Experimental Investigation of Composite Materials...

International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:18 No:01 64

173101-2929-IJMME-IJENS © February 2018 IJENS I J E N S

Abstract— This study focused on the dynamic behavior of pure

epoxy, random-fiber-glass-epoxy, and woven-kevlar-fiber-epoxy

at five angle of twist (4 ̊, 6 ,̊ 8 ̊, 10 ̊ and 12 ̊ ) by using torsional-Split-

Hopkinson-Bar (TSHB). A new mechanism of the clamp system is

used in this work for the pressure on the lower ends of the clamp

arms, which introducing a major simplification to (TSHP) design

requirements the samples of composites were prepared by three

volume fractions (25, 40 and 55%). The results indicated that

maximum shear strain, and shear stress increased with increasing

the shear strain rate for all types of these samples. The maximum

shear-strain, shear-stress, and shear-strain-rate are observed in

the 55% volume fraction of woven kevlar fiber-epoxy composite.

When the angle of twist increased from 4 ̊ to 12 ̊, the shear-strain,

shear-stress, and shear-strain-rate for 55% volume fraction of

woven kevlar fiber-epoxy composite are increased by 4.9%, 5.8%

and 6.7%, respectively. The composite samples with 55% volume

fraction (random fiber glass and woven kevlar fiber) give shear-

strain, shear-stress, and shear-strain-rate greater than samples

with 25 and 40% volume fraction.

Index Term— Hopkinson bar, high shear strain rate, composite

materials.

1. INTRODUCTION

High shear strain rate testing is encountered in many types of

operations processing, such as forming, machining, grinding,

penetration, and punching operations, processes, [1]. Air force

massive ordnance air blast bomb and air force delta II rocket

are examples of the military applications for the United States

Air Force that benefit from testing with split-Hopkinson-

torsion-bar [2]. The technique of Hopkinson was firstly

developed to compression test of materials, later was extended

to torsion and tension [3]. The torsional Hopkinson bar is a

reliable device for materials testing from to strain

rate. Hopkinson torsional bar is consisting of two bars

(transmitter and incident bars) propped by bearings which

permit them to rotate easily. A short-sample is fixed between

them, (thin wall tube). The torsional-wave is generated in the

incident-bar by the leave of the torque which is stored primarily

in the part of device (between loading-end and clamp system)

[4].

A. Gilat and C. S. Cheng (2002) [5] worked in the bases of

modeling the split-torsional-Hopkinson-bar experiments at

strain-rates above . Finite-element-analysis is used to

investigate the material response by split-torsional-Hopkinson-

bar apparatus in which strain-rate about is reached

through utilizing samples with very short gage length. Stress-

strain curves which are acquired in this method from measured

elastic-waves in the bars, by assuming uniform and pure

deformation and shear-stress in gage length, have some errors

but, supply a satisfactory estimation for the response of

material.

F. Barthelat et al. (2004) [6] presented experimental-

methodology for the dynamic torsion and quasi-static tests of

nano, and micro-structured coating for (Tungsten carbide cobalt

WC-Co, and Aluminum Titanium Oxide Powder - )

coating, at aluminum substrates utilizing speckle photograph.

The displacement field given by speckle correlation method

permitted the identification of mechanism of deformation, i.e.,

early crack bridging and crack propagation by ductile

aluminum-substrate. No big changes in strength and modulus

are founded between micro, and nano-structured material, but

final changes between micro, and nano-structured material are

maybe disguised by the first damage in coating, as observed

on Scanning Electron Microscope and optical pictures. A. Gilat

et al. (2007) [7] carried out a study to investigate the

mechanical response for (PR-520 and E-862) epoxy-resin in the

shear, and the tensile loading. On two sorts of the loading, two

types of resins are experimented at the strain-rate about 450 to

700, 2, 5x . Results, display that, toughened (PR-520)

can carry greater stresses than untoughened (E-862).

