FLIGHT SAFETY FOUNDATION

HELICOPTER SAFETY
Vol. 15 No. 6 For Everyone Concerned with the Safety of Flight November/December 1989

Helicopter Crashworthiness Part One

In this first of a two-part series, the author reviews efforts to study
and improve crash survival and helicopter accident.

by

Roy G. Fox
Bell Helicopter Textron

Perceived occupant safety is an important part of the tively, occurred in accidents resulting from material fail-
public acceptance of helicopter transportation. There are ures.
three levels of design effort involved in aviation safety:
(1) preventing the occurrence of an emergency, (2) mini- Airframe and engine manufacturers will certainly con-
mizing the effect of an emergency, and (3) minimizing tinue to try to eliminate material failures completely as
injuries in a crash. This paper discusses the present accident causes. Yet, even if they were to succeed,
status of civil helicopter crashworthiness activities in the occupants would still be injured by accidents resulting
United States for the third level crash survival. It from other causes; 74.2 and 65.5 percent of the occu-
combines information from References 1, 2, and 3 with pants involved in crashes of single and twin-turbine heli-
later unpublished information. copters, respectively, would still be seriously or fatally
injured. Obviously, there is good reason for continuing
to improve crash survival features of helicopters.
Why Improve Crash Survival?
Why should one consider improvements in crash sur- The Rotorcraft Airworthiness Requirement Committee
vival? The answer is that some accidents will continue to (RARC) of the Aerospace Industries Association of
occur regardless of aircraft design features. How do America (AIAA) established a Crashworthiness Project
these accidents affect the occupants? In the United States, Group (CPG) to develop and recommend realistic crash-
the National Transportation Safety Board (NTSB) and worthiness criteria for future civil helicopters. The
Federal Aviation Administration (FAA) collect data on members of the CPG represented Bell Helicopter Tex-
accidents. These data for U.S.-registered helicopters from tron Inc., Boeing Helicopters (formerly Boeing-Vertol
1982 through 1986 were analyzed in an evaluation of Helicopters), McDonnell Douglas Helicopters (formerly
configuration effects on occupant injuries. Hughes Helicopters), and Sikorsky Aircraft. The re-
sulting recommendations (2) were for energy-attenu-
Serious (major/fatal) injuries were categorized by the ating seats with shoulder harnesses and crash-resistant
initiating accident cause factor for helicopters with single fuel systems.
turbine engine and twin-turbine engines and it was found
that between 25.8 and 34.4 percent of the known serious The CPG limited its consideration to future civil helicop-
injuries in single and twin-turbine helicopters, respec- ters for which a type certificate application has not yet
 been made. This allows the designer to plan on larger ter that meets or exceeds the requirements in the FARs as
engines and any other changes needed to accommodate interpreted by the FAA. Having done so, the manufac-
the weight penalty of safety features. Thus, a balanced turer must then sell the helicopter in a competitive mar-
design is maintained if the safety improvements are in- ket to an operator who in most cases must make a profit
cluded in the initial concept. A large weight increase on in an even more competitive market. Thus, it is RARCs
a current helicopter can degrade the safety it already has position that safety requirements for future civil helicop-
by reducing hover and climb performance. ters should be improved to realistic and beneficial levels,
not to a mandated, overly severe level based on specific
military requirements.
Background
The historic regulatory approach has been to ensure that The crash conditions for civilian helicopters differ from
an aircraft is safe to fly by requiring a minimum static those for military helicopters, and therefore the criteria
strength level. U.S. Code of Federal Aviation Regula- for crash safety features also differ. The vertical heli-
tions (FARs) 14, Parts 27 and 29 specify these for heli- copter crash condition is the most important criterion
copters with gross weights under 6,000 pounds (2,722 that affects helicopter design. Figure 1 (4) shows the
kg) and over 6,000 pounds (2,722 kg), respectively. Heli- vertical velocity component of U.S. Army light fixed-
copters are designed to meet or exceed these minimum wing and rotary-wing aircraft involved in survivable ac-
static strength standards. The present FARs (27.561 and cidents. It indicates that 95 percent of those accidents
29.561) have only one specific crash requirement: that occurred with a vertical velocity component of 42 feet
the helicopter be able to endure a five-feet-per-second per second (12.8 m/s) or less. A more recent study
(1.5 m/s) drop impact without causing serious injury to limited to U.S. Army helicopters (5) indicated that the
occupants using safety equipment. helicopters involved in 95 percent of survivable acci-
dents had a vertical velocity component of 30 feet per
The present helicopter seat has a static strength require- second (9.1 meters per second m/s) or less (Figure 1).
ment of 4G forward, 4G downward, 2G lateral, and 1.5G The recent FAA civil helicopter crash scenario study (3)
upward for a 170-pound (77 kg) occupant. The manufac- indicates that 95 percent of survivable civil helicopters
turer must verify that a seat meets the static strength involved in accidents impact at 26 feet per second (7.9
requirements by analysis and by performing a static pull m/s) or less, as shown in Figure 1. Thus a realistic civil
test on a test seat for each of the specified directions. helicopter criterion for vertical impact is 26 feet per
second (7.9 m/s), and not the 42-foot-per-second (12.8
m/s) military requirement. It should be noted that at 26
Civil vs. Military feet per second (7.9 m/s) the kinetic energy that must be
The U.S. Army and others have analyzed accidents in- attenuated is 27 times that which must be attenuated at
volving Army helicopter and light fixed-wing aircraft to the present five-feet-per-second (1.5 m/s) impact stan-
determine occupant crash survivability factors and what dard of the FARs.
improvements should be investigated. The culmination
of these research activities was the Crash Survival De-
sign Guide, the latest revision of which is TR 79-22 (4).
The first major use of this military research was the
development, production, and installation of the crash-
resistant fuel system (CRFS) to prevent massive post-
crash fires in survivable army helicopter accidents.

