(4) The friction in limb joints should be set so that they barely restrain the weight of the limb when extended horizontally.
(5) The ATD should be placed in the center of the seat in as nearly a symmetrical position as possible. The ATD shou ld be placed in the seat in a uniform manner so as to obtain reproducible test results.
(6) The ATD's back should be against the seatback without clearance. This condition can be achieved, if the ATD's legs are lifted as it is lowered into the seat. Then the ATD is pushed back into the seatback as it is lowered the last few inches into the seat pan. Once all lifting devices have been removed from the ATD, the ATD should be “rocked” slightly to settle it in the seat.
(7) The ATD's knees should be separated about four inches.
(8) The ATD's hands should be placed on the top of its upper legs, just behind the knees. If tests on crew seats are conducted in a mockup that has airplane controls, the ATD's hands should be lightly tied to the controls. The ATD arms should be positioned such that they are not over the armrests, so as not to bear on the armrests and influence the collection of lumbar loads during the down test.
(9) The feet should be in the appropriate position for the type of seat tested (flat on the floor for a passenger seat, on control pedals or on a 45-degree footrest for flightcrew systems). The feet should be placed so that the centerlines of the lower legs are approximately parallel, unless the need for placing the feet on airplane controls dictates otherwise.
(10) If the system is tested in other than a “horizontal floor” position, it is recommended that the ATD be placed such that the hip joints are in essentially the same position relative to the seat as when seated with a 1g pre-load as shown in Figures 10-1 and 10-2. Achieving this position may require that the lap belt be very tight and a shim be placed behind the ATD’s back and pelvis.
(11) Auxiliary restraints may be required to ensure that each ATD will be in its proper position prior to the impact. The auxili ary restraint(s) must not interfere with the results of the test.
c. Seat Adjustment. Except as noted in paragraph 3b(4), to the extent that they influence the injury criteria, all seat adjustments and controls should be in the design position intended for the 50th percentile male occupant. If the seat restraint systems being tested are to be used in applications where special requirements dictate their position for landing or takeoff, those positions should be used in the tests.
d. Installation of Instrumentation. Professional practice should be followed when installing instrumentation. Care should be taken when installing the transducers to prevent deformation of the transducer body that could cause errors in data. Lead-wires should be routed to avoid entanglement with the ATD or test article, and sufficient slack 40
should be provided to allow motion of the ATD or test article without breaking the leadwires or disconnecting the transducer. Calibra tion procedures should consider the effect of long transducer lead-wires. Head accel erometers and femur load cells should be installed in the ATD, in accordance with the ATD specification and the instructions of the transducer manufacturer. The load cell between the pelvis and the lumbar column should be installed in accordance with the approach shown in Figure 7-1 of this AC or in a manner that will provide equivalent data (see paragraph 7b).
(1) If an upper torso restraint is used, the tension load should be measured in a segment of webbing between the ATD's shoulders and the first contact of the webbing with hard structure (the anchor point or a webbing guide). Restraint webbing should not be cut to insert a load cell in series with the webbing, since that will change the characteristics of the restraint system. Lo ad cells that can be placed over the webbing without cutting are commercially available. They should be placed on free webbing to minimize contact with hard structure, seat upholstery, or the ATD during the test. They should not be used on double-reeved webbing, multiple-layered webbing, locally stitched webbing, or folded webbing, unless it can be demonstrated that these conditions do not cause errors in the data. Th ese load cells should be calibrated using a length of webbing of the type used in the restraint system. If the placement of the load cell on the webbing causes the restraint system to sag, the weight of the load cell can be supported by light string or tape that will break away during the test.
(2) Since load cells are sensitive to the inertial forces of their own internal mass and to the mass of fixtures located between them and the test article as well as to forces applied by the test article, it may be necessary to compensate the test data for that inaccuracy if the error is significant. Da ta for such compensation will usually be obtained from an additional dynamic test that replicates the load cell installation but does not include the test item.
e. Restraint System Adjustment. The restraint system must not be tightened beyond the level that could reasonably be expected in use, and the emergency locking device (inertia reel) must not be locked prior to the impact. Automatic locking retractors must be allowed to perform the webbing retraction and automatic locking function without assistance. Care must be taken to assure that emergency locking retractors that are sensitive to acceleration do not lock prior to the impact test because of pre-impact acceleration applied by the test facility. If comfort zone retractors are used, they must be adjusted in accordance with instructions given to the user of the restraint system.
