AIN'T By Richard F. Chandler, Supervisor Protection and Survival Lab Civil Aeromedical Institute Mike Monroney Aeronautical Center P. O. Box 25082 Oklahoma City, OK 73125

use of a correctly designed and installed upper torso restraint could mean the difference between a healthy productive life or death or irreversible injury should that relatively rare occurrence spoil the pleasure of flying. Pelvic Restraint Systems

Introduction ± HE MAIN PURPOSE of this article is to acquaint you with a few of the basic design concerns regarding restraint systems so that you will be

able to understand the design, to install or to evaluate a system with some knowledge of its function and its possible malfunctions. Particular emphasis will be on systems suitable for crew restraint, but the general concepts will be applicable to passenger restraints as well. Before we get started, it would probably be of some use to review just why restraint systems were introduced in aircraft. The lap belt (or seat belt, whichever you prefer), has been the traditional basic restraint system in aircraft for many years. Its purpose is to keep the occupant closely attached to the seat during all the operational phases of the aircraft flight. It doesn't take much imagination to visualize the problems of trying to keep control of an aircraft flying through turbulence if the pilot

doesn't have his seat belt on ... and, for such obvious reasons, most pilots readily accept the need to wear the lap belt. Unfortunately, upper torso restraints (shoulder belts) are associated with a relatively rare occurrence in the operation of an aircraft, and an unpleasant one at that, the aircraft crash. Most pilots, being typical human beings, don't like to think of unpleasant things, particularly if

they are related to flying, and so don't think too much of torso restraint systems. Besides, we hear that they are hot, uncomfortable, a real mess to put on, interfere with the pleasure of flying, etc., etc. All these excuses should become insignificant to the rational pilot who understands that the proper

Let's look at the action of the lap belt during a crash (see Figure la). Typically, the crash forces tend to move the aircraft occupant down, forward and maybe to the side relative to the aircraft interior. If the crash deceleration is much above three or four Gs, even the most healthy occupant won't be able to resist these forces with his own strength. His body will pivot about the seat belt (this is called flailing) until it is stopped by hitting something inside the aircraft. Usually, the head and face violently hit the instrument panel, causing the head to rotate backwards on the neck as the torso goes forward until it too is stopped by impacting the control yoke. It's not the place of this article to describe the injuries which result. You can use your own imagination, or can consult any number of good reports on crash investigation (c.f. References 1 and 2). It should be obvious that some form of upper torso restraint would improve this situation. But before we discuss the upper torso restraint, let us consider some other problems which might occur with a lap belt in a crash, since lap belts are found in all aircraft restraint systems. Restraint system design looks simple, but it is actually fairly involved, with many factors begging consideration. Computer programs which consider

generally performs best if it acts at an angle of about 45° with the aircraft longitudinal axis. If the lap belt is installed so that it acts along a shallow angle, as in Figure Ib, it is likely to slip off the skeletal pelvis of the occupant, and apply loads to the abdomen. This is called "submarining". In a severe crash these loads will be high enough to cause injury to internal organs, a potentially life threatening situation. The flailing that results as the pelvis rotates under the belt and the upper torso rotates over the lap belt will also cause bending of the lower (lumbar) spinal column. The vertical crash loads will be transmitted through the bent spinal column, creating high stress in the front portion of the vertebrae which have been

brought together by the flailing action. The injury which may result, called an "anterior wedge fracture of the lumbar vertebra", may heal without serious impairment, but any injury to the spinal column runs the risk of involving damage to the spinal cord. Spinal cord injury is the major cause of paralysis resulting from crashes. If the lap belt is installed at too steep an angle, as in Figure Ic, it will be ineffective in resisting forward movement of the occupant. Such installations may result from rearward adjustment of the seat along its tracks by a long legged occupant (when the belt is anchored to the floor), or from an attempt to keep the belt out of the passageway or foot space of an aft seated occupant. Since the belt can carry only tension loads, it will allow the occupant to move forward until the belt is reoriented so that the tension load in the belt generates an adequate longitudinal component to

most of these factors have been developed to aid the restraint system designer so that compromises in design can be effective (References 3

resist further forward movement. Unfortunately, by this time the occupant may either have his knees in the instrument panel (knee or femur in-

and 4). But if the designer doesn't have access to these programs, he (or she) best follow certain "rules of

jury), or be so far forward in the seat that he slips off the front edge, allowing the belt angle to become shallow

thumb" which will yield a high probability of a good design. One of these rules has already been

with all the injury potentials discussed in the previous paragraph. This is a good place to point out two

illustrated in Figure la. The lap belt

other factors in restraint system deSPORT AVIATION 35

*The only alternative to an upper torso restraint would be total delethalization of the cockpit, a difficult task considering the space required

and the proximity of the instrument panel.