The first CRFS was installed by Bell Helicopter Textron

Inc. (BHTI) in the UH-1H in May 1970. Since CRFS
introduction, there have been very few thermal fatalities
in survivable Army helicopter accidents. This is an out-
standing safety accomplishment. However, the U.S. Army
accepted a large weight penalty to ensure the success of
the CRFS. The situation is different for the commercial
operator.

The military gives the helicopter manufacturer specific

requirements for a military helicopter application. The
military must get exactly what it specified or it will not
accept the helicopter. For commercial models, the manu- Figure 1
facturer designs, tests, qualifies, and produces a helicop-

2 F L I G H T SA F E T Y F O U N D A TI O N H E LI C O P T E R SAF E TY N OVE M B E R /DE CEMBE R 1 9 8 9

 The crash scenarios of U.S.-registered civil helicopters
were analyzed in Reference 3 to determine crash safety
criteria. The frequency of injuries, their severity per the
Abbreviated Injury Index, and the type of crash scenario
were combined to rank the crash scenarios by their injury
potential (Table 1). This study showed that the six most
serious crash hazards were:
Post-crash fires;
Excessive deceleration loads;
Unusual aircraft impact conditions due to wire
strikes;
Occupant restraint problems/clearances, allowing
striking of structure;
Lack of available shoulder harness, allowing strik-
ing of structure; and,
Injuries that prevented escape in water and led to
drowning.

With the exception of wire strikes, the remaining five Figure 2

hazards can be significantly reduced by an energy at-
tenuating (absorbing) seat with shoulder harness for each
occupant and a crash-resistant fuel system. Human Tolerance (Uninjured)

Lap Belt Only (Ref. 6) Harness & Lap Belt (Ref. 4)

Crash Survival Requirements Longitudinal (-Gx) 15 G (0.002 Sec) 45 G (0.1 Sec)

25 G (0.2 Sec)

Vertical (+Gz) 4 G (Uninjured) 25 G (0.1 Sec)

Crash survival has four basic requirements: Lateral (+Gy) 11 G (0.1 Sec) 20 G (0.1 Sec)

Maintain a livable volume;