Toughened (PR-520) has greater maximum-stress in the shear

and greater failure-stress in the tension, compared with

untoughened (E-862). M. N. Bassim (2007) [8] used Armor

materials to evaluate high strain rate. They include Rolled

homogeneous armor steel plate, 5083 aluminum alloys and

ceramics. It used both direct impact, Hopkinson-compressive-

bar, and split-torsional-Hopkinson-bar. Specimens were

examined to determine the microstructural evolution during

impact. Microstructural evolution in the materials during high

strain-rate deformation is also investigated. Occurrence of

adiabatic shear bands shows considerable influence on dynamic

stress-strain curves for the materials and cause failure at high-

strain-rates. Results refers to that the strain-rate in this study in

excess of ( ). The main objective of the study carried out

by R. S. Yatnalkar (2010) [9] is to investigate the plastic

deformation of 6.35 mm thick Ti-6Al-4V plate under various

loading conditions. Tensile, shear and compressive tests are

performed at quasi-static strain-rate ( ) to

Experimental Investigation of Composite

Materials Subjected to Torsional Stresses at

High Shear Strain Rate

1Asst. Prof. Dr. Fadhel Abbas Abdullah, M.Sc. student. Eng. Waad Adnan Khalaf

Mech. Eng. Dep., College of Engineering, Al-Mustansiriyah University ([email protected])



high-strain-rate ( ) to investigate strain-rate-sensitivity of

material. The constants of Johnson-Cook are found from

simulations of experimental data and found fitting for Johnson-

Cook expression. The different strain-rate experiments display

which, (Ti-6Al-4V) is sensitive alloy strain-rate. change in

strain-rates causes change strain-hardening behavior for (Ti-

6Al-4V). Increase in the temperature and strain-hardening

behavior led to decreases of yield strength also higher-

temperatures changes strain-hardening observation. High strain

rate effect on the shear-properties of (epoxy LY-556) with

(hardener HY-951) have been studied by N. K. Naik et al.

(2010) [10] used split-torsional-Hopkinson-bar machine for

experimental studies in shear-strain-rate range about (385–

880/sec). For comparison, investigations are conducted at

quasi-static device. It was observed which, shear-strength is

improved below high-strain-rates loading, compared to quasi-

static loading. Shear-strength improvement at shear-strain-rates

about 880/sec is 45% compared with that, at quasi-static

loading. G.-l. Nicolaescu et.al. (2015) [4] studied torsional

testing at the high-strain-rate using a kolsky bar. The test

specimen is made from CW614N Brass and is a short thin wall

tube. The results showed that the shear strain rate is about (

), the maximum shear strain in excess of (0.4) and the

maximum shear stress in excess of (300 MPa ). Also, the

maximum strain in excess of (0.2), the maximum stress in

excess of (615 MPa) and the strain rate in excess of ( ).

The aim of this study is design, and construction the split-

Hopkinson-torsion-bar machine, the most important

characteristic of the torsional-split-Hopkinson-bar machine is a

clamp system responsible of dynamic torsion process. Also, test

samples from different types of composite materials and

identify the dynamic torsional behavior for these materials

when subjected to shear stresses at high shear strain rates. In

addition, verify the experimental results by using Johnson-

Cook model.

2. EXPERIMENTAL WORK

2.1 Experimental Set-up:

In this work, the torsional-split-Hopkinson-bar is

designed for the dynamic behavior of various materials at the

high-shear-strain-rate. It consists of a clamp system, torque

generating mechanism, transmitter, and incident bars, the

brackets, fixing and data acquisition system that consist of: four

strain gages (foil type, 120 Ω, gage factor=2) bounded on the

incident and transmitted bars, Wheatstone bridge, amplifier

circuit, digital oscilloscope and personal computer. The TSHB

apparatus is shows in figure (1) and (2).

