Hpr 14 OS 10:07a

ESTABLISmNG A

SPINAL INJURY CRITERION FOR MILITARY SEATS

Martin Rapaport, Estrella Forster and Ann Schoenbeck

Naval Air Warfare Center

Aircraft Division

Patuxent River, MD

Leon Domzalski

ARCCA, Inc.

ABSTRACT

Peons Park, PA

INTRODUCTION

Currently, military specifications require use of anacceleration-time criterion to assess the physiologicalacceptabilityof crash resistant seats and restraint systems.With the evolution of crash test dummies which now

exhibit greaterbio-fidelic performance, military researchersare considering "direct force" measurements taken withinthe Hybrid III anthropomorphic test device (ATD) as analternate evaluation criterion. The Federal AviationAdministration(FAA) recently established such a criterionas part of a new dynamic test requirement for thecertification of airline passenger seats.I Prior to militarytri-service implementation of a similar requirement,validation of the methodolgy with the "aerospace" modelof the Hybrid III manikin was necessary. This paperdescribes the test program cOl1ductedwith Hybrid III crashtest dummies to establish a lumbar spinal load injurycriterion for military crash resistant seat compliancetesting. The effort was sponsored under the Naval AirSystems Command's (PMA-202) Advanced CrashworthyAircrew Survival Systems (ACASS) Program.

Dynamic testing was. conducted on the HorizontalAccelerator Facility at the Naval Air Warfare Center,Aircraft Division, Warminster, PA. The test programexamined the lumbar spinal response of Hybrid III testdummies when ex.posed to predominantly verticalhelicopter crash pulses under controJled laboratoryconditions. Sufficient data was acquired to conduct astatistical analysis of the Hybrid Ill's lumbar spineresponse to compressive loading (Le., + Gz ~eadwarddirecLion). The analysis supports the recommendation toemploy lumbar force as a primary physiological criterionfor military crash resistant seat compliance. Also, themerits of using this injury indicator during escape systemtestingarediscussed. .

Military specifications do not state perfonnaneerequirements in tenns of specific injury mechanisms orquantifiable injury threshold lcvc:ls.Typically, perfonnancerequirements for crash resistant seats and restraints are

"indirectly" validated against potential injury dUriri~qualification testing. For example, MIL-STD-1290stipulates an impact velocity change parameter for theaircraft's crash survivability requirement which is based onmishap data from operational aircraft accidents and crashtesting of specific aircraft models. Crash resistant systemsproposed by industry must meet structural strengthrequirements, and an occupant survivability criterion basedupon the Eiband Curv~3 when tested to the standard'sgeneric crash pulse. The Eiband Curve criterion essentiallydefines a maximum acceleration-time profile for Lheheadward (+ Gz) direction to which the scat/occupantresponse must comply (Figure t). However, since injurymechanisms are intrinsically related Lothe applied forcesustained by the body member, military crash safetyresearchers are now investigating the feasibility ofemploying a "direct" parameter. In the case of vertebralfractures, a "spinal force" parameter is proposed. Althoughthis investigation focused on lumbar forces to establish analtemative criterion to the Eiband Curve, the process canbe applied to other injury mechanisms during thedevelopment/qualificationphase of crash resistant systems.As part of the ACASS program, test methodolgy changesare under consideration for various injury mechanisms forthe head, brain, cervical neck, face, upper torso/thorax,lower limbs, and lumbar spine. Reference 4 reviewsinjury criteria which provide the basis for advancedmanikin use in the development of automotive restrainttechnology.

156

p.2

Report Documentation Page Form ApprovedOMB No. 0704-0188

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering andmaintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, ArlingtonVA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if itdoes not display a currently valid OMB control number.

1. REPORT DATE 1997 2. REPORT TYPE

3. DATES COVERED 00-00-1997 to 00-00-1997

4. TITLE AND SUBTITLE Establishing a Spinal Injury Criterion for Military Seats

5a. CONTRACT NUMBER

5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) 5d. PROJECT NUMBER

5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) Naval Air Warfare Center,Aircraft Division,Patuxent River,MD,20670

8. PERFORMING ORGANIZATIONREPORT NUMBER

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)

11. SPONSOR/MONITOR’S REPORT NUMBER(S)

12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited

13. SUPPLEMENTARY NOTES

14. ABSTRACT

15. SUBJECT TERMS

16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT

18. NUMBEROF PAGES

9

19a. NAME OFRESPONSIBLE PERSON

a. REPORT unclassified

b. ABSTRACT unclassified

c. THIS PAGE unclassified

Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18

Flf.'r.:'... 14 06 10:07a

~,

~1I

'.'j

.

