May 1988 DGIEM No. 88-TR-16 [611162]

May 1988 DGIEM No. 88-TR-16
EJECTION SYSTEMS
and
THE HUMAN FACTOR:
A GUIDE FOR FLIGHT SURGEONS
AND AEROMEDICAL TRAINERS
W.R. Sturgeon
i I C
ELECTEK
Defence and Civil Institute of Environmental Medicine S u 1 ~ 8
1133 Sheppard Avenue West
P.O. Box 2000
Downsview, Ontario M3M 3B9
DEPARTMENT OF NATIONAL DEFENCE -CANADA
T dammt im beft I~a
&W pub!. "too". =4 OR%
w -S

r~qVW%~w r~rr~qwrww , x" I~yr~~wrI~nPonrV~wVrbV~wT~wV V " 'I' V w
TABLE OF CONTENTS Pg
ABST ACT … …. …. … …. …. … …. …. … …. aV e
ABTRADCTION……………………………………………VI1
-Milestones in RCAF/CF Ejection Systems…………………….
BASIC EJECTION SYSTEM COMPONENTS…………………………. 3
-Head Rest and Canopy Breaker…………………………….3
-Seat Structure………………………………………. 4
-Harness Systems……………………………………… 4
-Ejaction Systems……………………………………… 5
-Seat Pack/Survival Kit………………………………… 5
-Parachute……………………………………………. 6
EJECTION STATISTICS…………………………………….. 6
DECISION PHASE…………………………………………. 8
-Unable to Eject……………………………………… 13
-Unperceived Ground Closure……………………………. 14
-Rapidity of Events…………………………………… 16
Non-Fatal "A" Category Mishaps………………………… 17
EJECTION PHI-SE………………………………………… 20
-Anterior Cuneiform Fractures………………………….. 22
-Comnminuted Fractures…………………………………. 23
EJECTION ENVELOPE……………………………………… 25
-Altitude………………………..25
-Bank Angle………………………………………….. 25
-Climb/Dive………………………………………….. 25
-Airspeed……………………………………………. 26
-Sink Rate……………………………………………. 26
DESCENT PHASE…………………………………………. 28
-Seat/Occupant Interaction…………………………….. 29
-Seat/Parachute Collision……………………………… 29
-Failure to Separate from the Seat………………………..29
-Failure of Parachute to Deploy…………………………..30
-Parachute -Pilot Entanglement…………………………..30
-Parachute/Seat Pack Entanglement…………………………31
-Parachute Canopy/Riser Entanglement………………………31
-Altitude………………………..31
-Deceleration………………………………………… 32
-Freefall……………………………………………. 3
-Terminal Velocity………………………………….. 34
.Spinning and Tumbling………………………………. 34
-Parachute Opening Shock………………………………. 35

LANDING ……………………… 36
SURVIVAL PHASE………………………………………… 37
SUMM RY … … … … … …. … … … … … … … 38
CONCLUSIONS…………………………………………… 40
ACKNOWLEDGEMENTS………………………………………. 41
BIBLIOGRAPHY………………………………………….. 42
TABLES………………………………………………… 44
FIGURES……………………………………………….. 57
*ANNEX A -"A" Category Accidents 1972-1987………………….82
* Strib3C
GRAcZ
Avaail~~–
1 oe
AvaV
4i

ABSTRACT
-This report outlines the Canadian Forces (CF) experience with
– ejection escape systems and the human factors involvement in their use
during the period 1952 through 1987. It is intended as an
introductory guide to Flight Surgeons and other aeromedical personnel
of the Canadian Forces investigating accidents or incidents involving
4 ejection systems, or for personnel responsible for training aircrew in
the use of ejection systems.
The report analyses human factors involvement in ejection
decision formulation and the medical/physiological considerations of
the ejection, descent, landing and survival phases of the escape
sequence. Case histories are used to illustrate the hazards involved
in ejection escape, and emphasis is placed on ejection statistics for
S the period 1972 through 1987.
4-
Vii1,
– ., –i- – '.~W .
N2

I
INTRODUCTION
During the Second World War, allied aircraft were typified by the
Hawker Hurricane, Supermarine Spitfire, Mosquito, Mustang, P40 Kittyhawk, and
.- Hawker Typhoon. These aircraft were characterised by piston engines that
could rarely exceed 400 mph, and by relatively "high" wing surface areas that
provided controllability at fairly low flying speeds of 70 to 80 mps (Figure
S1). As a result of these minimum control parameters, emergency egress
procedures were limited to "bail-out", i.e., the pilot undid his lap belt and
*. shoulder harness assembly, opened the canopy and jumped out, manually opening
*i his seat pack parachute.
* During the War, prompted by Allied successes at shooting down Axis
aircraft and pilots, the Germans experimented with the development of the
ejection seat. Busch carried out the first successful ejection at 300 mph
from a Heinkel aircraft in 1939, and in 1944 the Heinkel Ejection Seat was
fitted to the Heinkel Jet Fighter and twin-engine Messerschmitt. By the end
of the War, German pilots had made 60 successful ejections.
Other countries were quick to follow this technological advance. In
1943 Sweden prototyped their first seat into the SAAB 17B Bomber and SAAB
J21A Fighter, and their first ejection was recorded 29 July 1946. The United
States recorded their first ejection in November 1946, followed by France in
February 1948. The RAF tested their first "dummy" ejection in 1945 from a
Defiant aircraft, and on 26 June 1946, Lynch became the first UK ejectee from
a Gloster Meteor at 515 kph in a Martin-Baker seat.
see , introductiu of jeL aircraft with high minimum controllability air
speeds, high air speeds that precluded conventional "bail-out" techniques,
and high "sink rates" that arose with engine failure, necessitated the use of
* the ejection seat.
Milestones in RCAF/CF Ejection Systems
The first Canadian Forces (RCAF) operational ejection seat equipped
aircraft was the F86 Sabre jet, taken on strength in August 1950 (Mark I),
followed by the Mark II in February 1951. Improvements in design continued
through the Mark VI before the Sabres were replaced in 1967. The Sabre
ejection seat was a ballistic system, and initially it was manually operated.
That is, the pilot and seat were "explosively" blown from the cockpit after
the canopy had been manually jettisoned. Free of the aircraft, the pilot had
to undo his lap belt, push away from the seat, then manually deploy the
parachute before ground impact. Needless to say, many deaths resulted from
.. the length of time required to manually perform all the steps in the ejection
I sequence. Not all pilots trusted the system -many elected to force land the
aircraft rather than eject, that resulted in numerous fatalities. Canopy
jettison failure also accounted for fatalities, along with several failures
of the ejection gun. During the latter half of the 1950s the Sabre ejection
system was converted to one of szmi-automatic operation. After jettisoning
– ….- . ..
-,

* 2
,. the canopy, the pilot ejected from the aircraft. He was still required to
manually undo the lap belt and kick away from the seat, however, as he did
so, a parachute arming cable attached to the seat lap belt activated a
spring-loaded timer which fired two .22 calibre blanks. When they fired, gas
pressure pushed against a piston that withdrew the parachute pins allowing
the parachute to open. A number of fatalities continued to occur because
aircrew occasionally forgot to attach the arming cable to the lap belt during
seat strap-in. Fuel exhaustion, stalls on landing approach and mid-air
.5 collisions accounted for many deaths during the Sabre era. During this
time-frame, the Sabre ejection sysLem exhibited an approximately 80% success
rate.
In 1953 the CF100 Canuck was acquired by the RCAF. It was the first
two-seat ejection aircraft to be manned by a pilot and an Air Intercept
navigator. This system was the first Canadian fully automatic ejection seat,4. although it remained a totally ballistic impulse system (Figure 2). The
CF100 remained on inventory until 1981. The seat, built by Martin-Baker
* (UK), also incorporated the first leg retraction/garter anti-flail restraint
system used by the RCAF (Figure 3A, B). It was also the first system to use
the drogue gun/drogue parachute extraction method, virtually the same system
0 later incorporated into the CF188 aircraft ejection system. Ejection was
initiated by reaching upwards and pulling a face blind over the head which
* activated two firing cartridges -one to jettison the canopy and one to fire
* the seat. The seat was designed to operate at a minimum altitude of ground
level with 90 knots forward air speed. Use of the face blind by the back
seat navigator was not without hazard as the following incident illustrates;
A CF100 at Cold Lake 'got into a tuck' on exceeding Mach 0.84. The
aircraft became uncontrollable and dived. The navigator was ordered to
*eject, but when the pilot 'blew the canopy', the navigator while reaching for
the face blind above his head, had his arms sucked out into the slipstream,
and he was pinned there. The pilot managed to regain control and returned to
basE, unaware that the navigator was still onboard. The intercom had become
unserviceable following canopy ejection. The navigator received badly
frostbitten fingers (that subsequently required amputation) of both hands, a
fractured left humerus and dislocated left shoulder.
In 1955 a rear windshield was installed to assist in the deflection of
the slipstream to allow the navigator to reach the face blind ejection
curtain, and an alternate seat ejection ring was placed on the front of the
seat pan structure. Loss of aircraft control, engine flame-outs, and mid-air
collisions took a constant toll of CFI00s and their crews until its
retirement in December 1981. The CFlOO exhibited an ejection success rate of
approximately 70%.
In 1959 the CF101 Voodoo came on line, the first supersonic aircraft
– in the RCAF inventory, complete with after-burners (AB) and drag-chutes. The
* ejection system was still ballistic, but fully automatic, incorporating a
shoulder harness inertia reel, ballistically-opened lap belt, and positive
man-seat separation system (Figure 4). When the pilot operated the ejection
handles, the canopy was jettisoned, the rear ejection seat was fired and 0.5
0A

3
seconl.. later, the front seat was ejected. The system was designed to work
at -' feet above ground and 120 knots. Pitch-up and loss of aircraft control
contributed significantly to losses until the aircraft was retired in 1982.
The ejection success rate was approximately 91%.
The next milestone in RCAF ejection system development occurred with
the acquisition of the CF104 Starfighter in 1961. This was the first
aircraft to incorporate a rocket catapult (ROCAT) ejection system with both
arm and leg retention devices. The aircraft was also the first to have a
MASTER CAUTION light, an enunciator panel, stall warning (shaker) and
automatic pitch control (APC) system. The design of the system gave a ground
level, 90 knot capability. Ejection was initiated by pulling the "D" ring on
the front of the seat pan (Figure 5). On ejection, the legs were retracted
by metal cables attached to spurs" worn on the pilot's boots and two bars
* swung forward to encase the pilot's arms in webbing to protect against flail.
The CF104 ended its career in 1982 with an ejection success rate of 92%. Its
low-level, high speed role resulted in considerable losses due to bird
ingestion, engine failure, and controlled flight into terrain.
The most recent milestone in CF ejection systems occurred with the
acquisition of the CF188 Hornet. This aircraft has a "zero-zero" ejection
capability. It also incorporates a head-box mounted steerable parachute,
remote rocket mot(r (under seat pan), and a Simplified Combined Harness (SCH)
system (Figure 6). Upper and lower leg retention garters are used for flail
protection. Like the CF100, man-seat separation is by parachute extraction.
* To date, the CF188 has a 100% ejection success rate. The only other aircraft
with the same success rate is the CF116 (CF5) Freedom Fighter.
BASIC EJECTION SYSTEM COMPONENTS
An ejection seat must perform two functions. It must provide basic
seating and restraint, and it must provide for the mechanics of emergency
egress. The basic components of an ejection seat are (Figure 7A-C):
a. head-rest and canopy breaker;
b. seat structure;
c. harness systems;
d. ejection system; and
e. seat pack/survival kit.
Head Rest and Canopy Breaker. This provides minimal head buffet/flail
protection when the pilot is ejected into the slipstream. A canopy "spike"
(CF5), wedge (CF188) or bar (CTIl4, CT133) is incorporated on the head-box to
provide positive canopy penetration should the canopy fail to jettison on
ejection. The head-box on the CF188 seat also houses the 17 foot aeroconical

S.- 4
parachute and the 22 inch and 5 foot drogue chutes.
Seat Structure. The seat structure provides the physical support and
protection required on ejection, and serves as the platform to house the
various components of the ejection system (ballistic devices, high pressure
lines, restraint lines). The seat structure contains the seat kit that
houses the personal survival items (Figure 7).
Harness Systems. There are two distinct harness systems incorporated
into the ejection seat for all aircraft except the CF188. They consist of a
torso restraint system and a seat separation system. The restraint system
consists of a set of shoulder straps, a lap belt, and a negative-G strap, all
of which are connected to the Rocket Power Incorporated (RPI) buckle (Figure
8A, 8B). Also connected to the RPI buckle is the parachute arming key that
is attached by a cable to the MK10 barostat in the personal parachute. As
the seat/occupant separate following ejection, the RPI lap belt is separated
by gas pressure, and the pilot is positively ejected from the seat by
retraction of the rotary actuator (butt snapper). As separation occurs, the
parachute arming cable is pulled from the MKI0 barostat thus activating it,
and the shoulder straps, lap belt, and negative "G" strap are freed, allowing
the pilot to physically separate from the seat. The rotary actuator consists
of webbing routed under the seat pack, connected to the forward edge of the
seat pan at one end, and to a ballistic reel at the other end. At the
instant of seat separation, the opening of the RPI lap belt and the
retraction of the rotary actuator occur simultaneously. Separation occurs
due to the differential drag between pilot and seat.
The CF188 seat presents a different concept in the harness system
because the parachute is located in the head-box rather than on the pilot's
back. In this system, the pilot straps into the parachute harness, that in
turn is anchored to the seat at four positions: the ballistic inertia reel
(BIR), lower harness locks, sticker clips, and negative-G strap. This system
is the SCH, (Figure 9A-C). Following ejection from the cockpit (assuming all
-N seat parameters are met), seat separation occurs when the scissor shackle
opens allowing the aeroconical parachute to deploy from the head-box through
the action of the drogue chutes. The lower leg restraint pins, "T" handle
seat attachment (negative G strap), and BIR shoulder harness are released.
The aerodynamic drag of the parachute provides the positive separation of the
pilot from the seat.
, The CF188 seat also incorporates a leg restraint system. This
consists of a set of upper and lower leg garters attached by arrow fittings
to leg restraint lines. These restraint lines (Figure 10) incorporate a
"break link" to allow separation from the aircraft on ejection. The lines
are attached to a floor bracket by a quick release pin, pass through a
snubbing box and the upper and lower leg garter arrow attachments, and
connect to taper plugs fixed to the left and right lower seat pan sides. On
* ejection of the seat, the restraint lines retract the legs against the seat
pan, and the break link separates as the seat leaves he cockpit. The legs
are maintained firmly against the seat by the frict.in lock of the snubbing
unit. On seat separation, the taper plug is freed allowing the leg garters

5
to slip free of the restraint lines.
Ejection Systems. (Figures 11-15) The seat ejection system is
composed of two sub-systems: the ballistic catapult and the rocket motor.
Such an ejection seat is usually termed a ROCAT (rocket-catapult) seat system
(Fig. 11). All CF ejection seat aircraft possess essentially the same
mechanism for ejection. When the ejection handles are raised, a sear pin is
withdrawn from the ROCAT initiators allowing them to be fired by percussion.
Gas pressure is directed through hoses to the BIR to retract the pilot into
the seat, and to the canopy jettison system. After a pre-set delay of 0.3 to
1.0 seconds depending on aircraft type, (Tables I to 4) initiators fire the
rocket catapult. The ROCAT is a set of telescopic tubes (three in the CF188
catapult) attached vertically to the rear of the ejection seat, the inner4' tube attached at the top to the seat guiderails of the aircraft. The initial
ejection force is provided by the ballistic charges in the catapult, followed
by ignition of the rocket motor that provides thrust through a controlled
burn over 0.25 to 0.46 seconds depending on aircraft type. In all except the
CF188, the rocket thrust is directed rearwards from the ROCAT. The CF188
system differs in that the catapult and rocket systems are physically
separated (Figure 16). The telescopic catapult is located at the rear of the
ejection seat, but the rocket motor is located beneath the seat pan. As the
seat travels ballistically up the guiderails, a static line cable attached to
the floor fires the rocket initiator. Whereas the rocket motor nozzles of
other ejection seats are directed from the rear of the seat, the CF188 rocket
nozzles are directed from the sides of the seat. The 9A seat nozzles are
1P larger on the right side than on the left side and provide a seat trajectory
to the left. The 10A seat exhibits the opposite characteristics. Thus, on a
dual ejection, the front seat trajectory is to the right, and the rear seat
is to the left.
' Seat Pack/Survival Kit. (Figure 17) The emergency seat pack fulfils
three basic functions: provides a seat cushion during normal operation,
provides a relatively firm and stable base in the event of ejection, and
provides containment for the storage of emergency survival items. The seat
* pack is of contoured fibreglass and is held to the parachute harness by left
and right "airlock" fasteners. The pack has an actuating handle on the right
" side which when pulled, deploys the pack contents. The pack contents are
physically secured to the pilot via a nylon lanyard connected to either the
parachute harness or the life preserver when worn. The basic contents of all
ejection seat kits are standardised, although there is variation between
aircraft and squadrons as to additional items carried. The quantity and type
of items carried is dependent on the volume of the seat kit. Basically, all
kits contain a survival items packet, a food packet, life raft, first aid kit
and sleeping bag. Table 5 lists the various items found in the CF116 (CF5)
seat kit as an example. Following ejection, the seat pack is normally
deployed by pulling the actuating handle at an altitude of about 500 to 1000
feet above ground, unless descent is made into a wooded area, then it is
recommended that it not be deployed.
For all aircraft except the CF188, deployment of the seat kit occurs
by the retraction of –wo retainer pins on the seat kit bottom (Figure 17).
The withdrawal of the pins allows the canvas cover to open, spilling out the
………………………. . . . . . . . . . .
–– – – – – – – – – – – – –