From two arms, the clamp system was consist which

are connect together by a load release pin at the top. Two large

bolts are used to tightened the clamp by pushes the lower ends

arms of the clamp together. After the desired torque is loaded

between the clamp and the torque wheel, the pressure of bolts

was increased until the load release pin is broken and the stored

torque will be released. The shape of the clamp system was

shown in figure (3). The operating of the clamp system depends

on the bolts instead of hydraulic pump or linear actuator utilized

in most machines. This allows flexibility for the testing, easy

operating mechanism and releasing the stored torque for test.

The torque generating mechanism consists of a torque wheel is

made of cast iron. Cast iron bushing is used to fix the incident

bar, is machined and welded into the torque wheel. The torque

wheel tightened to the incident bar by ten screws. A shaft is

welded on the torque wheel to help in turning it. A two

hydraulic jack, the first is used to rotating the torque wheel and

the second is used to fixing the torque wheel and prevent it from

movement after apply the angle of twist; the shape of the torque

generating mechanism was shown in figure(4). Five angle of

twist (4,6,8,10,12 )̊ are used in this work.

The brackets are the structural member that is used in the

setup to hold the incident and transmitter bars and to avoid the

bending. The brackets are made of two plates (low carbon steel)

are welding together, teflon bushing is used to support the bar

is machined and fitted into the bracket. To fix the transmitter-

bar and stop it from moving, the fixing is used which is made

of two plates (low carbon steel) are welding together, low

carbon steel bushing is used to fix the transmitter bar, is

machined and welded into the fixing. The incident, and

transmitted-bars are all made of 7075 aluminum-alloy.



Fig. 1. Photographs of the Torsional Split Hopkinson Bar



Fig. 2. Schematic of the Torsional Split Hopkinson Bar



Fig. 3. The Clamp System. a) Photograph. b) Schematic

Fig. 4. The Torque Generating Mechanism. a) Photograph. b) Schematic

2.2 Material properties:

In present work, random fiber glass, and woven kevlar fiber all with 25%, 40% and 55% with epoxy resin are used to

prepared samples. Table (1) describes properties of fiber glass, and kevlar fiber. Table (2) displays properties of epoxy resin.

Table I

List of Properties of Kevlar Fiber, and Fiber Glass [11,12]

Properties Kevlar fiber Fiber glass

Density (gm/ ) 1.44 2.54

Tensile strength (MPa) 3600 3450

Modulus of elasticity (GPa) 83 72.4

% elongation 4 4.8

Table II

Properties of Epoxy Resin (Quickmast-105)

Density

(g/ )

Tensile-strength

(MPa)

Compressive-strength

(MPa)

Flexural-strength

(MPa)

1.1 25 70 45

a) b)



2.3 Specimens Molding:

1. Measuring the epoxy resin by electronically balance.

2. A glass mold of (18 mm diameter and 140 mm length) with unlocked end is used in this work. Lubricating the pipe

from inside by oil and placing the fibers in glass-mold in homogenous shape to cover most sample volume, as shown

in figure (5).

5. Adding the epoxy with the hardener inside the molding.

6. The composite was left in mold for (6) hours at room-temperature to harden.

7. Lastly, test samples will be submitted to notching, cutting, and drilling processes.

Fig. 5. The Fibers in the Glass Mold

2.4 Specimens Machining:

The specimens are cut, notch and drill by using a lathe machine to get the final shape and dimensions of specimens. The

shape of the notch is square and place of the notch in the middle. The dimensions of a thin-walled cylindrical sample in this work

are 18 mm outer diameter (Do), 8 mm inner diameter (Di), 12 mm gage mean diameter (Ds), 14 mm overall length (Lt), 3.5 mm

gage length (Ls) and 2 mm wall thickness (ts); the final shape of specimen is shown in figure (6). In the previous work of TSHB,

there are different dimensions of specimen [4,8,10]. In this work, the dimensions of specimens are based on this different for

dimensions of the previous work.

