|
The Consequence
of Tailplane Icing
Following the catastrophic loss of a modern turboprop airplane a few years ago there appeared numerous
studies trying to determine the exact cause of that accident. A review of other accidents which had occurred previously, there
emerged a startling similarity between the events. The probable cause of each of the accidents seemed to indicate the existence
of ice accumulation on the tail-plane in each instance. Several studies were conducted and it was determined that with only
a few tenths of a millimeter ice on the wing leading edge of a two-meter chord wing there can be an increased stall speed
by as much as 20 percent.
But the studies went on to determine that the same roughness, or amount of ice, on the tailplane
leading edge can result in a catastrophic flow separations along the lower surface of the tailplane. This phenomenon is especially
prevalent at large wing-flap angles. And this separation may result in forward stick forces the pilot cannot counteract.
The
same amount of ice accretion will probably be insignificant while in cruise flight. However, being aware of the large differences
existing between an insignificant drag effect in cruise and the very significant effect on stall speed during an approach
is of extreme importance.
During the past several years, several transport-category airplanes have dived steeply or
vertically into the ground from landing approach because of tailplane stall. Among these catatasrophies are:
Five four-engine
Vickers Viscounts - 58,000 pound turboprop airplanes without elevator boosters - crashed under similar conditions of weather
and airplane loading.
Several Russian-built Illyushin 18 (IL-18) aircraft - four-engine turbopropeller airplanes weighing
as much as 110,000 pounds, and like the Viscount, without elevator boosters - dived almost vertically into the ground when
the flight crew extended full landing flaps while operating in icing conditions.
These accidents resulted in much consternation
within the aviation community, and initially the causes of the accidents were not fully understood. However, studies soon
showed that all of the accidents occurred after the flight crews had selected a large flap angle for landing.
In January
1977 A Vickers Viscount, a four-engine turbopropeller airplane, crashed near Stockholm, Sweden. The flight path taken by the
airplane when flaps position 40 was selected at 400 feet above the ground resulted in a complete upset to a pitch-down attitude
of nearly 100 degrees. Engineers and safety experts estimate that, if the pilots had been able to pull the elevator yoke back,
the airplane would have been controllable.
In 1973 a Bulgarian IL-18 flying near Moscow entered into an upset maneuver
that was recorded on the airplane flight data recorder. During the approach phase, at the point where the pilots would normally
select the landing flaps position, the control yoke suddenly moved to the full forward stop, nose-down, position.
Earlier,
in 1963, there was an incident involving a Continental Airlines Viscount. In that incident, both pilots, working together.
managed to hold the yoke back until the flaps extension could be reduced.
The most significant characteristic of the
tailplane icing phenomenon is that the yoke suddenly moves forward by itself when flaps are extended. The flight crew reaction
must be to immediately pull it back on the yoke, and just as importantly, reduce selected flap angle. An increase in airspeed
does not help in this icing – tailplane stall – situation. Quite the opposite is true, increasing airspeed only makes things
worse!
Manufacturers have continuously refined their airplane design, including the efficiency of wing flaps. For the
wing to perform at the various conditions which the airlines demand, flaps have undergone radical improvements, including
a complete change in wing geometry from takeoff through cruise, and into the approach and landing phases. It soon becomes
evident that pilots fly entirely different airplanes that behave quite differently aerodynamically in different phases of
flight.
Lift generation through angle of attack can be accomplished with from 10 to 12 degrees less pitch when flaps
are extended than when the flaps are retracted.
When the aircraft's angle of attack is reduced, the tailplane receives
a greater amount of air blowing from above. This blow-down effect provides a large downward force on the tailplane, which
raises the aircraft's nose. Additionally, changing the tailplane angle of attack results since wing flaps usually extend only
along a portion of the wingspan. This phenomenon concentrates lift on the wing centerspan section and increases the down-wash
behind the flaps forward of the tailplane. This adds to the nose-up pitching moment.
Nose pitch-up moment is greater
than is the pitch-down moment on most transport category airplanes. This nose-up pitching moment is larger as a result of
the rearward shift of the wing center of lift. This means that pilots need to push forward on the yoke while trimming out
the nose-up pitch force. However, the local airflow change to the tailplane is so great for some airplanes with very high-efficiency
flaps that the tail-plane is designed with a bent-up leading edge or sometimes a leading edge slat. For example, the Fokker
27 tailplane leading edge is permanently curved to combat adverse icing considerations.