..

p.3

thespineof anAID couldbemeasuredwithoutadverselyaffecting the flexural characteristics of the lumbar spine.

The FAA's newly adopted seat dynamic perfonnancestandard which incorporates a lumbar spinal load injurycriterion is described in Reference I I. A maximwlIcompressive load of 1,500 pounds between the pelvis andthe lumbar spine of the Part 5728, 50th percentile maleATO has been established for the crash environment inwhich the predominant impact load component is directedalong the spinal column of the occupant. Figure 2 depictsthe installation of the pelvic lumbar spine load cell in a Part572B anthropomorphic dummy associated with the FAA'sstandard.

I,!I!,

III

,----'------." r~I ~ II . 1" ,'~IIIt~~~----

Figure 2: Part 572 Pelvic Segment with Lumbar Spineand Load Cell Assembly

A key element in the application of injury threshold levelsto the evaluation of a system's cra.<;himpact perfonnance isthe availability of a reliable crash test dummy capable ofproviding reproducible results in controlled testing. TheHybrid III is a crash test dummy developed by GeneralMotors Corporation in the 1970s to improve the biofidelityand injury prediction capability of test surrogates.Parameters related to impact injury, such as headacceleration, neck forces/moments, thorax accelerations,femur forces, and lumbar forces/moments can be measuredin currently available Hybrid III ATOs. Reference 12describes biofidelic enhancements to the Hybrid III designwhich support its use in predicting human injury duringsimulated automobile crash tests. Although a directcorrelation between the Hybrid Ill's biomechanicalresponse and the human has lIot been fully achieved, theHybrid III can serve as a viable test device to allow a

comparative evaluation of the injury mitigation capabilitiesof proposed military crash resistant systems. Modificationsto Hybrid III ATOs for use in the military tcst environmentare described in Reference 13.

157

I III... "...U II;;:~II... II..

::21

~. {,iiiil.I..I",: II ."""1"'''';;: 1 < (lIiljm'.lIldUIII'11i

~ I IUUItl

". 0hi":!! "'''"11

:; ~ ::~ h"hllli..~ ! .1 1I"1:::~':'III""

III. h., 111,"1 i..11 ClIII II II II IIIIMI

1.111 .111 .114.111.11 .11 .11.11 .1 .II J .1.1.11 I IIII

DVlAno. Of 'K~DI. ,mUtATION -SEe

Figure I: EIBAND Spinal Injury Tolerance Curve

Backeround

Minimal crash impact testing with volunteer subjects hasbeen conducted since Colonel StappSran his rocket sledtests in the 1950s and the Naval Air Medical ResearchLaboratory (NAMRL)6conducted head/neck accelerationtests with militaryvolunteersin the 1970s. Since then,military and automotiveresearchers have concentrated theirefforts on compiling biomedical data from cadaver testsand documented human accidents to derive injuryassessmentcriteria. The Society of Automotive Engineers(SAE) has fonnally published several documents7.R.9whichsummarize much of the biomedical injury data currentlyavailable. A prime example of an application of this data isthe National Highway Transportation SafetyAdministration's (NHTSA) Head Injury Criterion (HIC),currently used as a "pass/fail" marker for automotiverestraint systems during barrier impact compliance testingof new cars under Federal Motor Vehicle Safety Siandard208.

Reference 10 describes an initial attempt to modify Part572 and VIP.95 test dummies with lumbar load cells toestablish a methodology for measuring spinal force in crashtests. Data from the rigid seat tests did not exhibit anydifferences in accelerations or forces measured before orafter installationof the spinal load cells. However, a trendwas observed in the data from the energy absorbing (EA)seat tests. Modification of the manikins resulted in chestx-componentssignificantly higher during the tests with EAseats, indicating a possible alteration of torso flexuralcharacteristics. Also, x-components of the head indicatedthat some alteration of neck response characteristics mayhave been produced during the modification process. Theauthors, however, concluded that forces and moments in

Af'r 14 06 10:08a

Test Obiectives

Note that a direct correlation between the Hybrid Ill'simpact performance and human injury tolerance was notconsidered within the scope of this program. Test programobjectives were intended to support the establishment of anew Navy injury criterion based on direct measurements oflumbar spinal forces/moments taken during dynamic crashtests of candidate crash resistant systems. Specifically, thefollowing test objectives were established:

(a) To quantify the effects of various load paths withinthe Hybrid Ill's lumbar spine construction.

