Collings Foundation B-17G 909 Has Crashed :-( 10/2/2019

[sandiego89]

I was wondering if landing runway 24 or runway 33 might have been a faster return to the airport.....

Turn direction is also important to consider. A "faster" return is not always the best option, especially if it requires a turn into a dead engine. One does not want to tighten a turn into a questionable engine. Not speculating, we do not yet know what the engine(s) were doing, but there have been some reports of engine issues on perhaps #3 or #4 which are on the right/starboard side. I am not speculating, just perhaps trying to help inform your wondering.

[Saville]

I was wondering if landing runway 24 or runway 33 might have been a faster return to the airport.....

Turn direction is also important to consider. A "faster" return is not always the best option, especially if it requires a turn into a dead engine. One does not want to tighten a turn into a questionable engine. Not speculating, we do not yet know what the engine(s) were doing, but there have been some reports of engine issues on perhaps #3 or #4 which are on the right/starboard side. I am not speculating, just perhaps trying to help inform your wondering.

All their turns were to the right. Not debating your point about not turning into a dead engine. Just noting a fact.

[r1830]

Here is a data fact sheet on the B17 from the Air Force in 1949
http://www.wwiiaircraftperformance.org/ ... istics.pdf
On page 4 it shows a power off stall speed of 89mph at a combat weight. It doesn’t give a flap setting which would change the speed. It is the only factual baseline reference I have been able to find so far. It is a baseline starting point but not a definitive answer. Stall speed is dynamic and some factors that affect it are gross weight, configuration, power and wing loading.

With an engine(s) out, aircraft performance is certainly degraded. First you have asymmetric power, which then requires asymmetric flight control input which both of which result in drag. The flight control input cancels out the asymmetric thrust but has only decreased the overall total drag created with an asymmetric power condition and no control correction being applied. A term in aviation for flying with an engine out, is raise the dead. In level flight it means have approximately a 5° bank angle to counteract the asymmetric thrust. For example with a dead engine on the right side of an aircraft you would want to raise the right wing. Doing this however further increases the drag on the right side has the right aileron is down and the left aileron is up. To counteract this and the asymmetric power, the rudder is deflected to the left which results in a zero sideslip condition that helps minimize the drag the fuselage would have created had the aircraft controls not been used to counteract the asymmetric thrust. While this technique has minimized the drag created when flying with an engine(s) out, overall drag has still increased and performance is degraded.

The stall speed on the wing with two good engines is lower than the stall speed on the wing with the engine(s) out as the airflow is better over the wing with the good engine(s). Another factor affecting performance is whether the engines are feathered or unfeathered with more drag if a dead engine remains unfeathered. The further away on engine is from the centerline of the aircraft, the more asymmetric drag is produced, resulting in more control input required to counteract it and performance degraded.

Any turns made would result in a loss of lift and an increase in descent rate. It is a trade off that needs to be in the decision making process. To minimize loss of lift shallow wide turns are recommended. Turning into the dead engine only requires a releasing the asymmetric control input that is being applied. Aerodynamically the aircraft will naturally want to turn that way. The trouble becomes when you want to stop the turn, full control deflection will be needed and it will not respond quickly. The result being aircraft performance degrading the steeper the turn.

As lift is loss in a turn, the nose will need to be lowered to maintain best speed increasing the rate of descent. If the up elevator is applied to arrest the descent in a turn, speed will decay, wing loading will increase which in turn increases the stall speed. The potential for rolling the airplane over on its back increases as there is not enough control authority to keep the airplane upright with asymmetric power condition. This is known as a VMC roll over.

The most important variable to control is airspeed which is done by minimizing the drag created by control and power inputs.

There is another way to counteract asymmetric thrust by reducing power on the good engine. Obviously, that will degrade thrust and performance and the only way to maintain airspeed is to lower the nose an descend.

Each aircraft is different, some aircraft would take so long to turn away from the dead engine, the resulting time and space required to make that turn would prolong the time it takes to complete the turn. If the aircraft is descending while flying at best speed, it might not be able to make the airport. Potentially the fastest way to turn is use the natural direction the aircraft wants to go. Either way has consequences and a pilot just has to make a judgement call as to which way is best for any given situation.

I will refrain from making any judgement or comment as to what happened up in the cockpit since I wasn’t there. I am just trying to provide background factual aerodynamic information that affects aircraft performance with an engine(s) inoperative as I understand it. You are free to disregard my post and/or draw your own conclusions.

[phil65]

Does anyone know the Crew chief and his condition ?
Hopefully he can shine some light on what happened.

Phil

[r1830]

News article from a few days ago.

https://fox61.com/2019/10/09/injured-b1 ... -families/

[GarryW]

Alright...the thread is pointless guys.

To you, perhaps, but I'm finding out stuff here that isn't anywhere else online. Several pilots giving info I was unaware of and updates come here faster than anywhere else (if at all)...

