full power on the good engine when the left wing dropped.

I could be wrong, but it seems to me that he had reduced the power on the good engine in order to maintain control and then got too slow and stalled it.
Click to expand...

I noticed that as well. But in some aircraft the Vmc is very close to the clean stall speed. So he could have been mushing along right at Vmc. It looks like he only has a little bit of flaps in.

I would think it is possible to be above Vmc, but slow enough you are behind the power curve and can't climb OEI without first trading altitude for speed.

A search shows clean stall speed for a 310R at 79 IAS while Vmc is 80 indicated. If you were heavy, one engine feathered and one at full power at 5 kts above stall speed, you would probably be sinking like a rock, but still in controlled flight.

Based on that, a failure to maintain enough airspeed to climb OEI could explain what we see in the video
 
Last edited:
Henning said:
It's spinning, but at a much slower speed. It's most noticeable in the stills where you can see the blur arc of each blade. Both are rotating, but the arc of the left is shorter.
Click to expand...

Okay I see what you're saying, the left one is definitely slower.
 
Looking at the picture that Ren posted, the blades on the left engine do look sorta feathered, at least they look to be at a high AoA. The prop is still spinning but it's a free flow turbine so it can still spin.

I don't know much about ATRs but I'd still be surprised if he doesn't have autofeather. So it's rather unlikely for the prop not to be feathered as the autofeather needs to fail or be turned off and the pilot needs to forget to feather the prop as well.

I think the prop is probably feathered. Although it is rotating a little fast for a feathered prop...

Would love to see post crash pictures of the levers in the cockpit and the blades.
 
Busflyer said:
View attachment 37936
Click to expand...
If this data is anywhere near accurate, they got up above 1000' before starting to descend. Notice, when the descent starts, the airspeed decays.

They either did not/could not feather the bad engine or had two bad engines, or he pulled power off the good engine to avoid VMC roll and did not subsequently lower the nose.
 
Perhaps he was just overweight and couldn't maintain altitude on a single engine.
At this point we just got to wait for the report.
 
MachFly said:
Perhaps he was just overweight and couldn't maintain altitude on a single engine.
At this point we just got to wait for the report.
Click to expand...

Knowing what I know about the Taiwan economic methods, I was going to opine that possibly the plane was overloaded, and/or aft CG. Agreed that this would cause further issues in a SE condition. Will be interesting to see what comes of the investigation.
 
docmirror said:
Knowing what I know about the Taiwan economic methods, I was going to opine that possibly the plane was overloaded, and/or aft CG. Agreed that this would cause further issues in a SE condition. Will be interesting to see what comes of the investigation.
Click to expand...
If it was overloaded, it wasn't PAX. It was not a full flight by any means.
 
Yesterday morning when I first saw the video, I immediately grabbed the remote and backed it up to move it one frame at a time. It seemed pretty clear that the lower prop was not turning while the one on the high wing was still turning.

Those DVR gadgets are pretty handy sometimes.
 
Fearless Tower said:
Honest question Bob. How can you tell it is a VMC rollover vs the beginnings of a stall/spin?

With the initial rate of descent and nose-high attitude, I am not thinking he had anywhere near full power on the good engine when the left wing dropped.

I could be wrong, but it seems to me that he had reduced the power on the good engine in order to maintain control and then got too slow and stalled it.
Click to expand...

I can't. That is why I weasel-worded my post.

Bob
 
This is a rookie question and I'm looking for an honest answer:

"If you fly a twin and have to turn off the secondary engine to prevent a VMC roll why would you have a Twin in the first place" Wouldn't it just be better with a single engine?

I'm still trying to understand the benefits of flying a twin engine airplane.
 
FloridaPilot said:
This is a rookie question and I'm looking for an honest answer:

"If you fly a twin and have to turn off the secondary engine to prevent a VMC roll why would you have a Twin in the first place" Wouldn't it just be better with a single engine?

I'm still trying to understand the benefits of flying a twin engine airplane.
Click to expand...