(b) To determine the effects of the following variables:

- Hybrid III size/weight;

- Upper torso stiffness;

- Crash pulse parameters such as:

Peak accelemtion;

Pulse velocity-change.

The influence of these variables on the response of theHybrid III's lumbar spine to typical crash impulses wasstudied by an analysis of variance (ANOV A) statisticalmethod.

EXPERIMENTAL PROCEDURE

Test Facility

The dynamic tests were conducted at the Naval AirWarfare Center Aircraft Division Horizontal AcceleratorFacility previously located in Warrninster, PA. TheHorizontalAccelerator simulates typical decelerative crashforces associated with vehicle mishaps by reversing theorientation of the test article and accelerating the systemfrom an initial velocity of zero. It consists of three mainassemblies: (1) accelerating mechanism, (2) test sled, and(3) a set of guide rails, 100 feet long. The acceleratingmechanism is a 12-inch HYGE actuator which generates amaximumforce of225,OOOpounds of gross thrust.

Test Articles

Test Seat I Restraint System

The test article consisted of a rigid seat structure

configured with a standard Mil-S-58095 five-point restraintsystem. The Mil-S-58095 restraint's inertia reel wasreplaced with an adjustable fixed anchor fitting to precludethe possibility of inertia reel failures during this study.Since the restraint system would experience minimal crashloads under the predominant vertical impact vector duringthis test series, replacement of test restraints was madesparingly.

The generic seat system (non-energy absorbing) Wasmounted to the sled platfonn in an orientation to simulate

p~4

the vertical impact mode. The test seat was rotated 90

degrees to the vertical axis to align the manikin's spineparallel to the input crash pulse. Standard operational seatcushions were not used to preclude any adverseanlplification factor in the controlled testing.

Figure 3 shows a post-test side view of the 95th percentileATD seated in the test seat as located within the vertica]orientation test fixture.

Figure 3: PosHest Side View of 95th Percentile ATDShowing Vertical-OrientationTest Fixture:Run No. 95030

Anthropomorphic Test Devices (ATO)

Hybrid 1lI male ATDs (5th, 50th and 95th percentiles)instrumented with lumbar load cells were seated within thetest seat. The Hybrid III lumbar spine is a polyacrylatcelastomer member with molded-in end plates forattachment between the thoracic spine and pelvis. Acurved lumbar spine is incorporated in the automotiveconfiguration to replicate the typical automotive seateddriving position. However, the "erect" (i.e.. straight)lumbar spine and the articulated hip assembly of the"aerospace model" were selected for this test series topermit axial loading of the spinal column. In addition tothese design modifications, structural enhancements to theshoulder-clavicular assembly have been incorporated intothe aerospace configurations.

Prior to the test phase of this program, each of the ATDswere disassembled to identify alternate load paths presentwithin the upper torso. The abdominal insert assembliespresented a secondary load path for reacting a portion ofupper torso mass during compressive loading of the spinalcolumn. Each oCthe ATD's upper torso components werescale-weighed. In addition, the test dummies wereelectronically weighed "with and without" the abdominalinserts positioned to determine the 1g effect of the alternate

load path. Table II lists the component weights, totalweights, and delta measurements attributed to the insertsfor each of the test dummies.

158

..

14 OS 10:09a

TEST CONDITIONS

Thecrashpulse:parameters shuwn below (Table I) definethe impact profiles which were employed during the testprogram. The pulse shape selected was essentially a "half-sine" wavefonn, with peak acceleration and velocity-change the critical parameters defining the severity of thecrash pulse:. Pulse sevt:rity was based on the premise thatthe analysis would be conducted on a "fixed" scat structureand, therefore, would not provide any energyabsorption/load-limiting effect to the seated occupant.

TABLE I: SUMMARY OF IMPACT PARAMETERS

Figure 4 shows sample overlay plots of the test facility'scrash pulse waveforms selected foi this test program.Representative. pulse signatures for the three peakaccelerations at the 25 fps velocity change level are given.

