Aluminum Honeycomb Characterization and

Modeling for Fuze Testing D. Peairs, M. Worthington, J. Burger, P. Salyers, E. Cooper

May 16, 2012

1

This presentation consists of L-3 Communications Corporation general capabilities

information that does not contain controlled technical data as defined within the

International Traffic in Arms (ITAR) Part 120.10 or Export Administration Regulations

(EAR) Part 734.7-11.

2

Motivation

• Need: Rapid evaluation of fuze models in impact with honeycomb materials

• Large expense of fuze testing in actual penetration and launch environments.

• Airgun testing used to simulate environments

• Pulse characteristics controlled by honeycomb materials.

• Shell model of honeycomb is too computationally intensive for many iterations

Aluminum Honeycomb Material

• Primarily two current types used at L3-FOS:

“ Mitigator” and “Backstop”

3

Mitigator Backstop

Foil Layers Foil

Thickness

Density Crush

Strength

Mitgator >2 .006 in 0.0166 lb/in3 >80,000 lbs

Backstop 1 .002 in 0.0045 lb/in3 >1000 lbs

Mitigator after impact

1/8” pre-crush

Material Model Description

• LS-DYNA Explicit FEA • Equilibrium with applied forces not maintained - update stiffness matrix in small

steps

• Timestep controlled by element size and wavespeed

• MAT_026, Mat_126 (honeycomb, modified honeycomb) options • Separate stress-relative volume curves allowed for normal and shear stress

direction (3 normal, 3 shear directions)

• Mat_026 uncoupled, nonlinear behavior for normal and shear stresses

• Mat_126 can model off axis loading, shear and normal curves can be coupled

• Two almost independent phases: Not Compacted, Fully Compacted

• Extrapolated yield stresses should not be negative

4

Shkolnikov, 7th International LS-DYNA Users Conference

Crush Strength

EUncompressed

Efully compressed

Displacement/Strain

Forc

e/S

tress

Sample Preparation

• Cylindrical mitigator sectioned to determine

compressive, shear properties in primarily axial,

theta, radial directions

• Samples bonded to rigid face plates for shear.

5 6 Jan 2011

Sample Geometry Compression 1Adirection T= direction AA= LCA

L-3 Proprietary Information

4 in.

6 in.

0.5 in.

Axial configuration

load

6 Jan 2011

Compression specimen 3adirection W or L= direction BB or CC= LCB or LCC, primarily radial direction

L-3 Proprietary Information

0.27”2”

6”

1”

3.87”

load

2”

load

Axial

Primarily

Radial

Shear Sample Preparation

• Steel face plates

• Hysol EA 9360 adhesive • 5000 psi lap shear strength

6

6 Jan 2011

Shear specimen 1adirection WT or LT= direction AB or AC= LCAB or LCCA, primarily radial axial or theta axial

L-3 Proprietary Information

4”

6”

load

load

0.75”

2”1”

Example sectioning for shear (primarily

circumferential-axial direction)

Example shear specimen(primarily

axial-circumferential direction)

Testing

• Quasistatic loading • 6 Mitigator specimens, 2 Backstop specimens

7

Shear testing

• No mitigator shear tests reached crushing portion of response

• Backstop shear test did not reach fully compressed response

8

Test Data Reduction

• Same number of data points desired for

each load/relative volume relationship

9

Ensures that extrapolation

do not extend into

negative stress region

Used for

Uncompressed

modulus

Shear Data Estimation

• Ratio of Backstop shear/compressive properties used to estimate Mitigator Shear/compressive properties

10

Material Model Input

• 4 curves used to describe crush strength • Axial

• Primarily radial

• Primarily circumferential

• Shear (1 instead of maximum of 3)

• 4 uncompacted Moduli

• Compressive modulus and Poisson’s Ratio from Aluminum for fully compacted condition

11

FEA Model Set-up

• Solid elements

• Applied axial displacement applied to top surface

• 5 in. at constant rate

• BC’s – Nodes around center

hole at top and bottom fixed in transverse direction

– Bottom face fixed in axial direction

12

Initial Modeling Results MAT_26

• No precrushed section causes

crushing to initiate in multiple

layers • Precrushed layer not modeled

• Insufficient face constraints and

multiple crushing locations can

cause bending/buckling

• No strain rate dependence

included

13

Initial Modeling Results MAT_26

• Initial stress increase in entire structure similar to test results

• Early termination because of poor model stability

14

MAT_126 (Modified Honeycomb) Model

• Switch to nonlinear spring element (type 0) • Allows large deformations

• Increased stability

• Load curves input in strain instead of relative

volume

• Material response during and after crushing

follows load curve • Post crush material properties ignored

15

MAT_126 Model Results

• Response matches test curve (as expected)

16

High Speed Impact

17

Shell Model

• Very good correlation to test results

– Computationally intensive

18

MAT_126 results

19

• ~85% reduction in run time from shell element mitigator model

• Run time now controlled by elements in device under test

• Honeycomb on honeycomb contact modeled

Stiffened Mat_126

20

Model Results Comparison

21

-1600

-1400

-1200

-1000

-800

-600

-400

-200

0

200

0 0.0005 0.001 0.0015 0.002 0.0025 0.003

Nest Velocity by Mitigator Type

Shell

MAT_126 (w/Backstop)

MAT_126 (Stiffened)

-20000

0

20000

40000

60000

80000

100000

120000

140000

160000

0.00E+00 5.00E-04 1.00E-03 1.50E-03 2.00E-03 2.50E-03 3.00E-03

Mitigator Crush Force by Type

Shell

MAT_126 (w/Backstop)

MAT_126 (Stiffened)

Comparison with Test Data

22

Conclusion

• Honeycomb material model is part of

ongoing efforts at L-3 FOS to continuously

improve modeling capabilities of all factors

affecting fuze survivability.

• Material model significantly reduces runtime

while maintaining acceptable accuracy.

• Reduced run time allows increased model

iteration and increased understanding of

response.

23