4-0 6
seat kit contents that are all connected to a 26 foot nylon lanyard that is
in turn attached to the pilot. Withdrawal of the retainer pins also allows
the fibreglass seat kit container to fall away from the pilot's body. As the
kit contents fall the length of the lanyard, the one-person life raft is
automatically inflated as the C02 cartridge is activated by the weight of
the dangling sleeping bag and contents bag.
." .. Activation of the CF188 seat kit deployment handle produces similar
results. However, since the emergency oxygen cylinder is attached to the
underside of the seat kit lid, only the bottom half of the fibreglass kit
-falls away from the pilot, thus deploying the contents. The seat kit lid
remains attached to the simplified combined harness (SCH) throughout the
landing sequence. The 26 foot lanyard suspending the seat kit contents is
attached either to the life preserver when worn, or to a bandolier slung over
the shoulder.
Parachute. There are three types of parachute currently in use in CF
aircraft. The CF116 has a 28 foot diameter canopy, the CT133 and CT114 a 24
foot diameter canopy and the CF188 a 17 foot diameter GQ aeroconical
parachute. In the CF188, parachute opening is by two drogue "chutes" (60
0 inch and 22 inch diameter controller and retardation drogues). These drogues
extract the parachute from the head-box when the barostat opens the scissor
shackle at 13000 + 1500 feet and when seat deceleration drops below 2.5G.
Below 7000 feet MSL, the parachute is extracted 1.8 seconds after initiating
"-i ejection (the 'G' limiter is inoperative below 7000 feet). In all other
aircraft, the parachute opening is controlled by the Mark 10 barostat set to
open at 16000 + 500 feet MSL, or when below this altitude, one second after
p seat separation. Timings for the ejection sequences are listed in Tables 1
to 4. Only the CF188 parachute incorporates steering lines to control
forward descent speed. Descent velocities (with a 200 pound weight) are 19
fps for the 17 and 28 foot canopies, and 22 fps for the 24 foot canopy.
EJECTION STATISTICS
There are four distinct phases to any aircraft ejection sequence;
a. the decision phase that comprises
the time span from emergency
onset to formulation of the decision to eject;
b. the ejection phase, that consists of the time required to
* activate the ejection handles and be ejected from the cockpit;
c. the descent phase, that occurs from rocket motor burn-out to
ground/water landing; and
d. the survival phase, consisting of the period of time from
* landing to rescue.
'
00*
r. Po SK-'

7
Each of these four phases has inherent limitations and hazards that can
jeopardize a successful outcome. Each of ticese phases will be covered
separately in the following pages, analyzing their impact on CF ejection
statistics since the first ejection on 9 April 1952 from an F86 Sabre.
Table 6 lists the elementary breakdown of data gathered for this study
from a wide variety of sources. Initial data were obtained from Brent
(3,4,5,6) and Smiley (23,24,25,26,27), and supplemented by NDHQ'S Directorate
W of Flight Safety (DFS) Documents (8,9). Maximum use was made of the Medical
Boards and Boards of Inquiry held at the Medical Life Support Division at
DCIEM Toronto. Information on early accidents is sketchy; techniques of
investigation were not as well developed as they are today, and knowledge of
human factors was very limited. In many cases, cause factors assigned could,
by today's standards, probably be debated. The author of this report
disagreed with a number of findings in some Boards, however, in most cases,
deference to the official Board was maintained. In several cases, this
author would not accept some conclusions, and as a result, the data contained
within this report may not strictly conform to official statistics. For
example, there were several cases in which a completely successful ejection
was carried out but the individual died from the effects of drowning or
* exposure (hypothermia) before being rescued. Smiley documents such cases as
"fatal ejections". Clearly they are not fatal ejections in the context of
this report. They resulted from a failure in either the rescue system or in
the ability to survive until rescue. As a cautionary note, the accent of
this paper is not on accident cause factor analysis, but is rather an
analysis of the dividing line between a fatal and non-fatal accident. The
cause of an accident is rarely the cause of the fatality. In this important
respect, the accident analyses in this report differ from cav;e factor
analyses conducted officially by DFS. A total of 720 personnel involved in
ejection seat aircraft mishaps were identified (Table 7).
It is also necessary to define terms used in this report. Within the
context of this study, an "attempted ejection" occurs when an individual
makes the initial attempt to carry out the ejection sequence. It is
indicative that a decision was made to abandon the aircraft, of recognition
of an emergency as life-threatening. In older manual ejection systems,
separate jettisoning of the canopy indicates the initial attempt to eject.
* In current systems, the first action to eject is to raise the ejection
handles. This distinction can be important in the investigative process. In
a fatal accident, if it can be established that there was intent to eject,
some cause factors may be eliminated such as "incapacitation". Sometimes, it
is not readily apparent that the pilot had such intentions. For example, it
was originally concluded in the crash of a CT133 at Peggy's Cove in 1982 that
* there was no attempt to eject -the aircraft had abruptly bunted nose down at
a low altitude and crashed almost vertically with total disintegration.
Theories of incapacitation of the pilot were formulated to fit this
conclusion. Yet the evidence had been there, and a year later the Aerospace
Engineering Test Establishment found that the pilot had pulled the ejection
handles before impact. Evidence was obtained from examining the canopy
remover, drop-away tube and canopy frame and conducting an inch by inch
sectional analysis of the ejection gas lines.
V l
Ja

* 8
"Attempted ejection", as depicted in Table 6, is further sub-divided
into "completed" and "not completed" A "completed" ejection is one in
which the individual exits the cockpit. Included within this group are
"manual bail-outs" and "inadvertent ejections". There were five manual
bail-outs and three inadvertent ejections recorded. Of the manual bail-outs,
four were from CT133 aircraft and one from an F-86. Only one bail-out
(CT133) was unsuccessful. An "attempted ejection" that was "not completed"
includes those identifiable situations where the individual commenced theejection sequence (jettisoned the canopy in the manual systems or pulled theejection handle in the automatic system), but was not ejected from the
cockpit. Failure to eject was attributable to some failure of the ejection
system -the canopy would not jettison (manual system), the ejection seat
would not fire, or the individual could not work the system due to
interference from other factors (negative G forces, equipment hang-up).
There may have been cases where the ejection system was activated prior to
impact, but aircraft disintegration occurred before the seat catapult was
fired. In fatal accidents, intent to initiate ejection is difficult to
establish in situations where it is obvious that the seat was not ejected
from the cockpit. Even in those cases where the seat has been ejected, it is
necessary to ensure that ejection was not the result of crash forces at
impact.
Another term requiring definition is "no apparent ejection attempt".
This includes all accidents where there was no conclusive evidence that the
individual attempted to eject from the aircraft prior to impact. Over 30% of
V.- the individuals involved in "A" category accidents, as far as can be
determined, made no attempt to eject. Eighty per cent of these individuals
were killed. The causes for not making an attempt to eject will be analysed
later to determine why, and what can be done to decrease these losses. It is
important that the Flight Surgeon, as far as is possible, establish intent to
eject. Failure to do so may result in the establishment of incorrect
hypothesis for the accident causation, and worst of all, may result in a
conclusion of "Cause Undetermined".
DECISION PHASE
Making the decision to eject is the single over-riding factor
* influencing survival. For example, from January 1972 to December 1987, of
147 personnel involved in "A" category mishaps (Annex A), 45 or 30% did not
eject. Thirty-seven (82%) of them were killed. It is probable that most of
them did not even make the decision to eject. This figure may be as high as
'-" 27 of the 37. Not only is making the decision to eject necessary to
survival, but the decision MUST be timely. Failure to execute the decision
* due to delay has claimed many a pilot.
Arriving at a decision to eject from an aircraft in flight, in many
cases, is not easy. It is currently accepted that the role of the pilot in
any man/machine interface is that of an information processor inserted in the
* control loop. The pilot receives input from the environment through the
sensory systems (kinaesthetic, visual, vestibular) in analog fashion. The
sensory organs act as transducers, relaying the information to the brain.
%

9
Due to physiological limitations placed on the sensory systems, the brain
never receives a 100% accurate "picture" of its environment. Sensory input
is usually lacking in either quality or quantity. There are five basic steps
required in the formulation of any decision and therefore, five possible
error points. To begin, the pilot must acquire the information from the
environment through the sensory systems (subject to their limitations). The
information must be integrated in the brain, and the quality and quantity of
the information must be perceived as reliable and adequate. It is then
processed centrally until a limited number of decision alternatives are
formulated (eject, force land). The pilot then uses judgement to assess the
probability of success of his limited number of decisions and selects one
course of action. The selection of the decision is highly individual, based
on training, experience, personality, and physical and physiological state.
For the decade 1977 to 1986, 24% of the "personnel-pilot" cause factors
* involved faulty judgement. Lastly, the pilot must implement the selected
decision using learned skills. In the same decade, 30% of the "personnel-
pilot" cause factors were due to "technique" errors. CFP 135 Chapter 16
Annex D lists standardized formats for cause factor analysis. Table 8 lists
the analysis of Air Accident Personnel-Pilot Cause Factors for 1977-1986.
Of particular importance to the Flight Surgeon is the concept of
information processing/judgement as these two actions are so highly
individualized and subject to existing physiological/medical influence.
Judgement is the result of information processing; implementation, the result
of judgement. It is necessary to assess the cockpit environment, both
* physical and ergonomic, and the physiological state of the pilot when
analyzing a pilot's decision or lack of decision in the handling of an in-
flight emergency. The probability that the pilot's judgement in selecting a
decision response will be correct depends on:
* a. The quantity of sensory input used to select decision alterna-
tives. Too much and too little input are detrimental to decision
formulation, especially in rapidly changing situations. The rate
that information must be processed is a limiting factor in
decision selection. It is also generally accepted that a human is
a single channel sequential information processor, and can only
devote 100% attention to one novel event at a time. Competing
stimuli must be prioritized by perceived importance, a totally
cognitive and subjective function (Table 9). Recent accidents
involving heads up displays (HUD) on the CF188 give current
meaning to this axiom. On a recent accident (CF188), Air Command
Headquarters was prompted to explain:
I
"Information presented on the HUD or DDI (Digital Display
Indicator) during a spin … (is) displayed in brief flash
sequences … not long enough to be useful for any constructive
action. The rapidly changing aircraft attitudes far exceed the
ability of the human processing system to keep up and apply
correct control inputs as required … it takes approximately 500
hours to become comfortable with all aspects of the HUD displays.
This is at least our second episode where the loss of a CF188 may
have been facilitated by problems related to difficulties in

10
' -interpreting the information on the HUD."
b. The quality of the input and its reliability. It is a well known
fact that a pilot learns to disregard most sensory input during
flight, and to seriously question the rest. Instrumentation
packages are provided for conflict resolution, yet disorientation
* conflicts still persist (CF188717). Humans are "pattern
recognizers" and current advanced digital/analog display cockpits
do not present aircraft attitude changes in patterned formats that
can quickly be assimilated by the pilot;
" c. The manner in which the information is presented. This includes
ergonomics and instrument design and layout, and the use of visual
presentation (HUD, DDI, HDD, flashing lights) or auditory
presentation (buzzers, beepers). Transitioning from heads-up
displays for precision weapons delivery to heads-down displays for
aircraft orientation can create fatal delays in information
%I processing;
• d. Individual processing capability. This varies diurnally with
circadian rhythms, and is affected by numerous operants such as
personality, training and experience, physical and physiological
state at the time of the emergency, and medical condition. Most
of these factors influence autonomic arousal and there is an
inverted 'U' relationship between performance efficiency and
arousal (Figure 18). Generally, as arousal level increases,
performance efficiency also increases until at some point
performance is maximum. As arousal continues, performance begins
to fall. Note also, that when an individual is working at maximum
performance, a removal of stress, or a reduction in arousal, may
also result in a decrease in performance. The double curve also
illustrates that performance on complex tasks deteriorates before
that of simple tasks as arousal increases. This is known as the
Yerkes-Dodson Law (19).
In cases where an individual does not have sufficient information
* quality or quantity, the information may be interpolated or extrapolated to
produce a more subjectively complete picture. The degree of interpolation or
extrapolation depends highly on training and experience and the rapidity with
which information is changing. A pilot who makes a decision error based on
erroneously interpreted input commits a false hypothesis. Accepting the
false hypothesis may result in an error due to cognitive failure: commission,
0? or doing the wrong thing; or omission, doing nothing. A false hypothesis is
most likely to occur under conditions where arousal levels are distorted:
5,- a. when the individual has been pre-conditioned to expect an event to
occur or not occur;
b. when arousal levels are distorted;
-'•

c. following periods of high arousal;
d. when attention is diverted; and/or
e. when incorrect hypothesis or beliefs have been held for a long
period of time.
One most common form of false hypothesis is "expectancy", an
officially recognized personnel-pilot cause factor. Expectancy arises
whenever an individual is pre-conditioned to an event occurring or not
occurring, and reacts inappropriately when cue stimuli are presented. With
extreme expectancy, an individual may perceive cues that are not really there
(interpolate, extrapolate), or conversely, may not perceive cues that are
presented. The following histories illustrate cases of expectancy.
A CF16 test pilot departed on a Pulse Code Modulation check flight in
* an aircraft configured with three full 125 gallon fuel tanks and empty rocket
launchers with nose cones. Shortly after take-off, following landing gear
and flap retraction at 230K, the pilot noticed that "at least one" of the
', Fire Warning lights were illuminated. He then retarded throttles from MAX AB
to IDLE with a brief pause at MIL. Fire warning lights remained on. He then
advanced power to MIL, receiving normal thrust response, then back to IDLE
and to MIL again. Both times, when advancing from IDLE to MIL, the aircraft
yawed left. The aircraft at this time was at an altitude between 200 and 400
feet AGL. The pilot, believing that the fire warning and yaw control
problems were related, ejected from the aircraft. The Board of Inquiry (BOI)
determined that all aircraft systems were operating normally at aircraft
impact. Bulb filament analysis indicated at least one of the two fire
warning bulbs was not on at impact, and the yaw control problem was a normal
aircraft r.sp 2onse when configured in the heavy gross weight/high drag
configuration. The pilot did not follow the AOl's during the perceived
emergency, and as a result, ejected from an apparently serviceable aircraft
after assuming engine problems! (He did not eject external stores);
A CF188 pilot attempted to take off from Cold Lake on a deployment and
ferry mission with a 707 tanker support to CFS Goose Bay, and ultimately
Baden-Soellingen. As the aircraft accelerated to full military power
U followed by AB selection, the pilot attempted to rotate the aircraft at the
lift-off speed. The aircraft failed to rotate, and abort procedures were
initiated at the 7200 foot mark. The arrester hook was lowered but failed to
engage the cable and the aircraft continued to roll off the departure end of
the runway into the approach lights area and exploded. The pilot ejected
about one second before the fire-ball and suffered a sprained ankle and some
bruises and muscle strains. The briefed take-off procedure had called for a
stabilator setting of eight degrees NU (nose up). During a thirty minute
delay, the pilot completed his pre-take-off check and inadvertently set the
stabilator setting to eight degrees ND (nose down) on the DDI. Prior to
take-off each pilot called their stabilator setting as eight degrees NU.
This was the second time the pilot visually mis-read eight degrees ND as NU.
Additionally he elected to do a heavy weight take-off from the 2000 foot
mark, and delayed his ejection for eleven seconds after crossing the arrester
cable. The hook failed to engage due to nose down trim and excessive
braking.
…. 1k'N~k I.."

12
As previously mentioned, successful escape also depends on making the
timely action of initiating the ejection sequence following the decision.
Delay may occur while alternate actions are being addressed, by the need to
steer the aircraft away from an inhabited area, or even by a perceived need
to salvage the aircraft beyond reasonable limits. Among the factors the
Flight Surgeon must analyse if undue delay in initiating an ejection sequence
is suspected are:
a. physical and physiological reaction times (receptor-effector,
cognitive, perceptual);
b. personality (fear of implementing the decision, ego, attempting to
salvage the situation beyond rational limits);
c. temporal distortion (failure to appreciate "real time");
d. complacency, or lack of anxiety (a pilot accustomed to high speed,
low level flight may l]se or suppress the "low altitude anxiety
response", and fail to appreciate the rapidity at which the
aircraft is travelling);
e. lack of appropriate training (unfamiliarity with egress pro-
cedures, habit-transference, ambiguous AOI's);
f. self-sacrifice (turning the aircraft away from inhabited areas,
delaying ejection while a passenger ejects);
g. technique error (too many salvage attempts, trying to enhance
ejection parameters, "groping" or "fumbling);
h. control surface, control loop response delays;
j. miscellaneous delay-inducing effects of negative "G", fatigue,
stress, circadian desynchrony and transient situational
disturbances.
The following accident illustrates a case of excessive delay:
'V The pilot of a CF116 took off on a test flight following a double
engine change, to be followed by an air display practise. The aircraft was
configured with a refuelling probe and three pylons without tanks. Followingsatisfactory run-up and take-off, the pilot flew a practice air display over
a lake north-east of the airfield while waiting for his block time. His last
manoeuvre was to be a battle break followed by a full stop landing. The
aircraft ran in for the break at 430 KIAS and 80-90% rpm. Pulling up for the
break, the pilot throttled back to IDLE and selected SPEED BRAKES OUT.
During the break the MASTER CAUTION and RIGHT GENERATOR lights came on.
Assuming generator failure he rolled out at 1800 feet and lowered the landing
* gear and flaps. Getting no throttle response and noticing both engines had
decayed to 12% rpm, the pilot identified a double-engine flame-out. Two
"TIGER STARTS" were attempted, the throttles shoved into max AB and back.
The engines failed to relight and the aircraft was sinking rapidly. Deciding
IX.,