S'oI Q<o-,I.to

JU

,-,.',-1",,1..-l"i"'.'--J""l.,

~ ,..J

"...,ell.'

u

-lU

1m

Figure4: Sled Impulse Showing 25 fps Velocity Changeof 10,20, 30 Oz Peak Accelerations

During Phase I, the "20 G" peak Gx Sled (i.e., + Gzrelative to ATO) parameter was maintained at a mean of20.49 Gs with a standard deviation of 0.24. Velocitieswere also closely controlled: for example, the "20 fps"

10.5

parameter had a 95 percent confidence interval between20.07 to 20.66 fps.

TEST RESULTS

Discussion of Test Variables

The initial test series studied the effects of velocity, ATDsize, torso restraint, and the abdominal insert load path onthe Hybrid HI's lumbar spine response to compressiveloading. Test conditions were established to. study theinfluence on lumbar spinal loading by the folJowingindependent teSt variables:

- Velocity change of the crash pulse;- ATD percentile sizes;- ATD head! torso;-ATO abdominal insert;

(3 ea)(3 ea)

("restrained" or "free")Cl'resent" or "absent")

- Peak acceleration ofthe crash pulse; (3 ea)(incorporated during Phase II)

One repetition was conducted for eaeh test condition.Initially, peak acceler.ltion effects were not considered; alJtrials during the initial series were conducted at the same20 0 peak amplitude. Subsequenl to the initial assessment,a supplementaltest series (Phase II) was conducted toinclude theeffectsof peak acceleration,addingtrialsat 10and 30 Os. This was possible once the effects of torsorestraint and the abdominal insert load path were quantifiedas minimal during the initial series at the 20 G peakacceleration level. Phase II consisted of an additional 36tests, keeping the upper torso restrained and each of theabdominal inserts present.

Influence of Abdominal Inserts

Figure 5 shows the Hybrid III ATD with the abdominalinsert removed for the 95th percentile ATD.

Figure 5: Pre-test Side View of 95th Percentile ATD withAbdominal Insert Removed: Run No. 95969

159

-----------

Vector Peak Acceleration (G) Velocity Change (fps)Gz 10 20Gz 10 25Gz 10 30Gz 20 20Gz 20 25

f----o--.:20 30Gz

Gz 30 20Gz 30 25Gz 30 30

,AjOru 14 OS 10: 10a

A preliminary assessment of lumbar load test data

measuredduringearlierNavytest programshad indicatedthat the Hybrid Ill's lumbar spinal load cell did notmeasurethe total fQrcetheoretically generated by the uppertorso mass during crash impact testing. The influence ofthe abdominal insert was investigated relative to thisapparent anomaly.

Table II shows the delta readings between the normal scalemeasurements and a static 1 g measurement of the uppertorso as measuredby the lumbar load cell with and withoutthe abdominal inserts positioned. The influence of theabdominal insert is evident from the I g rotationweighings resulting in weight deltas ranging from aminimumof 6.8 pounds for the 5th percentile manikin to amaximim of 26.9 pounds for the 95th percentile manikin.The higher mass readings recorded with the abdominalinserts removed support the assertion that the insertsproduce an alternate load path to compressive/axial spinalloading.

Also of significance was the finding that the upper torsostructure of the 95th percentile ATO did not exhibit aproportional increase of mass as compared to thedifferential between the 5th versus 50th percentiles.Subsequent load cell measurements of the 95th-percentileATD with the insert installed reduced the "effective" uppertorso mass to 69.3 pounds.

This result coupled with what appears to be nonlinearscaling of the 95th percentile's upper torso mass presenteda set of test dummies which skewed the test results. Itshould be noted that at this time, essentially noconfiguration control of the aerospace model Hybrid 1IIATD exists. Caution should be exercised when comparingtest data from various facilities, or even within the sametest organization, using this particular configuration of thebasic Hybrid III design.

TABLE II: Hybrid III Aerospace ATD Weights

Upper Torso Mass consisted offollowinl'. components:- Head/neck assembly

-Upper torso assembly minus foreanns. Instrumentation (tri-ax accelerometers in head & thorax; loadcells in lower neck: & lumbar spine)

. Skin & abdominal inserts

ObservatiQos

Figure 6 displays the characteristic response of the HybridIll's lumbar spinal column when subjected to a typicalhalf-sine crash pulse as produced by this test facility. Thephase lags associated between the thorax and pelvisaccelerations typically recorded during this test series arealso evident.