(1) If manual adjustment of the restraint system is required, slack must be removed, and the restraint system should be snug but not excessively tight, about the ATD. For test 2, this can normally be dete rmined when two fingers fit snugly between the belt and the pelvis of the ATD. The re straint system shall be checked and adjusted just prior to the floor deformation phase of the test.
(2) If the system is tested in other than a “horizontal floor” position (see paragraph b(10) above), the restraint should be properly adjusted with the seat in the 41
“horizontal floor” position and the webbing transducers installed (if required). After sufficient time has elapsed to allow the cushion to reach an equilibrium position, the webbing should be marked to indicate the correct adjustment point. The seat and ATD should then be installed on the fixture in the appropriate dynamic test orientation and the restraint system again adjusted to that same point.
(3) An alternative method to impose a 1-g pre-load is to measure the position of the ATD hip joints relative to the floor as shown in Figure 10-1 below. The ATD is then depressed into the cushion to reproduce this relative position after the ATD and seat have been installed on the fixture, as shown in Figure 10-2. The lap belt may be tightened to maintain this position. This load may make it impossible to insert two fingers between the lap belt and the pelvis of the ATD, but it should not produce a cushion displacement in excess of that measured by placing the ATD on the seat in a 1-g orientation.
f. Floor Requirement. A floor is not required for Test 2, but if a floor is installed, it should not influence the behavior of the seat or unduly restrict the movement of the ATD's feet. This is a concern especially when floor distortion is applied. For consistency, a floor should be used for tests used to gather head path data.
g. Seat Components. Seats frequently consist of components that are both integral to and ancillary to the seat’s basic design and function. Historically, anything that is attached to the seat has been treated as part of the seat from a type design standpoint.
However, these items can be divided into several categories with different methods of substantiation in order to facilitate simplified compliance methods. In summary, these categories are as follows:
• Features of the seat that affect its dynamic performance,
• Operationally removable items attached to the seat,
• Small items that can be verified by inspection, and
• Lifevests.
Appendix 5 describes in detail methods to address compliance for each type of item.
Figure 10-1 surement of 1g Pre-load Mea 43
Figure 10-2 est Orientation—1g Position T 44
11. DATA REQUIREMENTS. The data should include charts, listings, and/or tabulated results, and copies of any photo instrumentation used to support the results. The following should be recorded:
• Impact pulse shape
• Head Injury Criterion (HIC) results for all ATDs exposed to head impact with interior components of the airplane or head strike paths and velocities, if head impact is likely but could not be evaluated by these tests
• Total velocity change
• Upper torso restraint system load, if applicable
• Compressive load between the pelvis and the lumbar column
• Retention of upper torso restraint straps, if applicable
• Retention of pelvic restraint
• Femur loads, if applicable
• Seat attachment (including structural damage)
• Seat deformation
• Seat attachment reaction time histories (if load cells are used)
• Retention of items of mass
• Evaluation of seat egress
12. DATA ANALYSIS.
a. General. All data obtained in the dynamic tests should be reviewed for errors.
Baseline drift, ringing, and other common electronic instrumentation problems should be detected and corrected before the tests. Loss of data during the test is readily observed in a plot of the data versus time and is typically indicated by sharp discontinuities in the data, often exceeding the amplitude limits of the data collection system. If these occur early in the test in essential data channels, the data should be rejected and the test repeated. If they occur late in the test after the peak data in each channel has been recorded, the validity of the data should be carefully evaluated, and the maximum values of the data may still be acceptable for the tests described in this AC. The HIC does not represent simply a maximum data value, but an integration of data over a varying time base. The head acceleration measurements used for that computation are not acceptable, 45
if errors or loss of data are apparent in the data during the time the ATD head is in contact with the airplane interior features during the test.
b. Impact Pulse Shape. Data for evaluating the impact pulse shape is obtained from an accelerometer that measures the acceleration in the direction parallel to the inertial response shown in Figure 3-1 of this AC. The impact pulses intended for the tests discussed in this AC have an isosceles triangle shape. These ideal pulses are considered minimum test conditions. Since the actual acqui red test pulses will differ from the ideal, it is necessary to evaluate the acquired test pulses to insure the minimum requirements are satisfied. The five properties of the idea l pulse that must be satisfied by the acquired test pulse are as follows (see Figure 3-1):
Pulse shape: isosceles triangle Greq: peak deceleration required by test condition Treq: rise time required by test condition V: total velocity change required by test condition Vtr: velocity change required during T req (Vtr = V/2) A graphical technique can be used to evaluate pulse shapes that are not precise isosceles triangles. Appendix 1 of this AC presents the graphical met hod of evaluating the acquired pulse (the recorded test sled acceleration versus time). For the acquired pulse to be acceptable, the following five criteria must be met:
(1) The magnitude of the peak value for the acquired pulse, Gpk, must be greater than or equal to Greq.