There isn't a 'secondary' engine. To prevent a VMC roll you just have to carry enough airspeed so the rudder can have the power it needs to overcome the asymmetric thrust.
 
I have to think there was some kind of mechanical malfunction related to the auto-feather and/or yaw damper systems.

That's a pretty modern airplane, losing an engine on takeoff should require only new underwear for the crew, it should not just nose up and roll.
 
FloridaPilot said:
This is a rookie question and I'm looking for an honest answer:

"If you fly a twin and have to turn off the secondary engine to prevent a VMC roll why would you have a Twin in the first place" Wouldn't it just be better with a single engine?

I'm still trying to understand the benefits of flying a twin engine airplane.
Click to expand...

Twins can be safer IF you fly then properly.

Above your minimum controllable airspeed for single engine (VMC) you have hope....either climb, maintain altitude or at least minimize rate of descent.

If you get below VMC on one engine, you have to get the nose down and reduce power on the good engine to get back to your VYSE.
 
FloridaPilot said:
This is a rookie question and I'm looking for an honest answer:

"If you fly a twin and have to turn off the secondary engine to prevent a VMC roll why would you have a Twin in the first place" Wouldn't it just be better with a single engine?

I'm still trying to understand the benefits of flying a twin engine airplane.
Click to expand...

You have to think about asymmetric thrust: The engine nacelles are offset from the longitudinal axis, just as a span of oxen is offset from the centerline of whatever they are hauling. If one ox falls down, the wagon (or whatever) will turn toward the fallen ox. Similarly, if an engine fails on a twin there is a momentary yaw toward the dead engine. I said momentary because the pilot's first action is to apply rudder to stop the yaw. The windmilling prop creates tons of drag, further retarding its wing. Virtually simultaneously, the airplane begins to roll toward the dead engine. Stop the yaw and you stop the roll. Certification standards require that the airplane be controllable with no more than 150 pounds of rudder pressure on the good engine side.

Single-engine airplanes have their thrust vector coincident with the longitudinal axis so there is no yaw when the engine quits.

Why a twin? I have had two inflight engine failures, and in both cases I was able to return to my departure airport using the remaining engine. Kinda makes sense, doesn't it?

Bob Gardner
THE COMPLETE MULTENGINE PILOT
 
bobmrg said:
Why a twin? I have had two inflight engine failures, and in both cases I was able to return to my departure airport using the remaining engine. Kinda makes sense, doesn't it?

Bob Gardner
THE COMPLETE MULTENGINE PILOT
Click to expand...

I realize this question might be kind of loaded with a lot of "it depends", but...

On average, about how much time does a pilot of a GA piston twin (think Aerostar, DA42, Baron, and the like) have to compensate with rudder for the yaw, identify, shut down, and feather the dead engine, and whatever else is needed to avoid a VMC roll at the most critical phase of flight, say departure?

Does all that have to be done split-second, a few seconds, 10 seconds, etc?

Does initial and recurrent training really drill this kind of thing into a twin pilot such that he/she can keep those skills almost like second nature on how to react in an engine out situation?
 
LoneAspen said:
Does all that have to be done split-second, a few seconds, 10 seconds, etc?
Click to expand...

Depends on the airspeed at the time of failure. The closer to Vmc the more timely the correction inputs need to be in order to avoid departing controlled flight.

LoneAspen said:
Does initial and recurrent training really drill this kind of thing into a twin pilot such that he/she can keep those skills almost like second nature on how to react in an engine out situation?
Click to expand...

I'm not a twin owner, but it doesn't look like a great track record for the *recreational owners (*don't say 'hobby' around here, the butt hurt pitchforks come out).
 
FloridaPilot said:
This is a rookie question and I'm looking for an honest answer:

I'm still trying to understand the benefits of flying a twin engine airplane.
Click to expand...

Psychological comfort, that's all. 2 engines equal twice the chance of failure.
 