-S
13
to eject, the pilot reached for the handles and missed. Straightening the aircraft, the pilot then looked down, grabbed the handles, and ejected at 400
feet. The investigation revealed that the pilot had been reluctant to eject
initially because of a perception that the Base Commander and Squadron CO
would not appreciate losing one of their aircraft. As a result he had wasted
valuable time at extremely low altitude attempting two relights which, even
if they had been successful, would not have saved the aircraft from its
descent trajectory. The ejection and crash occurred in a valley below normal
ground level. If it were not for this, the delayed ejection would have in
all probability been fatal.
A Flight Surgeon investigating a fatal accident where ejection was not
initiated faces a difficult challenge in trying to determine why ejection did
not take place, especially in situations where all aircraft components are
determined to have been functioning normally at impact. In such cases, in-
flight incapacitation or distraction are always highly speculative,
especially if witness statements indicate there was no observable attempt to
avoid ground collision. Physiological collapse, disorientation, inattention,
and self destruction would also account for such an accident. In all such
cases, deductions must be made from evidence gleaned from personal medical
* documents, autopsy (coronary artery disease, cardiovascular accident),
toxicological screening, psychosocial/psychiatric investigation
(psychological autopsy), and in-flight factors that may have been operating
prior to the accident (hypoxia, "G" forces, etc.). In addition to
physiological or psychological impairment, ejection failure may also occur
from physical factors that make ejection impossible such as severe negative
Cz or ejection system malfunction.
As stated earlier, there were at least 37 fatalities arising from
failure to eject during the period January 1972 to December 1987. These
fatalities may be loosely categorized into three groupings that account for
failure to eject (Table 10): unable to eject, unperceived aircraft to ground
closure (controlled flight into terrain) and rapidity of events (one case of
unknown cause).
Unable to elect. Eight cases (21.6%) may have arisen through some
* physical or physiological factor(s) preventing ejection. This group may
include such diverse factors as severe negative G forces (which inhibit
reaching the ejection handle), violent aircraft manoeuvres, in-flight break-
up, mid-air collision, ejection system failure, loss of consciousness, and
psychological impairment (hypoxia, drugs). The following three case
histories illustrate accidents in this category:
The first case involved a CF104 aircraft in an air-to-air gun attack
on a towed target (DART). During the third successive attack, the aircraft
entered into a 15 second 4.5 -5.0 Gz after-burner turn, then abruptly went
30 to 40 degrees nose low, continuing on to impact. After examination of all
*the facts, the BOI concluded that the most likely causes of the accident were
either acute severe barotrauma "G" induced loss of consciousness (LOC). The
pilot had a resolving upper respiratory tract infection and mild chronic
sinusitis with post nasal drip. Additionally, he had been on a self-imposed
%

14
dipt for an undetermined time pre-accident, was a long distance runner, and
had only 5.5 hours of sleep the previous evening. All are known G-tolerance
reducers. Finally, rapid, successive +Gz loadings have a cumulative effect
on lowering arterial oxygen saturation.
The second case involved a CF104 dual on a basic fighter manoeuvre
training mission in defensive counter-measures. During the second air-to-air
gun attack against another CFl04, the aircraft performed a defensive roll-
under followed by a nose-down extension manoeuvre. The aircraft entered
cloud at 2000 feet in a 20 to 30 degree nose down attitude and impacted the
frozen surface of a lake at high velocity. It was felt by the BOI that
during the -Gz defensive manoeuvre, both pilots' seat packs had dislodged
from their seat pans, jammed the control column forward and prevented the use
of the ejection "D" rings on the front of the seat pan.
The third illustrative case involved a T-33 with two pilots that took
off on a full card air test. One minute after take-off the aircraft captain
radioed that he was experiencing control difficulties. The aircraft was
laterally unstable and remained in left bank as a result of what was later
diagnosed as mis-rigged ailerons. After several circuits about the airfield
* the decision was made to eject. The rear seat pilot ejected first. The
front seat pilot then flamed the aircraft engine out and attempted to eject.
The seat would not fire as a result of the ejection handle jamming in the
'partially raised position due to excessive play in the linkage and a design
fault. The pilot was unable to re-light the engine due to low rpm. Twenty-
five seconds after the rear seat pilot had ejected, the front seat pilot
transmitted with great finality that he was unable to eject. Twenty-six
seconds later, the aircraft stalled, rolled inverted and crashed into a stand
of trees killing the pilot.
Unperceived Ground Closure. This is also called "controlled flight
into terrain (CFIT)" and "situational unawareness". Nineteen of the thirty-
seven fatal cases (51.4%) fit into this category. In all cases the pilot,
apparently in complete control of the aircraft, flew into the ground or
water, or entered into some aerobatic manoeuvre such as a dive, loop or roll
without sufficient air space to complete the action. CFIT may result from
visual illusion, inattention or distraction, willful self-destruction
* (deliberate sabotage) or physical factors such as insufficient "G" for bank.
In all accidents in this category, there did not appear to be any
physiological or psychological impairment accounting for a lack of ground
avoidance procedures. Four aircraft and six aircrew flew flight paths into
water, and 11 aircraft and 13 aircrew flew into the ground. There are manyinstances of CF104 aircraft flying into the ground as a result of the pilot
" attempting to maintain visual meteorological conditions (VMC) in
deteriorating weather while flying over rising ground. The following case
history illustrates a "controlled flight into terrain" accident.
Kiwi 17, a CT133 with two pilots onboard took off from Baden-
* Soellingen on a planned low level navigation mission. Twenty minutes into
flight, their track converged with that of a Belgian F-16 also on a planned
navigation route. The F-16 pulled up and turned left in a mock attach on the

U
15
*CT133. The CT133 pilot apparently responded with a defensive steep left 4G
– 180 degree turn. Rolling out of the turn, the aircraft struck a stand of
trees on a ridge. The abrupt decelerative force probably rendered both
pilots unconscious, and the aircraft continued descent to ground impact where
it exploded. The most probably cause was channelized attention in that both
pilots were probably concentrating their attention on the attacking aircraft
to the exclusion of clearing their flight path.
Of interest is a 1983 USAF study on distraction and vigilance (Table
11). Using experienced A1O instructors, the study found that a simple TACAN
channel change required as much as 15 seconds with a corresponding aircraft
altitude loss due to distraction of up to 100 feet (average 38 feet). More
complicated secondary distractions such as fuel computations resulted in
* altitude losses as great as 900 feet. Transitions between the exterior of
the cockpit to the interior and back consumes valuable time and creates
momentary "loss of situational awareness".
Another CFIT accident involved a CF104 dual on a cross-country visual
* navigational trip in Germany. The final leg of the trip was discovered to be
*marginal or below visual meteorological conditions (VMC), and the pilot
.. apparently deviated right of his track in an attempt to acquire better VMC.
Turning right, he apparently passed through a patch of low lying cloud.
* Diving from the cloud, the aircraft levelled off and continued to fly a
shallow descent until it impacted the crest of a ridge at five to ten degrees
* right-hand bank and ten to twenty degrees nose up pitch. After initial
impact, the aircraft "skipped" into the air and the fuel tank exploded
accelerating debris and the pilots' bodies over 1,000 feet. Both pilots were
probably killed by the initial decelerative force at impact. Lack of
characteristic fractures of the front seat pilot's right hand indicate that
he may have had his hands on the ejection "D" ring at the moment of impact.
A similar situation occurred during another CF104 visual navigation
sortie in marginal weather conditions. For undetermined reasons the pilot
attempted to maintain VMC and deviated 3.8 nm left off his track over gently
rising ground. While attempting to accurately pin-point his geographical
position, he may have momentarily diverted his forward vision to his map.
The aircraft, flying at 450 knots, crashed into the top of a dense stand of
snow-covered trees at the crest of a hill in a descending flight path. The
-, aircraft and pilot disintegrated through the trees.
Disorientation may also be involved in a number of accidents in this
category. As a contributing factor, disorientation must be deduced by a
process of elimination. Reconstruction of the aircraft flight path and
flight manoeuvres may provide evidence to the possibility of disorienting
conditions. Witness statements may provide valuable information on any last
minute aircraft attitude changes which may be explained by disorientation.
The following case history outlines one accident wherein disorientation was a
probable cause factor.
A CF188 pilot proceeded from Bagotville to Summerside to participate
%* V

16
in an air show. As the weather was unsuitable for flying, he spent the day
with his aircraft on static display. Later that afternoon, he taxied to the
runway for the return trip. On take-off he did a steep AB climb into cloud
based at 300 feet and 4000 feet thick. Twenty-five seconds later the
aircraft descended from the clouds and impacted the water of Malpeque Bay off
the end of the runway. Information from the data recording system supported
the theory that the pilot experienced a somatogravic illusion during climb,
pushed forward on the control column and induced the classic inversion
illusion.
Rapidity of Events. Low-altitude high speed tactics leave little
margin for error. In many cases, the time margin may be less than human
reaction time. In critical situations where time and air space are marginal,
,i ~the time required for reaction may be the precipitating factor in a mishap.
Total minimum reaction time has been calculated to be as long as nine seconds
in a complex man/machine system where a decision must be formulated from
• alternates. This time would significantly increase whenever arousal levels
are distorted (stress), or when multiple simultaneous emergencies occur
(workload). To minimize decision times, aircrew must predetermine their
reactions to any given set of conditions. However, if the situation is
misinterpreted, the wrong procedures may be initiated! Nine fatalities
(21.6%) probably resulted from a lack of time to react.
Accidents in this category resulted from the occurrence of two primary
situations; loss of power on take-off, or a lack of sufficient air space to
complete an initiated aircraft manoeuvre such as a dive, turn, or roll. To
illustrate: A CF116 dual with a student and instructor crashed on the runway
following a touch-and-go approach. The aircraft had initially touched down,
then after a short roll the student lifted it off with excessive nose up
attitude. At 50 feet above ground the aircraft began oscillating, then
abruptly rolled right and crashed with 135 degrees right bank and slightly
nose down attitude. The Board concluded that the student had allowed the
aircraft to become prematurely airborne with excessive nose up attitude,
thereby creating an aerodynamic stall. Ejection was not initiated due to a
S lack of response time as a result of extremely low altitude and high sink
rate.
In some instances, the personality of a pilot may contribute to a
mishap. Strong desires to succeed stemming from egotistical, compulsive, or
borderline personality disorders may over-ride rational decision-making
process in an emergency situation, and indeed may even be responsible for the
emergency situation arising, e.g. "press-on-itis". Underlying personality
may have directly contributed to four fatalities and the loss of three
0 aircraft.
".- .-
'p
0
% % Z

17
" The first illustrative case involved a CT1I4 student pilot on a clear
hood solo flight. He was authorized to practice stall procedures and slow
* flying in the area, and then some touch-and-go landings in the circuit.
After take-off, he deliberately deviated from his flight plan, proceeding to
his aunt's farm where he made two low passes. Following this, he flew east
ft. to his cousin's farm and made a low pass. On the second pass, he flew over
in a left bank, then lined up for another pass. As he went by, he entered a
'. left roll that continued to aircraft inversion. The nose pitched down 45
degrees and the aircraft struck the ground and exploded. During his flying
training, his instructors had consistently used the word "over confident"
when describing him. His last flight was a pre-planned "pretentious display"
of "showing off", that allowed him to get into a situation from that his lack
of experience would not allow him to recover.
* The second incident is similar in most respects. This rather
inexperienced CT133 pilot filed a visual flight rule (VFR) flight plan
* knowing that enroute weather was below VFR minimums. He deliberately
deviated northeast of his track to a small civilian aerodrome where he had
previously done some sky diving, and made two high-speed low level passes
f' xdown the runway, executing a crisp right roll on the second pass. Banking
right, he appeared to be preparing to rejoin his original flight track, but
suddenly made a tight 90 degree left bank to return to the spectators that
had gathered. During this fly past, he executed a "sloppy" right roll and
began a left roll. At 45 degrees into the roll, the aircraft nose pitched
– down 20 degrees, and the aircraft flew into the ground, exploded, and
* disintegrated. During the investigation, the pilot was described as 'self-
critical", and "underconfident" during his flying training. The BOI
concluded that his personality contributed to his desire to "show off",
attempting flight procedures for that he was grossly inexperienced.
The last case involved fatalities because of a conscious delay in the
decision to elect. This aircraft accident involved a CT114 with a student
and instructor pilot on an extra dual training mission. On the final leg of
i! the return trip to home base, the pilot transmitted an unspecified emergency.
For 20 seconds, the instructor managed to stabilize the aircraft in a wings-
level, shallow descent. At 8.5 NM from base and 50 feet AGL, a slow speed
stall occurred. The aircraft abruptly banked 33 degrees right, pitched 21
degrees nose down and crashed. Why had both pilots not ejected from the
. aircraft during the final 20 seconds? It is interesting to note that the
instructor had previously been commended for recovering a CT114 after an
engine failure in flight.
Non-Fatal "A" Category Mishaps. Four aircraft were involved in
mishaps where the occupants did not eject, but survived. In three cases, the
accident occurred after the aircraft had landed, breaking the vertical
descent rate, and in one case the accident occurred on the take-off roll.
The following two case histories illustrate accidents in this category.
The first accident involved a CF101 on a radar square pattern approach
to a full-stop landing. Due to pilot fatigue, post-alcohol syndrome, and a
visual illusion, the pilot allowed the aircraft to touch down 90 feet short
%

* 18
of the runway threshold. It continued onto the threshold where the right
main landing gear collapsed and separated from the aircraft. Continuing to
slide, the aircraft departed the right side of the runway with the right%wingtip and aileron dragging, and the navigator ejected. After departing the
runway, the nosewheel sheared off, the aircraft slewed right collapsing the
left main wheel inboard, and the left wing dug into the ground flipping the
aircraft over. The pilot, who had remained with the aircraft, was later
rescued from the inverted cockpit suffering only a fractured left clavicle
and some contusions. The navigator received two superficial bruises to the
head.
The second accident involved a CF116 during a TACAN straight-in
approach. The runway had not been cleared of snow, leaving a windrow
ploughed down the centreline. On touch-down, a large hole was torn through
the floor of the front cockpit allowing snow to enter forcing the centre
console to the left trapping the pilot's left leg and fracturing both legs.
Snow pressure actuated the mechanical tripper, firing the RPI lap belt and
butt snapper, forcing the pilot up and forward, enhancing his leg injuries.
The back-seat pilot received no injuries, and egressed the aircraft after it
came to a stop. The front seat pilot was unable to extricate himself until
assisted by rescue personnel.
In all fatal no-ejection accidents, the Flight Surgeon must conduct
the investigation by listing all possible cause factors that could account
for the failure to eject; then through exclusion, arrive at the most probable
cause(s). Close liaison with other BOI members to compare facts and theories
is mandatory. The following should not be overlooked as potential sources of
information:
a. Flight Path Reconstruction. Instrumentation analysis, flight data
.0 recorders, flight briefing data, witness statements, strip maps,
and weather reports provide clues to the flight path and flight
environment and may provide evidence for the existence of
– disorienting or illusory conditions or support hypotheses of
physiological collapse. Fly the exact route under the same
S,'conditions, if appropriate.0
-" b. CF2034, Look for any evidence of a medical nature that may
corroborate hypotheses of physiological collapse or psychological
dysfunction such as: medications, recent illnesses, psychological
or psychosomatic complaints su-h as fatigue or depression.
c. Witness Interviews. In some cases, this may be the only evidence
on that to formulate accident cause theories. Spouses, friends,
relatives, or other squadron personnel may provide clues to
personality, intentions, psychosocial stress, fatigue, workload,
or substance abuse that may support or confirm other evidence.
d. Training Records. Course reports and performance assessments can
"?,
%

– -~Wiw 1VW~ WVWV WV7
19
show deficiencies in training or technique and even provide
insight into character or personality traits such as "immature",
"over confident", "aggressive", or "lacks confidence". This is a
good place to gain insight into currency and proficiency of the
individual or individuals involved.
e. Aircraft Operating Instructions. Determine if the individual
adhered to the AOIs. Are the AOIs current and complete?
f. Human Engineering. Does the position of switches, warning lights
and instruments lead to possible disorientation or distraction
during certain phases of flight? Can instruments be
* misinterpreted? Could there have been any habit transferrence
during an emergency (in this regard was the individual qualified
on two different aircraft types, or had the pilot recently flown
different aircraft)? Is the flight data presented in a way that
allows for significant and rapid assessment by the pilot (for
example, the CF18 HUD presents rapidly changing information that
is difficult for the pilot to integrate quickly)?
g. Laboratory Analysis Instrumentation settings, power settings,
switch positions, warning light bulb filament analysis and engine
inspection provide evidence of flight conditions at impact, and
may provide clues to events occurring within the cockpit prior to
impact.
h. Wreckage Distribution Plot This gives a picture of the aircraft
attitude and approximate speed at impact.
j. Toxicological Drug Screen Analysis of bod -luids may show the
presence of a foreign substance.
k. Autopsy Look for evidence to support incapacitation -coronary
artery disease, aneurysm. For multiple fatalities, x-rays of
hands and feet may provide clues as to who was flying the aircraft
at the time of impact.
m. Aircraft Maintenance Records These log books may show recent
problem areas or recurring mechanical failure patterns that may
fit the accident profile.
n. Pilot's Flying Log Book. This diary shows the individual's
experience, currency and proficiency. It may indicate possible
* flying fatigue, and in rare cases, even personality. For example,
one pilot involved in a mid-air collision had been using his log
book as a personal diary, entering off-handed demeaning and
sarcastic remarks about his superior officers and other squadron
Zo ,,

!o 20
personnel. This helped the BOI to evaluate his personality
profile.
o. Flight Recordings Analyse gun camera film or data recording
system information whenever available.
p. Psychological Autopsy Hobbies, music collection, marital and
sexual history, type of car (bumper stickers, custom plates),
clothing, pets, lifestyle and locker contents all provide clues to
personality and mental state.
EJECTION PHASE
Between January 1952 and December 1987, at least 481 individuals
\ -3. attempted to eject from RCAF/CF aircraft. Of these, 450 completed the
ejection sequence or bailed out; 74 were killed following egress. This gives
an approximate 35 year ejection success rate of 80.5% (Table 7).
Normally, the ejection phase immediately follows the decision phase,
though not necessarily. When it does, it is usually preceeded by a time
delay. The delay is equal to the individual's reaction time plus the system
reaction time. Tables 1 to 4 show that the system reaction time is
mechanically fixed from 0.3 seconds for CF188 seat ejection to 1.5 seconds
for CT133 seat ejection. Human physiological time is variable. By far, the
greatest number of fatalities arising following ejection initiation were
4." because of ejecting outside the design limits of the system, or "out of the
envelope". Modern ejection systems are one-step, irreversible and fully
-' automatic. Only a system failure will prevent ejection once the handle is
pulled. Figures 12 through 15 illustrate the automatic sequence for current
CF aircraft. A generalized ejection sequence is shown in Table 12.
Once the ejection handle or "D-ring" is pulled, the ejection sequence
has begun. The ejection phase ends with rocket motor burn-out. Although the
ejection phase is extremely short, it accounts for many of the injuries that
occur. Included within the ejection phase are the effects of explosive
decompression, positive Gz acceleration and windblast. Since the Flight
Surgeon is responsible for analysing injuries occurring on ejection, it is
important that there is an understanding of the events that may produce
injury during this sequence.
Activating the ejection sequence results in immediate jettison of the
aircraft canopy with cockpit depressurization to ambient. This exposes the
ejectee to all the attendant risks of altitude including: hypoxia,
decompression sickness, barotrauma and cold. Several cases of conjunctivitis
have been reported because of dust and dirt from the cockpit floor entering
* the eyes during depressurization. Since 1972, all ejections have occurred
below 16000 feet MSL (Figure 22) however, before this ejections tended to be
.. higher (Table 13). There are numerous cases of F86 and CF100 ejections above
Z3, -1A

