J B DA Component Cushion Test Development Presented

J B DA Component Cushion Test Development Presented at the Fourth International Fire and Cabin Safety Research Conference November 18, 2004 Lisbon, Portugal Steve Hooper, Ph. D and Marilyn Henderson J. B. Dwerlkotte Assoc. , Inc. Wichita, KS USA 316 -269 -6970, ext 14 Email: shooper 474@earthlink. net November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 1

J B DA • • • Outline Motivation to develop component test Review dynamic seat certification requirements Background – Review foam material properties • • • Describe a dynamic component test method Material test results Correlate with dynamic sled (seat) test results Present proposed component test method Summary and conclusions November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 2

J B DA • Motivation to Develop Component Test Seat Cushions are a significant maintenance item – Typical life 3 -5 years – Cushions identical to the original cushions not always available – Full-scale seat tests are destructive • Airlines are not willing to obtain / destroy seats for full-scale tests – Bottom cushion design strongly influences lumbar load • During 14 -G download tests • Component tests are useful in validating dynamic simulations of seats / restraints / and occupants November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 3

J B DA • Download Test Description Test Conditions – 14 g triangular pulse, tr = 0. 08 s (peak g), DV = 35 fps – Applied 90 deg to flight path vector • Pass / Fail Criteria – Lumbar Load £ 1500 lb. [Amdt. 25 -64, 53 FR 17646, May 17, 1988] November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 4

J B DA • Background – Foam Material Properties Gibson and Ashby describe foam compression as – Three significant regions – Two significant material properties: e. D and spl * • Some mat’ls exhibit a “work hardening” response STRESS, s – As shown on the right STRAIN, e November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. e. D 5

J B DA • Background – Foam Material Properties Upper R. H. portion of curve is an artifact of the test – Not a Material Property November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 6

J B DA • Background – IFD Test ASTM D 3574 Indentation Force Deflection (IFD) Test – Measure Resistive Force at 25 and 65% deflection of cushion thickness – 7 ½-in (190. 5 -mm) Diameter Specimen – Static loading – No unloading measurements Perforated Platen Foot Self Aligning Fixture November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 7

J B DA • Background – Other Efforts Lim method developed during the FAA / NASA AGATE 1 Program – Addressed rate effects • Hooper and Henderson developed dynamic component test – Included diaphragm in the test article definition • Lim’s method not documented in the public literature – Not validated as a robust method • Hooper and Henderson’s method – Too expensive (diaphragm issue) • These investigators identified the significance of the densification strain on lumbar load during dynamic seat tests November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 8

J B DA • • Description of Dynamic IFD Test Dynamic test based on ASTM D 3574 -03 (IFD Test) Utilize high-rate servo hydraulic test stand – – – • 220 kip load frame 110 kip actuator 10 kip piezoresistive load cell MTS Testar-IIm Controller MTS Multi-Purpose Testware (MPT) software Sampling rate: 12, 288 samples/sec November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 9

J B DA Fixture Design Dimensions in inches November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 10

J B DA Test Stand Performance Position Velocity Unfiltered Data November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 11

J B DA • • Test Description Test articles fabricated and supplied by three aircraft seat cushion suppliers Monolithic (nonflotation) cushions – – – • Laminated (flotation) cushions – – • 4 polymers 3 densities (3. 1 – 4. 4 lb/ft 3) 3 thicknesses 3 laminates • 2 polymers comfort foam • 3 polymers flotation foam 3 densities (comfort foam) • Same mat’l as used in monolithic specimens Flotation foam thickness established to satisfy TSO-C 72 c • 3 densities (1. 4 – 2. 6 lb/ft 3) 3 thicknesses Nonflotation specimen The entire test matrix was not tested Flotation specimen November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 12

J B DA November 18, 2004 Test Results – Monolithic Materials J. B. Dwerlkotte Assoc. , Inc. 13

J B DA November 18, 2004 Test Results – Laminated Materials J. B. Dwerlkotte Assoc. , Inc. 14

J B DA November 18, 2004 Stress- Strain Curve - Monolithic Foam J. B. Dwerlkotte Assoc. , Inc. 15

J B DA • Stress-Strain (cont. ) Shifting the unloading curves to a common stress value produces a common stress-strain curve November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 16