Apparently my sarcasm didn't come across well. My apologies! I agree with you 100%

[GarryW]

Here is a data fact sheet on the B17 from the Air Force in 1949
http://www.wwiiaircraftperformance.org/ ... istics.pdf
On page 4 it shows a power off stall speed of 89mph at a combat weight. It doesn’t give a flap setting which would change the speed. It is the only factual baseline reference I have been able to find so far. It is a baseline starting point but not a definitive answer. Stall speed is dynamic and some factors that affect it are gross weight, configuration, power and wing loading.

With an engine(s) out, aircraft performance is certainly degraded. First you have asymmetric power, which then requires asymmetric flight control input which both of which result in drag. The flight control input cancels out the asymmetric thrust but has only decreased the overall total drag created with an asymmetric power condition and no control correction being applied. A term in aviation for flying with an engine out, is raise the dead. In level flight it means have approximately a 5° bank angle to counteract the asymmetric thrust. For example with a dead engine on the right side of an aircraft you would want to raise the right wing. Doing this however further increases the drag on the right side has the right aileron is down and the left aileron is up. To counteract this and the asymmetric power, the rudder is deflected to the left which results in a zero sideslip condition that helps minimize the drag the fuselage would have created had the aircraft controls not been used to counteract the asymmetric thrust. While this technique has minimized the drag created when flying with an engine(s) out, overall drag has still increased and performance is degraded.

The stall speed on the wing with two good engines is lower than the stall speed on the wing with the engine(s) out as the airflow is better over the wing with the good engine(s). Another factor affecting performance is whether the engines are feathered or unfeathered with more drag if a dead engine remains unfeathered. The further away on engine is from the centerline of the aircraft, the more asymmetric drag is produced, resulting in more control input required to counteract it and performance degraded.

Any turns made would result in a loss of lift and an increase in descent rate. It is a trade off that needs to be in the decision making process. To minimize loss of lift shallow wide turns are recommended. Turning into the dead engine only requires a releasing the asymmetric control input that is being applied. Aerodynamically the aircraft will naturally want to turn that way. The trouble becomes when you want to stop the turn, full control deflection will be needed and it will not respond quickly. The result being aircraft performance degrading the steeper the turn.

As lift is loss in a turn, the nose will need to be lowered to maintain best speed increasing the rate of descent. If the up elevator is applied to arrest the descent in a turn, speed will decay, wing loading will increase which in turn increases the stall speed. The potential for rolling the airplane over on its back increases as there is not enough control authority to keep the airplane upright with asymmetric power condition. This is known as a VMC roll over.

The most important variable to control is airspeed which is done by minimizing the drag created by control and power inputs.

There is another way to counteract asymmetric thrust by reducing power on the good engine. Obviously, that will degrade thrust and performance and the only way to maintain airspeed is to lower the nose an descend.

Each aircraft is different, some aircraft would take so long to turn away from the dead engine, the resulting time and space required to make that turn would prolong the time it takes to complete the turn. If the aircraft is descending while flying at best speed, it might not be able to make the airport. Potentially the fastest way to turn is use the natural direction the aircraft wants to go. Either way has consequences and a pilot just has to make a judgement call as to which way is best for any given situation.

I will refrain from making any judgement or comment as to what happened up in the cockpit since I wasn’t there. I am just trying to provide background factual aerodynamic information that affects aircraft performance with an engine(s) inoperative as I understand it. You are free to disregard my post and/or draw your own conclusions.

Thank you so much for your input. This is the kind of information I relish not having the actual flight training but the thirst for knowledge. I think that's kind of the general sentiment of this thread. No one is coming out with "x is the reason why" or "y is at fault". Most of what I see is people trying to understand how this may have happened and waiting to hear from the NTSB or other credible sources for confirmation. I for one am pleasantly surprised by everyone's candor thus far. There is no finger pointing or declaration of "this is what happened", just a group of people trying to make sense of how this could have happened. Factual information is simply that...facts. And I thank you for those.

[r1830]

I found a flight manual for the B17 that was produced by EAA.
https://www.eaa.org/~/media/files/eaa/f ... manual.pdf

On page 101 it has charts for the stall speed of the aircraft in two different configurations. Each chart than contains the variables of gross weight and bank angle.

For example,
Power off, flaps and landing gear down.
Zero bank angle and 55,000 pounds, the speed looks to be around 93 miles per hour.

Power off flaps and landing gear up.
Zero bank angle and 55,000 pounds, the speed looks to be around 105 miles per hour.

It is a graphical representation of how weight and bank angle affect the stall speed.

Page 113 is the beginning of normal and emergency procedures checklist.
Page 135 is the expanded procedures for the emergency checklist.

A lot of good information in there on how to handle an engine(s) failure and the challenge such a failure presents.

[Saville]

I found a flight manual for the B17 that was produced by EAA.
https://www.eaa.org/~/media/files/eaa/f ... manual.pdf

On page 101 it has charts for the stall speed of the aircraft in two different configurations. Each chart than contains the variables of gross weight and bank angle.

For example,
Power off, flaps and landing gear down.
Zero bank angle and 55,000 pounds, the speed looks to be around 93 miles per hour.

Power off flaps and landing gear up.
Zero bank angle and 55,000 pounds, the speed looks to be around 105 miles per hour.