-0. – w ,rw r – -i. W-4 P-u ,
421
20000 feet MSL, and several of these were above 25000 feet. At least four
ejections were over 30000 feet, the most recent being from CF100789 in May
1971. Annex B presents some ejection parameters for CF ejections since 1972.
A few cases of hypoxia following ejection have been recorded. These appear
to be confined to high altitude F86 ejections where the oxygen bottle and
mask were lost on ejection, resulting in varying degrees of hypoxia during
– descent and in one case, loss of consciousness (F86 23278 May 1955). No
recorded cases of decompression sickness were found. Aerodynamic suction
also occurs on canopy jettison because of Bernoulli's principle. This
suction has been known to pull an individual off his seat cushion because of
slackness in the harness restraint system, thereby increasing theaccelerative "jolt" on the spine at the time of seat ejection.
Injury analysis of 67 successful ejections from January 1975 to
December 1987 (Table 15) show that there is over an 86% chance that some form
of injury will occur during ejection. During the ejection sequence, all of
the following injury mechanisms have been known to occur:
.
a. knee contact with the instrument panel to produce fracture and/or
|* laceration;
b. knee contact with the canopy bow to produce punctures, lacera-
tions and fracture;
c. knee and thigh contact with arm rests or ejection handles to
produce contusion;
d. elbow contact with the canopy sill;
e. collision with loose objects in the cockpit;
f. shoulder contusion because of ballistic inertia reel harness
retraction forces; and
4
g. posterior thigh contusion from seat pack contact.
As the ejection seat ascends the guide rails, positive Gz force is
created. Table 14 provides some physical data on current CF ejection seats.
Accelerative forces vary between 14 and 20 Gz, with onset rates from 180 to
300 Gz/sec/sec. The actual loading on the body depends on the occupant's
*weight and the slackness in the restraint system. The CF116 is the worst
case scenario. For a 5% body weight male, peak "G" will be in the vicinity
of 20 Gz. For a 5% female, peak "G" may reach 28 Gz. All ejection seats
4 except the CF188 are designed for the 5 to 95 percentile individual; the
CF188 seat is designed to accommodate the 3 to 98 percentile (130 to 225
_ pounds) individual. Lighter individuals tend to experience higher "G"
loading, a faster ride up the ejection rails and greater seat tumbling.
4

22
Conversely, heavier individuals would experience lower "G" loading, a slower
ride up the rails and a decreased seat height trajectory.
Figure 19 illustrates a "typical" acceleration versus time pulse of a
ROCAT seat. A ballistic charge or charges explodes creating pressure that,
contained within telescoping tubes, propels the pilot/seat system up the
guide rails. As it does so, a mechanical tripper or wire fires the rocket
motor that provides a fairly constant acceleration over the next 0.25 seconds
or so. The initial impulse thrust is approximately in the Gz plane, varying
from 10 to 22 degrees rearwards of the vertical, depending on seat type, to
allow instrument panel knee clearance. Since the centre of gravity of the
head and torso lies anterior to the vertebral column, the head is rotated
forward and down creating flexion and strain of the cervical spine and
musculature. In addition, the inertia of the viscera connected to the rib
cage and spine increases the load on the vertebral column. The vertebral
column is a compound "S" curve that can be divided into four regions of
alternating curvature:
a. cervical spine, anteriorly convex;
b. dorsal (thoracic) spine, posteriorly convex;
c. lumbar spine, anteriorly convex; and
d. sacrococcygeal spine, posteriorly convex.
Effectively, however, the spine may be divided into two at T12 to form the
cervicodorsal column and lumbosacral column.
–During the ballistic phase of seat ejection, the spine and body act in
semi-rigid fashion (intervertebral disc compression), but the fleshy buttocks
and seat pad compress, allowing the ejection seat to build up a relative
velocity before striking the occupant. The travelling seat strikes the
stationary occupant with a "jolt" that may exceed vertebral tolerance (18 to
20 Gz). This is known as "dynamic overshoot". If the occupant has adopted
poor posture such as flexing or slumping, a shear force through the area of
greatest curvature (thoraco-lumbar) may result in fracture. Loose restraint
harnesses also magnify accelerative jolt. When the upward travelling seat
strikes the stationary buttocks, the occupant is accelerated to a greater
velocity than the seat and is momentarily extended into the shoulder harness.
The looser the harness, the greater the extension. As the harness
decelerates the upward travel of the occupant, the accelerating seat impacts
the buttocks again, creating an extremely high impact pulse that may be in
* .-. the order of 500 G/second. There are two primary vertebral injuries that may
occur during the ejection phase as a result of excessive forward flexion of
the spinal column because of poor posture or loose restraint systems:
Anterior Cuneiform Fractures These are caused by forward flexion and
vertical compression forces. Most frequently, these are characterised by
N.. anterior wedge fracture of the upper and/or lower vertebral lips and a
…..%"-.W

23
decreased height of the vertebral body with or without protrusion into the
spinal canal; and
Comminuted Fractures These are caused by obliquely acting forces that
drive the anterior inferior edge of the superior vertebra into the upper
surface of the inferior vertebral body. The intervertebral disc is usually
torn and ligaments ruptured. The vertebral body is shattered and the
fragments expelled. As the angle of the ejection seat back increases,
anterior cuneiform fractures tend to become comminuted. In addition,
increasing the seat back angle creates greater flexion of the thoracic spine
necessary for optimum head attitude, that increases the risk of spinal
fractures in the mid-thoracic region. The USAF reports 15 to 20% of all
ejection vertebral fractures are not accompanied by clinical manifestations.
Thus, the most searching examination cannot provide sufficient evidence to
eliminate with certainty a fracture of the spinal column, and a radiological
examination is always prudent.
Spinal injury is rather common during ejection. From 1975 through
1987, DCIEM aircraft accident files indicate six cases of vertebral fracture
occuring during ejection out of a total of 67 ejections. This gives an
incident rate of 9%. In addition, there were eight more instances of
vertebral fracture attributed to improper landing. Distribution of these 14
cases shows all fractures were incurred in the T6 to L5 region (some patients
presented with multiple fractures):
a. T6 -1 e. Ll -10
b. T10 2 f, L2 -2
c. Tll -2 g. L3 -1
d. T12 5 h. L5 -1
Most of the compression fractures were mild, characterised primarily
by tenderness in the area on palpation. Compared to other aircraft types,
the CT114 (Tutor) appeared to have the highest incidence of associated
vertebral compression fracture while the CF104 predominated with cervical
sprain/strain. Spinal sprains/strains are also somewhat common. During the
same time period, 15 cervical sprains were recorded. Table 15 documents
ejection injuries occurring during 1975 to 1987.
Another factor to consider during the ejection phase is windblast.
The seat and occupant are subjected to dynamic air pressure, or windblast
often called "Q" force, that is proportional to the square of aircraft
velocity and inversely proportional to altitude. Thus, the worst "Q" forces
occur during high speed low level ejections. The stagnation frontal area of
the occtpant/seat combination creates a ram pressure that acts to decelerate
the seat to "zero" horizontal velocity. During this deceleration, air flow
over, under and around the seat system creates an erosive effect that may
lead to loss of protective equipment (helmet, mask), tearing of exposed skin
,.'N V 'r "eo AM XMAU.A

0 24
(eye lids), and limb flail. There have been seven cases of windblast helmet
loss since 1975, all from CF104 and CF101 crew ejecting into more than 300
KIAS. This represents a 10.5% loss rate.
Windblast contributes to the greater number of injuries encountered
during ejection, and are largely limited to superficial facial abrasions and
lacerations caused by mask and helmet interactions. A study of injuries by
Noble and Olsen (Table 16) records several cases of internal organ damage by
the ram pressure of windblast. Of 67 non-fatal ejections 1975 through
December 1987, only nine received no injuries. This figure has not
significantly changed since 1952 when Smiley reported a 75% to 80% injury
rate during the decade 1952 to 1961. Since 1972, all CF ejections have been
below 500 KIAS, with 50% of these being below 250 KIAS. There have been two
possible ejections from the CF104 in the 450 to 500 knot range at 90 feet and
200 feet above ground. The pilot of CF104859 ejected at 90 feet inverted
following a mid-air collision with a light aircraft. He lost his helmet,
mask and spectacles from windblast and sustained a concusion and amnesiz from
severe head buffeting. In the second case, the pilot of CF104769 ejected at
200 feet after colliding with trees following a bomb delivery. He lost his
helmet and lost consciousness following ejection. He awoke on the ground
with only minor injuries.
I. A bizarre case relating to windblast occurred to the pilot of
CT13321551 on 30 May 1961. This pilot lost control of the aircraft at high
speed through speed-brake failure. Ejecting at 1000 feet in a spiral dive at
high airspeed, ram air pressure is suspected to have caused enough flexion of
the parachute pack that the retainer pins were pulled allowing the parachute
to deploy prematurely before seat separation. This resulted in severe
maceration injuries to the pilot's pelvis and lower legs resulting in death
five hours later.
A 1973 USAF Aeromedical Research Laboratory study (Brinkley) concluded
that the incidence of limb flail following ejection increases exponentially
with airspeed, being "significant" in the 300 to 400 KIAS range averaging 32%
in the 400 to 500 KIAS range) and reach 100% in the 600 KIAS range (Figure
21).
In general, the effects of windblast may be responsible for:
-!. a. flailing of limbs causing dislocation and fracture;
* b. head buffeting producing concussion and/or amnesia (there are at
least four recorded cases since 1972, and all occurred at ejection
airspeeds above 400 KIAS);
c. failure of equipment (tearing of protective clothing; loss of
protective equipment such as helmet, mask, gloves and boots);
d. contusion and rupture of internal organs;
% 0 P
'p..S

25
e. petecchial haemorrhage;
f. ocular haemorrhage, edema, conjunctivitis;
g. laceration of eyelids, nostrils and lips (if mask lost); and
h. sinus, bronchial and gastrointestinal inflation (if mask lost).
0EJECTION ENVELOPE
The "ejection envelope" is a set of defined physical parameters within
that an ejection may be successfully executed. It is primarily an
interaction of two independent sets of parameters: the physically designed
characteristics of the particular ejection system, and the dynamics of the
aircraft flight profile at the moment of ejection. Although not a part of
the ejection phase per se, it is an operating pre-condition to ejection
success, and will be covered with "ejection". Each ejection system has a
minimum fixed operating time from ejection to parachute deployment varying
-from 3.5 seconds plus canopy inflation time for the CT133, to 1.8 seconds
plus canopy inflation time for the CF188. Full canopy inflation may take an
additional two to three seconds. The minimum time available for a successful
ejection must be at least equal to pilot reaction time plus system operation
time. The dynamics of the aircraft flight profile that will affect ejection
success are: altitude, bank angle, pitch angle, airspeed, and sink rate. In
most situations, all these flight parameters are operating simultaneously,
Uand their net effect determines ejection success. The interaction of these
factors is normally presented graphically in the aircraft technical manual.
An example is given at Figure 20.
Taken in isolation, each of the above factors affect ejection success
by either increasing or decreasing ejection time available (ETA) as follows:
a. Altitude. The greater the altitude above ground level, the longer
the descent time. Most CF ejections since 1972 have been under
2000 feet (Figure 22);
b. Bank Angle. Increasing the bank angle reduces the vertical
component (i.e. height) of the ejection seat thrust. At a 30
degree bank, the vertical component is reduced 14%; at a 60 degree
bank there is a 50% reduction; and at 90 degrees, a 100%
reduction. To illustrate using CF188A data (60 degree dive, 130
kts), to be successful a zero degree bank requires a minimum
ejection altitude of 420 feet AGL; a 60 degree bank, 480 feet AGL;
and a 90 degree bank, 550 feet AGL;
c. Climb/Dive. In a climb, the vertical component of the aircraft
velocity vector will add to that of the ejection seat. As a
result, a higher ejection trajectory is attained and ETA is
increased. In a dive the opposite occurs and the downward
%, 111' N-

jq
26
velocity vector is subtracted from that of the ejection seat; ETA
is decreased. To maintain ETA, a greater altitude is required.
To illustrate using CF188A data (0 degree bank, 130 kts), a 30
degree dive requires 120 feet AGL for successful ejection; a 60
degree dive, 420 feet; and a 90 degree dive, 620 feet;
d. Airspeed. Airspeed has no effect on the vertical ejection
component in straight and level flight, but it does affect the
horizontal travel of the seat. In a climb, increasing the
airspeed will increase the upward velocity, hence altitude
attained, of the ejection seat. In a dive the opposite occurs.
,A. In a bank, airspeed may become an important factor. As bank angle
increases, greater forward airspeed is required to maintain level
flight: 60 degrees bank requires +2Gz; 75 degrees bank requires +4
Gz; and a 90 degree bank needs about +llGz. Thus, when an
aircraft loses power, it is extremely hazardous for it to enter a
bank condition, since without enough power to maintain the
required Gz loading, the nose will drop and a sink rate will
develop. The time to ground impact will be proportional to its
height, h, above ground
and the horizontal distance travelled will be proportional to
forward velocity (airspeed), v,
9.8
e. Sink Rate. This refers to the vertical descent velocity. In
freefall, following a power loss during level flight, sink rate
would be proportional to the acceleration because of gravity, "G",
(9.8 meters/sec/sec). If the aircraft is in a climb, the vertical
acceleration constant would be subtracted from "G"; if the
aircraft is in a dive, the vertical acceleration constant would be
added to "G". Thus, high sink rates may overcome the vertical
' , acceleration of an ejection seat, producing a net downward
acceleration of the man and seat.
Table 17 lists the ejection fatalities that occurred between January
1972 and December 1987. Eleven fatalities occurred on impact as a result of
lack of time available for the system to operate fully (technically, one
individual was fatally injured because of cervical fractures incurred when he
4 struck a high tension wire after ejection). One fatality, a CT133 passenger,
was killed on water impact when his system did not function properly because
of improper strap-in and posture at the time of ejection.
Three fatalities arose because of deliberate delays in starting the
ejection sequence. When the ejections were finally begun, the flight
parameters had drastically changed, and they were then unfortunately "out of
'nvelope". A CT114 pilot and student who experienced engine flame-out due to
ird ingestion lost valuable time while attempting to steer the aircraft away
from a residential area at a civilian airport. Both ejections were made at
low altitude with left bank nose down attitude and high sink rate. The pilot
of a CT114 pilot that experienced engine failure on take-off because of fuel
"%

I
27
pump malfunction delayed ejection for two seconds after the left hand pilot
had successfully egressed. By then, however, the aircraft was out of the
i C-envelope with severe right bank and nose down pitch at low altitude. One
CT133 pilot commenced ejection during a severe nose-over bunt at low
altitude, but the aircraft impacted the ground just after canopy separation.
The following case histories illustrate two "typical" out-of-envelope
ejections;
a. The first accident involved a CT114 with two pilots executing an
ILS touch and go at a civilian airport at night. Following
touchdown, full power was applied and a normal overshoot procedure
commenced. Shortly after lift-off, at about 150 feet AGL, the
engine flamed out because of main fuel pump failure. As RPM was
rapidly decaying the decision was made to bail out. Two seconds
after transmitting "bailing out", the left-hand seat pilot started
the ejection sequence. As the canopy and left seat departed, the
aircraft began to pitch down and roll right. Two seconds later,
with the aircraft under 100 feet AGL, rolling through 120 degrees
4 of right bank and pitching through 30 degrees nose down, the
right-hand seat pflot ejected. One second later the aircraft
impacted the runway. One and one-half seconds after ejecting, as
the lap belt opened and the "butt snapper" began operation, the
pilot impacted the runway head first, killing him instantly. The
surviving pilot suffered a 25% compression fracture of the T12
vertebra on ejection and a mild sprain to the right knee on
landing. It is unknown why the second pilot delayed his ejection
for two seconds after the first pilot bailed out. Unfortunately,
by then, the aircraft had dropped out of the ejection envelope.
b. The second accident involved a CFI01. This aircraft, with a pilot
and navigator, was scrambled on a flush alert during a base EDP
exercise. Immediately after becoming airborne, the aircraft
experienced a hung main landing gear. The pilot began "fumbling"
for the circuit breaker, and while "head down" allowed the
*t aircraft to enter a left turn with decaying airspeed and a high
* roll rate. It suddenly pitched up 45 degrees and entered a
steeply descending right-hand turn with after-burners on. At
approximately 355 feet AGL with high right bank, the pilot
ejected. About 0.5 seconds later with more than 90 degrees right
bank and at approximately 224 feet AGL, the navigator ejected. At
four to six feet above ground, the pilot separated from his seat,
* and his parachute began streaming. He was killed on ground
impact; the navigator was just separating from his seat at impact.
The pilot's aggressive personality may have been secondarily
involved in this accident; he had several recent admonishments
regarding his impulsive and overconfident flying behaviour.
At this point it is historically interesting to note that there have
been five manual bail-outs to date:
I

• 28
a. CT13321078 -7 July 1954 -successful (19000 feet);
b. CT13321390 -14 December 1954 -fatal (altitude unknown);
c. CT13321460 -13 January 1957 -successful (20000 feet);
d. F8623673 -28 August 1957 -successful (24000 feet); and
e. CT13321611 -31 March 1966 -successful (20000 feet).
To illustrate, the pilot of CT133611 experienced a mid-air collision
with an F-4 (Phantom) at 20000 feet. The aircraft immediately entered into a
severe tumble, pulling alternating positive and negative G. The rear seat
pilot ejected first. The front seat pilot experienced difficulty grasping
the ejection handles. He was finally able to do so, and pulled. Being an
ex-CFl04 pilot who had recently ejected from a CFI04, he forgot that in
addition to pulling upwards o. the ejection handles to eject the canopy, the
triggers on the handle had to be squeezed to fire the seat. (Note: triggers
have since been removed from ejection seats.) Sensing that perhaps there was
a seat failure, the pilot then undid his seat belt and was thrown clear of
the cockpit where he manually deployed his parachute and made an uneventful
landing.
In addition to the five manual bail-outs, there were three cases of
inadvertent ejection:
a. F8623243 -28 February 1957 -successful (12000 feet);
b. CF100321 -25 September 1958 (11000 " ); and
c. CF100747 -7 November 1962 – (15000 "
To illustrate, CF100747 with pilot and navigator on board, was
undergoing an air test at 15000 feet. While pulling negative G, the
navigator and his seat went crashing through the canopy into space because
service technicians had failed to anchor the seat to the floor after re-
* installing it following maintenance. The pilot's seat was also not anchored
to the floor, and it too rose up the rails, but the pilot managed to prevent
himself from penetrating the canopy by holding tight to the control stick.
%e. The navigator, finding himself falling through space, managed to separate
. from his seat and deploy his parachute, landing without injury. The pilot
" managed to land successfully back at base.
DESCENT PHASE
Even though an ejection occurs within the ejection envelope,
fatalities have continued to arise because of problems encountered during the
descent and survival phases. The ejectee is totally dependant on the system
to function automatically. The typically low altitudes that ejections
0
N%. %%~. V.%.'%~~ >~v-:*;,4 ,

