Amit Ramji – A4 – University of Hertfordshire
Composite Lab Report
Composite Testing Lab undertaken on 22nd
October 2013
Group E
Amit Ramji
10241445
University of Hertfordshire - Aerospace Engineering
Year 4 – Mechanics and Properties of Materials - 6ACM0003
30th
October 2013
Amit Ramji – A4 – University of Hertfordshire
Introduction	
  
Composites in engineering are an alternative solution to traditional materials, where man-made fibres and natural fibres are
bonded into a matrix to provide overall directional strength, improved tensile and fatigue resistance due to the
imperfection removal processing involved when creating fibres. Firstly the fibre volume ratios are the main consideration
where one would investigate the overall area of fibres and volume of resin required to encapsulate the fibre into the
matrix.[1] The next consideration is made toward the fibre length where a critical length is established by the use of small
section analysis.[2] Further to fibre length are decisions made on alignment of fibres to the principal stress axis of the
application in question, this involves detailed stress analysis based on geometry of the part and loading conditions. Later
followed by configuring the laminate layup and fibre/multiple fibres of different materials and its resin/plastic matrix
based on the stress and stiffness requirements of the application.
The Rule of Mixtures is a quick approach to determine the composite layup required for the particular application. It is
generally a conservative approach which uses an estimate of 10% of the normalised composite strength in a direction of
loading at angles to the principal stress axis using the composite strength from the idealised aligned loading.[3] This
experimental investigation aims to verify the 10% Rule of Mixtures by physical testing of samples of differing layups of
16 ply’s of Carbon Fibre Reinforced Epoxy (CFRE-T800 / 0.4xEpoxy-924 & 0.6xCarbon Fibre).
Procedure	
  Test	
  A	
  –	
  3	
  Point	
  Bending	
  Test	
  
Firstly 4 samples of various layups of composites are selected in order to obtain experimental evidence of the 10% Rule of
Mixtures. The layup of samples selected are [0]8s, [90]8s, [+/-45]4s, [0 90 +45 -45]2s all of which contain flat laminar
layers. Secondly measure the dimensions of the sample in length, width and thickness in a minimum of 3 separate
locations and record the average. Next set up a tensile test apparatus in a reversed pull direction with a jig to enable the
samples to be tested under 3 point bending as shown in Figure 2 & Figure 1 below.
	
  
	
  
	
  
	
  
	
  
	
  
	
  
Load the sample as shown in Figure 1 and ensure to remove any backlash (movement in the direction of loading), zero the
load and deflection output and program the rate of loading [1mm/min]. Begin loading the sample until the deflection is
1mm from starting position and record the load being applied, repeat the loading in increments of 1mm until 5mm. Finally
repeat the above procedures and test for the remaining 3 composite layup combinations.
Procedure	
  Test	
  B	
  –	
  Tensile	
  Test	
  
Firstly record the dimensions of each sample in order to later obtain the cross sectional stress.
Later load a new sample of each composite layup into a tensile test apparatus while ensuring the jaws of the fixture are
pre-loaded in order to hold the sample under test. Begin loading of the samples behind the protective shield and test until
failure, record the failure load and record the type of failure. Repeat this procedure for the remaining 3 layups.
Results	
  Test	
  A	
  –	
  3	
  Point	
  Bending	
  Test	
  
Deflection
(mm)	
   [0]8s	
   [90]8s	
   [+/-­‐45]4s	
   [0	
  90	
  +	
  -­‐]2s	
  
	
  
[0]8s	
   [90]8s	
   [+/-­‐45]4s	
   [0	
  90	
  +45	
  -­‐45]2s	
  
0	
   0	
   0	
   0	
   0	
   Fibre	
  Direction	
   0	
   90	
   +45	
  -­‐45	
   0	
  90	
  +45	
  -­‐45	
  
1	
   73	
   4	
   9.4	
   28.8	
   Slope	
  p/v	
  (N/mm)	
   75	
   4	
   10	
   30	
  
2	
   147.4	
   8.2	
   19	
   58.6	
  
Supported	
  Length	
  
L	
  (mm)	
   125	
   125	
   125	
   125	
  
3	
   223	
   12.4	
   29.2	
   88.6	
   Width	
  b	
  (mm)	
   26.01	
   25.78	
   25.3	
   25.31	
  
4	
   298.4	
   16.6	
   38.6	
   118.2	
   Thickness	
  d	
  (mm)	
   2.08	
   2.07	
   2.24	
   2.24	
  
5	
   372	
   20.6	
   48.4	
   148.6	
  
Modulus	
  E	
  
(N/mm2)	
   155672	
   8847	
   16588	
   51062	
  
	
  
