OARS - Open Access Rewards System
DOI : 10.2240/azojomo0207

The Low Velocity Impact Response of Foam Core Sandwich Structure

Hazizan Md. Akil and W.J. Cantwell

Copyright AD-TECH; licensee AZoM.com Pty Ltd.

This is an AZo Open Access Rewards System (AZo-OARS) article distributed under the terms of the AZo–OARS https://www.azom.com/oars.asp which permits unrestricted use provided the original work is properly cited but is limited to non-commercial distribution and reproduction.

AZojomo (ISSN 1833-122X) Volume 2 May 2006

Topics Covered

Abstract

Keywords

Introduction

Experimental Procedure

Energy Balance Model

Results and Discussion

Rate Sensitivity of the Skin and Core Materials

Work of Fracture of the Foams

Indentation Characteristics of the Sandwich Structures

Impact Response of the Sandwich Structures

Energy Partitioning

Failure Modes in the Sandwich Structures

Conclusions

Acknowledgements

References

Contact Details

Abstract

The low velocity impact response of a range of foam based sandwich structures has been investigated using an instrumented falling weight impact tower.  Initially, the rate-sensitivity of the skin and core materials was investigated through a series of flexure and indentation tests.  It was shown that the flexural modulus of the skins and all 11 foam materials did not exhibit any sensitivity to crosshead displacement rate over the conditions studied here.  In addition, it was shown that the indentation response of the sandwich structures could be modelled using a simple indentation law, the parameters of which did not exhibit any sensitivity to loading rate.  Low velocity impact tests on the sandwich structures resulted in a number of different failure modes.  Shear fracture was found to occur in the PVC/PUR systems based on brittle core materials.  In contrast, buckling failures in the uppermost composite skin were observed in the intermediate modulus systems, whereas initial damage in the higher modulus PVC/PUR systems took the form of delamination within the top surface skin.  It has been shown that a simple energy balance model based on the dissipation of energy during the impact event can be used to successfully model the elastic response of foam based sandwich structures.  The energy balance model is particularly useful since it can be used to establish the partition of energy during the impact process.

Keywords

Mechanical Properties, Foams, Adhesion, Impact Response, Sandwich Structure

Introduction

Sandwich structures based on strong, stiff composite skins bonded to a low density core material are finding increasing use in aerospace, offshore and marine industries.  One of the main drawbacks of these high performance structures is their relatively poor resistance to localised impact loading [1 4].  A number of workers have investigated the impact response of sandwich structures and an excellent review is given in Ref. [2].

Horrigan et al. [1] conducted experimental and theoretical investigations on a Nomex honeycomb sandwich structure with glass fibre reinforced epoxy skins.  They showed that a soft, compliant projectile results in shallow crushing of the core whereas hard bodies create deeper damage that conforms to the shape of the projectile.  Charles and GuedraDegeorges [5] showed that the dent depth around the impact point increases with impact energy until a maximum value is reached.  Rhodes [6] conducted impact tests on a range of systems and showed that enhancing the crush strength of the core material can serve to increase the impact resistance of the sandwich structure.

Mahfuz et al. [7] conducted through thickness split Hopkinson pressure bar tests on a range of sandwich structures in order to characterise the strain-rate-sensitivity of these structures and to investigate the influence of core properties on the resulting failure modes.  They showed that the strain-rate-sensitivity of sandwich composites increases with the density of the core material.  They also showed that much of the impact energy is absorbed by the core material, with there being little evidence of debonding at the skin-core interface.

