Scary VMC roll video, question for multi-pilots

JimNtexas

JimNtexas

Pattern Altitude
Joined
Dec 21, 2006
Messages
2,259
Location
Austin, Texas
Display Name
Display name:
Jim - In Texas!
I suspect that a lot of pilots are like me when they hear about an accident. I/We immediately look for the thing the accident pilot did that I/we would never, ever do.

I was watching this excellent video from an FAA accident investigator covering what he thinks are the ten most common causes of GA accidents.

The accident that really took me back was at 28:13 into the presentation:

http://youtu.be/Zr6iLRIN234?t=28m13s

This is a reconstruction of a vmc roll in a turboprop twin right after liftoff. I'm looking at this trying to figure out what the accident pilot did wrong. The roll happened so fast I don't think I could have reacted fast enough to counter it either.

I have a theory, but I 'd like to ask pilots with more than than my little bit of multi-engine experience.

Experienced multi-engine pilots, what could this pilot have done differently to avoid this tragic accident?
 
Currently working on my multi and have less than 10 hrs in twins, but looks like he rotated and was airborne prior to achieving redline. From what I've been taught that isn't the best plan.
 
James_Dean said:
Currently working on my multi and have less than 10 hrs in twins, but looks like he rotated and was airborne prior to achieving redline. From what I've been taught that isn't the best plan.
Click to expand...

Blue line.
Dan
 
Engine failure the the worst possible time.
 
Looks like did a quick high AoA rotation before Vmc. I do not know what his configuration, weight, and CG was, but those could have been contributing factors. Given that he couldn't have done anything to avoid that engine failure, looks like he rotated too soon (bellow Vmc) and too abruptly.
What he should have done in that situation was cut power (both engines), drop the nose to regain airspeed, and roll left, all of this needed to be done simultaneously.
I don't know the skill and experience of this pilot, so what I'm about to say does not apply to him directly. I see a lot of GA pilots that like to fly twins for increased safety but the last time they practiced single engine operations was on their ME checkride. If you want to be safer in a twin, you need to be very proficient.


Would sure like to see the NTSB report.
 
James_Dean said:
Currently working on my multi and have less than 10 hrs in twins, but looks like he rotated and was airborne prior to achieving redline. From what I've been taught that isn't the best plan.
Click to expand...
In most light twins Vr is at least redline or greater. There are some higher powered twins where VMC/redline is so high that you have to liftoff before redline.
 
Dan Thomas said:
Blue line.
Dan
Click to expand...
No, red line -- Vmc. Waiting to blue line (Vyse) to rotate is going to cause a host of other problems, starting with, in most cases, wheelbarrowing because the plane really wants to fly and you're holding the nose down on the runway with forward stick/yoke. Other than Part 25 aircraft with computed Vr's (which are always at/above Vmc, and never anywhere near Vyse), we usually rotate twins at Vmc+5 or something like that, and that's well below blue line in every light twin I've flown.
 
Skylane81E said:
Well had he at least made redline it wouldn't have flipped over like that.
Click to expand...
Not necessarily. Redline Vmc as marked on the airspeed indicator is computed with the airplane in up to a 5 degree bank into the good engine -- and I don't know of any manufacturer who uses less than the maximum 5, since that lowers the book Vmc as much as they can using bank. If your wings are level, Vmc can be a good bit higher -- about 9 knots higher, for example, in the Grumman GA-7 Cougar (70 vs 61, by my flight testing). So, if you rotate at redline and one quits, you may not have enough speed to control the yaw because the wings are level.
 
Fearless Tower said:
In most light twins Vr is at least redline or greater. There are some higher powered twins where VMC/redline is so high that you have to liftoff before redline.
Click to expand...

I was taught VMC+5 or faster. You still might not be able to climb away from a failure but at least you won't flip over
 
James_Dean said:
The video clearly showed him rotating and lifting off before achieving red line.
Click to expand...
Yes, it does, and now you see why, in the absence of book performance data for computing Vr, we usually teach rotation at Vmc+5, which usually results in liftoff at Vmc+10.