Restrain the occupant; Table 1
Keep occupant crash loads within human nonin-
jury tolerance; and, In the forward direction, eyeballs out (as in a head-on
Provide means and time to escape. nose impact), the human body is capable of withstanding
about 15Gs with a lap belt only, assuming that the area in
Livable Volume front is clear of obstructions. However, the addition of a
shoulder harness to restrain the upper torso increases the
The first requirement is to maintain a livable volume uninjured tolerance in this direction to about 45Gs, a
around the occupants throughout the crash sequence. It three-fold increase.
is quite meaningless to invest in special seats and re-
straints only to have the roof structure crush down to the The most important impact direction for a helicopter
floor. Some fuselage deformation into the cabin is ac- occupant is vertical (or eyeballs downward). A shoulder
ceptable as long as it does not significantly interfere with harness increases human tolerance without injury for the
the occupants living space. The fuselage should be a vertical direction from 4Gs to 25Gs, an improvement
rugged structure that will accept some localized defor- factor of six. Laterally, the shoulder harness increases
mation. tolerance by a factor of two.
Occupant Restraints Not only does the shoulder harness prevent many dis-
Injury studies of U.S. Army helicopter occupants in (Ref- abling or fatal injuries during many impact sequences,
erence 4) shows that life-threatening injuries occurred but it also permits the occupant to remain conscious and
primarily to the head, neck, body torso, and spine (Figure coherent by preventing a severe head impact. This can
2). Tolerance to impact loads can be changed signifi- be important if the time to escape is short, as in a water
cantly by the method of restraint. The tolerances shown crash. The reason a shoulder harness is so effective in
in Table 2 are for a young male aviator in good physical the vertical impact or vertical component is that it holds
condition. They can vary with factors such as age, muscle the upper torso upright, keeping the spine aligned prop-
tone, rate of onset, and duration. The tolerances for erly and allowing it to carry much higher crash loads.
civilian occupants would be lower, but the relative im-
provement due to restraints should be comparable. The position of the occupant at impact is important. The
occupant must sit upright and against the seat back or a

F L I G H T S A F E T Y F O U N D A T I O N H E L I C O P T E R SA F E T Y N OV E M B E R / D E C E M BE R 1 9 8 9 3
 shoulder harness cannot function properly. Seat strength accident data of Reference 3. Table 3 shows the use of
is important, but increasing the vertical seat strength restraints by occupants of front and rear seats in civil
without ensuring that a shoulder harness is used is a helicopter accidents.
meaningless waste of time, money, and weight. The
tolerance of the human, with lap belt only, to vertical Severity of Injury With Upper Torso Without Upper Torso
Restraint (%) Restraint (%)
impact is 4Gs, above which body flailing is expected.
Thus, a 20G seat with only a lap belt would be a weight Moderate 84.6 60.0
AIS = 1 or 2
penalty that made no improvement in injury prevention.
Severe 9.6 34.3
AIS = 3 or 4