Kenny and Torre [8] developed and tested a novel corrugated sandwich panel and showed that such structures offer a superior energy-absorbing capacity to that exhibited by traditional sandwich structures.  A number of workers have investigated the perforation resistance of sandwich structures [9-11].  Mines et al. [9] conducted low velocity impact tests on square panels based on polymer composite sandwich structures based on Coremat (propriety material made from non-woven polyester felt containing 50% resinimpregnated glass microspheres) and aluminium honeycomb cores.  The authors showed that much of the incident energy of the projectile is absorbed in crushing the core material within a localised region immediate to the point of impact.  Mines et al. [9] also undertook a series of quasistatic perforation tests and found that the perforation resistance of these structures increases with strain rate, an effect that was attributed to a change in the geometry of the deformation zone and rate effects in the composite skins.  Following low velocity impact tests on a number of foam core and balsa core sandwich structures, Cantwell et al. [11] concluded that skin shearing was the primary energy absorber under localised impact conditions.

A number of workers have attempted to model the dynamic response of sandwich structures [2, 12, 13].  Abrate [2] outlined an energy-balance model in which the incident energy of the projectile is equated to the energy stored in the target in bending and shear deformations and the energy used to create local deformations in the contact region.  Ambur and Cruz [12] used first order shear deformation theory to model the low velocity impact response of composite panels.  Sun and Wu [13] used shear-deformable plate finite elements to model the composite skins in an impact loaded sandwich structure.

The aim of the present work was to develop and apply an energy-balance model similar to that outlined by Abrate [2] in order to predict the impact response of a foam-based sandwich structure.  The accuracy of the model has been investigated by varying the incident energy of the falling impactor as well as the properties of the foam core.

Experimental Procedure

Eleven sandwich structures were evaluated during the course of this research.  The composite skins were based on a woven glass fibre phenolic resin (PN90044-43 from Stesalit Ltd, Switzerland).  Individual prepreg plies with dimensions of 200 mm x 190 mm were cut into the shape of the mould using a sharp razor blade.  The plies were then wrapped in polyethylene film and placed in a picture frame mould.  The composite was cured in a hot press at 125°C for 120 min under a pressure of 0.7 bar to yield plates with an approximate thickness of 1 mm.  Following manufacture, one side of the laminate was roughened using gritted paper in order to improve adhesion to the foam core materials.  The skin was then cleaned and washed using ethanol to eliminate any dust generated during sanding process.

Eleven different foams were supplied in 10 mm thick sheets with dimensions of 1500 mm x 1000 mm.  The foams were supplied in three main grades these being linear Polyvinyl chloride (PVC) foams, polyetherimide (PEI) foams and PVC/PUR foams.  Details of a number of key properties of the foams are given in Table 1.  The data clearly indicate that the tensile and shear moduli of these foams strongly depend on density.  The tensile modulus of the foams varies from 28 MPa for the 40 k g/m3 PVC/PUR system to 178 MPa for the 200 kg/ m3 PVC/PUR foam.  Similar variations are apparent in the shear modulus (G), which are varied from 11 to 75 MPa.  The foam panels were cut to the same dimensions as the composite face sheets using a band-saw and a compressed air supply was used to remove any dust generated during the cutting procedure.  The composite skins were bonded to the polymeric foam cores using an Araldite LY5082 epoxy resin with an Araldite AY103 hardener.  The adhesive was applied to both the skin and core materials and the resulting structure was placed in a cold press for 12 h.  The cured sandwich panels were then cut into coupons with dimensions 190 mm X 20 mm in preparation for testing.

Table 1. Summary of the properties and characteristics of the foam materials.

Structure

Foam

Nominal Density (kg/m3)

Tensile Modulus (MPa)

Shear Modulus (MPa)