LoneAspen said:
I realize this question might be kind of loaded with a lot of "it depends", but...

On average, about how much time does a pilot of a GA piston twin (think Aerostar, DA42, Baron, and the like) have to compensate with rudder for the yaw, identify, shut down, and feather the dead engine, and whatever else is needed to avoid a VMC roll at the most critical phase of flight, say departure?

Does all that have to be done split-second, a few seconds, 10 seconds, etc?

Does initial and recurrent training really drill this kind of thing into a twin pilot such that he/she can keep those skills almost like second nature on how to react in an engine out situation?
Click to expand...

In cruise, lots of time, it's no factor as you are clean with plenty of extra speed and altitude. On rotation with the gear still in transit, you better work fast, or just chop the throttles and land into what's ahead. That is something you figure out on every take off independently. In the phase of flight these guys were at with the gear up and a thousand feet, you have more than sufficient time to calmy and purposefully go through the caging drill and whatever other reconfigurations are left. Even the lowest performance recip twins out there will get you back to the runway from where he was is flown with the skills you passed your ME checkride with.

This one just has all the earmarks of a disassociative reaction to an emergency.
 
nosehair said:
Psychological comfort, that's all. 2 engines equal twice the chance of failure.
Click to expand...

Yep, but I landed the Travelair on a runway each time, and made repairs for <$100 in a couple of hours and was flying again. Had I been in a single, those oil lines would have cost me an engine at minimum, and likely forced an off field landing. I really like 2 engines at night or when I can't see where I'm gliding into. Day VFR, meh, I'll fly whatever.
 
Henning said:
Yep, but I landed the Travelair on a runway each time, and made repairs for <$100 in a couple of hours and was flying again. Had I been in a single, those oil lines would have cost me an engine at minimum, and likely forced an off field landing. I really like 2 engines at night or when I can't see where I'm gliding into. Day VFR, meh, I'll fly whatever.
Click to expand...

I haven't lost an engine in a single or twin yet, but have experienced things like vacuum pump failures, alternator failures and such that were complete non-events in the twin but would have been potential emergencies in a single.
 
Fearless Tower said:
If this data is anywhere near accurate, they got up above 1000' before starting to descend. Notice, when the descent starts, the airspeed decays.

They either did not/could not feather the bad engine or had two bad engines, or he pulled power off the good engine to avoid VMC roll and did not subsequently lower the nose.
Click to expand...
This is my thought as well. If they still had/regained power in the RH side looks to me like they might have tried to get more clearance beyond the bridge and rolled it. No real proof of that, just a hunch.
 
I could be wrong but I believe all the new atr's have fadec engines with auto feather. Just from the video this looks like a vmc roll due to pilot screw up...assuming the remaining engine was developing full power.
 
LoneAspen said:
I realize this question might be kind of loaded with a lot of "it depends", but...

On average, about how much time does a pilot of a GA piston twin (think Aerostar, DA42, Baron, and the like) have to compensate with rudder for the yaw, identify, shut down, and feather the dead engine, and whatever else is needed to avoid a VMC roll at the most critical phase of flight, say departure?
Click to expand...

Easy answer.... The rest of your life....
 
LoneAspen said:
I realize this question might be kind of loaded with a lot of "it depends", but...

On average, about how much time does a pilot of a GA piston twin (think Aerostar, DA42, Baron, and the like) have to compensate with rudder for the yaw, identify, shut down, and feather the dead engine, and whatever else is needed to avoid a VMC roll at the most critical phase of flight, say departure?

Does all that have to be done split-second, a few seconds, 10 seconds, etc?

Does initial and recurrent training really drill this kind of thing into a twin pilot such that he/she can keep those skills almost like second nature on how to react in an engine out situation?
Click to expand...
Here's the thing: on departure, if you fly the airplane properly and follow your training, VMC should never be an issue.

Most piston twins do not lift off below VMC. Even in the DC-3 (1200 HP per side) we don't lift off below VMC. The only piston airplane I have flown where you do fly it off below VMC is the B-25, but that is a very different animal.