... ~; -- -II III 'ZlIfI<;l5,"01_' "u: I..: "r.1I1716 -3l.mr., .....51.'

.v 15.' ~.,.,.....

1.~v.

po,E:;

~

.\"

..i

-1Ncc14'.-I...'.,, z......

] "'.1'"m 4 u.....

""-'''''''-~'--'''''-'-

....~....~~...!.~ ....~.~~.....

1

:;'/I .I ,

,"..."\...A "j''''''

.............-......oZ.' ~-_.

..

Figure 6: Typical Vertical Response Parameters of 50thPercentile Hybrid III ATO at 20 G Peak;25 fpsVelocity Change

During the Phase II test series, analysis of the 50thpercentile ATD response produced a mean dynamic"effective" upper torso mass of 71.8 pounds (standarddeviation = 6.4) for the 12 runs with the 50th percentileATO (torso "restrained" and abdomen "present") ascompared to tbe static scale weight value of 90.5 poundsand 84.3 pounds witb the abdominal insert positioned.Evidently, the insert produced a dynamic delta of 18.7pounds compared to the static delta value of 12.0 pounds.This infers that the abdominal inserts produce an unloadingeffect on tbe Hybrid Ill's spinal column duringcompressive loading. Consideration should be given tothis effect whenever measured lumbar loads within theHybrid III are comparedto ultimate vertebral failure levels.

160

ATD Total Upper Upper Upper Mean Abd.Size Weight Torso . Torso Torso Delta Insert

Weight Weight Weight Weight Weight("/0) (Ibs) (Ibs) ID1h Abd (lbs) (Ibs)

Insert Abd.(Scale) (Scale) (Ibs) Insert (Load

(Load (lbs) Cell)Cell) (Load

Cell)5th 160 78.6 72.5 79.3 6.8 1.250th 179 90.S 84.3 96.0 12.0 1.495th 209 97.5 69.3 96.2 26.9 5.6

Ap~_!~_~! lO:lla

TABLE III: Test Results; Means (Phases I & II )

DeltaPki 5th 5th 5th5Qth:::~Qd,,~Oth 95th 95th 95th

&;~)~:dP~;. ~~r. L~~.;~~;. .~~'¥~f' P~~.T~~r.L~:..(O's) eG's) (G's) (Jbs)q~~) t(:t~) (II's): (g's) (G's) (lbs)

16.9 1056 :<n~:z:z.1;7<J$n 16.4 20.3 141533.3 2314:J9:S1:2.?+Mt 28.9 48.6 248745.3 25305(f652:&>~J$5 44.4 42.7 3127

L1123J8AJ\M:I$:H 16.5 20.9 1373266447:V A~;?:::~2~O33.2 54.0 2100340~6~,s. :71.5::4~~ft 50.9 55.6 4297

Test Conditions'

- Upper Torso Restrained-Abdominal Insert Positioned

-Table represents 54 Trials (18 [phase I] + 36 [phase II]);"Combined Effects" of Delta V, Gpk & ATD Size

illM-l

A General Linear Model (GLM) analysis addressed theeffect of manikin size and velocity change on lumbar loadwhere the head and torso were restrained and the abdomenwas in place. Lumbar load was affe.ctedby manikin size(F'" 104.64, p= .000) and velocity change (F'" 48.99, p=.000), where there was an interaction (F= I I, p= .001)between these variables. Figure 7 represents thisinteraction. The nonlinear effect of manikin size on thelumbar load variable was somewhat unexpected andprompted the mass survey of the test manikins' uppertorsos shownin Table II.

2000 -t--5%

I50%

manikin

--;95%

Figure 7. Effect of Manikin Size and Sled Velocity Change

p.?

QLM.2 .

A General Linear Model analysis addressed the effect ofmanikin size and abdomen placement on lumbar loadwhere the head and torso were restrained and the velocitywas maintained at 25 fps. Lumbar load was affected bymanikin size (F= 55.29, p= .000) and abdomen placement(F= 7.65, p= .024); but there was no interaction betweenthese variables. The influence of the abdominal insertwhen present, shows an unloading effect on lumbar force(Fz) levels. Lumbar force measurements (Fz) for the threetest manikins selected for this test series again indicates anonlinear dynamic characteristic. The results appearconsistentwith the measured weights of the upper torsos ofthe three test specimens as shown in Table II.