Of course, if you lose one in a Part 23 light twin with the gear still down and below Vyse, you probably won't be able to accelerate to Vyse before the ground comes up and smacks you in the belly, but if you have the speed to control the plane, at least you'll be under control when that happens, and that's a lot more survivable accident than the one in the video.
 
Important to note is that the red and blue lines are computed in a specific configuration, this configuration is not likely the one you are in, and so the speeds are different than simply what is shown on the dial.
 
Not sure what model of airplane the actual accident plane was, but (assuming Vr was = to or greater than Red Line), the second thing wrong was what appears to be a high AoA on rotation which only exacerbated the problem.

There are some cases where you may need to rotate before VMC extreme short field or aircraft design: B-25 comes to mind), but if you value your life, in those initial moments after rotating/liftoff, the number one priority is smooth and rapid accelaration to blue line before you start climbing out. From the reconstruction, it looks like the accident pilot did not do that.
 
MachFly said:
What he should have done in that situation was cut power (both engines), drop the nose to regain airspeed, and roll left, all of this needed to be done simultaneously.
Click to expand...
I suspect that due to the early rotation and high AoA, when the failure happened, it was already too late to successfully react.
 
ja_user said:
Important to note is that the red and blue lines are computed in a specific configuration, this configuration is not likely the one you are in, and so the speeds are different than simply what is shown on the dial.
Click to expand...

IIRC they are computed for the most adverse condition, so it may be possible under some conditions to fly OEI below redline, not that I'd want to try it:no:
 
Skylane81E said:
IIRC they are computed for the most adverse condition,
Click to expand...
Close. They are computed for a specified set of conditions, some of which are worst case (e.g., aft-most cg), some of which are arbitrary (e.g., 5 degrees bank into the good engine). Bank angle is the one which is probably least understood, although the article by the late Prof. Mel Byington is probably the best discussion about that.

https://www.aopa.org/Pilot-Resource...ne-Out-Booby-Traps-for-Light-Twin-Pilots.aspx

...it may be possible under some conditions to fly OEI below redline...
Click to expand...
In fact, many conditions, and in some cases safely (but not something you want to do other than under controlled conditions with a lot of space between you and the ground).

...not that I'd want to try it:no:
Click to expand...
Since you have a ME rating, you have probably done it already. The Vmc demo is a required Task for that rating, and unless you do it at sea level :eek: if you do it per the PTS, you will be below redline when the yaw occurs. At a typical 3500-5000 AGL, the higher-than-SL density altitude means less than full power on the good engine and that lowers Vmc below the redline on the airspeed indicator. Of course, that assumes 5 degrees bank into the good engine, which if you've read Byington's work, you won't be doing. The PTS calls for bank for "best control and performance", and 5 degrees is a long way from the typical nine degrees for best performance and significantly off the 2 degrees for best control. OTOH, 2 degrees is much better performance and not much farther from best control), and at that attitude, at 2 degrees, you still may be below redline when the yaw break happens. Either way, you just have to remember to pull power on the good engine immediately upon occurrence of the yaw once you run out of rudder.
 
Last edited:
Fearless Tower said:
I suspect that due to the early rotation and high AoA, when the failure happened, it was already too late to successfully react.
Click to expand...

Very likely, I'd still try to correct for it though. In the reconstruction they show that port engine was still providing full power.
 
Skylane81E said:
IIRC they are computed for the most adverse condition,
Click to expand...

Ron Levy said:
Close. They are computed for a specified set of conditions, some of which are worst case (e.g., aft-most cg), some of which are arbitrary (e.g., 5 degrees bank into the good engine). Bank angle is the one which is probably least understood, although the article by the late Prof. Mel Byington is probably the best discussion about that.

https://www.aopa.org/Pilot-Resource...ne-Out-Booby-Traps-for-Light-Twin-Pilots.aspx
Click to expand...

Just to summarize a list of Vmc required conditions.