Modulus	
  E	
  
(GN/m2)	
   156	
   9	
   17	
   51	
  
Table 1 - Raw Data and Modulus Calculation
From the above Table 1, the modulus has been calculated by;  𝐸 =
!
!
!!
!"
!!!
!"
=
!
!
!!
!!!! , this is later compared by the Rule of
Mixtures for two layups in a single loading direction in the Analysis section.
Figure 2 - Test Set-up
Figure 1 - Specimen Set-up
Amit Ramji – A4 – University of Hertfordshire
Figure 3 - Collated Table of Raw Data
Results	
  Test	
  B	
  –	
  Tensile	
  Test	
  
[90]8s:  𝑤 = 17.92𝑚𝑚;   𝑡 = 2.06𝑚𝑚;   𝐴 = 𝑤𝑡 = 36.915𝑚𝑚!
;  𝐿!" ≈ 1𝑘𝑁   ∴ 𝜎 !" !!
=
!!"
!
= 27.1  𝑀𝑃𝑎	
  
[+/-­‐45]4s:  𝑤 = 10.14𝑚𝑚;   𝑡 = 2.17𝑚𝑚;   𝐴 = 𝑤𝑡 = 22.0𝑚𝑚!
;  𝐿!" ≈ 3.9𝑘𝑁   ∴ 𝜎 ±!" !!
=
!!"
!
= 177.2  𝑀𝑃𝑎	
  
[0	
  90	
  +/-­‐45]2s:	
   𝑤 = 9.98𝑚𝑚;   𝑡 = 2.15𝑚𝑚;   𝐴 = 𝑤𝑡 = 21.45𝑚𝑚!
;  𝐿!" ≈ 13𝑘𝑁   ∴ 𝜎 !  !"±!" !!
=
!!"
!
= 605.8  𝑀𝑃𝑎	
  
[0]8s:	
   𝑤 = 10.15𝑚𝑚;   𝑡 = 2.27𝑚𝑚;   𝐴 = 𝑤𝑡 = 23.04𝑚𝑚!
;  𝐿!" ≈ 57𝑘𝑁   ∴ 𝜎 ! !!
=
!!"
!
= 2473.9  𝑀𝑃𝑎	
  
Analysis:	
  From the results:	
  
Discussion	
  
Experimental	
  Errors:	
  From the testing, it is conclusive that the Rule of Mixtures when used against test data is accurate
and is generally conservative. Errors for the two layup’s and pull directions shown are minimal and is conclusively a fast
approach to calculate alternative layups/loading directions based on application requirements. Errors such as the 8.9%
found in the [+/-45]4s can be due to experimental errors in setting up the apparatus/sample, overshooting the deflection
selection, sliding of the measurement device within the tensile test apparatus. An alternative may be to use strain gauges
on a repeated investigation, which would provide a change in length at the extreme fibres, thus the stress and modulus can
be calculated.	
  
Layup	
  Inconsistency:	
  The composite layup plays a significant factor for inconsistencies, however repeated tests would
outline these errors. Manufacturing of composites involves layering of laminates in various fibre forms, which could leave
pockets of air and impurities hence causing inter-laminar failure during loading.	
  
Conclusions:	
  Using ABD matrices one can also see the same results as the test and 10% rule, however is a time consuming
task and requires information on detailed material properties of the matrix and its fibre interaction. Overall the Rule of
mixtures allows the composite designer to predict and size the composite based on the load requirements.	
  
Comparison	
  to	
  other	
  materials:	
  In comparison to other materials, composite solutions play an advantage in weight saving,
as the strength can be directionally oriented to that of the principle stress axis. Fatigue resistance is also much greater and
allows for manufacturing of components that usually could not be manufactured in single pieces.	
  
References	
  
[1] Messiry, M.E., Theoretical analysis of natural fiber volume fraction of reinforced composites. Alexandria Engineering
Journal, 2013. 52(3): p. 301-306.
[2] McGrath, J.J. and J.M. Wille, Determination of 3D fiber orientation distribution in thermoplastic injection molding.
Composites Science and Technology, 1995. 53(2): p. 133-143.
[3] Kim, H.S., On the rule of mixtures for the hardness of particle reinforced composites. Materials Science and Engineering: A,
2000. 289(1–2): p. 30-33.
Figure 4 - Failure Types from
Tensile tests
Amit Ramji – A4 – University of Hertfordshire