Figures 8 and 9 represent these results.

3500I3000

...I...I

2500

2000 .

N abdomeni

Y

Figure 8: Effect of Abdomen Insert Placement

2000

5%I

50%manikin

I

95%

.....-

figure 9: Em~ct of Manikin Size

Lumbar Force Tolerance Criteria

Figure 10 provides a lumbar force tolerance curvenonnalized for occupant total weight. This curve wasdeveloped for the Automatic Encrgy Absorber (ABA)

161

3500I

,r# . '....,

3000l ,t' ., VeloCity

# "'/ ".' 30:I j ,'/

.25- .

2500 L / ----- ------ 20

3500

3000

...I...I

2500

ApI" 14 06 10: 11a

Program14 currently investigating the possiblity ofproducing a load-limiter which optimizes perfonnancerelative to lumbar load response as opposed to the existingcriterion of seat/pelvic acceleration (Le., Eiband Curve).

Lumbar .Tal...nc. CU.ve

un

2800 "'''''''''''''''''''''''''' . ............ . , ..' ... ... .

h..f ; ...j..' i""+"""'i'''''''~' "hh~' .

-; 2100..,~ "00 .

~ 1200;;.. ICOO

j ; ,. ..,......

... . "'"'''''''''', . .i . '

)""'(""'r"

"'''('''i

, . ." ..... . ,

. '''0....:: litO

1400 L.. . . .. . . .

. :"'''::1::':::::1::::'.1+, i ,-1--'-+

100 11& In 146 11.

. . .

:...":::::::::;. . .. . ., . .. i . i j-, i ' i '

111 110 20f no UI 150

1200

1000

oeoup.nt .daM (lbol

Figure 10: Proposed Lumbar Load Tolerance Curve

The parameters of the sled's impact pulse (Table I) wereselected for comparability to known load-limiting (energyabsorbing) seat design responses.IS This approach wasfollowed because it was not practical to employ a load-limiting device for each test condition. As a consequenceof this limitation, lumbar forces measured within theHybrid Ills actually reached levels considered beyondvertebral fracture levels at the more severe impact levelsstudied. The related pelvic and thoracic accelerationsmeasuredat the more severe pulses are consistent with the"high" lumbar force levels.

Tolerance levels for the three manikins employed duringthis test series were 1940 pounds (5th AID = 160 pounds),2170 pounds (50th ATD = 180 pounds), and 2500 pounds(95th ATD ""210 pounds), respectively. Clearly, even thethe 5thpercentilemale AID would not have compliedwiththe tolerance criterion under the higher impulse severitylevels (i.e., 20 and 30 Gpk). Similarly, the 50th percentileATO would not have met the criterion at the upper impactseverity levels without some means of load limiting. Theabove results are explained by the selection of test impactparameters and are essentially in agreement with theperformance of standard non-energy absorbing seatdesigns.However,the response of the 95th percentileATDpresented misleading results, indicating compliance at theless severe impact severities. As previously indicated, thisis directly related to the nonproportional mass of the 95thaerospaceATO's upper torso.

The rationaleto support use of a lumbar force criterion forejection seats is based on the similarity of lumbar spinalinjury mechanisms associated with the two acceleration-time profiles. The inertial loading vector to the aviator'sspine is essentially similar whether ejected under

accelerativeforces or decelerated to zero'velocity in avertical crash scenario. Historically, crash resistant seatshave been designed to essentially the same injury tolerancecriteria applied to ejection seats (i.e.. Eiband Curve andDRI). Since lumbar injuries are directly related to theultimate strength of spinal vertebrae, a lumbar forcecriterion appears valid for both seat applications.

CONCLUSIONS

The analysis performed on the test data supports thefollowing conclusions:

1. The Hybrid III anthropometric test device(ATD) in the 5th and 50th percentile sizes can be used as atest surrogate during "vertically-oriented" (i.e., alongspinal axis of occupant) impact tests. This conclusion isrelated to the "straight" spinal segment of the Hybrid IIIATD aerospace configuration.

2. The 95th-percentile ATD selected for thisinvestigation exhibited a nonproportional upper torsoweight distribution which resulted in lumbar load responselevels below those of the 50th-percentile manikin. The 5thpercenlile male ATD's response appeared commensuratelybelow that of the 50th percentile male ATD.