(2) The actual rise time, Tr = T2 - T1, must be less than or equal to Treq.
(3) The result of integrating the acquired pulse during the interval from t = T1 to t = T3 must be equal to or greater than Vtr, one-half of the required velocity change for the specified test. If the magnitude of the acquired pulse is greater than the ideal pulse during the entire interval from T1 to T3, this requirement is automatically met.
(4) The result of integrating the acquired pulse during the interval from t = T1 to t = T1 + 2.3 (Treq) must equal or exceed the required test velocity change, V, of the test condition. If the acquired pu lse returns to zero g's at t = T4 < (T1 + 2.3 (Treq)), the end of the interval of integration is reduced to t = T4;
(5) If the magnitude of the acquired pulse is greater than the ideal pulse during the entire interval of t = T1 to T2 and the parameters of paragraphs (1) through (4) above are satisfied, then the acquired pulse is acceptable.
(6) If the magnitude of the acquired pulse is not greater than the ideal pulse during the entire interval t = T1 to 1.33( T3- T1), the difference between acquired pulse and the ideal must be no greater than 2.0 g's at those times when the acquired pulse is less 46
than the ideal. The parameters of paragraphs (1) through (4) above must also be satisfied for the acquired pulse to be acceptable.
c. Total Velocity Change. Impact velocity can be obtained by measurement of a time interval and a corresponding sled displacement that occurs just before or after (if appropriate) the test impact and then dividing the displacement by the time interval.
When making such a computation, the possible errors of the time and displacement measurements are used to calculate a possible velocity measurement error, and the test impact velocity should exceed the velocity shown in Figure 1 by at least the velocity measurement error. If the sled is not ch anging velocity during the immediate preimpact or postimpact interval, the impact velocity is the total velocity change. If the sled is changing velocity during the immediate pre-impact or post-impact interval or if the facility produces significant rebound of the sled, the total velocity change can be determined by integrating the plot of sled acceleration versus time, as described in Appendix 1. If this method is used, the sled acceleration is measured in accordance with Channel Class 60 requirements.
d. Head Injury Criterion (HIC).
(1) Data for determining the Head Injury Criterion (HIC) need to be collected during the tests discussed in this AC only if the ATD's head is exposed to an impact on airplane interior features (not including the floor or the ATD's own leg) during the test.
The HIC is calculated according to the following equation:
t2 1 2.5HIC =[(t2 - t1 ) [ ∫ a(t) dt ] ]max(t - t ) t2 1 1 Where t1 and t2 are any two points in time (in seconds) during the head impact, and a(t) is the resultant head acceleration (expressed in g's) during the head impact.
(2) The HIC is invariably calculated by computer based data analysis systems, and the discussion that follows outlines the basic method for computation. The HIC is based on data obtained from three mutually perpendicular accelerometers installed in the head of the ATD in accordance with the ATD specification. Data from these accelerometers are obtained using a data system conforming to Channel Class 1000, as described in SAE Recommended Practice J211. Only the data taken during head impact with the airplane interior need be considered; this is usually indicated in the data by a rapid change in the magnitude of the acceleration. Film of the test may show head impact that can be correlated with the acceleration data by using the time base common to both electronic and photographic instrumentation. Simple contact switches that do not significantly alter the surface profile could also be used to define the initial contact time.
(3) In many cases, a full system sled test to evaluate specific occupant injury conditions may not be needed to evaluate a redesign of the seat system that affects only HIC. In such cases, the photometric head path data can be gathered and used to ensure no contact will occur or to define the head angle and velocity at impact. It is acceptable to consider seat back rotation or deflections that occur under the dynamic load condition in an analysis to show that there is no head contact with the seat forward. Adding 3 inches to the 50thpercentile ATD head path is typically acceptable to account for the 95% occupant head path (see Appendix 4). These data can be used in a component test of severity comparable to the whole system sled test. Other factors, such as the inertial response of the impact target, must be accounted for in the component test condition so that the impact condition is representative. Component testing methods must be demonstrably comparable to whole system sled tests as a HIC measure, and the specific methodology used will require approval by the FAA.