  • Critical engine failure.
  • Bank angle does not exceed 5 degrees.
  • Vmc must not exceed 1.2 Vs1, and that Vs1 must be for maximum takeoff weight.
  • Most unfavorable weight.
  • Most unfavorable CG.
  • Out of ground effects.
  • Each engine set to max power before the critical engine is rendered inoperative.
  • Trimmed for take off.
  • Flaps in a take off position.
  • Landing gear retracted.
  • Prop levers in a take off position.
  • Heading must stay within 20 degrees.
  • Must not be necessary to reduce power on the operational engine.
  • Maximum 150lb on the rudder pedal.
 
JimNtexas said:
I suspect that a lot of pilots are like me when they hear about an accident. I/We immediately look for the thing the accident pilot did that I/we would never, ever do.

I was watching this excellent video from an FAA accident investigator covering what he thinks are the ten most common causes of GA accidents.

The accident that really took me back was at 28:13 into the presentation:

http://youtu.be/Zr6iLRIN234?t=28m13s

This is a reconstruction of a vmc roll in a turboprop twin right after liftoff. I'm looking at this trying to figure out what the accident pilot did wrong. The roll happened so fast I don't think I could have reacted fast enough to counter it either.

I have a theory, but I 'd like to ask pilots with more than than my little bit of multi-engine experience.

Experienced multi-engine pilots, what could this pilot have done differently to avoid this tragic accident?
Click to expand...

He pulled it off the ground at too low of speed. With the 310 one can add the vortex generators and bring Vmc below stall speed to prevent similar occurrence.
 
In a DA 42 it is reasonable to actually remain on the ground to Vyse. It's a little squirrelly but it works.

In a PA34, you need to remain close in to ground effect after you break ground at about 75 mph (redline is 80), and then I remain in ground effect and don't suck the gear up until 105 (blue line).

ONLY if my spreadsheet shows V1=Vyse (no margins), I will suck them up right at 80 to max out acceleration at the expense of simplicity if I must close both throttles.

The answer is be ready to slam the throttles shut at the first hint of a yaw...at least until the point when landing straight ahead is less surviveable than flying it around and bringing it back.
 
For us single engine guys trying to follow along and trying to learn something - I do understand the concept of how this roll occurs, but I don't know the terminology:

Blue line vs red line?
 
Matthew said:
For us single engine guys trying to follow along and trying to learn something - I do understand the concept of how this roll occurs, but I don't know the terminology:

Blue line vs red line?
Click to expand...

Blue line, painted on the ASI gives one Vyse, best climb on single engine. Red Line, also painted on the ASI gives one Vmc, the minimum speed to maintain control on a single engine in the above defined condition in post 23.
 
I know there are some King Air pilots here, and I know that latter model King Airs (and I'm sure other turboprops) have yaw dampers and auto-feathering systems.

Would an auto-feather system have saved this airplane?
 
JimNtexas said:
I know there are some King Air pilots here, and I know that latter model King Airs (and I'm sure other turboprops) have yaw dampers and auto-feathering systems.

Would an auto-feather system have saved this airplane?
Click to expand...

Below Vmc? Maybe, maybe not, reaction time has a lot to do with it. What we don't know is what type of reactor the guy was. Some people react to these types of events by disassociating, they will die regardless of how much time is available, reference AF 447, they had over 3 minutes to save it in a fully functional aircraft.
 
JimNtexas said:
I know there are some King Air pilots here, and I know that latter model King Airs (and I'm sure other turboprops) have yaw dampers and auto-feathering systems.

Would an auto-feather system have saved this airplane?
Click to expand...


Nothing that I'm aware of that would have cut power to the operating engine which is the only thing that might have saved this guy. I think all bets are off if you pop the nose up 20* while five knots below red line.
 