That said, if you respect that and utilize balanced fields for departure, VMC should not be a big hazard. If you climb out at blue line or higher, and reduce your angle of attack should you develop an engine problem, you have plenty of time to go through the identify/verify/feather routine.

In my opinion, the greater hazard with OEI flight is stalling with asymmetric thrust. The majority of twin crashes where there was an engine failure seem to happen during the approach where you aren't using full power on the good engine, but the pilot gets low and slow and loses it.
 
ChemGuy said:
Easy answer.... The rest of your life....
Click to expand...

It would actually be more of an infinity thing, but yeah.

Seriously, it takes less than 3 seconds in most planes to go through the process deliberately. If you are spring loaded for it like on a take off, I can have every thing done in the 310 in less than a second since there are no flaps or cowl flaps to deal with, just gear and feather.

The important thing is you have to get the nose down to accelerate to get onto 'the front side of the power curve' to climb.

Someone has in their sigfile, "To go up pull back, to go down, pull back further." These guys were in the 'further' realm of pitch.
 
Fearless Tower said:
The majority of twin crashes where there was an engine failure seem to happen during the approach where you aren't using full power on the good engine, but the pilot gets low and slow and loses it.
Click to expand...

Probably these pilots are used to flying relatively flat, power on approaches all the time. With one engine out, flying the same approach profile takes a lot more throttle than they're used to.

Little behind the plane, getting slow and low, realizing their glide path puts them short of the runway, then giving it a good handful of throttle and wham, Vmc roll.
 
There was an article "life after V1", which advocated (in piston singles) to takeoff something like this:

1. Right hand on throttles, accelerate to Vr, rotate.
2. Positive rate, hand moves to gear lever, retract gear,
3. Hand moves from gear lever back to prop controls, ready to feather bad engine.

The argument was that with an engine failure at low altitude you do not have time to go through the "verify" step. Simply identify and feather. Setting the heading bug on runway heading would help to more quickly identify the failed engine (bug moves toward running engine). The ball can swing around in turbulence and make the "identify" step take longer.

What do you guys think about this technique?
 
FWIW- I was flying with a guy today that used to fly ATR's for American Eagle. Said if the auto feather didn't work and/or they never got the prop feathered, the chance of climbing was nil. Imagine a 12ft diameter, 6 blade prop windmilling... It doesn't look like the prop is feathered but hard to determine from the dashcam video.


JimNtexas said:
I have to think there was some kind of mechanical malfunction related to the auto-feather and/or yaw damper systems.

That's a pretty modern airplane, losing an engine on takeoff should require only new underwear for the crew, it should not just nose up and roll.
Click to expand...
 
hkyplr18 said:
FWIW- I was flying with a guy today that used to fly ATR's for American Eagle. Said if the auto feather didn't work and/or they never got the prop feathered, the chance of climbing was nil. Imagine a 12ft diameter, 6 blade prop windmilling... It doesn't look like the prop is feathered but hard to determine from the dashcam video.
Click to expand...

Ya but you know these airlines are like mo' money than talent pilots: "We didn't buy this airplane to fly undergross..." :rofl::rofl:
 
Well, latest on CNN is that both engines had issues and the stall horn was going off most of the brief flight.

Sounds like they were attempting to restart the left engine when the right started having issues. I can't help but wonder in their rush to get the left restarted if they did something that affected the right.
 
The blackbox shows that at 1000', the right engine went into "idle" but it still had oil pressure, and then the left engine was shut down manually, so that the aircraft thereafter had no thrust. That's from the BBC.

Edit, a better written story is cited by fearless tower in the preceding post. It does look like the crew stopped the fuel to the wrong engine, leaving them without power.
 
Last edited:
Latest info on CNN is that they had problems with the first engine but then discussed shutting down the other one.

So my money is on these dumforks shutting down the wrong engine.
 
You must log in or register to reply here.
Back
Top