1

Linear PVC

50

37

11

2

Linear PVC

80

56

21

3

Linear PVC

140

90

37

4

PEI

60

37

14

5

PEI

80

52

22

6

PVC/PUR

40

28

13

7

PVC/PUR

55

45

22

8

PVC/PUR

75

63

30

9

PVC/PUR

90

81

38

10

PVC/PUR

130

109

50

11

PVC/PUR

200

178

75

Impact tests were conducted using an instrumented falling-weight impact tower.  Here, a 1.98 kg carriage with a 10 mm diameter hemi-spherical indentor was released from heights of up to 1.0 m.  The vertical guides of the impact tower were lubricated to minimise any friction generated during the descent of the carriage.  Incident impact energies in the range of 0.1-1.94 J were achieved by varying the drop height.  The specimens were supported on two 10 mm diameter steel cylinders positioned on movable right angle supports, as shown in Figure 1.  After impact, the carriage was caught manually in order to avoid secondary impacts.  The impact force history during the test was measured using a piezo-electric load-cell (Kistler 5011) located just above the impactor tip.  The signals from the load-cell were then fed to a strain gauge amplifier and recorded by computer.

Figure 1. Schematic of the impact test arrangement.

It is likely that impact-loaded sandwich structures will absorb significant energy in contact deformations local to the point of impact.  In this study, a series of indentation tests were carried out by using an Instron 4505 mechanical test machine at crosshead displacement rates of 0.2, 1, 10 and 100 mm/min.  The indentation tests were conducted on sandwich beams with dimension of 200 x 20 x 12 mm3 positioned on a solid steel support.  The contact displacement was measured by using an extensometer attached to the 10 mm diameter hemispherical indentor and upper surface of the sandwich structure The sandwich specimens were loaded up to a force of similar magnitude to that recorded during the impact test.  The applied load P during the indentation test and the resulting indentation, α, were assumed to obey a generalised indentation law of the form

P = Cαn

where C and n are constants.  The load-indentation data were fitted to this relationship to yield the values of n and C for each crosshead displacement rate.

The rate-sensitivity of the 11 foam materials was investigated by conducting three point bend tests on samples with dimension of 200 X 20 mm2 this being the same configuration as that used during the impact tests on the sandwich structures.  The beams were supported on 10 mm diameter cylindrical rollers positioned 175 mm apart as shown in Figure 2(a).  The foam beams were then loaded centrally at crosshead displacement rates of 1, 10, 100 and 1000 mm/min and the flexural modulus, Ef, was determined as a function of the central deflection, δ, the specimen width, b, the support span, L, and the beam thickness, t, using

Figure 2. Schematic drawings showing (a) the three point test set-up and (b) the SENB specimen.

Finally, the fracture properties of the foams were investigated using the single edge notch bend (SENB) sample shown in Figure 2(b).  Beams with dimension 120 x 25 x 10 mm3 were cut from panels using a handsaw.  A 12 mm long pre-crack was introduced at the mid-span as shown and sharpened by passing a fresh razor blade along the root of the notch.  The specimens were positioned on 10 mm rollers positioned 100 mm apart and loaded at a crosshead displacement rate of 10 mm/min.  Following the SENB tests, the load-displacement curves were integrated to give the energy dissipated during fracture.  These values were then normalised by the resulting area of fracture to yield the work of fracture for each system.

Energy Balance Model

The impact response of the sandwich structures was modelled using an energy-balance model.  Here, it is assumed that the target responds quasi-statically during the impact event and that the kinetic energy of the target is absorbed in bending, shear and contact effect deformations such that

           (1)

where the subscripts b, s and c refer to energy dissipation in bending, shear and contact effects, respectively.  The force-displacement relationship, P - δ, for a sandwich beam is given as Ref. [14]

          (2)

where G is the shear modulus of the foam core, L is the span, D is the flexural rigidity of the skins and A, a geometrical parameter that depends on the thickness of the core and skin materials and the beam width [14].  The first term in the brackets accounts for bending of the skins and the second term for shear effects in the core.

The energy absorbed in bending and shear effects at maximum displacement, δmax, under the maximum lood, Pmax is equal to

      (3)

Contact effects were modelled using the Meyer contact law which relates the indentation, α, to the applied load, P, through

     (4)

where C and n are constants which were determined experimentally in this study.