Additionally, a seat may be designed for use in multiple locations where head contact against a range on unknown bulkhead targets is anticipated (for example, front row seats). For these seats, HI C may be measured using a representative impact target mounted in front of the seat at the installation setback or range of setbacks. This target will represent typical fixtures, such as galleys, partitions, lavatories, and closets, and its stiffness will be representative for those monuments. If contact occurs, the HIC must not exceed 1000.
e. Upper Torso Restraint System Load. The maximum load in the upper torso restraint system webbing can be obtained directly from a plot or listing of webbing load transducer output. If a three-axis load transducer fixed to the test fixture is used to obtain these data, the data from each axis is combined to provide the resultant vector magnitude.
If necessary, corrections are made for the internal mass of the transducer and the fixture weight it supports. This correction will usua lly be necessary only when the inertial mass or fixture weight is high or when the correction becomes critical to demonstrate that the measurements fall below the specified limits. Restraint load data may be reduced by the ratio of the actual peak pulse versus the required peak pulse up to a maximum of 10 percent of the measured load. For example, a restraint system load of 1800 lbs that had a peak pulse of 16.9g, could be reduced to 16/16.9 of 1800, or 1704 lbs.
f. Compressive Load Between the Pelvis and Lumbar Column. The maximum compressive load between the pelvis and the lumbar column of the ATD can be obtained directly from a plot or listing of the output of the load transducer at that location. Since most load cells will indicate tension as well as compression, care should be taken that the polarity of the data has been correctly identified. As with restraint system loads, lumbar load data may be reduced by the ratio of the actual peak pulse versus the required peak pulse up to a maximum of 10% of the measured load. For example, a lumbar load of 1590 lbs for a Test 1 (14g down test) that had a peak pulse of 15g could be reduced to 14/15 of 1590, or 1484lbs.
g. Retention of Upper Torso Restraint Straps. Retention of the upper torso restraint straps on the ATD's shoulders can be verified by observation of photometric or documentary camera coverage. The straps must remain on the ATD's shoulder until the ATD rebounds after the test impact and the upper torso restraint straps are no longer carrying any load. The straps must not bear on the neck or side of the head and must not slip to the upper rounded portion of the upper arm during that time period.
h. Retention of Pelvic Restraint. Retention of the pelv ic restraint on the ATD's pelvis can be verified by observation of photometric or documentary camera coverage.
The pelvic restraint must remain on the ATD's pelvis, bearing on or below each prominence representing the anterior superior iliac spine, including until after the ATD rebounds after the test impact. Provided that the pelvic restraint remains on the ATD's pelvis, trapping of the belt between the leg and the pelvis is acceptable.
Movement of the pelvic restraint above the prominence is usually indicated by an abrupt displacement of the belt onto the ATD's soft abdominal insert which can be seen by careful observation of photo data from a camera located to provide a close view of the belt as it passes over the ATD's pelvis. This movement of the belt is sometimes indicated in measurements of pelvic restraint load (if such measurement are made) by a transient decrease or plateau in the belt force as the belt slips over the prominence, followed by a gradual increase in belt force as the abdominal insert is loaded by the belt.
i. Femur Load. Data for measuring femur load s can be collected in the tests discussed in this AC, if the ATD's legs contact seats or other structure. Data need not be recorded in each individual test, if rational comparative analysis is available for showing compliance. For large clearance installations (distance from seat SRP to strike target is more than 40 inches nominally), no data is necessary to substantiate femur loads.
13. PASS/FAIL CRITERIA. The dynamic impact tests must demonstrate that:
a. The seat system remains attached to the test fixture at all points of attachment, the occupant restraint system remains attached at all points of attachment, and the primary load path remains intact.
(1) For the purpose of showing compliance with the structural requirements of § 25.562, acceptable damage to the load-carrying structural elements includes bending, deformation, tension deformation, compression crippling, and shear buckling. Cracking of structural elements and the shearing or separation of rivets and minor delamination of composite panels is allowed, provided a continuous load path remains between the occupant and the seat attachments.