Comparing major and fatal injuries in survivable acci- Life Threatening 5.8 5.7
AIS = 5 or 6
dents for civil and military helicopters. Note that the
injury percentages are fairly similar for the different type
aircraft, except for the spine injuries. This indicates that
more than 30 percent of the major and fatal injuries in
survivable civil helicopter accidents were related to the Table 3
spine. This is nearly twice as high as for the military The severity of injuries sustained by occupants of civil
occupants. One of the differences is believed to be the helicopter accidents given in Reference 3 data shows the
difference in age of the occupants. Reference 3 indicates effectiveness of the shoulder harness. Table 4 shows the
that the average age of the military pilots was 26 years, percentage of occupant injury grouped by moderate, se-
whereas the average age of general aviation pilots was 38. vere, and life-threatening injury severity by Abbreviated
Injury Index (AIS) codes.
The effect of age on the spinal injury tolerances indi-
cated that the amount of load to cause spinal compres- This shows that only 9.6 percent of those wearing a
sion fracture (on cadavers) decreased dramatically as shoulder harness had severe injuries, compared with 34.3
the persons age increased (7). Civil helicopter occu- percent of those using lap belt only.
pants, being older, have less tolerance to impact than
military occupants. Although each accident is unique and shoulder harness
effectiveness may vary for a particular impact condi-
The RARC Crashworthiness Project Group recommended tion, the shoulder harness can significantly reduce in-
to the FAA that a shoulder harness be required for all juries for the helicopter fleet. It should be noted that
occupants for future helicopter designs. They also rec- even with a shoulder harness, some injuries still occur.
ommended that the torso restraint system specification Thus another design feature is still needed for the
SAE AS-8043, developed by the Society of Automotive attenuation of energy. [See also Shoulder Harness
Engineers (SAE), be used in either the dual or diagonal Restraints Considered by Allen K. Mears in the July
shoulder belt configuration. SAE AS-8043 was compat- 1989 issue of Flight Safety Foundations Accident Pre-
ible with dynamic seat testing recommended by the Gen- vention. Ed.]
eral Aviation Safety Panel for helicopters and for light
fixed-wing aircraft. The SAE AS-8043 would double the Energy Attenuation of Crash Loads
lap belt loop strength from 3,000 pounds (13,345 new-
tons N) to 6,000 pounds (26,689 N) and provide a Impact dissipates the kinetic energy of a moving mass
2,500-pound (11,121 N) shoulder belt. The FAA later to bring it to a stop. Designs for aircraft crash protec-
created a new Technical Standard Order, TSO C114, tion aim to manage this energy dissipation and thereby
Torso Restraint System, which included SAE AS-8043. limit the load transmitted to the occupant to a tolerable
The comment, A passenger would not wear a shoulder noninjurious level. This is typically called energy
attenuation. (Some people refer to it as energy ab-
Restraint System Component Front Seats (Percent) Rear Seats (Percent) Total (Percent)
sorption.) Energy attenuation is analogous to a hy-
Lap Belt
draulic pressure regulator that takes an input of various
Availability rate 97.8 100.0 98.2
Usage rate* 100.0 98.0 99.7
high pressures from a pump but allows only an output
of a constant lower pressure. If the constant lower
Shoulder Harness
Availability rate 42.5 11.1 36.6
pressure is equivalent to a load that the spine can
Usage rate 83.6 100.0 84.6
tolerate without serious injury, the system is referred to
as an energy attenuating system. Thus an energy at-
Table 2 tenuating seat will accept the kinetic energy of the oc-
cupant and seat in a crash, but limits the vertical crash
harness if he had one, is typically voiced by people loads transmitted to the occupant. It spreads the load
opposed to shoulder harness installations. The effective- over a long period by slowing down the occupant over a
ness and use of shoulder harnesses were determined from distance (the seat stroke). The effect of stroking is