Figure (6): The Final Shape of Test Specimen. a) Photograph. b) Schematic

3. CALCULATION OF THE SHEAR-STRAIN, SHEAR-STRESS, AND SHEAR-STRAIN-RATE:

The analytical solution to calculate shear-stress, shear-strain and shear-strain-rate of samples are [8]:

The expression for the specimen shear stress:

………………………. (1)

Where D is diameter of bar, G is modulus of shear of the bar, is gage thickness of the sample, is mean diameter of

gage of the sample and is the shear strain at the transmitter bar.

The expression for the specimen shear strain:

………………………. (2)

………………………. (3)

Here is shear wave speed in bar material, is the length of gage of the sample, is the density of the bar and is

the shear strain reflected to the incident bar.

The expression for the specimen shear strain rate:

…..…………………… (4)

a) b)



Thus, the shear stress- shear strain behavior of specimen is determined simply by measurement made on the elastic wave in

torsional-split-Hopkinson-bar. The above equations relate strain gauges measurement to shear strain shear stress behavior in the

specimens deforming and require that the wave within the bar must be one dimensional and the specimen must deform uniformly.

4. RESULTS AND DISCUSSION

4.1 Shear Stress, Shear Strain, and Shear Strain Rate: Figure (7) shows the relation between the dynamic shear strain and shear stress for different shear strain rates. From this figure,

it can be seen that, when the angle of twist increased from 4 ̊ to 12 ̊, the shear strain, shear stress and shear strain rate increased by

27.8%, 42% and 35.9%, respectively. These increases reason is due to the increasing in the stored energy at the bars section between

the systems clamp and the torque of wheel which led to increase in transmitted (released after the clamp system) energy to the

sample.

Figure (8) shows the effect of volume fraction of the random fiber glass-epoxy composite on the shear stress-shear strain

behavior at maximum and minimum angle of twist. From this figure at Ө=12 ̊, when the volume fraction increased from 25% to

40% , the shear strain rate, shear strain, and shear stress are increased by 18.9 %,12.4 %, and 12.5% respectively, as well when

the volume fraction increased from 40% to 50% , the shear strain rate, shear strain, and shear stress increased by 6.4%, 8.7%

and 5.5%, respectively. It is evident that, when the volume fraction increasing, the magnetic force between atoms will be increasing,

and the sample will be becoming more stiffer and strength. These results indicate that improvement of the dynamic shear behavior

and the properties of materials occur when the samples subjected to torsional load at high shear strain-rates.

Fig. 7. Stress-Strain curves of Epoxy

Fig. 8. Stress-Strain curves for Random Fiber Glass-

Epoxy-Composite with different Volume-Fraction



The shear stress versus shear strain curves for woven kevlar fiber -epoxy composite samples at minimum and maximum angle

of twist is shown in figure (9). It was appeared in the shape of the curves of shear strain -shear stress, the non- linearity increased

when the shear strain increases. From this figure, at θ=12 ̊, the percentage incremental in shear strain and shear stress with respect

to increase in volume fraction from 25% to 55% are 15.9% and 10.3%, respectively. Also, the incremental percentage in shear

strain rate with respect to incremental in volume fraction from 25% to 55% is 14.4%. The reason for these increases is due to

the increasing of the volume fraction which causes enhancement in the stiffness of samples.

Figure (10) shows the effect of maximum shear strain rate on the variations of maximum shear stress and maximum shear strain

for epoxy. From this figure, the increase in maximum shear strain and shear stress with shear strain rate is demonstrated clearly

because the increase in angle of twist led to an increase in transmitted energy. In addition, this material exhibited the maximum

shear strain and shear stress values at maximum shear strain rate of 425.2/s. The maximum shear strain and shear stress are 4.02 %

and 44.1 MPa respectively.