29
generally occur would preclude manual separation from the seat and activation
of the parachute should the automatic function fail. The descent phase
begins with rocket motor burn-out and ends with ground or water landing.
During the descent phase, the individual must separate from the
ejection seat through the positive action of the rotary actuator or by
parachute extraction (CFl8), followed by parachute deployment and landing.
The successful function of these events depends on sufficient time remaining
for the automatic features to operate, and the absence of adverse precluding
., factors such as improper strap-in. A review of CF ejections since 1952 has
*. shown that injuries or fatalities may occur through any of the following
*' mechanisms:
a. seat/occupant interaction during separation;
b. seat/parachute collision/entanglement;
c. failure to separate from the seat;
I d. failure of parachute to function;
e. parachute/pilot entanglement;
f. parachute canopy/parachute riser entanglement; and
g. parachute/seat pack interaction.
* Seat/Occupant Interaction. During separation from the seat, the differences
in aerodynamic drag between the occupant and the seat usually provides for
* good physical separation; the ejection seat, decelerating faster, trails
behind the ejectee. There have, however, been ample instances of the pilot
contacting or being suspected of contacting, the seat -usually as a result
of seat/occupant tumbling at the time of separation, from a so-called "death
grip" (holding on to the ejection handle during operation of the rotary
actuator), or from equipment snags (CF104778 June 1964- incomplete cable cut
resulted in leg contact injury). Injuries range from abrasions to
concussions to fractures.
Seat/Parachute Collision. Except for the CF188, ejection is followed within
one second by seat separation, and if ejection is below parachute opening
altitude (16000 plus or minus 500 feet), parachute deployment is virtually
4simultaneous with seat separation (one second). There is, therefore, always
' a probability of the ejection seat colliding with the deploying parachute at
these lower ejection altitudes. Instances exist of pilot chute -seat
entanglement (CT133524 May 1964), parachute shroud line -seat entanglement
(CT114086 December 1971) and parachute canopy -seat entanglement (CT133076
June 1962). There have been at least five fatalities resulting from
I seat/parachute interaction since 1952.
Seat Separation Failure. Historically this was a common occurrence on the
iZ

030
F-86 and CT133 manual systems in the 1950s. Even with the conversion of
these systems to semi-automatic (canopy jettison was still a separate manual
'p. step), there was the occasional failure to separate from the seat, for
example, through failure of the lap belt opening (CF104794 September 1963,
CT13321169 June 1958). Since the acquisition of totally automated ejection
systems such instances are virtually non-existent. There have, however, been
a few instances of partial failure of the rotary actuator since 1957, but
none contributing to fatality. For example, the pilot of CF104635 who
ejected in May 1968, experienced severe parachute opening shock as the
opening parachute snatched him from the seat following failure of the rotary
actuator to provide good separation.
Parachute Deployment Failure. Such instances are rare. Those that have
"2 occurred were the result of faulty maintenance or accidental damage. The
pilot of CT133527 ejected at 1000 feet on 25 May 1962 and was killed when his
parachute failed to deploy due to improper maintenance. The pilot of
CT133286 ejected at 11000 feet on 9 April 1963 and had to manually deploy his
parachute after it failed to open because of improper maintenance. A tragic
A' incident occurred 30 July 1969 when the pilot of CT114157 successfully-Ap.- ejected and separated from the seat, but fell from his harness on parachute
opening because the quick release box (QRB) was in the UNLOCK position.
* Laboratory evidence showed that the pilot was suffering from acute carbon
monoxide poisoning (14% COHb). More recently, a pilot and navigator of
CF1019 ejected on 12 August 1973 at 9000 feet following an in-flight
breakup. Both aircrew passed through the aircraft fireball that burned both
parachutes. The navigator's parachute deployed normally, but the pilot's
parachute was fused by heat and only opened slowly after a free fall of 7000
feet.
From 1952 through 1987, there are at least 14 fatalities resulting
from parachute failure:
a. mechanical failure to deploy (dtsci – d automatic cable, defec-
tive barometric release, damaged ripcord) resulted in eight
deaths;
b. destruction by burning -one fatality;
c. canopy shredded on deployment at high speed -one fatality;
d. failure to open because of flat spin freefall -one fatality; and
e. fell out of parachute -three fatalities.
Parachute/Pilot Entanglement. These cases are rare, and the few instances
reported involved unusual circumstances such as tumbling during parachute
deployment. For example, the pilot of CF101452 on 10 November 1962 collided
with a Viscount aircraft that had taxied onto the runway while the CF101 was
* taking off. Lifting off, the main gear of the Voodoo sliced through the
Viscount. Control was lost and the pilot and navigator ejected successfully
at 700 feet, 200 knots and 90 degrees bank. During seat separation and
0 %
4ik

31
parachute deployment, the pilot's right arm and foot entangled in the shroud
lines and the ejection seat snagged in the parachute. He managed to free
himself before landing, and suffered only lumbar strain, thigh and shoulder
bruises.
Parachute/Seat Pack Entanglement Premature deployment of the seat pack
contributed to one fatality. On 13 May 1964, the pilot of CT133524 became
lost and flew to fuel exhaustion. Unfortunately, he had neglected to connect
the seat pack airlock fasteners to his parachute harness and during ejection
* the free-hanging seatpack, attached to the pilot by the lanyard, fouled in
the parachute shroud lines. The pilot parachute fouled in the ejection seat,
and the main parachute was unable to inflate.
Parachute Canopy/Riser Entanglement. These cases have involved risers
splaying over the parachute canopy to produce a "brassiered" effect. These
fouled risers may be manually worked off the canopy during descent, or if
severely snared, may be cut. Fortunately such instances are rare.
During descent to landing, an ejectee may be forced to survive a
number of potentially fatal physiological stresses:
a. altitude;
b. deceleration;
c. freefall;
d. tumbling and spinning;
e. parachute opening shock; and
f. landing impact.
Altitude. The effects of altitude per se on the individual followingCejection have been rarely encountered. Most ejections have been reasonably
low enough that hypoxia and decompression sickness were not a significant
hazard. There are at least 22 high altitude ejections recorded that
necessitated significant freefall to parachute opening (Table 18).
The longest ew-rosure to altitude following ejection was sustained by
the crew of CF100 b9 who ejected following acute loss of control in a CB
cloud. The pilot and navigator ejected at 31000 feet in a steep dive at an
unknown airspeed. During ejection, windblast tore off both helmets and while
freefalling, both were affected by hailstones, snow, freezing rain and rain.
On parachute deployment, both were subjected to up and down drafts,
prolonging the descent, and the navigator was struck by lightning that fused
and melted his parachute in several areas. The time from parachute
deployment to landing was recorded as 20 minutes for the pilot and 25 minutes
for the navigator. The pilot landed on a hardened surface and sustained a
% .Z.

M W WOO~r vw'y -,* _ _ ____ __ 19rTR.'W71- -. , C-A IV W% r W-. .WVW. rr
32
-* large hematoma on his left elbow, numerous punctate bruises on the face,
haemorrhage of the left eye, and a compression fracture of T-11. Tie
navigator sustained mild compression fractures of T-ll, T-12 and L-l;
multiple abrasions to the face and extremities from hail; and frozen hands,
with a loss of sensitivity present one week after ejection.
% Hypoxia was recorded on two of these high altitude ejection cases
(F8623278, CT13321600). The pilot of 278 lost consciousness while descending
in his parachute since he had lost his emergency oxygen bottle with his
seatpack during ejection. Hypoxia may have contributed to the death of the
pilot of F862333 who ejected 13 August 1956 following a mid-air collision
with another Sabre (23543, fatal). While apparently hypoxic, this individual
pulled his parachute ripcord while still strapped in his seat following
ejection. Hypoxia may alo have contributed to at least six other fatalities
by incapacitating the aircrew before ejection (CF10018146 10 November 1953,
CF10018122 17 August 1954, F8623654 20 August 1956, and CT13321235 30
December 1957). Mild hypoxia has also been mentioned on several lower
altitude ejections.
Deceleration. Following ejection, the seat and occupant are decelerated from
the airspeed at the time of ejection to zero horizontal velocity. This
decelerative force is dependent on altitude and airspeed and may be measured
as a "G" loading. The increase in "G" load as a function of airspeed varies
as the 2.47 power of the velocity (12). For example, a 15% increase in
indicated air speed (IAS) results in a 50% increas, in "G" load. At 600 KIAS
under 10000 feet, "G" load is approximately 30G (1200 pounds per square
foot). These decelerative forces may result in equipment loss (erosion) or
failure, or some degree of physical injury. Injuries that may occur are
essentially those of negative G and comprise:
a. head and face congestion;
b. periocular edema;
c. subconjunctival haemorrhage;
d. retinal haemorrhage;
e. leg abduction;
f. sinus haemorrhage and congestion;
* g. pulmonary compression with decreased arterial oxygen saturation;
h. cyanosis;
J. petecchial haemorrhage; and
* k. mental confusion.
As ejection altitude is increased, deceleration time is prolonged.
J"J..%0
18:

PUM W T N W5-1 T – 1. TWR- .- WT .-Wvd W
33
Deceleration time at altitude is inversely proportional to the square root of
the ratio of the altitude air density over sea level density. For example,
it takes approximately twice as long to decelerate from 35G to lOG at 40,000
feet as it does at sea level.
The following case histories illustrate four CF high speed, low level
ejections. In all cases, helmets and masks were lost, and personal injury
occurred. Amnesia and disorientation were prevalent.
a. CF101: (Ejection at 350 KIAS, 5000 feet AGL); severe right knee
flail injury. "Just after the smoke cleared he (pilot) yelled
"eject, eject, eject!". So I pulled the handle and went. I
hesitated for a second, disbelief and shock. I could not believe
this was happening to us … it took the canopy ages to go. The
canopy was really slow it seemed … I felt the windblast for a
second then I blacked out. When I came to I was under the chute,
I was out of my seat already, the chute was open. Initially I
could see KP (the pilot). I was having trouble focusing my eyes
beyond a certain distance … On the way down I went to check my
i •oxygen mask and realised that it was gone, and I did not have a
helmet on … My leg had hurt from when I regained consciousness,
as soon as I realised where I was, there was a lot of pain in the
leg…"
b. CF104: (Ejection at 400 KIAS, 3,500 feet AGL); 20% anterior
compressior fracture to T7 vertebra. "I looked down to make sure
of getting the handle, grasped it firmly with both hands, and gave
.% a sharp pull. Just before pulling, I attempted to get a good
position but know that I was not in fact in an ideal ejection
position. What appeared to be about one second after pulling the
handle there was a great roar (probably the canopy going) and
almost immediately a tremendous blast effect. This was followed
shortly by a jolt that I assume was seat separation, and what
seemed like two to three seconds later my chute opening (the force
of that was quite severe). Taking stock of the situation, I found
that my helmet, mask, and gloves, were gone (chin strap had been
*fastened but visor was not down). I have no actual recollection
of losing these items and was in fact completely disoriented until
chute opening … about this time I checked my chute and saw that
… it did have several ripped gores".
* c. CF104: (Ejection at 430 KIAS, 1000 feet AGL); concussion and
amnesia because of head flail. "My recollection is of reaching
for the handle and realising that the aircraft attitude was
critical, but I don't remember what it was and the remainder was
hazy, so much so that if anyone had told me that I had not ejected
I would have believed them. It's not firm at all, I just don't
* have it. But I wasn't surprised when I was in the hospital, when
I found out that there was a mid-air … I remember nothing of the
impact itself or of what occurred before it, other than the fact
that I had a vague recollection of rolling to the right and
%I.'

1 .. 34
* reaching for the handle and not being happy with the attitude of
" the aircraft (Note: ejection was at 270 degrees bank.)
d. CF104: (Ejection at 500 KIAS, 90 feet AGL); concussion and
% amnesia, cervical and lumbar strain, dislocated right shoulder
% from flail. "As it (the aircraft) came through the inverted
position to level flight I consciously slowed down the roll rate
… I pulled the handle. I recall the tremendous rush of wind as
the canopy separated and noted that I was still sitting in the
aircraft. The next thing that I distinctly remember was sitting
aboard the helicopter … I don't recall anything specific except
for the fact that it was suggested by one of the crew members that
I lie down, and it was about that time that I realised that I had
done something to my right arm and I was having some difficulty in
lying down."
Freefall. When an ejection occurs above parachute opening altitude, freefall
* must take place. Parachute opening altitude is controlled by the MK1OA
Barostat that is pre-set to 16000 + 500 feet ASL for all aircraft except the
CF188 (that is set at 14500 to 11500 feet ASL). After ejection from the
cockpit, the pilot is separated from the seat by the action of the "butt
snapper", and must stabilize his freefall to parachute opening. For the
CF188 aircraft, 0.5 seconds after ejection, a 22 inch controller drogue chute
and 60 inch stabilizing and retardation drogue chute are deployed from the
headbox. One point five seconds after the pilot/seat system has descended
through 14,500 feet, the scissor shackle opens and the harness release system
is operated. This frees the drogue chutes, the parachute mechanical lock,
the BIR shoulder straps, and the lower restraint harness. The drogue chutes
withdraw the parachute from the headbox that, in turn, extracts the pilot
from the seat.
For all successful ejections between January 1972 and December 1987,
none were over 15000 feet AGL, therefore no freefalls were recorded. Many
ejections occurred at and below 2000 feet.
For situations where freefall occurs, the physiological problems of
concern are:
a. Freefall (Terminal Velocity). Figure 23 illustrates the relation-
ship between altitude and true air speed (TAS). Freefall velocity
varies directly with altitude, being balanced by acceleration
* because of gravity and the "drag" force of air resistance. Wind
drag may cause buffeting, flailing, and injury to exposed skin
surfaces from thermal and friction effects.
b. Spinning and Tumbling. During freefall, tumbling and/or spinning
may occur that, if uncontrolled, may reach extremely high
". rotational rates upwards of 400 rpm. Centrifugal forces compel
-, arms and legs to move outwards cartwheel fashion, and generate
-.1

35
stagnant hypoxia through blood pooling in the periphery. Vertigo,
nausea, and vomiting may also ensue. Spinning during automatic
parachute deployment could result in twisting of the risers with
increased canopy deployment time.
Freefall has been implicated in causing one known fatality. On 28
August 1957 following a mid-air collision at 24000 feet, Lhe pilot of
F8623669 ejected. Seat separation was apparently unremarkable, except that
following separation, the pilot adopted a "back to the earth" freefall
posture. It is felt that the dynamic air pressure against the parachute pack
prevented the parachute from deploying, resulting in a fatal ground impact.
Parachute Opening Shock. All CF parachutes except for the CF188 (that
*deploys at 13000 plus or minus 1500 feet) are designed to automatically
W deploy at or below 16000 feet ASL. They may, however, be manually deployed
at higher altitudes by pulling the "D" ring located on the left-hand side of
the parachute. (The exception is the CF188 parachute that is located in the
headbox and cannot be manually deployed.) The opening characteristics of a
parachute canopy depend on TAS, that in turn varies inversely as the square
root of air density, i.e., TAS increases with altitude. Generally, a
parachute must travel six to eight times its diameter before being fully
inflated, therefore, since TAS is greater at altitude, a parachute opens
faster at higher altitudes. Coupled with a faster terminal velocity, this
may result in high decelerative "G" loads (Figure 24). High opening shocks
may damage the escape system (tear the parachute canopy and risers), cause
loss of life support equipment and clothing (boots, gloves), and result in
ib personal injury. For example, parachute deployment at 30000 feet exerts
about 18 G on the body, while automatic opening at 16000 feet is equivalent
to an 8G load.
Premature deployment of the parachute may result in parachute failure
(tearing, shredding), loss of personal equipment (gloves, boots), and
personal injury (groin abrasion, contusion, hematoma). At least two
fatalities have resulted from premature parachute deployment. The first case
occurred 30 June 1959 when the pilot of CF100762 became severely disoriented
at 42000 feet and lost control of the aircraft. The navigator ejected at an
estimated altitude of 35000 feet and manually deployed his parachute where it
shredded to pieces and he was killed on impact. The pilot crashed with the
aircraft. The second case occurred to the pilot of F8623333 who ejected at
an unknown altitude following a mid-air collision 13 August 1956.
Apparently, while hypoxic, he deployed his parachute will still strapped into
the ejection seat.
dIn addition to injuries and fatalities, there are at least four
recorded cases of boot loss during parachute deployment and numerous
instances of spectacles, gloves and watch loss. On 10 March 1960, the pilot
of F8623646 ejected at 8000 feet following engine seizure at 17000 feet. The
ejection occurred at 300 KIAS, and on parachute opening, the pilot lost his
boots, a pair of Wellingtons. He also lost his helmet, seatpack and dinghy
(not connected to parachute), and landed in the Rhone River where he had to
swim to shore. He was rescued eight hours later and was treated for
lacerated feet incurred while walking through stubble fields.