Figure 1 - Lap belt restraint. The action of lap belt restraint systems in a crash is illustrated by stick-figures representing the occupant. The "sticks" are centerlines of the body segments. Soft tissue is not shown, and cushion thickness is not

shown.

sign. The action which was just discussed involves elongation (stretching) of the belt webbing and the resistance of the seat pan to vertical loads. The restraint system designer should keep in mind that webbing is typically a very elastic material, elongating as much as 17% at its breaking load (usually 4000 Ibs. for nylon). Long lengths of webbing stretch more than short lengths (elongation is a percentage of length) and allow more movement of the occupant towards the interior of the aircraft. If the belt in the previous example was short, as it would be if it were anchored to the seat instead of the floor, the movement of the occupant wouldn't be nearly as much, and the risk of serious injury would be less. Also note

that the seat cushion is compressed by the belt pulling down as well as back on the occupant, even if there is no vertical crash force. A soft seat cushion may appear to be comfortable (actually, hard seats can also be comfortable, but comfort isn't the topic of this article), but a soft cushion will do little to resist the movement of the occupant in a crash. As the occupant

moves, it becomes more difficult for

the restraint to keep up with the occupant to limit further movement. Even with an optimum seat and restraint system which didn't have 36 JANUARY 1985

much compression or stretch, the body of the occupant will compress or stretch, so that a certain amount of movement will be inevitable. If the seat should break during the crash, as shown in Figure Id, it will allow the body to move down and change the relative orientation of the lap belt. This promotes the kind of injuries discussed for shallow seat belt angles. If there is something hard under the seat, such as a wing spar, when the seat breaks and if the body comes into violent contact with the hard structure, serious injury can result from this secondary impact. This injury is usually some form of vertebra fracture or pelvic fracture, or both, with the possibility of spinal cord involvement. This discussion should have made it evident that, even with a simple restraint system like a lap belt, many

problems can turn up if the designer doesn't consider the various factors influencing performance. The addition of upper torso restraint, while

necessary to reduce crash injury*, can add more problems if not done right. In the next section, consideration of

these problems for dual shoulder belt systems and single (diagonal) shoul-

der belts will be considered. These two types make up most of the aircraft

upper torso restraints available.

Upper Torso Restraints The usual design for dual shoulder belts results in a restraint system with all segments joined by the buckle in the center of the lap belt. This is simple and allows quick egress from the restraint by only releasing one buckle. (Quick egress is important if the occupant must escape from the aircraft after a crash because of fire, the aircraft sinking in water or some other environmental hazard.) However, if the lap belt is installed at a shallow angle, as shown in Figure 2a, the shoulder belts can pull the lap belt up, off of the pelvis, into the abdominal region. From the earlier discussion about shallow lap belt angles, the injuries which can result should be apparent. It may not be so apparent that, as the lap belt moves up, it introduces slack into the shoulder belts. As the torso moves forward into this slack the head is exposed to increased risk of injury from impact with the instrument panel or control yoke. One way to correct this problem is to change the installation angle of the lap belt, as shown in Figure 2b. Note that the lap belt length was kept at a minimum by moving the belt anchor points to the seat pan area. Here, the steep angle of the lap belt allows it to react to the force of the upward pull of the shoulder belts, and the short lap belt length limits stretching, so the movement of the occupant is better controlled. Another way to correct this problem is through the use of a "negative-G strap" as shown in Figure 2c. Negative-G straps used to be called crotch straps, but that nomenclature is both inaccurate and discouraging. The negative-G strap, when correctly installed, is attached at one end to the buckle, and at the other end to the front edge of the seat pan. Its length is such that there is no slack in the strap when the lap belt is properly positioned on the pelvis. In this position, it acts with the lap belt to react against the upward pull from the shoulder belts. The forward location of the lower attachment point of the strap is important to prevent genital injuries. This restraint concept can be very effective, and has been adopted for many military and commercial pilot crew seats. If none of these approaches is satisfactory, one alternative is to attach the lower ends of the shoulder belts in the same general area as the ends