4 F L I G H T SA F E T Y F O U N D A TI O N H E LI C O P T E R SAF E TY N OVE M B E R /DE CEMBE R 1 9 8 9

 shown by the formula for the average load during decel- strikes the cabin floor. At this point, the occupants
eration, assuming a triangular pulse: body will try, unsuccessfully, to deform the cabin floor.
G avg (Velocity) 2 This will result in extremely high and intolerable (proba-
bly fatal) loads. A rigid seat of high strength would only
2 gs prolong the time before the seat fails and the occupant is
Where: once again in free-fall to the cabin floor. If an energy
G avg = average load applied as a multiple of the attenuating seat is used in this same impact, the landing
object weight. gear and fuselage would see the same loading but the
Velocity = initial impact velocity expressed in feet seat will allow the occupant to be slowed to a stop while
per second (final velocity is zero) applying a controlled tolerable load. As long as the seat
g = 32.2 feet per second2 squared (gravitational pan loads do not exceed 23G for more than 5 millisec-
constant) onds, the occupant with a shoulder harness restraint should
s = stopping distance in feet. not be seriously injured (8). Since, in this case, the
Note: Peak load would be twice the average G. energy attenuating seat was stroking at a constant 14.5G
load, the occupant loading did not go into the hazardous
For example, if an object impacts at 20 feet per second zone. The FAA has recently established a limit of 1,500
(6.1 m/s), and stops in 0.25 inch (0.64 cm), it has an pounds (680 kg) or less measured in a lumbar load cell of
average G load of 298G. If it strikes at the same velocity a Part 572 instrumented dummy (9). It should be noted
but stops in four inches (10.2 cm) of uniform stroking, its that an energy attenuating seat cannot function unless a
average G load is 18.6G. The latter G load is within shoulder harness is used and the occupant is against the
human tolerance of non-injury loading, whereas the for- seat back.
mer is not. In the same way, an energy attenuating seat
allows the occupant and seat pan to decelerate over a Energy attenuating seats were first required by the U.S.
stroking distance while limiting the vertical loads trans- Army in the Utility Tactical Transport Aircraft System
mitted to the occupant. Average G loading is often used (UTTAS), which was to become the Sikorsky UH-60A.
as a limiting criterion in the design of energy attenuating The UH-60A energy attenuating crew seat has a maxi-
seats. The goal is to use the shortest stroking distance mum of 16 inches (40.6 cm) stroking distance. This
that limits the load on the occupant to a noninjurious required a hole in the floor to allow seat pan motion if the
level. full stroke was needed. The seat, which uses an inverted-
tube energy attenuator, entered service in 1978.
One must be careful when using average G loads in rela-
tion to an aircraft crash. The G loading experienced at Boeing Helicopters, under NASA contract NAS1-14637,
each point in the aircraft is unique and determined by its developed prototype energy attenuating passenger seats.
stopping distance. For instance, in the case of an aircraft The floor-mounted seat had a wire/roller energy attenu-
that collides head-on with an obstacle, the seats are sub- ator mounted in the center leg braces. A ceiling-mounted
jected to progressively lower G forces the farther aft they seat design had a wire/roller energy attenuator mounted
are located due to the energy absorbing effect as the fuse- to the top of the seat back corners. As the seat stroked
lage crushes. Average G load, a meaningless and mislead- down, it pulled the wires through the seat-mounted roll-
ing concept for an accident, should be used only to de- ers, thereby absorbing energy by bending wire. Boeing
scribe the unique loading conditions at a specific point. Helicopters developed this latter seat design under U.S.
Army contract, and it was subsequently used as the UH-
Figure 5 shows a simplified version of the crash loads 60A troop seat.
history that occurs in a severe, flat impact with a high
vertical deceleration component for a typical helicopter. The first civil aircraft to go into production with an
The skid gear absorbs crash energy until the fuselage hits energy attenuating seat for each occupant was BHTIs
the ground. If the helicopter has typical oleo wheel gear, Model 222, first delivered in 1980. In a crash with high
this may not absorb much energy before it fails, because vertical forces, the crew seat slides down vertical guide
oleo motion is rate sensitive. (Specially designed oleos rails. This drives down a center tube, which crushes a
can be made to function at the higher crash impact speeds.) carbon epoxy tube that is an energy attenuator. The
Likewise, if the gear is retracted at impact or the impact passenger seats use a crushable rigid foam block beneath
terrain (such as water) precludes landing gear function- the seat pan to crush at the same 14.5G level for a 170-
ing, the landing gear will absorb no energy. pound (77 kg) occupant as does the crew seat.

Since the fuselage is fairly rigid, the G loading on the BHTIs Models 214ST and 412 share a common energy
floor would peak out at a very high load. The standard attenuating crew seat design. Like the 222 crew seat, it
aircraft seat will fail a few Gs higher than its design uses a carbon epoxy energy attenuator, but it has more
criteria, thus allowing the occupant to free-fall until he stroking distance available. The original Model 412

F L I G H T S A F E T Y F O U N D A T I O N H E L I C O P T E R SA F E T Y N OV E M B E R / D E C E M BE R 1 9 8 9 5
 passenger seat was an energy attenuating seat that used 6. A. M. Eiband, Human Tolerance to Rapidly Applied
the wire/roller concept. It was similar to the ceiling- Accelerations: A Summary of the Literature, NASA
mounted seat, except for being mounted to a frame and Memorandum 5-19-59E, (U.S. National Aeronautics and
having an antirebound feature. Like those of the 222, the Space Administration, 1959).
energy attenuators are designed for 14.5G for a 170-
pound (77 kg) occupant. These seats have the greatest 7. E. I. Stech and P. R. Payne, Dynamic Models of the
vertical energy attenuation available in a civil aircraft Human Body, AMRL -TR-66-157, (Aerospace Medical
crew and passenger seat. Research Laboratory, 1969).