0 36
A.- LANDING
Although part of the descent phase, landing may be appropriately
addressed as a separate topic. In preparation for landing, an individual
may be faced with a number of circumstances: tree landings, hard ground,
high winds, water, and obstacles such as power lines. During ground
landing, the individual must convert linear momentum to angular momentum
and decelerate his velocity to zero over the greatest distance to yield low
frequency acceleration pulses, if injury is to be avoided. This is
normally done by executing a parachute landing fall (PLF) to distribute
deceleration forces over the foot, shin, thigh, hip and chest. Since
aircrew do not normally receive instruction and practice in executing
proper PLFs, it should be expected that each ejection carries a high risk
factor of injury on landing due to incorrect technique, but this does not
appear to be the case. Injuries that occur strictly due to landing show a
rare incidence. Instead, those that have occurred appear to be directly
related to seat pack retention. A previous study by Smiley (1964)
indicates that there is also a tendency for injuries to be associated with
unfavourable drift. He reported that 74% of those drifting backwards or
sideways, and 57% of those drifting forward were injured. Although purely
subjective, 50% of 195 ejectees (April 1952 to January 1964) reported their
* landing as "severe".
Severe winds present a hazard; the pilot may be injured while being
dragged in the parachute, or drown following water landing. for examplethe pilot oF CF116760 who ejected 2 May 1976 was dragged for 1200 feet by
high winds. The adoption of the "T" handle harness release on the CF188
parachute allows for quick one-step release and should prevent dragging
problems in the future.
Landing in trees presents a particular hazard. Injuries can occur
during tree landing and during attempts to descend after landing. For
example the pilot of CF104804 landed in a tree following ejection on 23
July 1971. While trying to descend from the tree, he fell to the ground
sustaining crush fractures to T12 and L2. To minimize injury while landing
a' into trees, it is therefore standard procedure not to deploy the seat pack.
Landing vertebral fractures are almost always anterior cuneiform or
comminuted due to relatively vertical impact force and hyperflexion of the" dorsolumbar column, especially in the area of the T12 to Ll, with sharp
predominance at Ll. There have been at least eight cases of vertebral
landing injury since 1975. Strong hyperflexion may also result in chin
contact with the quick release box (QRB) and myocardial contusion (CT114123
11 May 1976). Intervertebral disc/rupture has also been reported
.d(CT13321020, 29 May 1957). Instances of fractured legs (F-86 19619),
ankles (F-86 23152), sprained ankles (F-86 23086), fractured ribs (F-86
23193), and miscellaneous contusions and abrasions have been encountered.
a' The following cases illustrate landing injuries. The first occurred to the
*pilot of CT114123 on 11 May 1976, and the second to the pilot of CT133442
on 14 February 1981.
4.
0w
%F 4. % %F %. .J%4 Fa'~-

37
CTll4. (Bird strike engine failure, ejection at 2,900 feet AGL, 130 KIAS;
failure to deploy seat pack.) Medical report on pilot: "The injuries
% sustained are typical of those seen in previous Tutor ejections being
sustained on impact with the ground … (the pilot) sustained …
compression fractures of T12, Ll, L2 and L3 with approximately 25% loss of
anterior margins. He developed a regional ileus after 12 hours and this
cleared slowly over the following four days … hematuria was present for
24 hours and cleared spontaneously … a small pleural effusion was
demonstrated on the third day and this cleared slowly without surgical
intervention … It is proposed to eventually use a Taylor Brace or plaster
jacket cast for three months … It is concluded … these injuries were
sustained at the time of landing in as much that (he) failed to deploy his
seat pack and in so doing fell heavily on the pack. (The pilot) recalls
his head snapping forward when he landed … (which) would cause the chinbto contact the Quick Release Box hard enough to split his chin open."
*. CT133. (Fuselage pump failure, engine flame-out on final approach for
landing; ejection at 450 feet AGL with high sink rate; failure to deploy
seat pack). Injuries occurred when pilot landed on seat pack. The pilot
sustained "… a compression fracture of Ll with at least 50% loss of the
* volume of the body of Ll. The fracture extends into the posterior elements
and there is some widening of the pedicles noted on the A.P. view. As
well, there is slight posterior displacement of the fracture fragments into
the spinal canal (Note: the body of Ll restored to its previous height over
48 hours) … the compressed lumbar vertebra protruded about 1/4 inch into
the spinal canal. He can be expected to be treated with bed rest, with his
back in hyperextension for the next two months, at which point it is
expected that he will be placed in a body cast for a further period …
even though there is no neurological involvement, there is a 50% chance of
recurring back problems as a consequence of this injury …"
Water landings present a particular hazard, especially with the
possibility of entanglement in risers, ingestion of water while being
dragged, and difficulties in releasing the parachute. Injuries are
particularly hazardous to survival following water landing. There have
been at least five drownings following successful ejection since 1952.
Historically, difficulties were often encountered with finding and
* activating the life-preserver inflation toggle following water landing.
These problems were corrected by installing an automatic inflation device
(AID) on the life-preservers. Exposure following cold water landing is
also an ever present hazard.
* SURVIVAL PHASE
% The survival phase commences with ground or water landing and lasts
until rescue. The possibility of having to face long survival times in the
-, wilderness following ejection was very real. Canada was less developed in
* the 1950's. "Stations" were more isolated, search and rescue facilities
were rudimentary, locator devices were crude. Modern locator devices
(locator beacons, crash positions indicators, SARSAT) and highly efficient
search and rescue facilities have virtually precluded long survival periods
N" %

%S
38
following ejection. Consequently, it can be expected that data will show a
continuing reduction in rescue times since the 1950's -and this is indeed
the case.
Since 1952, there have been eight instances of overnight survival,
and two of survival over 48 hours. The record for length of survival
following ejection occurred on 10 December 1956 when the pilot and
navigator of CF10018411 ejected following double engine flame-out and
electrical failure. Both ejections were made through the canopy into a
cold winter wilderness. They were rescued by a search team after 43 hours
(two nights). The last overnight survival occurred following ejection from
CT13321614 on 18 November 1965 in the Vosges mountains of FI.ance. During a
TACAN let-down in zero visibility, the aircraft struck trees. The pilot
zoomed the aircraft and twice ordered the second pilot to eject. Both
ejected, but the aircraft captain, who delayed his ejection, was killed on
ground impact. The second pilot survived for 14.5 hours (overnight) until
rescue.
For historical interest, the longest survival period since the war
is apparently credited to Flight Lieutenant McKenzie who ditched his
Gloster Meteor EE311 in Helen Barr Lake in Northern Ontario on 29 June 1946
while on a trip from the Winter Experimental Establishment (WEE) Flight in
Namao. Flight Lieutenant McKenzie survived 27 days in the bush before
being picked up by fishermen. The Gloster Meteor, like the Vampire, was
not equipped with ejection seats.
The greatest hazard to survival is injury, especially burns and
fractures. Two Sabre pilots who ejected over water drowned before rescue.
One had received a basilar skull fracture with bruised larynx during the
ejection phase. A CT133 pilot who ejected following an in-flight fire, and
endured a night in the bush, was unable to open his ration pack because of
severely burned hands that subsequently required skin grafting.
Lastly, there have been a few cases where the rescue attempt has
been too late. On 18 April 1952, the pilot of F8619181 ejected over the
North Sea following a mid-air collision with F8619177 (which crashed in the
sea killing the pilot). He did not have a dinghy as part of the survival
kit and as a result, succumbed to exposure before rescue. Similarly, on 14
April 1960, the pilot of F8623229 ejected for unknown reasons over the
water of Mirimichi Bay. His ejection was successful, but he died of
exposure. His body was discovered three months later, floating in the
* life-preserver.
SUMMARY
This report has outlined RCAF/CF ejection experience from 1 April
1952 to 31 December 1987, with emphasis on ejection statistics covering the
d period January 1972 to December 1987 (Annex A). One hundred and three
ejections from 107 aircraft were analysed. Ninety-two of these ejections
S
-p.

U
39
were successful which represents a success rate of 89.3%. Forty-five
individuals did not eject for various reasons, of which 37 perished in the
crash. The greatest cause for these fatalities appears to be due to
unperceived ground closure, or controlled flight into terrain (Table 10).
Included in this category is the phenomenon of "late awareness",
essentially a form of CFIT, but a realization at the last moment that
ground contact is inevitable. In this context, late awareness has some of
th, elemeL~ts of "rapid*ity of events", but differs in that a serviceable
* aircraft was "flown into the ground" while under complete control as a
result of an error in pilot judgement.
Of the 67 successful ejections since 1975, nine were completed
without injury. The most common form of ejection injury was due to* windblast pressure on the mask and helmet, resulting in facial lacerations,
abrasions or contusions. The second most frequent injury pattern observed
was thigh contusions from seat pack or ejection handle contact during the
ejection of the seat from the cockpit, and vertebral fractures from
improper positioning either on ejection or on landing.
The majority of ejections occurred at very low altitude, all under
15,000 feet. Airspeed varied from 0 to 500 knots with the majority of
ejections occurring between 200 to 250 knots. There appeared to be no
r. correlation between severity of injuries and airspeed; however, windblast
effects appeared to increase as airspeed surpassed 350 to 400 knots. In
'3 this range injuries became more numerous through the action of aerodynamic
forces on the helmet and mask, and were limited to facial abrasions,
contusions and lacerations. Helmet losses appeared within this airspeed
range, and as airspeed increased above 400 knots, the incidence of
dislocations increased (elbow, shoulder and knee).
Table 6 summarises RCAF/CF ejection experience over the period
January 1952 through November 1987. These data were extracted from four
separate sources, and the data trends were consistent. The overall success
rate for attempted ejections has steadily increased while the total number
of ejections has decreased. A noticeable turning point in low altitude
ejection success occurred throughout the 1960s, and may be correlated with
the introduction of the rocket assisted ejection seat of the CF104 aircraft
and the incorporation of an automatic opening parachute. A "new awareness"
among aircrew as to the importance of timely ejection may have also
manifested itself with the supersonic era ushered in by the CF1OI and
CFl04, perhaps coupled with the phase-out of wartime fighter pilots and the
"ditch or bail-out" mentality of the Hurricane era. Unfortunately, the
success rate for low altitude ejections does not appear to have continued
to increase through the decade of the 1970s despite fully automated
ejection systems. The reasons for this are not clear, but may be due to
the employment of supersonic aircraft as low level tactical fighters. This
would effectively lower the margin for ejection success as flight
parameters approached the mechanical limits of the ejection system, and
success became more dependent on the human factor. DCIEM studies (22, 29)
have implicated workload, critical time sequencing, rapidity of events,
pilot inexperience, and lack of information presentation when "heads up"
during high speed low level flight or tactical weapons delivery, as
4%

*40
overriding factors in low altitude no-ejection fatalities. If this is
true, then improvements in ejection systems alone will not increase
survival rate. What will be required is a closer matching of the aircraftdesign to its operational role requirements.
S' _CONCLUSIONS
Aircrew flying ejection seat aircraft must be regularly briefed and
reminded of the necessity to make rapid and correct responses to
emergencies at low altitude, and to recognize the limitation to human
information processing during high speed low level flying. In 21st century
aircraft such as the CF188, this should also include the effects of
"information saturation", as opposed to "information lack" characterised by
century series airplanes, and the limitations to cognitive processes
engendered by digital/analog cockpit displays.
Aeromedical briefings on ejection physiology should continue to
emphasize the importance of posture during ejection, and the mechanics of
back injury. Airspeed and windblast effects should also be covered in
detail, as well as the importance of releasing the seat pack before ground
* landing. Lastly, aircrew must thoroughly understand the minimum operating
parameters of their escape system under all conditions. Approximately 10%
of all fatalities (11 out of 103) arose due to ejection outside the
-" envelope".
As older and slower ejection systems are replaced by the "zero zero"
seat, the CF ejection success rate should increase to well over 90%. The
limiting factor will always be the human information processing system.
% %"

41
ACKNOWLDGEMENTS
*This report is the result of two years of res'arch through
historical reports and Boards of Inquiry. The author is indebted to many
individuals who proof read the drafts of the manuscript and offered
constructive comments on both format and technical content, notably: Mr.
J.A. Firth who explained in great detail the operation of the CF188
ejection sysem; Warrant Officer J.C. Steffler and Sergeant A.N. MacLeod
who provided historical details on early systems; and Diane E. Spence who
passed uncounted hours typing drafts and making corrections. Thank you
all.
f.%
q

. 42
BIBLIOGRAPHY
1. ANDERSON, I.H.; A Survey of Major Accidents Incurred Enroute During
High Speed Low Level Operational Training or Practice Missions; CFIEM
Report 69-TM-i, July 1969 (Restricted)
2. AUFFRET, R. et al; Spinal Injury After Ejection; AGARD-AR-82 February
%. ~. 1975.
3. BRENT, H.P.; RCAF Operational Experience With Ejection Escape Systems
1 April 1952 -30 March 1956; RCAF IAM Toronto #56/3 1 September 1956.
4. BRENT, h.P.; RCAF Opoerational Experience With Ejection Escape Systems
. 1 April 1956 -30 March 1957; RCAF IAM Toronto #57/6 1 December 1957.
5. BRENT, H.P.; RCAF Operational Experience With Ejection Escape Systems
1 April 1957 -31 March 1958; RCAF IAM #58/7, 31 December 1958.
6. BRENT, H.P.; RCAF Operational Experience With Ejection Escape Systems
1 April 1958 -31 March 1959; RCAF IAM #59/3, 25 June 1959.
7. BRINKLEY, J.W. et al; A study of Aerodynamic Forces as a Cause of
. Injury During Emergency Escape; AFAMRL Report TR-73-55,1973.
-8 Canadian Forces Ejection Experience 1962 -1971; CFHQ DFS
9. Canadian Forces Ejection Experience 1972 -1979; CFHQ DFS
10. Accident Boards (CF211) (DCIEM Toronto copy) 1972 -1987.
11. DELAHYE,. R.P. et al; Physiopathology and Pathology of Spinal Injuries
in Aerospace Medicine; AGARD-AG-250 (Eng), February 1982.
12. ERNSTING, J.; Aviation Medicine: Physiology and Human Factors; Tri-Med
Books, London, England, 1978.
13. GOEBEL, G.M.; How Long It Really Takes; Air Clues, January 1984.
14. HARRISON, W.D.; Aircrew Experiences in USAF Ejections 1971 -1977;
16th Annual SAFE Symposium, 8-12 October 1978.
15. McMEEKIN, R.R.; Aircraft Accident Investigations; Fundamentals of
Aerospace Medicine; R.L. Dehart (Ed); Lea and Febiger, Philadelphia,
U.S.A. 1985.
16. NESTLE, A.J.; Aerodynamic Forces Experienced During Ejection; AFAMRL
0 Report TR-80-16; March 1981.
17. NOBLE, R.E., S.W. OLSEN; Accident Statistics Relevant to Windblast;
4-… AGARD Conference Proceedings No. 170, Toronto, 1975.
18. NUTTAL, J.B.; Emergency Escape from Aircraft and Spacecraft; Aerospace
Medicine 2nd Ed; H.W. Randel (Ed); Williams and Wilkins Co.,
Baltimore, 1971.
0,

43
19. POULTON, E.C.; Environment and Human Efficiency; C.C. Thomas, publisher,
Springfield Illinois, U.S.A.; 1972.
20. PREBBLE, C.; How Low Can a Pilot Get? Flying Safety, January 1983.
21. ROWE, K.W., C.J. BROOKS; Head and Neck Injuries in Canadian Forces
Ejections; Aviation, Space, and Environmental Medicine, April 1984, pp
313-316.
22. RUD R.C., D.F. LEBEN; Human Factors in High Speed -Low Level
Accidents -a 15 Year Review; AGARD Conference Proceedings No. 267; pp
23-1 to 23-6.
23. SCHLEEDE, R.L.; Application of a Decision-Making Model to the Inves-
tigation of Human Error in Aircraft Accident;ISASI Forum, Winter 1979;
I* pp 62-78.
24. SMILEY, J.R.; RCAF Operational Experience with Ejection Escape Systems
1 April 1958 -31 March 1960; RCAF IAM #60/2, 17 June 1960.
25. SMILEY, J.R.; RCAF Operational Experience With Ejection Escape Systems
4- 1960; RCAF IAM 61-RD-I; 15 May 1961.
26. SMILEY, J.R.; RCAF Operational Experience With Ejection Escape Systems
1961; RCAF IAM 62-RD-3; 7 September 1962.
27 SMILEY, J.R.; RrAF Ejection Experience 1962-1966; CFIAM 67-TM-B;
fDecember 1967.
28. SMILEY, J.R.; RCAF Ejection Experience 1952-1961; RCAF IAM 64-TM-1;
January 1964.
29. THATCHER, R.F., P.J. DEAN; Pilot Workload and Work Capacity in Low
– Level High Speed Roles, DCIEM 1973.
30. WHITE, R.P.; Development of a Mechanical Analog of a Human Spine and
Viscera; Proceedings SAFE 23rd Annual Symposium; December 1–5, 1985;
pp 201-209.
31. YEAGER, R.R.; Gross Behaviour of the Body's Gluteal Region; Technology
Inc; Dayton, Ohio, December 1971.
32. ZELLER, AF.; Human Reaction Time; USAF Safety Journal, May 1973.
A!
d –…
– l.'-'

0, 44
Table 1. CT133 Election Time Sequence
ELAPSED TIME (sec) ACTION
0 (R) Right pulled, Canopy Jettison and BIR activation
1 Rear seat ejected
S1.5 Front seat ejected
2.0 Rear rotary actuator lap belt separation
2.5 Front rotary actuator lap belt separation
3.0 Rear parachute begins deployment
3.5 Front parachute begins deployment
6.2 Rear parachute stable
6.7 Front parachute stable
Table 2. CT114 Timing Sequence
ELAPSED TIME (sec) EVENT
• 0 Handles pulled
0 Canopy initiator
0 BIR activation
0.5 Seat catapult, bail-out tone, UHF and transponder
0.65 Seat at top of rails, rocket fires
0.90 Rocket burns out
1.50 Lap belt released, rotary actuator
2.5 Parachute deployment (below 1600 + 500 ft)
5.0 Stable Parachute
Table 3. CF116 Election Time Sequence
ELAPSED TIME (sec) ACTION
0 Handles pulled, canopy jettisoned and BIR actuated
0.3 Seat ejection
1.3 Rotary actuator, RPI lap belt
2.3 Parachute deployment (below 16000 + 500 ft)
5.05 Stable parachute
Table 4. CF188 Election Time Sequence (SJU-9/A)
E APSED TIME (sec) ACTION
0 "D" ring pulled, canopy jettisoned, BIR activated
0.3 Seat ejection
0.8 Drogue chutes deployed
1.8 Parachute deployment (14500 -11500 ft)
–––––––––––
N.
%04

4
45
Table 5. Equipment Check List -CF5
1. Booklets: a. How to use the Silva Prospector Compass Magnetic Model
b. CFP 222 Land and Sea Emergencies
2. Vinyl Single blade boat paddle (2)
3. Yellow rubberized boat bailer (1)
4. Leak stopper set (1)
5. Emergency fishing kit (1)
6. One-man inflatable life raft -1800 psi compressed gas cylinder (1)
7. Tinted protective eyeshield (1)
8. Seawater de-salter kit (1)
9. Flat bastard file (1)
10. 1.5 lb. single bit axe (1)
11. Flint (1)
12. AN/PRQ 501 radio beacon (1)
13. ll.2v dry cell battery (1)
14. Magnetic compass (1)
15. Dimenhydrinate anti-seasickness tablets
% 16. Insect repellant stick
17. Water purification tablets
18. Survival hunting knife (1)
19. Sponge (1)
20. Survival first aid kit
21. Polyethylene plastic bag (2)
22. Aluminum foil 18" x 24" (1)
23. insect head net (1)
Q24. Unlined work type mittens (I pr)
25. Unlined leather work type mittens (1 pr)
26. Wool/nylon socks (1 pr)
27. Whistle (1)
28. Sunburn ointment tube (1)
29. Facial tissue (1 pk)
30. Sleeping bag (1)
31. 800 calory Food packet (1)
32. 400 cal Food packet (1)
33. Fuel tablets (I pk)
34. 25 feet 0.028 dia. brass wire
35. Watertight matches (1 pk)
4 36. Signal kit a. hand-held projector (1)
b. signal pyrotechnics (7)
c. signal mirror (1)
d. red cotton signal panel
L.