A B

Fig. 10. Variations of Maximum Shear Strain rate for Epoxy in: (A- Maximum Shear Stress, B- Maximum Shear Strain)

Fig. 9. Stress-Strain curves for Woven Kevlar Fiber-

Epoxy-Composite with different Volume-Fraction



The change in maximum shear strain and maximum shear stress for random fiber glass at different volume fraction show same

behavior of other cases, which is shown in figure (11). From figure, the shear strain and shear stress increased as the shear strain

rate increased due to increases in the stored energy which will be causes increasing in applied torque of the specimen. Moreover,

the maximum shear strain and shear stress values at maximum shear strain rate of 594.7/s, are obtained in the 55% random fiber

glass, the maximum shear stress and maximum shear strain are 60.8 MPa and 5.12 %, respectively.

1 2

Figure. 11. Variations of Shear Strain rate for Different Volume Fraction of Random Glass Fiber -Epoxy Composite in:(1- Shear-Stress, 2- Shear-Strain)

Figure (12) shows the relationship between the maximum shear stress, shear strain and rate of shear strain for woven kevlar fiber

composite. It is clear that, the increase in the volume fraction led to increases in shear strain rate due to the change in bonds and

morphology between atoms will be strong. Also, the maximum shear strain and shear stress values at maximum shear strain rate of

693/s, is obtained in the 55% kevlar fiber, the maximum shear strain and maximum shear stress are 5.83 %, and 80.4 MPa

respectively.

1 2

Fig. 12. Variations of Shear-Strain-rate for Different Volume Fraction of Woven Kevlar Fiber-Epoxy composite on: (1- Shear-Stress, 2- Shear-Strain)

4.2 Comparison between Types of Materials:

Figure (13) shows the relationship between the shear-strain-rate and maximum shear stress for pure epoxy composite, random

fiber glass-epoxy-composite and woven levlar fiber-epoxy-composite for 25, 40 and 55% volume fraction. It is noted that, the

increase of maximum shear strain, and shear stress with shear strain rate was demonstrated clearly because the increase in angle of

twist led to an increase in applied torque.



A B

C

Effect of shear-strain-rate on the maximum shear-strain for pure epoxy composite, random fiber glass-epoxy-composite and

woven kevlar fiber-epoxy-composite for 25, 40 and 55% volume fraction are shown in Figure (14). It is clear that, the 55% volume

fraction have higher shear strain and shear strain rate compared with those of 25 and 40% volume fraction. So, it is clear that, the

increase in volume-fraction led to increases in shear strain and shear strain rate.

A B

C

The shear stress versus shear strain curves for the epoxy in the present work and the epoxy in N. K. Naik et.al. [10] are shown

in figure (15).

Fig. 13. Comparison of Epoxy and Composite samples for

Effect of Shear Strain rate on Maximum Shear Stress at: (A-

25% Volume Fraction, B- 40% Volume Fraction and C-

55% Volume Fraction)

Fig.14. Comparison of Epoxy and Composite samples for

Effect of Shear-Strain-rate on Maximum Shear Strain at:

(A- 25% Volume Fraction, B- 40% Volume Fraction and C-

55% Volume Fraction)



The results show that, a reasonable agreement between the shear properties of epoxy in the present work and the shear properties

of epoxy in N. K. Naik et.al. [10]. In addition, the epoxy in the present work has similar behavior to that of the epoxy in N. K. Naik

et.al. [10].

5. CONCLUSIONS

1. Tests have been carried out with split-Hopkinson-bar machine specially designed for the dynamic torsional tests. The design

and construction of the TSHB with its new clamp system mechanism (two bolts) proved to be successful and permitted

conducting tests in an easy and simplified method.

2. The composite samples with 55% volume fraction of (random fiber glass, and woven kevlar fibers) give shear stress, shear

strain and strain rate greater than samples with 25 and 40% volume fraction.

3. The 55% volume fraction of the woven kevlar fiber-epoxy composite, and radom fiber glass-epoxy-composite improved the

shear stress by 82.3%, and 37.9%, respectively compared with pure epoxy at θ=12 ̊.

4. When the angle of twist increased from 4 ̊ to 12 ̊, shear strain, shear stress, and shear-strain-rate of (55% volume fraction)

woven kevlar fiber-epoxy composite are increased by 2.9%, 5% and 5.9%, respectively.