J B DA • 14 -g Sled Tests 14 -g sled tests were conducted of a limited number of cushions November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 17

J B DA Correlation of Sled Results w/ Mat’l Properties 2 -in. monolithic • • 3. 25 -in. monolithic Mat’l plotted in black produced highest lumbar load in every test Blue and red curves (mat’ls) always in same relative position on s -e curve and always produced lower lumbar loads. November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 18

J B DA • Comparison of Static and Dynamic Test Results Measurable rate effects may include: – Increased plateau strength – Reduced apparent densification strain November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 19

J B DA Correlation of Sled Results w/ Mat’l Properties (cont. ) 1155. 8 lb. 1794. 5 lb. 2136. 7 lb. 1253. 8 lb. 1698. 5 lb. November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 20

J B DA Effect of thickness on lumbar load • Caution: this trend is probably material specific • Points to importance of Densification Strain November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 21

J B DA Correlation of Sled Results w/ Mat’l Properties 2 -in. laminated • • 4 -in. laminated Mat’l plotted in black produced highest lumbar load in every test Blue and red curves (mat’ls) always in same relative position on s -e curve and always produced lower lumbar loads. November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 22

J B DA • Physical Explanation of Results 14 -g tests of an ATD installed in a Rigid “iron seats” with no cushion produce lumbar loads <1000 lb. – The F – d curve of this steel cushion lies above all of the foam cushion curves • Consider the performance of a very soft cushion that is installed on the “iron seat” – This cushion is so soft that it is completely consolidated under the ATD’s 1 -g preload – But, the F – d curve to the right of the consolidation strain is nearly as stiff as the F – d curve for steel • Therefore, the lumbar load for this cushion will approach 1000 lb. as well November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 23

J B DA Regression Analysis • Outlying data point in second plot due to difference in 1 -g cushion deflection under ATD load • Quadratic curve selected as the Criterion Curve November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 24

J B DA • Criterion Curve Definition: Criterion Curve – The load-deflection curve for a specified cushion thickness that produces the largest lumbar load November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 25

J B DA November 18, 2004 Criterion Stress – Log Strain Curve J. B. Dwerlkotte Assoc. , Inc. 26

J B DA • Proposed Component Test Method Perform dynamic component test of Certified Cushion Specimen – Compute criterion curve for specimen thickness • From stress – log strain curve • • Compare F - d data for certified cushion with criterion curve If above, then show replacement cushion in Usable Region Monolithic or Laminated Cushions November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 27

J B DA • Proposed Component Test Method (cont. ) If below, then show replacement cushion in Usable Region Monolithic Cushions Only November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 28

J B DA • Evaluation Region The Usable Region is determined by analyzing the position of the F - d curves in the Evaluation Region November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 29

J B DA November 18, 2004 Ineligible Material J. B. Dwerlkotte Assoc. , Inc. 30

J B DA • Static Requirement Replacement cushion specimen displacement under 130 -lb. Load must be equal to, or greater than, the corresponding displacement of the original certified cushion November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 31

J B DA • Acknowledgements This research was funded by the FAA William J. Hughes Technical Center by a subcontract through the National Institute for Aviation Research at Wichita State University. – Mr. Timothy G. Smith, FAA COTR – Dr. John Tomblin, WSU P. I. • The authors would like to acknowledge the contributions of – – • Ms. Lamia Salah and Wadii Benjilany of Wichita State Mr. Rick De. Weese and David Moorcroft of FAA CAMI Mr. Mike Thompson of the FAA Transport Aircraft Directorate Mr. Matt Riggins and Mr. Habtom Gebremeskel of JBDA The late Van Gowdy, who contributed to the design of this research. November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 32

J B DA • Summary & Conclusions High-rate material tests were performed of typical flotation and nonflotation seat cushions – Seat cushions exhibit a measurable rate sensitivity • • The material properties from these tests were qualitatively correlated with lumbar loads measured during dynamic seat tests These results point to the existence of a F - d curve that maximizes the lumbar load for a specified cushion thickness – These results show that a replacement cushion can be designed that will produce a lumbar load that is equal to, or less than, the lumbar load produced by an original certified seat cushion – The replacement cushion does not have to exhibit properties that are identical to the original certified cushion November 18, 2004 J. B. Dwerlkotte Assoc. , Inc. 33