46
Table 6. RCAF/CF Ejection Seat Mishaps -April 52 to Dec 87
Personnel Involved
72
No Apparent Ejection
Attempt Unknown-Fatal *Attempted Ejection
223 12 485
Non-Fatal Fatal Completed Ejection Not Completed
*46 177 452 33
I;
*Non-Fatal Fatal Non-Fatal Fatal
378 74 12 21
System-Related Out of Envelope Unknown
26 47 1
1', ejection seat -3
.4.. canopy jettison -3
Sparachute -14
inertia reel -1
seat/parachute entanglement -5
*includes 5 bail-outs and
4~44~ -. 3 inadvertent ejections
*includes 2 cases of fatal
hypothermia and 3 drownings

p 47
6
0 0
5 . ' 5– ffo +, + 61 fn 0 0 en tn
+..+.+.,-,+ .,+- ,- –, ,'41
00 C 0 uq–- .\.
-~ W- 0
000
-,0xo '4+ –
*l N
X0 -0
alll ItpN, NI'' I

48
Table 8. Air Accident/Cause Factor:Personnel-Pilot 1977-1986
A Breakdown of Personnel-Pilot cause factors. 1986 cause factors
are subject to revision since some cases are still under investigation.
Factors 1977 '78 '79 '80 '81 '82 '83 '84 '85 '86 Total
Information/communication 1 1 2
Human Engineering 1 1
Acceleration effects 1 1 2
Disorientation 11 1 1 4
Fatigue 1 1
Noise, Vibration/Buffet 1 1
Visual Illusions/Limitations I 1 1 1 4
Physical/Physiological -other 1 1
Carelessness 1 1 1 1 1 1 1 7
Channelized Attention 1 1 1 1 4
Complacency 1 1
Distraction 1 1 2
Expectancy 1 1
Human Information Processing 1 1 1 3
Inattention(*) 8 4 8 2 3 5 6 3 1 2 42
Judgement(*) 12 3 8 5 4 7 5 1 45
Motivation 1 1 2
Non-compliance with orders 2 2 1 1 6
Technique(*) 7 5 13 3 2 7 7 6 2 3 55
Training 1 1
Psychological -other 1 1
TOTAL 30 15 34 14 12 24 25 12 9 10 186
V.
{,,Z'.5-Zty._1L.-,kt;bM

49
Table 9. Effects of Stress on Information Processing
From: Aviation Medicine; G. Dhenin of al;
Wiliam Clowe & Sons Limited London; 1979; A 71.
PSYCHOLOGICAL MECHANISM Notes asd Examples
* Omimion The subject simply does not respond to a situation. The undercarriage
3 of an aircraft may not be lowered during the final approach because
the pilot is either fatigued or overloeded.
Error A subject may respond incorrectly to a given stimulus. He ey select
the wrong lever and raise the undercarriage instead of the flap.
Queuing This Is a process of sequential delaying. The operator realizes that he
has several things to do, but because of the pressure of work he delays
certain actions until the workload falls to a more acceptable level. A
pilot may delay fuel-state checks whilst negotiating uncomfortable
turbulence,
Filtering Instad of queuing. the subject may resort to the rejection of certain
tasks in order to compensate for the workload. The rejection is a con-
scious process and might be accompanied by verbal phre like "I've
got enough on my plate without worrying about …..
Approximatlon In order to produce rapid results, eircrew may approximate, either con-
* sciously or subconsciously, in calculations or in flying technique.
Coning of attention As the stropl increases the environmental field to which man pays
attention decreases. It is as if man were travelling down a cone towards
the vertex. The further down the cone he goes, the lo he 'sees'.
.q Inability to Integrate This is related to the coning of attention. Whereas the normal pilot
information from adequately scans several instruments and integrates the information
Svarious source from each of them to obtain a mental picture of the aircraft a a whole,
the severely stressed pilot may concentrate on fewer and fewr instru-
ments and even then ho difficulty in integrating the information that
they give.
Regremion Human beingl under stress will often regress to a pattern of behaviour
that was learned at an earlier time. A stressed pilot may confuse the
location of a switch or trim control in one aircraft with the Iocation of
the control in an aircraft he flew at a much earlier date. The groping
for something which is not there may last for a vital second or two,
before the pilot realizes what is wrong.
muse tevmien, tremor. These are some of the physiological accompaniments of stress which
or freezing can occur even to experienced pilots. Freezing might explain tm of
the so-called 'suicide' flights, dubiously lbelled a such because no
other rational explanation som possible at the time.
SEcepe The ultimael response is to seek refuge from the tak by the rejection
of stimuli or suppression of response in favour of avoidence. The
operator may simply 'give up'. panic, or a sometimes occurs in aircraft
accidents, be unable to do anything whatsoever, that is, freeze.
100".4ti belavlour: The detrimental efflcts of Stress do not suddenly disappear once the
; reliving the experience, stres has been removea. It is common for People to have disturbed
'0_ ietallization, leep becaus they relive the stres in soms way. Often this process
* ln rsev WON do, involves thinking what one should have done or having imaginary con-
s. m versations with the other people involved in the stressful situation, or
simply excusing one's own behavour. The disturbed slep caused by
* thme mental processn could give rise to human error.
W
a1
%
)%

05~50
Table 10. 37 Fatalities/Failure to Initiate Ejection -1972-1987
Reason for Failure to Aircraft No. of No. of Predisposing
Initiate Ejection* Type Aircraft Fatalities Factors
–
A. Unable tc CF104 4 5 control jam-2?
eject mid-air -2
GLOC -1 ?
21.6% CF116 1 1 GLOC ?
CT114 1 1 mid-air
CT133 1 1 system failure
B. Unperceived CF100 1 2 illusion?
* Aircraft/Ground
Closure Rate CF101 1 2 distraction
CF104 7 8
51.4% CF116 2 2 disorientation?
CT133 1 2 inattention
CF188 2 2 disorientation
CT114 1 1 distraction?
––––––––––––––––––––––––––
% C. Rapidity of CFI01 1 2 stall
Events
• 24.3% CT114 2 3
" CF116 2 3
CT133 1 1 personality
D. Unknown CF116 1 1
" °"" .2.7%
* Not official cause factorM:
N.%
i U11

51
Table 11. Time Intervals and Altitude Deviations
During Routine In-Flight "Distractions" (Goebel)
Time Used (Seconds) Altitude Changes (Feet)
Task Min. Max. Average Max. Average
TACAN Change 2 15 6.3 100 38
Weapons Switching 4 20 9.7 300 38
UHF Change 5 16 9.6 100 37
IFF Change 4 22 9.8 300 80
Checklist Reference 01 54 28.8 600 197
Letdown Book Access 7 46 26.3 500 169
Letdown Book Reference 12 80 26.6 200 84
Bingo Calculation 5 69 27.4 900 214
Fuel/Distance Calculation 17 120 44.4 300 144
Table 12 General Ejection Sequence
1. Handle pulled
2. Zero delay initiator(s) fired
3. Canopy jettison, BIR activation
4. Seat ejection
I +
CFII6, CT14, CT133 CF188
5. Rotary actuator activation 5. Drogue chutes deployed
and seat separation
6. Freefall to 1600 + feet 6. Freefall to 13000 + 1500 feet
7. Barostat activation 7. Barostat activation
8. Parachute deployment 8. Scissor shackle opened
Parachute deployment
Harness release
Seat separation
9. Parachute descent 9. Parachute descent
10. Seat pack deployment 10. Seat pack deployment
11. Landing 11. Landing
12. Release parachute 12. Release parachute
.%

-• 52
Table 13. RCAF/CF Ejection Experience -January 52 to April 1986
Altitude 1952 1961* 1962 -1971** 1972 -1986***
(Feet AGL) Attempted Success Attempted Success Attempted Success
,0'- 0 140 7 0 14 7 10 4
,.150 -240 2 0 4 4 5 3
250 -340 4 0 1 0 3 3
350 -440 6 4 6 6 3 3
450 -950 26 13 16 13 10 9
1000 -9500 88 81 58 58 54 53
10000 19500 47 42 15 14 12 12
20000 32 25 6 6 0 0
% Success Rate 77.9 90.7 89.7
Unknown 6 9 1
0 Data Sources:
* S/L Smiley JR; RCAF Ejection Experience 1952 -1961;
– RCAF IAM Tech Memo 64-TM-I; January 1964
-** CF Ejection Experience 1962 -1971; CFHQ DFS Publication
CF Ejection Experience 1972 -1979; NDHQ DFS Publication
• ** DCIEM/MLSD Accident Files
Table 14. Ejection Seat Characteristics
Aircraft Type CF188 CF116 CTI14 CT133
Catapult Forces
"G" Onset (G/sec 180-210 300 178 178
Max "G" 14- 16 20 14 14
Velocity (fps) 65 81 50 60
Rocket Time (sec)
Burn Time (sec) .25 .35 .46 .25
Thrust (Ib) 4500 4600 6631 U/K
Max Ejection Airspeed (kts) 600 500 420 450
Min. Ejection Parameters 0, 0 0, 70 0, 60 0, 70
(alt, airspeed)Miminum Operating Time (sec) 1.5 2.3 2.5 3.5
Seat Angle (deg) 22 10 17 14
,%
.
V.-,

V
4 53
Table 15. Injury Analysis -67 Successful Ejections -1975-1987
4, WNote: Figures not mutually exclusive, one ejectee may present
with several injuries.
Aircraft Type CFI01 CF104 CT114 CF116 CT133 CF188 Total
……………………………………………………………-
* No. of Types 7 15 13 8 3 5 51
No. of Ejectees 13 17 19 8 5 5 67
No. Injured 11 15 16 7 5 4 58
lip WSuperficial Injuries
Head/Neck 9 7 7 4 4 1 32
Torso – -1 – – -1
Arms -1 3 4 8
Hands – – -I -1
Shoulders 1 2 2 2 – 2 9
* Legs – – 3 – 2 -5
Thighs 6 5 4 1 1 17
Groin 4 4 6 1 – – 15
Knees 3 1 – 2 1 7
Feet 1 1 1 – – 3
Sprain/Strain:
Cervical 2 7 3 2 1 15
Shoulders 3 – – – – 3
Ribs -1 – -1
Ankles – 1 1
Knees 1 1 1 3
Thoracolumbar – – -1
Fracture:
Ulnar 1 -1
Vertebral -1 8 2 1 12
Clavicle 1 – – – -1
Knees – – -1
Nose 1 – -I
-) Other:
Concussion/Amnesia/LOC 3 2 1 3 9
Dislocated shoulder -1 – -1
Lung contusion – -1 1
Myocardial contusion -1 1
Barotrauma -1 – –
I
4
4~~ % –
% %S . ,%.

,% 54
Table 16. Non-Fatal Canadian Forces Ejection Injuries -1966-1974
No. of Type of
Average Q Force (psi) Ejections Injury
MINOR INJURIES (N -63)
.56 15 Facial1.3 17 Facial
2.4 12 Facial
4.5 4 Facial/Muscular Aches
7.5 2 Facial/Muscular Aches
Unknown 13 Facial
SERIOUS INJURY (N -19)
.56 5 Contusion to kidney
* Compression Fracture to T-10,T-12
Compression Fracture to T-4,T-6
Fractured ribs/torn bladder
Burns
-: 1.3 6 Fractured skull
Compression Fracture T-II,T-12
Compression Fracture T-12,L-I
Compression Fracture T-10,T-II
Compression Fracture T-8
Compression Fracture D-9,10,11,12
2.4 2 Compression Fracture T-8
Burns
4.5 2 Compression Fracture T-12,L-2
Fracture upper arm, broken ribs
Compression Fracture L-1
7.5 2 Burns
Unknown 2 Compression Fracture T-11
Compression Fracture T-10,T-II
––-
…
%0 %

* v I J! WU W%7 -JU'%1 1 –
55
Table 17. Post-Ejection Fatalities -1972-1987 *
Type Personnel Fatalities Cause Ejection Attitude
CF101017 2 2 Pitch-up, stall 229 -355 ft AGL
after take-off > 90 deg RH bank
CT114010 2 1 Fuel Pump Failure < 100 ft AGL
on take-off; 35 deg ND
delayed ejection 120-135 deg RH bank
CT114028 2 2 Birdstrike after < 50 ft AGL
take-off; 20-36 deg Li bank
delayed ejection 15-42 deg ND
CT114118 1 1 In-flight break-up 40-50 ft GL
during airshow 60 deg ND, 90 deg
LH bank high sink rate
CT133363 2 1 Vertical fin stall 500 ft AGL, 30-45
deg ND, 30 deg LH bank
CT133639** 1 1 Undetermined (fire?) ground level; 70-80 deg ND
CF104864 1 1 Pitch-up 150 ft AGL
CT114127 1 1 Landing, lost control 40 ft AGL
–––––––––––––––––––––––––––––-
CT114136 2 1 Loss of control low
Total- 9 14 11
• Total ejections for this period is 103. Thus, ejection success rate is
89.3%.
** Technically the pilot of 639 had pulled the ejection handles, the
canopy had jettisoned, but ground impact occurred just prior to the
ejection gun firing. It can therefore be classified as an out-of-
envelope attempted ejection.
-V-

56
Table 18. Ejections Requiring Freefall to Parachute Opening -1952-1987
(including manual bailouts)
Aircraft Type Date Ejection Altitude (ft) Disposition
CT13321078 7 Jul 54 19000 bailout-successful
.3,
21252 21 Oct 54 20000 successful
21460 13 Jan 57 20000 -pilot bailout-successful
20000 -t'pilot successful
21600 7 Jul 60 20000 -pilot successful
20000 -2'pilot successful
21646 5 Dec 61 31000 -pilot successful
* 31000 -:dpilot successful
3- 21611 31 Mar 66 15000 -pilot bailout-successful
20000 -2apilot successful
F86 23278 27 May 55 27000 successful
23384 9 Jan 57 28000 successful
23413 9 Jan 57 28000 successful
23514 12 Jun 57 27000 successful
23669 28 Aug 57 24000 fatal
– 23556 10 Feb 58 35000 successful
23546 1 Aug 60 28000 successful
CF10018762 30 Jun 59 35000 -nay only fatal
18789 8 May 71 31000 -pilot successful
31000 -nay successful
CF10412884 24 Feb 65 19000 successful
12738 13 Aug 65 20000 successful
C%"

* 57
Figure 1. Aircraft Airspeed -1940-1985
I ' . -U Z– – – – – – – –
wla
8q _
,z.- – – -cvo
* I
/NO 19' 1o 1* 96 95 f 43 /M "
1–EA-
%4 6

* 58
"-'" Figure 2. The CF1O0 Election Seat (1967)
(EQO-55-50 -2C)
,.
0
A,
.' V
A%.

1 59
* Figure 3.a. Leg Restraint System
SHOUDER TRAPQUICK-RELEASE BOX
SNUBBING UNIT
0D RING
/ SNUBBING UNIT
CAFSTRAPS
RIE
/ STRAPSSRIVE
UNI PIT FL O/I E
izjBRACKEET
LEG~STAP RERANISRPVOEITFORTRCE
S14UBIN

S 60
"-i Figure 4. CFI01 E.jection Seat (T.O. 13A5-18-3)
.= 'p
"%.
• .,
:*:::i:
.
'p.