The energy absorbed in contact effects is shown to be

     (5)

The energy-balance for the sandwich structure is therefore

  (6)

Once the properties of the skin and core materials have been determined, this equation can be used to predict the maximum impact force, Pmax, for a given impact energy.  In this study, a spreadsheet was used in which the impact force on the right-hand side of Eq. (6) was increased until both sides of the equation were equal.

Results and Discussion

Rate Sensitivity of the Skin and Core Materials

Before undertaking the impact tests on the sandwich beams, the rate-sensitivity of the composite skins and foam cores was assessed by conducting flexure tests on simple beam-like samples.  Clearly, this geometry is not ideal for investigating rate effects since the strain rate varies along the length of the sample.  However, three point tests are simple to undertake (particularly on foam materials) and the test configuration is similar to that adopted for the low velocity impact tests on the sandwich structures.  The variation of the flexural modulus of the composite skin with crosshead displacement rate is shown in Figure 3.  It is clear that the flexural modulus of the glass fibre reinforced skins does not vary with loading rate with the value of Ef averaging approximately 29 GPa at all crosshead displacement rates.  Similar observations have been made elsewhere following tests on glass fibre reinforced composites [15].  These results suggest that the quasi-static value of flexural modulus can be used in an energy-balance model to predict the impact response of a sandwich structure.

Figure 3. The variation of the flexural modulus of the composite skins with crosshead displacement rate.

The rate-sensitivity of four of the foam materials is shown in Figure 4.  The standard deviations have been removed for clarity.  An examination of the data indicates that all four systems exhibit a rate-insensitive response with the flexural modulus remaining constant over the three decades of loading.

Figure 4. The variation of the flexual modulus of four foams systems with crosshead displacement rate.

Flexure tests on the remaining seven foam systems yielded similar trends to those apparent in Figure 4 with the flexural modulus of all systems remaining constant over the range of conditions considered.  A comparison of the data in Figure 4 and Table 1 indicates that the flexural modulus of the 200 kg/m3 PVC/PUR system is over 20% higher than its tensile modulus.  The reason for this is not clear although some variations in the measured and quoted values of density were observed.  Figure 5 shows the variation of the flexural modulus of the foams with the actual (measured) density at a crosshead displacement rate of 1 mm/min.  It is interesting to note that all of the data appear to lie roughly on one line regardless of the material type.  This suggests that it is the foam density, P, that detennines the flexural modulus of these foams and that foam-type has only a secondary effect on this mechanical property.  A linear curve fit was applied to the data in Figure 5, yielding a relationship of the form

Ef = 0.96ρ - 24.3 (MPa)

Figure 5. The variation of the flexural modulus of the polymeric foams with density at a crosshead displacement rate of 1 mm/min.

Work of Fracture of the Foams

SENB tests were conducted to characterise the toughness characteristics of the 11 materials.  It was observed that all of the linear PVC and PEI samples failed in a stable mode of fracture with the crack propagating at a controlled rate through the foam sample.  In contrast, all of the PVC/ PUR samples failed in an unstable mode yielding a load displacement curve with a saw-tooth appearance.  Attempts to analyse the SENB data using a linear elastic fracture mechanics approach proved fruitless due to the high level of non-linearity observed in the load-displacement traces and difficulties in satisfying the specimen size requirements outlined in ASTM D5045-91.  Instead, the toughness of all the systems was characterised by determining the work of fracture, Wf, and Figure 6 shows the properties of the 11 foam systems.  It is apparent that there is a significant variation in the fracture properties of the foams with Wf varying from approximately 50 J/m2 for the lowest density PVC/PUR system to 3750 J/m2 for the linear PVC with a density of 140 kg/m3.  It is interesting to note that the work of fracture appears to increase in a linear fashion with increasing foam density.  From the figure, it is also apparent that the linear PVC foam out performs the two other systems with the PVC/PUR blend exhibiting the poorest fracture properties.

Figure 6. Summary of the work of fracture of the 11 foam materials.