Part Two of this series will be presented in the next issue. It 8. Seat System: Crashworthy, Non-Ejection, Aircrew,
will discuss crashworthiness tests and proposals to improve General Specification for, MIL-S-58095(AV), (U.S. De-
crash safety features for future rotorcraft designs. partment of Defense, Washington, 1971).

9. Occupant Restraint in Normal and Transport Category

Rotorcraft, Notice of Proposed Rulemaking (NPRM), Docket
References No. 25287; Notice No. 87-4, (U.S. Federal Aviation Ad-
1. R. G. Fox, Relative Risk, The True Measure of Safety, ministration, Washington, 1987).
Flight Safety Foundation. 28th Corporate Aviation Safety
Seminar, April 1983. About The Author
Roy G. Fox directs the System Safety Engineering Group
2. R. G. Fox, Realistic Civil Helicopter Crash Safety, as Chief Safety Engineer for Bell Helicopter, Textron.
American Helicopter Society National Specialists Meet- He joined Bell in 1966 immediately after graduation with
ing on Crashworthy Design of Rotorcraft, April 1986. a Bachelor of Science Degree in Mechanical Engineer-
ing from New Mexico State University. In addition to his
3. J. W. Coltman et al., Analysis of Rotorcraft Crash work in safety engineering, Fox has been deeply involved
Dynamics for Development of Improved Crashworthiness in crash survival of military and civil helicopters. In this
Design Criteria, DOT/FAA/CT-85/11. (U.S. Federal Aviation field, he has directed the Crashworthiness Project Group
Administration, 1985.) for the helicopter industry.
4. S. P. Desjardins et al., Aircraft Crash Survival Design Fox is a member of the General Aviation Safety Panel for
Guide, USARTL-TR-79-22, (U.S. Army Research and seat restraints and post-crash fire protection, and par-
Technology Laboratories, 1980). ticipates in the SAE committees on seat and restraint
requirements. He is a lecturer on crash survival and
5. D. Crist and L. H. Symes, Helicopter Landing Gear human performance at the Bell Pilot Training School
Design and Test Criteria Investigation, USAAVRADCOM- and the U.S. Federal Aviation Administration Helicopter
TR-82-15, (U.S. Army Research and Technology Labo- Safety and Accident Investigation course.
ratories, 1981).

Whats Your Input?

Flight Safety Foundation welcomes articles and papers for publication. If you have an article proposal, a completed manuscript or a
technical paper that may be appropriate for Helicopter Safety, please contact the editor. Submitted materials are evaluated for
suitability and a cash stipend is paid upon publication. Request a copy of Editorial Guidelines for Flight Safety Foundation Writers.

HELICOPTER SAFETY
Copyright 1989 FLIGHT SAFETY FOUNDATION, INC. ISSN 0898-8145
Articles in this publication may be reprinted in whole or in part, but credit must be given to Flight Safety Foundation and Helicopter Safety.
Please send two copies of reprinted material to the editor. Suggestions and opinions expressed in this publication belong to the author(s) and are
not necessarily endorsed by Flight Safety Foundation. Content is not intended to take the place of information in company policy handbooks and
equipment manuals, or to supersede government regulations. Manuscripts must be accompanied by stamped and addressed return envelopes if
authors want material returned. Reasonable care will be taken in handling manuscripts, but Flight Safety Foundation assumes no responsibility for
material submitted. Subscriptions : $50 U.S. (U.S. - Canada - Mexico), $55 Air Mail (all other countries), six issues yearly. Staff: Stephanie
F. Yoffee, production coordinator; Jacque Edwards, word processor; Arthur H. Sanfelici, consultant Request address changes by mail a nd
include old and new addresses. Roger Rozelle, editor, Flight Safety Foundation, 2200 Wilson Boulevard, Suite 500, Arlington, VA 22201-3306
U.S. tel: 703-522-8300 telex: 901176 FSF INC AGTN fax: 703-525-6047

6 F L I G H T SA F E T Y F O U N D A TI O N H E LI C O P T E R SAF E TY N OVE M B E R /DE CEMBE R 1 9 8 9