3. Lumbar force provides a suitable spinal injurymeasurement within the Hybrid III ATD for comparativeanalysis of manikin response during evaluation of militarycrash resistant seat systems and ejection seat systems.Lumbar force can serve as an additional injury parameter inthe assessment of spinal injury prediction, sUPPQrtingthecurrent indicators, such as the Eiband Curve and theDynamic Response Index (ORI). Assessment of lumbarforce data in this test series would have indicated thepotential for spinal injury had the test seat employed beenunder evaluation. However, as noted this result is related'to the expedient use of a nonstroking seat mechanism in asevere crash impact environment. Current crash resistantseat designs employ load-limiting mechanisms to managethe seat occupant's response to vertical impacts at theseverity levelsemployed.

4. The abdominal inserts provide an alternate load

path to the spinal column of the Hybrid III ATD and canattenuate the level of lumbar force measured by the lumbarload cell. The etTect of the abdominal insert wa.~shown to

be statistically significant, but no attempt was made toquantity the unloading effect relative to manikin size. Theanalysis identified the existence of this alternate load pathin each of the three manikins tested. Allhough these datashow that the abdominal inserts atTected lumbar load

response, calibration test methods could be employed toquantify the effect, thus permitting the usc of lumbar forceas an effective comparative criterion for assessing thepotential for spinal injury.

162

.. - ,,- - -'

POHf:I

,

;I

Ii!i

~J

Apr 14 OS

~~\' ...

10: 12a

5. All threeindependentvariableshad a significanteffecton lumbarload. Lumbarloadsas measured within

the Hybrid III ATD were responsive to the peakacceleration level and velocity change parameters of thecrash pulse, and manikin size. The analysis of variancedetennined that a three-way interaction existed betweenmanikin size, velocity change, and sled peak acceleration(F = 6.27, P = .000).

6. The etTect of upper torso restraint was studiedand found to have a statistically significant effect onlumbar load. However, the effect was minimal for the 5thand 50th percentile ATDs, i.e., approximately, a 100 pounddifference between the mean levels.

RECOMMENDATIONS

I. Lumbar force measurements are recommendedasa spinal injury assessmentcriterion, both for crash resistantseats and ejection seats.

2. A configuration study should be conducted todetermine the variance of the 95th percentile aerospaceATD's upper torso mass within the Navy's inventory oftest dummies.

3. To support use of a lumbar force criterion,additional calibration testing of both the "aerospace" and"automotive" versions of Hybrid III ATDs should beconductedto assess the following factors:

a. Response ofthe 5th percentile femaleaerospaceHybrid III ATD.

b. Effect of the "curved" automotive lumbar

spinal element on lumbar force.

c. Repeatability (i.e., similarity of results of asingle ATD under identical test conditions) andreproducibility (i.e., variability between ATDs)coefficients should be established for the + Gz

headward loading direction.

REFERENCES

I. Code of Federal Regulations; 14 FAR 25.562,JanuaryI, 1989.

2. MIL-STD-1290A(AV), Military Standard, LightFixed And Rotary-Wing Aircraft, Crash Resistance, 26September1988.

3. Eiband, A. M., "Human Tolerance to RapidlyApplied Accelerations: A Summary of the Literature,"NASA Memorandum 5-I9-59E, National Aeronautics andSpaceAdministration,Washington, DC, USA, June 1959.

-

10.9

4. "Human Tolerance to Impact Conditions asRelated to Motor Vehicle Design," SAE J885, JUL 86,Handbook Supplement HSJ885 JUL 1990 Edition, Societyof Automotive Engineers, Warrendale, PA.

5. Stapp, J. P., "Human Exposure to LinearAcceleration," The forward Facing Position andDevelopment of a Crash Harness, WADC TR-5915,Wright-PattersonAir Force Base, Ohio, 1951.

6. Ewing, C. L., and Thomas, D.1., "Human HeadAnd Neck Response to Impact Acceleration," NavalAerospace Medical Research Laboratory, 1972.

7. Pike, Jeffrey, A., "AUTOMOTIVE SAFETY,Anatomy, Injury, Testing, Regulation," Copyright 1990SAE, Warrendale, PA.

8. SAE PT-43, "BIOtv1ECHANICS of IMPACTINJURY and INJURY TOLERANCES of the I-IEAD-

NECK COMPLEX," Edited by Stanley H. Backaitis,1993,society of Automotive Engineers, Warrendale, PA.