Indentation Characteristics of the Sandwich Structures

Figure 7 shows typical load-indentation plots for four foam systems tested at 0.2 mm/min.  Clearly, the slope of the load indentation force increases with foam density with the effective gradient of the curve for the 200 kg/m3 system being approximately four times that of the 50 kg/m3 linear PVC material.  It is interesting to note that, following the initial indentation phase the load-indentation traces for three of the foams are relatively linear over the testing regime considered.  In contrast, the slope of the lowest density system decreases with increasing indentation suggesting that some crushing or failure could be occurring under the indentor during the test.  The initial region of nonlinearity in several of the load-indentation traces is attributed to the compliance of the test machine and a slight lack of fit in the test fixture.  The data from the load-indentation curves were subsequently used to determine the indentation constants C and n in the Meyer indentation law.

Figure 7. Typical load-displacement traces following indentation tests at 0.2 mm/min.

Table 2 summarises the contact parameters C and n at a crosshead displacement rate of 0.2 mm/min.  An examination of the table indicates that many of the values of n are close to unity suggesting that the load-indentation traces were relatively linear for a large number of the systems examined in this study.  It is interesting to note, however, that several of the lowest density systems (Structures 1, 4 and 6) offer values of n that are significantly below unity.  The values of C in Table 2 show a strong dependency on the type of foam and its density.  In general, the value of C was higher for the PVC/PUR foams, tending to increase with increasing foam density.

Table 2. Summary of the contact properties of the foam materials.

Structure

Foam

C(N/mn)

n

1

Linear PVC

5x104

0.86

2

Linear PVC

1.3x105

0.92

3

Linear PVC

1.8x105

0.93

4

PEI

4.1x104

0.76

5

PEI

1.4x105

0.91

6

PVC/PUR

3x105

0.77

7

PVC/PUR

3.1x105

1.09

8

PVC/PUR

3.4x105

1.05

9

PVC/PUR

2.5x105

0.99

10

PVC/PUR

4.3x105

1.04

11

PVC/PUR

9.4x105

1.09

Figure 8 shows the rate-sensitivity of the value of n for four of the sandwich structures.  The standard deviations have been omitted to enhance the clarity of the figure.  From the data, it is apparent that the values of n do not appear to exhibit any apparent sensitivity to crosshead displacement rate.  Indeed, when the standard deviations are included, it is evident that the trends are flat suggesting that this parameter is insensitive to loading rate.  Similar observations were made when the rate-sensitivity of the parameter C was evaluated, leading to the conclusion that the indentation response of all of the foams was generally rate-insensitive over the range of testing conditions considered here. Once again, this evidence suggests that the statically-determined values of the Meyer constants, C and n can be used in an energy-balance model.

Figure 8. Summary of the rate-sensitivity of the contact parameter n for four foams.

Impact Response of the Sandwich Structures

Figure 9 shows the variation of the maximum impact force with impact energy for two of the linear PVC foams.  The data for the 50 kg/m3 foam show an initial rise in force with increasing impact energy before reaching a plateau at approximately 170 N.  A post-impact examination of the samples indicated that for energies above 0.4 J, all specimens failed as a result of a top surface bucking failure of the skin directly under the point of impact.  It is likely that the low density (modulus) foam did not offer significant support to the composite skin under impact loading conditions.  As expected, an increase in foam density to 140 kg/m3 resulted in an increase in the measured impact force for a given incident energy.  Clearly, the curve corresponding to the 140 kg/m3 system continues to rise over the range of energies examined here.  Initial failure in the form of top surface fibre buckling occurred at an impact energy of approximately 1.4 J.  The solid lines in Figure 9 correspond to the predictions offered by the energy-balance model employing the mechanical properties determined at 1mm/min.  Agreement between the model and the experimental data is very good for the higher density system, even at energies close to the damage threshold.  In contrast, agreement between the model and experimental results quickly deteriorates in the 50 kg/m3 system.  This poor correlation results form the low damage threshold energy associated with this structure.