It has been demonstrated that
the moving stabilizer which is used for trimming for pitch changes is more efficient when dealing with tailplane icing. Let’s
look at a couple of examples that demonstrate this factor.
Consider that you are making an approach, flying the numbers
at 1.3 times the velocity of stall at each configuration until decaying your airspeed for touchdown. Let’s begin the approach
at 200 knots until necessary to slow for flaps configuration speed of 120 knots.
As you begin slowing from 200 to 120
knots in the clean configuration the angle of attack must increase from approximately 2.5 degrees to as much as 11 degrees
just to maintain the lift necessary to hold altitude. With the application of the first flaps position extension the angle
of attack may be decreased to around 6 degrees and the resultant lift remains sufficient to maintain level flight.
Now
it is time to slow for the final approach airspeed of 100 knots. As the airplane slows to the final flaps-setting airspeed
while maintaining altitude the angle of attack must again increase to maintain altitude, and pitches up to about 11 degrees.
When you select the final flaps position and to counter the pitch-up tendency you push forward on the yoke to achieve an angle
of attack of approximately 3 degrees.
From this configuration you fly the airplane until you increase the angle of
attack to further slow the airplane for landing.
This ideal scenario is seldom present when the airplane has been flying
in icing conditions. If ice accumulates on the tailplane, the yoke becomes increasingly heavy, and can suddenly snap forward.
If the altitude is low, as it would normally be when selecting landing flaps, the airplane could impact the ground.
One
NASA test pilot and engineer intimately involved with the tail-plane icing study said, "If the tailplane stalls, the primary
problem is the elevator hinge moment and the ensuing stick force, not the tailplane's ability to create the needed download
for aerodynamic balance."
A Swedish domestic airline crew flying a Convair Metropolitan with a defective deicing system
on the tailplane experienced a sudden pitch-down while approaching to land. When the captain extended final landing flaps,
the nose pitched down 45 degrees. The quick-thinking captain correlated the sudden maneuver with the flaps extension, and
immediately retracted flaps to the previous setting. This action allowed him to regain airplane control.
If you experience
a tailplane icing stall, you may be in one of three levels of danger:
In Level 1 you or you and the other pilot can
hold the elevator controls back, preventing the airplane from diving. If ice is suspected of accumulating on the tailplane
leading edge, both pilots should be prepared for a tailplane stall.
Level 2 cases finds the tailplane cannot provide
the needed download to maintain the desired trim speed even with the elevator in full nose-up position. This is due to the
separation of airflow on its surface. A steep dive will develop fairly slowly, however, and you will have time to recognize
the situation and take corrective action by partially retracting the flaps. The Convair Metropolitan incident described earlier
belongs to this level of danger. The captain commented that, “. . .I am sure we kept the yoke fully aft. In such a situation
you become very strong."
Level 3 results in a situation where you cannot prevent the yoke from moving full forward.
The time available for reducing flap is reduced to only a couple of seconds. Following a discussion of the crash of a Vickers
Viscount near Stockholm in 1977 a famous old Swedish pilot, Count von Rosen, commented, "No one can be such an idiot as to
use full flaps if there is the slightest chance of ice on the tailplane." He thus pointed to the problem of transferring experience
from an older generation to a younger one.
The pilot of a Piper Cherokee, weighing less than one ton, would have no
problem keeping the yoke back. However, the Vickers Viscount weighs 29 tons, and two pilots working together barely managed
to control it.
In summary then, if the tailplane stalls, the primary problem is not the tailplane's ability to create the
required download for aerodynamic balance, but rather the elevator hinge moment and the ensuing stick force. The consequence
of this upset of balance depends on the weight of the aircraft and on whether it has elevator boosters servos.
Aerodynamically,
balancing the elevator perfectly for the no-ice case acts completely in the wrong direction when ice has accumulated on the
tailplane. The elevator will them deflect of its own volition, pitting its power against that of the pilots.
Portions
of this page have come to me from many sources, most of them I have NO IDEA who they are. I take this opportunity to give
credit where it is due. If you see something you created, I thank you, and all of my friends thank you. The only pieces of
which I claim credit will have my name associated with it.
----------------------------------------------------------
|