(2) Damage to seat belts, such as scuffing, fraying and breakage of fibers is considered acceptable. The seat belt should not be cut or torn by features of the seat or the belt adjuster mechanism. Cuts or tears should be investigated as to their cause and appropriate corrective action taken, although a retest may not be necessary.
b. If the ATD's head is exposed to impact with interior features during the test, a HIC of 1,000 is not exceeded.
c. Where upper torso restraint straps are used, tension loads in individual straps do not exceed 1,750 lbs(7.78 kN). If dual straps are used for restraining the upper torso, the total strap tension load does not exceed 2,000 lbs(8.90 kN) .
d. The maximum compressive load measured between the pelvis and the lumbar column of the ATD does not exceed 1,500 lbs (6.67 kN).
e. Where installed, the upper torso restraint straps remain on the ATD's shoulder during impact.
f. The pelvic restraint remains on the ATD's pelvis during impact.
g. Where leg contact with seats or other structure occurs, the axial compressive load in each femur does not exceed 2,250 lbs (10.0 kN).
h. The seat permanent deformations are within the quantitative limits of Appendix 2 of this AC and will not significantly impede an occupant from releasing his restraints, standing, and exiting the seat. In no cas e should deformation of the seat cause entrapment of the occupant, whether or not the defined limits referenced in Appendix 2 are exceeded.
NOTE: With the exception of seatbacks, it is assumed that the maximum seat structural deformation will result from the structural evaluation (that is, single row Type 1 or 2 test). Once this is accomplished, it would not, therefore, be considered necessary to repeat deformation measurements after the injury criteria (multiple row) tests, unless the structural and injury criteria tests were combined into one test. Maximum deformation to the seatback usually occurs as a result of impact by the occupant to the rear of the seat.
i. All deployable items must remain stowed, unless it can be shown that they do not impede egress or cause serious injury (see Appendices 2 and 5).
14. TEST FAILURES VS. RETEST.
a. As noted in paragraph 13, a variety of failures can result in an unsuccessful test.
Failures can range from structural separation of the seat from the tracks to deployment of items that impede egress. All such failure s should be addressed and corrective action taken. However, the necessity of repeating tests following corrective action is the same decision process as that used to determine which tests are conducted initially.
b. Failures in any part of the primary load path, including the seat attachment to the track or restraint system attachment to the seat, will almost certainly require a retest.
Failures in (secondary) internal structure may be able to be addressed analytically. For example, failures in members for which analytical substantiation is acceptable when making the test article selection (using the procedures outlined elsewhere in this AC), may not require a retest. However, each case should be assessed individually and a determination made that the failure point would not simply be transferred to another part of the load path. In general, members for wh ich the failure mode is not catastrophic (for example, compressive failures in a forward leg as opposed to a tension failure in an aft leg for a 16g forward test) are less likely to warrant retest. The extent to which a secondary load path(s) can carry the load is a factor in determining the pass/fail of a structural test.
c. Special attention to the seat structure prior to the removal of floor warpage is advised. Structural failure can occur as a cons equence of removal of floor warpage. If it can be determined that the damage or seat deformation occurred solely as a result of removing the floor warpage, it is not considered a failure.
d. Similarly, the evaluation of the seat attachment should be made before the seat tracks are straightened (unwarped). The pr ocess of straightening the seat tracks may result in a seat attachment becoming detached. This is not a test failure. The assessment for seat attachment should be made after the restraining force on the pitch-and-roll fixture has been released. It is not necessary to retu rn the floor to a flat condition to evaluate the seat attachment. Once the evaluation for seat attachment has been completed, the floor may be returned to a flat state in order to take deformation measurements (if applicable).
e. Cuts or tears in a restraint system may not require a retest, if it can be demonstrated that the corrective action will be effective and if all other pass/fail criteria were met on the test in question.
f. Failures of attachments of items on the seat may be addressed analytically, provided that the corrective action does not impact the primary load path of the seat/occupant system or occupant injury criteria. However, the seat must be shown to be able to carry its full weight, including any attached items. Similarly, items that deploy should not require retests, if the corrective action does not affect the dynamic behavior of the seat or occupant.
g. If a test exceeds the minimum test conditions and results in a failure, an assessment of the test conditions and the failure mode must be made, and a rational basis for retest without a design change must be presented to allow a retest without modification.