of the lap belts, as shown in Figure 2d. With this installation, the lap belt does not carry any load from the shoulder belts, and is more likely to remain on the pelvis during a crash. However, releasing the buckle on the lap belt does not release the shoulder belts. If the upper ends of the shoulder belts are attached to the aircraft through an emergency locking retractor (commonly called an inertia reel, and used with the shoulder belts to allow movement of the occupant during normal flight, to increase comfort, and to stow the belts when not in use), and if that retractor automatically unlocks the belts after the crash, this may not present a serious egress problem. The shoulder belts can be slipped off just as if taking off a jacket. The emergency locking retractor will also solve the problem of shoulder belt length adjustment that exists with this type of installation. Without a retractor, this system can still be used, but with more difficulty (see Reference 5). In this case, both shoulder belts are furnished with manual adjusters which allow the restraint to be fitted to the occupant. Each lower end of the dual shoulder belt passes through a slot in the corresponding lap belt attachment hardware (end fitting), and is then stitched to the lap belt, near the buckle. During the crash, tension in the shoulder belt segments is reflected through the lap belt end fittings and then to the lap belt and buckle. When the buckle is opened for egress, the lap belt no longer holds tension on the lower ends of the shoulder belts. They can be pulled through the slots in the lap belt end fittings until the stitched area is reached. This allows slack in the shoulder belts so they can be slipped off the shoulders. While not as easy as in some of the other designs, egress can still be fairly rapid. The dual shoulder belts which have been discussed are generally thought to give better crash protection than single diagonal belts. The dual belts distribute the load over a larger area of the body so contact pressure is lower and their symmetric nature means that the body is held more symmetrically, with less tendency to twist. They also provide more consistent restraint for crashes involving lateral forces. However, it is often impossible to find suitable structure behind the occupant to attach the upper end of a dual belt system. Diagonal shoulder belts have been used in automobiles for a number of years, and the crash experience which has resulted indicates that they perform surprisingly well (see, for example, Reference 6 and 7). Most automobile restraint systems are the result of careful planning, extensive testing,

Figure 2 • Dual shoulder belts for upper torso retraint used with a lap belt for pelvic restraint.

and considerable experience. Without that background, the designer of a system for aircraft should proceed with caution. A good typical installation is shown in Figure 3a. The diagonal shoulder belt is positioned so that it passes over the midpoint of the shoulder, with the lower end fastened well to the side of the occupant's hip. The lower end can be attached to the lap belt, in which case the portion of the belt common to both the lap belt and shoulder belt must be strong enough to carry loads from both segments, or it can be attached separately as long as the geometric arrangement is maintained. Problems can result from use by occupants of greatly differing size or by incorrect location of the belt attachment points. For example, Figure 3b shows an installation where the shoulder belt bears against the neck or the side of the head, as may be the case if the occupant were short or if the upper attachment point were located too far inboard. This is not only aggravating, it can be dangerous. If the crash involves severe vertical loads as well as longitudinal loads, the body will move down into the seat and forward, and the belt can force the head violently to the side. This action may cause fracture of the vertebrae in the neck with involvement

of the spinal cord and irreversible injury. On the other hand, if the occupant is too tall or if the upper attachment point is too far outboard, the belt may tend to fall off the shoulder, as shown in Figure 3c. Again, this is aggravating and potentially injurious. If the belt falls below the center of mass of the upper body, the body will rotate around the shoulder belt. If the crash is severe, the resulting body motion may be enough to allow the head to contact the instrument panel, and the combined twisting, bending and compression of the spinal column may cause its injury. A similar result may occur if the shoulder belt is simply attached to a lap belt buckle located near the center of the body, as shown in Figure 3d. With this installation, the shoulder belt passes to the side of the center of mass of the upper torso, so that in a severe crash the torso may twist around the belt, and even slide out of the belt. The risk of head injury by contact with the instrument panel is increased, and the twisting of the spinal column will increase chances of injury to the spinal cord. It should be apparent that the selection of an anchorage point for the upper end of a dual or single shoulder belt restraint system is important. SPORT AVIATION 37

can be made. However, as a starting point, "typical" loads measured in a controlled laboratory test considered

Figure 3 • Single diagonal shoulder belt for upper torso restraint used with a lap belt for pelvic restraint.

This has been discussed in many publications, including an Advisory Circular issued by the FAA (Reference 8)*. One concern, in addition to those already presented, is to avoid compression of the spinal column by the shoulder belt. This may be encountered when the upper end of the shoulder belt is mounted low, as in Figure 4. In this case, the shoulder belt would pull down and back on the torso as it resists the forward motion of the occupant. The downward component of force will place the spinal column in compression, and will add to the stress on the column caused by the vertical component of the crash force. This can be avoided if the angle of the

page 50, and the October 1980 issue, page 18.) *This Circular also gives advice on other aspects of restraint installation, some of which may be subject to misinterpretation. In particular, don't count on seats being strong enough to carry restraint loads unless they were specifically designed to carry the loads, don't depend on local reinforcement at restraint attachment points to carry the restraint load into primary aircraft structure, and don't expect a 500 Ib. test load on the shoulder belt attachment point to represent the true loads in the restraint during a crash. ••:••;•. •••••