* 61
Figure 5. CF104 Ejection Seat (EO 55-50A-3)
HANDLE, AUXILIARY CABLE HARS
SPACER BLOCK
SHOULDER HARNESS
OXYGEN HOSE
LAP BELT HOSE
BALLISTICS
SWITCH SEAT ACTUATOPFOTRM
-. DISCONNECT, CANOPY EJECTION
BALLISTICS SYSTEM
6%

* 62
Figure 6. SJU-9/A CF188 Ejection Seat (left-hand view)
1. Parachute container
2. Drogue gun
3. Oxygen hoses and mic/tel lead
4. Sticker strap
5. Lap strap
6. Seat Bucket
7. Go-forward lever N
8. Locating block 2
9. Rocket motor
10. Seat height actuator switch
11. Leg restraint line (2)
12. Lower garter buckle (2)
13. Upper garter buckle (2) 18
14. Survival kit
15. Seat firing handle .
16. Back rest 17
17. Parachute risers 19
18. Head pad
19. BIR
20. Oxygen gauge
21. Canopy breaker
22. Trip rod to drogue gun1 zz
23. T handle 16
24. Initiators 1
25. Cross strap
26. rocket nozzle
15
:-.::•3
1g: .( / r
S… .. / S
,.""6
0
75

63
Figure 7.a. BASIC COMPONENTS OF CONVENTIONAL EJECTION SEAT Showing-
canopy breaker (1), headrest (2), seat frame(3), ROCAT attachment (4),
seat bucket (5), rotary actuator (6), shoulder harness (7), RPI lap
belt (8), seat kit (9), negative-G strap (10), ejection handles (11)
and ballistic hose (12).
a4
a1
I-

64
Figure 7.b. CT114 Ejection Seat Schematic showing Rotary Actuator Webbing
*%.

65
Figure 7.c. CF188 Ejection Seat Components
1. Seat bucket 6. Catapult/guide rails
2. Seat kit 7. Rocket motor initiator
3. Head box/parachute 8. Drogue gun
4. Ballistic Inertia Reel 9. Barostatic time delay/C limiter/
5. Main Beams shackle release
10. Rocket Mat-or
U8

* 66
Figure 8.a. RPI Lap Belt Buckle showing: parachute arming key (1)
right shoulder harness (2), left shoulder harness (3) and
* negative-G strap (4).
00
Figure 8.b. RPI Lap Belt JinL kPoiinso ngbals ccbe(1
00
Fiue8b P a eti okPsto hwn:blitccbe(
00
I,,
-L l

67
Figure 9.a. Quick-Release Fitting -Don Position
Figure 9.b. Quick-Release Fitting -Unlocked Position

* 68
Figure 9.c. CF188 Simplified Combined Harness (SCH) showing: seat
attachment points (4- ); risers (1), BIR attachment (2), shoulder strap
adjust (3), T-handle (4), leg strap adjust (5), sticker clip (6), lower
harness lock lug (7), seat pack connect (8), V-strap (9), negative-G
strap (10), and velcro patch for oxygen regulator (11)
0@

69
Figure 10. CF188 Leg Restraint system showing: Taper plug (1). upper
garter attachment (2), lower garter attachment (3), snubbing box (4),
release tab (5), break link (6), and floor attachment (7) (right-hand
* side)
W
710~~~~ N-0- N NN,7 .

6 70
Figure 11. Conventional Rocat System
showing: ejection seat guiderail (1), catapult (2), rocket (3),
main beams (4) and venturi (5).
0
ii 0

71
Figure 12. CT133 Election System Schematic
Command Ejection
Front Seat
(R) Handle
M3A1 M3A1
Initiator Initiator (2)
BIR Initiator Panel M5A2
i Front Seat Canopy Thruster
M0I 1431 M5A2
2 sec 1 Sec 0 delay M3A2
* delay delay Initiator
M7 2 MAl
Initiator Canopy Remover
0.5 sec delay
M31 BIR
Initiator Rear Seat
1 sec delay
-p Front Seat L Rear Seat Rear Seat
ROCAT ROCAT Handle
i 4 M3A 1
M32A1 M32AI Initiator
1 sec delay I sec delay
*Lap Rotary Lap Rotary
Belt Actuator Belt Actuator
Seat Seat
Separation Separation
6r
.- ~p

* 72
Figure 13. CT114 Election System Schematic
(R) Seati
(L) Handle (R) Handle
1 Jr
M27 M27
Gas M72 M31
Generator Initiator Initiator
* 0.5 sec delay 1 sec delay
MIA3 MIAl
Canopy Canopy
.- Extractor Remover
M5A2 Seat
Canopy Booster ROCAT
Initiator
Canopy Mechanical
Jettison Tripper
: CL) Seat4
Handle M27 M32 Han lLap Belt Initiator
1 sec delay
Lap Belt Rotary Actuator
M3A
0';' Canopy
Emergency
Jettison Seat/Pilot
Se pa ra t i o n

q
73
Figure 14. CF 116 Ejection System Schematic
Front Seat
ir I
(L) Handle (R) Handle
M27 M27
Initiator Initiator
M26 M26
0.3 sec delay 0.3 sec delay
Gas Canopy Internal
Generator Quick Disconnect C opy JettisonM27
BI* M25Al.
Canopy Thruster M27
Initiator
Canopy Jettison I1rnal
Canopy
Jettison
MX37
ROCAT -> Seat EjectionI
Mechanical
Tripper
P otary Actuator
M32AI 1
1 sec delay PILOT/SEAT
SEPARATION
MA5 Lap Belt

74
Figure 15. CF188 Ejection System Schematic
"D" Ring
Initiators
DlyDelay Jtio
-CATAPULT
SEAT EJECTION
(R) Trip (L)
Rod Rod
Barostat Rocket Motor Drogue Gun
Time Release Initiator Delay Initiator
1.5 sec 0.5 sec
*ROCKET MOTOR DROGUE CHUTES
0.25 sec
Harness Scissor Parachute
RlaeShackle Unlock
PARACHUTE
DEPLOYMENT
SEAT/PILOT
SEPARATION
r0 .

75
Figure 16. SJU-9/A Election Seat
1. Headbox
2. Time release mechanism
16 3. Shoulder harness reel
4. Manual override handle
15 5. Seat safety handle
6. Pin puller
7. Lower attachment bracket
8.15 8 Trombone tubes
9. Seat structure
10. RH ballistic delay
11. Canopy jettison connector$ zo –- 12 .Catapu lt13. Trip rods
14. Upper attachment bracket
, 15. Parachute mechanical lock
16. Catapult initiator/mani-
fold valve assembly
17. Rocket initiatorZ 18. Rocket motor
19. Shackle release plunger e 20. Drogue gun
21. Catapult secondary cartridges
".- 14
.." 12
w4
110
-A'21
" 15
7k

" 76
N' Figure 17. CT114 Seat Kit (A), and Details of Bottom Lock Mechanism (B)
% showing: lanyard (1), airlock connector (2), fibreglass kit (3),deployment handle (4), canvas closure (5), pins (6) and pin cover (7).
0 0
A
%.2 .. %
.1%0;.2
.2-'-8
0,–
0,..
N.%-%X' '.

Lrrr,"j W ww% , M'UK RV,- ~- M w.wwvv-tPu r-1 v-w.,-l
77
Figure 18. The Relationship Between Performance and Arousal
HIGH -PRACTICE=RIGHT SHirT-
z
U.%%U.
w
C-)
LOW
LOW HIGH
-AROUSAL LEVEL

%? Ir wvw 1. 11 -Y.- U -.1w – 'J ~- ~ " .. Jw ' ult– nF TW ZV, FY? iJ W v w Y -.IT " V WT W' 1 I- -l' xr -7
~0 78
Figure 19. Ejection Loading as a Function of Time for a Typical Rocat Seat
Line AB, catapult acceleration phase; BC rocket phase; dotted line
occupant acceleration (modified after White 1985).
.%
0.1 o. 0 .3 o.4 o ..
" a.TIME (SEC)
.0
..°

4 79
Figure 20. Minimum Ejection Altitude (CF188A)
2000 –- -……….. …… ….. ……..
1300- -: …. … ..
.. .__…___…_ ..
1400-. ..
14 AA IT
~~~~~~~~~~~ :::I–––––––……..
o 0––––––
-o700 –
3003
n o. … ………..
……..
.. .. … …. … …. …. .. .. .. .. .
…….
………. …
loc

80
Figure 21. USAF Ejection Flail Injury Versus Airspeed/1964-1972
(Brinkley 1973)
I06
0.
100 zo oo30 00 400 560 600 TOO
SKIAS
20Figure 22. Altitude Distribution of 87 Successful Ejections -1972-1986
(Data extracted from Annex B)
37,
Jq30
4 -A°" O
* ALTI1-LAFE X 10.C0 FEET

. 481
Figure 23. The Relationship Between Altitude and True Air Speed
At Terminal Velolcity in a Free-Falling Man (Ernsting 1978)
to -I A
200 _ _
* 1~~~~00…….
SL 1 0 20 30 1 0 Fe IQ 0 so0 O
Figure 24. Parachute Opening Shock at Various Altitudes of Deployment.
(Values shown are typical of those obtained at the terminal velocity
of man using a 28 foot canopy.)
25-
S20–
0 S 10 is 20 25 30
Opening stiok G'.
4
… . / …\-N

82 ANNEX A
"A" CATEGORY AIRCRAFT ACCIDENTS
1972 -1987
Code: NENF -No Ejection, Non-Fatal NEF -No Ejection, Fatal
CFIT -Controlled Flight Into Terrain ROE -Rapidity of Events
UE -Unable to Eject UK -Unknown Cause
-. Crew Aircraft No. Date Deaths Eject. Fatal Eject. NENF NEF Cause *
2 CF100788 17 Oct 73 2 0 0 0 2 CFIT
2 792 3 Mar 72 0 2 0 0 0
…………………………………………………………………….-
2 CF1O1007 22 Jun 84 0 2
2 016 14 May 75 0 2
2 017 1 Dec 77 2 2 2
2 018 29 Nov 79 0 2
2 019 12 Aug 73 0 2
2 023 14 Jan 78 0 1 1 –
2 026 19 Apr 82 0 2
2 029 15 Sep 83 0 2
2 033 25 Sep 80 0 0 2 –
2 039 4 Jul 75 0 2
0 2 049 7 Jul 73 0 2
2 055 19 Feb 80 2 0 2 ROE
2 061 5 Jul 76 2 0 2 CFIT
2 062 14 Feb 73 0 2
2 CF104631 7 Nov 74 0 2 –
1 640 3 Dec 82 1 0 1 UE
1 647 3 Dec 82 1 0 1 UE
2 649 17 Nov 77 2 0 2 UE
2 651 24 Jun 80 0 2 –
2 656 4 Mar 77 2 0 2 CFIT
2 665 16 Mar 81 0 2 –
2 666 5 Mar 75 0 0 2 –
1 705 11 Dec 81 0 1 –
1 714 12 Feb 76 1 0 1 CFIT
1 715 11 Dec 74 0 1 –
1 720 17 Aug 73 0 1
1 732 30 Apr 82 0 1 –
1 1 744 18 May 83 1 0 1 UE
1 754 21 Feb 79 1 0 1 CFIT
1 762 9 Jun 81 0 1 –
1 769 4 May 73 0 1 –
1V 1 772 18 Apr 73 1 0 1 CFIT
1 775 15 Nov 77 1 0 1 CFIT
* 1 779 9 Sep 75 0 1 –
1 789 27 May 74 0 1 –
1 1 807 27 Nov 80 0 1 –
1 813 22 May 83 0 1 –
1 821 10 Jan 83 0 1
1 827 29 Jul 82 0 1
0 1 829 19 Aug 78 0 1
– –––––––––––––
V

T-
83 ANNEX A
Crew Aircraft No. Date Deaths Eject. Fatal Eject. NENF NEF Cause*
1 CF104830 16 Jun 83 0 1
1 838 7 Mar 78 0 1
1 840 25 Jul 78 1 0 1 CFIT
1 857 22 Oct 73 1 0 1 CFIT
1 859 27 Aug 80 0 1
lp1 864 10 May 73 1 1 1
1 872 11 Mar 74 0 1
1 892 4 Jun 82 0 1
1 895 14 Mar 74 0 1
1 CF116706 13 Dec 79 0 1
1 711 3 Jan 74 0 1 –
1 720 12 Nov 87 0 1 –
1 72S 11 Jul79 0 1 –
1 731 20 May 77 0 1 –
1 735 26 Feb 81 1 0 1 UE
1 741 2 Mar 76 0 1 –
1 755 7 Jun 77 1 0 1 UK
1 1 756 12 May 74 0 1 –
1 760 2 May 76 0 1
1 761 12 Feb 79 0 1
1 770 10 Aug 77 1 0 1 UK
1 771 20 Jan 83 1 0 1 CFIT
1 816 7 Mar 73 1 0 1 CFIT
2 817 22 Dec 83 0 0 2 –
2 820 11 May 76 0 1 1 –
2 844 30 Apr 82 2 0 2 ROE
2 CT114007 3 Apr 78 2 0 2 ROE
2 010 26 Jun 85 1 2 1
1 016 19 Dec 73 0 1
2 028 31 May 76 2 2 2
2 029 12 Aug 75 0 2
2 057 24 Nov 78 0 2
2 074 21 May 75 0 2
1 082 16 Jul77 0 1
1 1 088 16 Jul77 0 1
1 117 16 Apr 80 1 0 1 CFIT
1 118 4 May 78 1 1 1
1 122 30Oct79 1 0 1 ROE
2 123 11 May 76 0 2 –
1 125 13 Jul78 0 11 127 20 Mar 72 1 1 1
1 129 17 Jun 86 0 1
1 132 24 Jan 77 0 1
2 136 22 Aug 73 1 2 1
2 137 26 Feb 74 0 2
2 138 14 Sep 76 0 2
%%
,LI

0 84 ANNEX A
Crew Aircraft No. Date Deaths Eject. Fatal Eject. NENF NEF Cause
1 CT114158 15 Nov 79 0 1
2 165 23 Sep 79 0 2
1 179 14 Jul 73 0 1
1 183 10 Jun 72 1 0 1 UE
2 CT133069 21 Sep 82 1 1 1 UE
2 315 7 Apr 87 2 0 2 CFIT
1 349 1 Feb 76 1 0 1 ROE
i 2 363 14 Sep 84 1 2 1
2 405 20 Aug 80 0 2
1 442 14 Feb 81 0 1
1 520 19 Sep 73 0 1
1 603 19 Sep 73 0 1 0
1 639 26 May 82 1 1 1
1 CF188715 12 Apr 84 1 0 0 1 CFIT
1 717 24 May 86 1 0 0 1 CFIT
1 721 21 Sep 87 0 1 0
1 737 4 Jun 85 0 1 0
1 761 20 Oct 87 0 1 0
* 2 919 4 May 87 0 2 0
148 107 48 103 11 8 37
* Not to be confused as official cause factor. These are the most
probable causes for failure to eject from the aircraft as subjectively
evaluated by the author from reading the Board of Inquiry.
".
.

SECURITY CLASSIFICATION OF FORM
(highest classification of Title. Abstract, Keywords)
DOCUMENT CONTROL DATA
iSecurity e'assification of title, body of abstract and indexing annotation must be entered when the overall document is ciaslifiedl
R (the name and address of the organization prepi , d t 2. SECURITY CLASSIFICATION
Organzations :or whom the document was prepared, e.g. EStablishment sponsoring n overall security classificution of the documenta contractor's report, or tasking agency, are entered in section 8)including special warning terms if applicable)
1XIE-M -Fop o0 tAN)CLAS
T 7TLE ithe compiete document title as indicted on the title page. Its classification should be indicated by the appropriate
bbreviation (S,C.R or U) in parentheses after the title.)
EJECTIv SST- AJD 1HE HuLMAM FACAOR -A 6tIDE' FOR FLJ6FT
ctGcM-s "D~T AE-eeiM~bIcAL -rRAINMES
4 AUTHORS (Last name, first name, middle initial. If military, show rank, e.g. Doe, Maj. John E.)
lipST-,LIGE1,J CAPT WA YNE 2 1
DATE OF PUBLICATION (month and year of publication of Ea1. NO. OF PAGES (total 5b. NO. OF REFS (total cited in
ccument) containig information. include document)
FE-7E Iqg>g Annexes, Appendices, etc.)
88
C DESCRIPTIVE NOTES (the category of the documetit, e g. technical report, technical note or memorandum. If appropriate, enter the type of
report, e.g. interim, progress. summary, annual or final. Give me inclusive dates when a specific reporting period is covered.)
TE C1 M)I CA L R EPo..–
5 SPONSORING ACTIVITY (the name of the department project office or laboratory sponsoring the research and development. Include the
address.)
SPROJECT OR GRANT NO. (if appropriate, the applicable research 9b. CONTRACT NO. (if appropriate, the applicable number under
and development project or grant number under which the document which the document was written)was written. Please specify whether project or grant)
NA NA
10a ORIGINATOR'S DOCUMENT NUMBER (the official document lob. OTHER DOCUMENT NOS. (Any other numbers which may
number by which the document is identified by the originating be assigned this document either by the originator or by the
activity. This number must be unique to this document) sponsor)
NIL(.
DCUMENT AVAILABILITY (any limitations on further dissemination of the document, other than those imposed by security classification)
I Unlimited distribution
Distribution limited to defence departments and defence contractors; further distribution only as approved
i Distribution limited to defence departments and Canadian defence contractors; further distribution only as approved
,"Distribution limited to government departments and agencies; further distribution only as approved
Distribution limited to defence departments; further distribution only as approved
Die, Iplease specify):
2 DOCUMENT ANNOUNCEMENT (any limitation to the bibliographic announcement of this document. This will normally correspond to
re Document Availabilty (11). However, where further distribution (beyond the audience specified in 11) is possible, a wider
announcement audience may be selected.)
WJNCLAS
SECURITY CLASSIFICATION OF FORM
0CDr, 2i06 P
% W N.

MJCLFt-5
* SECURITY CLASSIFICATION OF FORM
1 3 ABSTRACT I a brief and factual Summary of the document I1 may also appear elsewhere in the body of the document itself. It is higii,
desirable that the abstract of classified documents be unclassified. Each paragraph of the abstract shall begin with an indication of the
security classification of the information in the paragraph (unless the document itself is unclassified) represented as (S), (C), (9), or IU .
It is not necessary to include here abstracts in both offical languages unless the text is bilingual)I.
THiS Doct.4-ieN)r De7,LS -7;4E RC/IF/CF LCXPk-',irACI5- kirrb- AFIE-c7-10,J gy-rcMS
hbM /9&2~ AJ'i7-H SPE-C/AL E?-)AsASIS OAJ HU,6AJ F4CFORqS //'VOLVCC /AI
MA)<JA)- -74E- D6EcIs/oAJ 7-o EU-EC-7
%0
–I'14. KEYWORDS. DESCRIPTORS or IDENTIFIERS (technically meaningful terms or short phrases that characterize a document and could be
helpful in cataloguing the document They should be selected so that no Security classification is required. Identifiers, such as equipment
model designation, trade name, military oroject code name, geographic location may also be included. If possible keywords should be selected
from a published thesaurus. e.g. Thesaurus of Engineering and Scientific Terms (TEST) and that thesaurus-identified. If it is not possible to
select indexing terms which are Unclassified, the classification of each should be indicated as with the title.)
EJECTb 0M A
E JECJA) %Y
xrPACKS
SLARIJVA L
% ~ ACrClDEDQJ IVST-1AT-,.Q
ECiu'J 4Ak'M&
V EJ~cT~4J iCf&P
I-kit"
SECU.RITY CLASSIFICATION OF FORM
0J