Figure 9. The variation of the maximum impact force with impact energy for two linear PVC foams.  The arrows indicate the damage threshold energy and the solid lines represent the predictions of the energy-balance model.

The trends in the maximum force data for the two PEI foams, Figure 10, are similar to those observed in the linear PVC foams.  Here, the 60 kg/m3 plateaued at approximately 300 N (at energies of 1 J and above).  Both systems failed as a result of a buckling failure in the top skin.  This will be discussed further in the following sections.  Here again, agreement between the model and the experimental data is very good at energies below that required to initiate damage in these structures.

Figure 10. The variation of the maximum impact force with impact energy for two linear PEI foams. The arrows indicate the damage threshold energy and the solid lines represent the predictions of the energy-balance model.

Figure 11 shows the impact force data for two of the PVC/PUR foam systems.  The maximum impact force in the 55 kg/m3 system increases steadily until approximately 1.4 J when a large shear crack propagated through the thickness of the core material.  Higher incident impact energies also resulted in a similar mode of failure at a maximum impact force of approximately 300 N.  In contrast, the impact force measured in the 200 kg/m3 PVC/PUR sandwich structure continued to rise over the range of energies considered.  The solid lines in the figure again indicate that a simple energy-balance model is capable of predicting the maximum impact force for a given incident energy.

Figure 11. The variation of the maximum impact force with impact energy for two linear PVC/PUR foams.  The arrows indicate the damage threshold energy and the solid lines represent the predictions of the energy-balance model.

Figure 12 shows the maximum impact force recorded in all 11 foams following a 0.4 J impact as a function of the shear modulus, G, of the foam core.  The data in the figure indicate that the maximum force tends to increase with increasing shear modulus before tending to at higher values of G. Indeed, the maximum impact in the three systems with the highest density (and highest shear modulus) were roughly equivalent.  Included in the figure is the energy-balance prediction of the variation of P with G assuming values of C and n of 3.1 x 105 N/mn and 1.09, respectively.  Clearly, the impact data appear to loosely  follow the trend predicted by the impact model.

Figure 12. The variation of the maximum impact force with shear modulus following a 0.4 J impact.  Included in the figure is the prediction assuming values of C and n of 3.1 x 105 N/mn and 1.09, respectively.

Energy Partitioning

In order to gain a greater understanding of the impact event, the energy-balance model was used to partition the energy in the sandwich structure.  The fraction of energy dissipated in bending, shear and contact in Eq. (6) was determined for a range of impact energies.  Figure 13(a) shows the energy absorption profiles corresponding to the PVC/PUR foam with a nominal density of 55 kg/m3.  Approximately 50% of the incident impact energy is absorbed in shear effects within the low density core.  Note that only 16% of the impact energy of the carriage is absorbed in bending of the sandwich beam during the impact event.  It is also clear that energy absorption profiles are relatively flat over the range of impact energies, suggesting that the distribution of the incident impact energy does not vary with impact conditions.  In contrast, contact effects represent the predominant energy-absorbing process in the higher density, 200 kg/m3 PVC/PUR system, Figure 13(b).  Top surface indentation effects account for between 45 and 50% of the impact energy of the projectile whereas shear effects represent only 21-25% of the incident energy.  Similar energy dissipation profiles were obtained for all of the foam systems investigated in this study and this information was used to investigate the effect of varying foam density on the energy absorption profiles.  Figure 14 shows energy dissipation profiles for the six PVC/PUR foams subjected to an impact energy of 0.4 J.  The data has been plotted as a function of the measured density rather than the manufacturer's data.  Clearly, there is some scatter in the data, an effect that is attributed to small variations in the skins thickness and variations in the indentation constants.  The figure clearly shows that energy dissipation in core shear decreases rapidly as the foam density is increased.  The energy dissipated in bending increasing steadily as the foam density increases.  The evidence also suggests that energy dissipation in contact effects also increases slightly with increasing foam density.