9. SAE PT-186, "BIOMECHANICS andMEDICALASPECTSof LOWER UMB INJURIES,"Symposium on Biomechanics, Oct. 29-30, 1986, SAE,Warrendale, PA.

10. Laananen, D. H., and Coltman, 1. W.,"Measurement of Spinal Loads in Two ModifiedAnthropomophic Dummies," Simula Inc., TR-82405, FinalReport, US Anny Aeromedical Research Laboratory, FortRucker, AL, May 1982.

II. Chandler,Richard,F., General,Aviation SafetyPanel, Crashworthiness Working Group, Federal AviationAdmistratlon, 1984.

12. "Hybrid III: The First Human-Like Crash TestDummy," S~ PT-44, 1984, Society of AutomotiveEngineers, Warrendale,PA.

13. Frisch, Georg, D., et aI., "Structural IntegrityTests of a ModifiedHybrid III Manikinand SupportingInstrumentationSystem,"

14. "A Third Generation Energy Absorber For CrashAttenuating Helicopter Seating," American HelicopterSociety 50th Annual Forum, dated May 11, 1994, LanceC. Labun, Simula, Inc., Phoenix, AZ., Martin Rapaport,Naval Air Warfare Center, Warminster, PA.

15. Aircraft Crash Survival Design Guide, Volume II- Aircraft Design Crash Impact Conditions and HumanTolerance, USAAVSCOM TR-89-D-22B, Dee 1989.

163

Apr 14 06 10: 12a

BIOGRAPHIES

ANN SCHOENBECK Ms. Schoenbeck has beenemployed by the Naval Air Warfare Center AircraftDivision in Patuxent River, MD, since 1996. She isinvolved in the Advanced Crashworthiness AircrewSurvival Systems (ACASS) team and the Cockpit AirbagSystem program. Ms. Schoenbeck received a B.S. inMechanical Engineering from the University of Missouri -Columbia in 1992 and a M.S. in Mechanical andAerospace Engineering from the University of Virginia in1996 with research in automotive crashworthiness fordisabled persons.

ESTRELLA FORSTER Ms. Forster is the principalinvestigator for the Navy Smart Aircrew Mission SupportSystem (SAMSS) and the Aircrew Integrated Life SupportSystem (AILSS) programs. Her thirteen years ofexperience include acceleration physiology, +Gz-inducedloss of consciousness (G-LOC), aircrew perfonnance,statistical applications in research, and programmanagement. This experience was gained in Brooks AFB,Texas (1984-1989) and the Naval Air Warfare Center.Aircraft Division (1989-present). She is currently pursuinga Ph.D. degree at Drexel University and is a member ofvarious associations addressing aviation medicine andtechnology. Ms. Forster has a M.S. in Statistics ITomDrexel University. Philadelphia, PA, and a B.S. inPhysiology from the University of Houston, Houston, TX.

MARTIN RAPAPORT Mr. Rapaport was employed bythe Naval Air Warfare Center Aircraft Division from 1990until 1997. He served as a crash safety engineer involvedin supplementaryrotary wing restraint systems, such lISth~Cockpit Airbag System, and crew seating. He currentlyholds a position with Volvo Cars of North America, Inc.Mr. Rapaport received a B.S. in Physics ITom CarnegieMellon University in 1985 and a M.S. in AppliedMechanicsfrom Polytechnic University in 1990.

LEON DOMZALSKI Mr. Domzalski has thirty-six yearsexperience in the field <:ifcrash safety engineering. He iscurrently employed by ARCCA, Inc., providingengineeringsupport services, and previously held a projectengineering position with the Aircraft and Crew SystemsTechnology Directorate at the Naval Air DevelopmentCenter,Warminster,PA. While employed at the Naval AirDevelopment Center, Mr. Domzalski held projectengineering responsibility for several important crashsafety programs, namely, Inflatable Body And HeadRestraint System (IBAHRS), Advanced CrashworthyAircrew Survival System (ACASS), V-22, and theAdvancedEnergy Absorber (AEA). He is a member ofeheSociety of Automotive Engineers (SAE) and received aB.S. in Physics from Saint Joseph's College in 1958.

nPH" 1 0

'jI!j

164

ii,.:i......-