ABBREVIATIONS

Symbol Description Units

A Static yield stress MPa

B Hardening modulus MPa

C Rate sensitivity constant -----

Shear wave speed m/s

D Diameter of bars mm

Inner diameter of a sample mm

Outer diameter of thin walled specimen mm

Gage mean diameter of thin walled specimen mm

G Shear modulus in the bars GPa

Gage length of thin walled specimen mm

Overall length of thin walled specimen mm

n Hardening coefficient -----

wall thickness of thin walled specimen mm

Volume fraction -----

The strain %

The strain rate 1/s

Equivalent plastic strain -----

The reference strain rate 1/s

The dimensionless strain rate 1/s

The angle of twist Degree

The shear strain %

Fig. 15. Comparison of Epoxy in the present work and Epoxy in

N. K. Naik et.al. [10] for Shear Stress- Shear Strain

Behavior



The shear strain rate 1/s

Shear strain reflected to the incident bar %

Specimen shear strain %

Specimen shear strain rate 1/s

Shear strain in the transmitted bar %

The density of the bars 3g/cm

The stress MPa

The shear stress MPa

Specimen shear stress MPa

FSW Friction-Stir Welded -----

TSHP Torsional Split Hopkinson Bar -----

6. REFERENCES [1] K. Howard and M. Dana, "Mechanical Testing and Evaluation", ASM Handbook, vol. 8, ASM international, 2000.

[2] L. Fuentes, J. Gomez-Leon, and C. Trujillo, "Torsional Split Hopkinson Bar Optimization", B.Sc. Thesis, Department of Mechanical and Materials Engineering, Florida International University, USA, 2012.

[3] W. Chen and B. Song, "Split Hopkinson (Kolsky) Bar: Design, Testing and Applications", Springer Science & Business Media, LLC, 2011.

[4] G.-I. Nicolaescu, G.-A. Stefan, C. Enache and M. V. Carmaci, "Torsional Testing at High Strain Rates Using A Kolsky Bar", Land Forces Academy Review, vol. 20, no. 2, pp. 233-242, 2015.

[5] A. Gilat and C.-S. Cheng, "Modeling Torsional Split Hopkinson Bar Tests at Strain Rates above 10,000 ", International Journal of Plasticity,

vol. 18, no. 5, pp. 787-799, 2002.

[6] F. Barthelat, K. Malukhin and H. Espinosa, "Quasi-Static and Dynamic Torsion Testing of Ceramic Micro and Nano-Structured Coatings Using

Speckle Photography", Recent Advances in Experimental Mechanics, pp. 75-84, 2004.

[7] A. Gilat, R. K. Goldberg and G. D. Roberts, "Strain Rate Sensitivity of Epoxy Resin in Tensile and Shear Loading", Journal of Aerospace

Engineering, vol. 20, no. 2, pp. 75-89, 2007. [8] M. N. Bassim, "High Strain Rate Evaluation of Armor Materials", DRDC Valcartier CR, vol. 266, 2007.

[9] R. S. Yatnalkar, "Experimental Investigation of Plastic Deformation of Ti-6Al-4V under Various Loading Conditions", M.Sc. Thesis, The Ohio

State University, 2010. [10] N. K. Naik, R. Gadipatri, N. M. Thoram, V. R. Kavala and V. Ch, "Shear Properties of Epoxy Under High Strain Rate Loading", Polymer

Engineering & Science, vol. 50, no. 4, pp. 780-788, 2010.

[11] D. Gon, K. Das, P. Paul and S. Maity, "Jute Composites as Wood Substitute", International Journal of Textile Science, vol. 1, no. 6, pp. 84-93, 2012.

[12] G., Lubin, "Handbook of Composites", Van Nostrand Reinhold

Experimental Investigation of Composite Materials...

Documents

Transcript of Experimental Investigation of Composite Materials...