Figure 13. Predictions of the breakdown of the absorption of energy in the PVC/PUR foams with densities of: (a)55 kg/m3: (b) 200 kg/m3.

Figure 14. Energy dissipation profiles with varying foam density.  The subscripts c, b and s refer to contact, bending and shear.

Failure Modes in the Sandwich Structures

Following impact, the failure modes in the samples were elucidated using a low power optical microscope.  Table 3 summarises the predominant failure modes in samples impacted at energies just above the damage initiation threshold.  From the table, it is apparent that three modes of failure were observed in the 11 sandwich structures.  Initial failure in all the three sandwich structures based on the linear PVC foams took the form of a buckling failure in the uppermost skin of the sandwich structure.  A similar mode of failure was observed in the two PEI-based sandwich structures, materials 4 and 5.  Initial failure in structures 6 and 7, i.e. those based on the lowest density PVC/ PUR foams, resulted from shear cracking through the depth of the core material.  This is not surprising since these two foam cores offered the lowest values of work of fracture (55 and 130 J/m2 for the 40 and 55 kg/m3 core materials, respectively).  Failure in the higher density PVC/PUR systems occurred as a result of delamination in the top surface skin immediate to the point of impact.

The trends in the Table 3 suggest that failure in these sandwich structures is governed by the fracture properties of the core material and the degree of support that the core affords the composite skin (i.e. its elastic modulus).  If the core is very brittle, initial failure is likely to take the form of shear cracking through the thickness of the foam.  If the toughness of the foam is sufficiently high so that shear cracking can be suppressed, initial failure occurs as a result of a buckling failure in the top skin.  In this case, the elastic modulus of the foam is relatively low and the skin receives very little support against buckling.  For higher density foams which exhibit a higher elastic modulus, buckling failure in the skin is less likely to occur and failure, in the form of delamination within the top surface skin, occurs around the impactor as a result of the higher impact forces.

Table 3. Summary of the failure modes in the sandwich structures.

Structure

Foam

Failure Mode

1

Linear PVC

Buckling failure in top skin

2

Linear PVC

Buckling failure in top skin

3

Linear PVC

Buckling failure in top skin

4

PEI

Buckling failure in top skin

5

PEI

Buckling failure in top skin

6

PVC/PUR

Shear cracking in core

7

PVC/PUR

Shear cracking in core

8

PVC/PUR

Buckling failure in top skin

9

PVC/PUR

Buckling failure in top skin

10

PVC/PUR

Delamination in top skin

11

PVC/PUR

Delamination in top skin

Conclusions

The low velocity impact response of 11 sandwich structures based on low density polymeric foams has been studied.  Initial tests to characterise the mechanical properties of the skin and core materials have shown that the elastic modulus of all the systems was insensitive to crosshead displacement rate over the range of test conditions investigated in this study.  Subsequent indentation tests on the 11 sandwich structures have shown that the load-indentation response of these systems can be characterised using a Meyer indentation law.

Low velocity impact tests on the sandwich structures have shown that the dynamic response of these systems depends on the elastic properties of the foam core material.  For a given impact energy, the maximum impact force, Pmax, tend to increase with increasing shear modulus, G. A subsequent optical examination of the impacted samples highlighted three prominent modes of failure.  Initial damage in the low density PVC/PUR sandwich structures took the form of shear cracking in the brittle foam core.  Failure in the intermediate modulus systems occurred as a result of fibre buckling close to the point of impact.  Finally, failure in sandwich structures with a high modulus core took the form of delamination in the top surface skin.

Finally, it has been shown that a simple energy-balance model based on the dissipation of energy during the impact event can be used to predict the low velocity impact response of sandwich structures in the elastic regime.  In addition, it has been shown that the impact response of these foam-based sandwich structures can be successfully modelled using mechanical properties determined at quasistatic rates of strain.