Matthew said:
For us single engine guys trying to follow along and trying to learn something - I do understand the concept of how this roll occurs, but I don't know the terminology:

Blue line vs red line?
Click to expand...
Officially, from the regulations, "VMC means minimum control speed with the critical engine inoperative." In more practical terms, that's the speed at which you run out of rudder trying to keep the plane from yawing/rolling into the dead engine. The "official" Vmc in the POH is computed as Bruce described above, and marked on the airspeed indicator with a red line (much like Vne up at the top end). Vyse is the best rate of climb speed with one engine inoperative, and Vyse under a specific set of conditions is marked on the airspeed indicator with a blue line. Those conditions for book Vyse (established in 14 CFR 23.69), where the blue line will be placed on the airspeed indicator, are maximum weight and sea level, plus:

(1) The critical engine inoperative and its propeller in the minimum drag position;
(2) The remaining engine(s) at not more than maximum continuous power;
(3) The landing gear retracted;
(4) The wing flaps retracted; and
(5) A climb speed not less than 1.2
0.C7A!OpenElement&FieldElemFormat=gif


Note that changes to those conditions (e.g., unfeathered prop, higher density altitude so the good engine puts out less power, lower gross weight, etc) can change the actual Vyse off the book figure/blue line on the airspeed indicator.

Usually Vyse is a good bit higher than Vmc, so in a real engine out situation, you would not be operating anywhere near Vmc unless:

  1. the engine failed below Vyse, or
  2. you're about to land, or
  3. you did something wrong.
In addition, it is most unwise to lift off before Vmc, because if you do, if one engine quits, you'll be unable to control the roll/yaw other than by immediately chopping the throttle on the good engine. Failure abide by those principles results in what happened in the video.
 
Feathering lowers the Vmc speed. My plane has a Vmc of 128 windmilling and 100 feathered. I rotate at 106 so an AutoFeather system is required.


BTW, Vyse is 140 KIAS.
 
Captain said:
Feathering lowers the Vmc speed. My plane has a Vmc of 128 windmilling and 100 feathered. I rotate at 106 so an AutoFeather system is required.


BTW, Vyse is 140 KIAS.
Click to expand...

That makes sense, thank you.

In fact, thank everyone who commented, I've learned a lot from this thread.
 
bbchien said:
In a DA 42 it is reasonable to actually remain on the ground to Vyse. It's a little squirrelly but it works.
Click to expand...

Oh yeah.
Normally rotate at 80, Vyse is 85.


If you guys want a twin where you can rotate above blue, go for the TwinStar :D
 
One of the things that I think every prospective Multi Engine pilot should have to do during training is watch one of the videos (either the Queen Air is SE Asia or the Baron in South America ) of an actual VMC roll and see just how quickly it develops. The way we teach VMC demos does not give the student an appreciation for how serious it is IMHO. It wasn't until a few years after I got my ME rating that I really understood how bad it was.
 
For the most part sims do not exist for the light twins. In my limited experience FTD's are not able to really duplicate controllability problems. The FTD is good for procedures, primary instrument students or perhaps the new multi pilot.
I am one that believes sim training is very important especially for the ME pilot. It is a shame that sims are not available for more make and models. Cost prevails.
 
ronnieh said:
For the most part sims do not exist for the light twins. In my limited experience FTD's are not able to really duplicate controllability problems. The FTD is good for procedures, primary instrument students or perhaps the new multi pilot.
I am one that believes sim training is very important especially for the ME pilot. It is a shame that sims are not available for more make and models. Cost prevails.
Click to expand...

FSI and SimFlight have a good 421 sim that reprograms for the 310 pretty well.
 
I would not consider a 421 a light twin. I am not familiar with that particular equipment. Do they use FTD's for the 421 or sims. I have only one experience of "reprogramming" for different models and it was not positive. In SimCom the Cheyenne III training used a Cheyenne I/II and reconfigures it for the Cheyenne III. Was not impressed at all. The actual Cheyenne III sim that FS had at Lakeland (now owned by SimCom) was very superior.
Years ago I did recurrent Navajo at RTC. They used a FTD and I never went back. Just was not what I was looking for. Any training is useful, sim or FTD. JMO
 
You must log in or register to reply here.
Back
Top