It is a graphical representation of how weight and bank angle affect the stall speed.

Page 113 is the beginning of normal and emergency procedures checklist.
Page 135 is the expanded procedures for the emergency checklist.

A lot of good information in there on how to handle an engine(s) failure and the challenge such a failure presents.

Thanks for providing the link. I note this on 2 engine failure on takeoff:

"There is a critical airspeed of approximately 115-120 MPH, below which the airplane will not climb. It also will not accelerate on two engines at or below that critical airspeed, regardless of how much additional power is utilized. For this reason, the only possibility of making this maneuver succeed -if the failures occur at less than critical airspeed - is to descend in order to attain the critical airspeed. "

According to the Flight Track Log plots they got to 115. But after the second "climb" they began losing speed and certainly could not maintain 115. This may explain the rationale for the dive when they were only (roughly) 1/3 the way on the downwind. During the constant ~300 foot AGL altitude they were losing speed. So perhaps the ideas was to dive to maintain controllability.

Excerpt from the single engine failure instructions:

"If the failure occurs immediately after liftoff, use any power the engine may be developing to attain the three engine climb airspeed of approximately 115 MPH at the weights this aircraft is operated. "

[r1830]

According to the Flight Track Log plots they got to 115. But after the second "climb" they began losing speed and certainly could not maintain 115. This may explain the rationale for the dive when they were only (roughly) 1/3 the way on the downwind. During the constant ~300 foot AGL altitude they were losing speed. So perhaps the ideas was to dive to maintain controllability.

Do be careful using the speeds on flight aware as a reliable indication as to what the indicated airspeed was. Speed presented on flight aware is a ground speed. So if they were flying into a headwind the speed would look slower than actual indicated, conversely if it was a tailwind the speed would be faster than indicated.
According to the plot, the speed did change with a turn, but it is impossible to tell with the data available if that was in indicated airspeed change or a result of the direction change the in relation to the wind.

Also, The wind speed measurement of calm at the surface is just that, a measurement at the surface and not a reflection of the wind at 500 or 1000 feet.

[Saville]

According to the Flight Track Log plots they got to 115. But after the second "climb" they began losing speed and certainly could not maintain 115. This may explain the rationale for the dive when they were only (roughly) 1/3 the way on the downwind. During the constant ~300 foot AGL altitude they were losing speed. So perhaps the ideas was to dive to maintain controllability.

Do be careful using the speeds on flight aware as a reliable indication as to what the indicated airspeed was. Speed presented on flight aware is a ground speed. So if they were flying into a headwind the speed would look slower than actual indicated, conversely if it was a tailwind the speed would be faster than indicated.
According to the plot, the speed did change with a turn, but it is impossible to tell with the data available if that was in indicated airspeed change or a result of the direction change the in relation to the wind.

Also, The wind speed measurement of calm at the surface is just that, a measurement at the surface and not a reflection of the wind at 500 or 1000 feet.

Your point about indicated vs Ground is taken.

But assuming they took off into a headwind then after the 180 degree turn to downwind, we should expect the ground speed to increase assuming constant indicated. But it was decreasing which indicates trouble and, as I wrote above may suggest a reason for the dive.

[r1830]

But assuming they took off into a headwind then after the 180 degree turn to downwind, we should expect the ground speed to increase assuming constant indicated. But it was decreasing which indicates trouble and, as I wrote above may suggest a reason for the dive.

I agree with your assessment if they had taken off into a headwind, then the turn onto down wind would be into a tailwind. However in my previous post, I was trying to point out that the wind on the ground which was calm is in no way a reflection of the wind up in the air. For instance, if they had taken off in calm wind on the ground and climbed into a increasing tailwind condition, then the turn from crosswind to the downwind leg would’ve been into an increasing headwind which would’ve resulted in the slower a ground speed. My apologies for not clarifying my previous post.

Again I feel I don’t have enough facts to make an educated guess as to what happened on that specific flight, I would like to see the wind data. Looking at the radar data the decrease in altitude occurs with a sharper turn which could be related to a steeper bank angle or on increase in drag for reasons yet to be determined. As I had discussed in my previous post about aerodynamics, an increase in the bank angle will result in a decrease in overall performance which also could explain the decent. Or, the descent could be from increased drag from a dying engine and lowering the nose to maintain airspeed. I wasn’t there and don’t have enough facts to make a conclusion as to what transpired, however this topic has resulted in a great review of engine out aerodynamics which I have found them personally worthwhile to review. I do agree with your application of the principles being discussed.

Kind regards.

[Randy Haskin]

FlightAware is as valid an accident investigation tool as Wikipedia is a valid research tool.

[aerovin]

You can’t trust anybody. The NTSB Preliminary report says the airplane was manufactured in 1944 vs. actual date of April 1945.

[Chuck Giese]

FlightAware is as valid an accident investigation tool as Wikipedia is a valid research tool.

OK Randy, I'll bite. I know Wikipedia is unreliable, at best. What's the issue with flightaware? It appears to be 15 second interval snapshot of the adsb data. Is my assumption wring!