Acknowledgements

The authors acknowledge the financial support of the University Sains Malaysia.  The authors are also grateful to Dr Lukas Berger for supplying the foam materials and Mr Marcus Erath of Stesalit Ltd, Switzerland for supplying the composite materials.

References

1.       D. P. W. Horrigan, R. R. Aitken and G. Moltschaniwskyj, “Modelling of Crushing due to Impact in Honeycomb Sandwiches”, J. Sandwich Struct. Mater., 2 (2000) 131-51.

2.       S. Abrate, “Localized Impact on Sandwich Structures with Laminated Facings”, Appl. Mech. Rev., 50 (1997) 70-82.

3.       O. Ishai and C. Hiel, “Design of Highly Damage-Tolerant Sandwich Panels”, 37th Int. SAMPE Symp., Exhibit., (1992) 1228-42.

4.       A. T. Nettles and A. J. Hodge, “Impact Testing of Glass/phenolic Honeycomb Panels with Graphite/epoxy Face Sheets”, Proc. 35th Int. SAMPE Symp. Exhibit., (1990) 1430-40.

5.       J. P. Charles and D. Guedra-Degeorges, “Impact Damage Tolerance of Helicopter Sandwich Structures”, Proc. 23rd Int. SAMPE Conf., (1991) 51-61.

6.       M. D. Rhodes, “Impact Fracture of Composite Sandwich Structures”, Proc. ASME/AIAA/SAE, 16th Struct., Struct. Dyn. Mat. Conf., (1975) 311-6.

7.       H. Mahfuz, W. Al Mamurn and S. Jeelani, “Effect of Core Density and Implanted Delamination on the High Strain Rate Response of Foam Core Sandwich Composites”, Sandwich Construct., 5 (1992) 597-606.

8.       J. M. Kenny and L. Torre, “Impact Testing and Simulation of Sandwich Structures”, 44th Int. SAMPE Symp., (1999) 2368-81.

9.       R. A. W. Mines, C. M. Worrall and A. G. Gibson, “Low Velocity Perforation Behaviour of Polymer Composite Sandwich Panels”, Int. J. Impact Engng., 21 (1998) 855-79.

10.   S. R. Reid and H. M. Wen, “Perforation of FRP Laminates and Sandwich Panels Subjected to Missile Impact”, in “Impact Behaviour of Fibre-reinforced Composite Materials and Structures”, Woodhead Publication Ltd, Cambridge, (2000) Chapter 8.

11.   W. J. Cantwell, C. Dirat and H. H. Kausch, “Comparative Study of the Mechanical Properties of Sandwich Materials for Nautical Construction”, SAMPE J, (1994) 30:45-5 1.

12.   D. R. Ambur and J. Cruz, “Low Speed Impact Response Characteristics of Composite Sandwich Panels”, Proc. 36th AIAA/ASME/ASCE/AHS/ ASC Struct., Struct. Dyn. Mat. Conf., (1995) 2681-9.

13.   C.T. Sun and C.L. Wu, “Low Velocity Impact Response of Composite Sandwich Panels”, Proc. 32th AIAA/ASME/ASCE/AHS/ASC Struct, Struct. Dyn. Mat. Conf., (1991) 1123-9.

14.   H. G. Allen, “Analysis and Design of Structures Sandwich Panels”, Oxford: Pergamon Press, 1969.

15.   W. J. Cantwell, M. Busser and H. H. Kausch. Computer Engineering, (1991) 1:293307.

Contact Details

Hazizan Md. Akil

School of Materials and Mineral resource engineering, Engineering Campus
University Sains Malaysia

E-mail: [email protected]

WJ Cantwell

Department of materials Science and Engineering
University of Liverpool

Published in print form in “Advances in Technology of Materialsand Materials Processing Journal, 6[2] 276-285 (2004)”.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this article?

Leave your feedback
Submit