Composite Lab Report

  • 1.
    Amit Ramji –A4 – University of Hertfordshire Composite Lab Report Composite Testing Lab undertaken on 22nd October 2013 Group E Amit Ramji 10241445 University of Hertfordshire - Aerospace Engineering Year 4 – Mechanics and Properties of Materials - 6ACM0003 30th October 2013
  • 2.
    Amit Ramji –A4 – University of Hertfordshire Introduction   Composites in engineering are an alternative solution to traditional materials, where man-made fibres and natural fibres are bonded into a matrix to provide overall directional strength, improved tensile and fatigue resistance due to the imperfection removal processing involved when creating fibres. Firstly the fibre volume ratios are the main consideration where one would investigate the overall area of fibres and volume of resin required to encapsulate the fibre into the matrix.[1] The next consideration is made toward the fibre length where a critical length is established by the use of small section analysis.[2] Further to fibre length are decisions made on alignment of fibres to the principal stress axis of the application in question, this involves detailed stress analysis based on geometry of the part and loading conditions. Later followed by configuring the laminate layup and fibre/multiple fibres of different materials and its resin/plastic matrix based on the stress and stiffness requirements of the application. The Rule of Mixtures is a quick approach to determine the composite layup required for the particular application. It is generally a conservative approach which uses an estimate of 10% of the normalised composite strength in a direction of loading at angles to the principal stress axis using the composite strength from the idealised aligned loading.[3] This experimental investigation aims to verify the 10% Rule of Mixtures by physical testing of samples of differing layups of 16 ply’s of Carbon Fibre Reinforced Epoxy (CFRE-T800 / 0.4xEpoxy-924 & 0.6xCarbon Fibre). Procedure  Test  A  –  3  Point  Bending  Test   Firstly 4 samples of various layups of composites are selected in order to obtain experimental evidence of the 10% Rule of Mixtures. The layup of samples selected are [0]8s, [90]8s, [+/-45]4s, [0 90 +45 -45]2s all of which contain flat laminar layers. Secondly measure the dimensions of the sample in length, width and thickness in a minimum of 3 separate locations and record the average. Next set up a tensile test apparatus in a reversed pull direction with a jig to enable the samples to be tested under 3 point bending as shown in Figure 2 & Figure 1 below.               Load the sample as shown in Figure 1 and ensure to remove any backlash (movement in the direction of loading), zero the load and deflection output and program the rate of loading [1mm/min]. Begin loading the sample until the deflection is 1mm from starting position and record the load being applied, repeat the loading in increments of 1mm until 5mm. Finally repeat the above procedures and test for the remaining 3 composite layup combinations. Procedure  Test  B  –  Tensile  Test   Firstly record the dimensions of each sample in order to later obtain the cross sectional stress. Later load a new sample of each composite layup into a tensile test apparatus while ensuring the jaws of the fixture are pre-loaded in order to hold the sample under test. Begin loading of the samples behind the protective shield and test until failure, record the failure load and record the type of failure. Repeat this procedure for the remaining 3 layups. Results  Test  A  –  3  Point  Bending  Test   Deflection (mm)   [0]8s   [90]8s   [+/-­‐45]4s   [0  90  +  -­‐]2s     [0]8s   [90]8s   [+/-­‐45]4s   [0  90  +45  -­‐45]2s   0   0   0   0   0   Fibre  Direction   0   90   +45  -­‐45   0  90  +45  -­‐45   1   73   4   9.4   28.8   Slope  p/v  (N/mm)   75   4   10   30   2   147.4   8.2   19   58.6   Supported  Length   L  (mm)   125   125   125   125   3   223   12.4   29.2   88.6   Width  b  (mm)   26.01   25.78   25.3   25.31   4   298.4   16.6   38.6   118.2   Thickness  d  (mm)   2.08   2.07   2.24   2.24   5   372   20.6   48.4   148.6   Modulus  E   (N/mm2)   155672   8847   16588   51062     Modulus  E   (GN/m2)   156   9   17   51   Table 1 - Raw Data and Modulus Calculation From the above Table 1, the modulus has been calculated by;  𝐸 = ! ! !! !" !!! !" = ! ! !! !!!! , this is later compared by the Rule of Mixtures for two layups in a single loading direction in the Analysis section. Figure 2 - Test Set-up Figure 1 - Specimen Set-up
  • 3.
    Amit Ramji –A4 – University of Hertfordshire Figure 3 - Collated Table of Raw Data Results  Test  B  –  Tensile  Test   [90]8s:  𝑤 = 17.92𝑚𝑚;  𝑡 = 2.06𝑚𝑚;  𝐴 = 𝑤𝑡 = 36.915𝑚𝑚! ;  𝐿!" ≈ 1𝑘𝑁   ∴ 𝜎 !" !! = !!" ! = 27.1  𝑀𝑃𝑎   [+/-­‐45]4s:  𝑤 = 10.14𝑚𝑚;  𝑡 = 2.17𝑚𝑚;  𝐴 = 𝑤𝑡 = 22.0𝑚𝑚! ;  𝐿!" ≈ 3.9𝑘𝑁   ∴ 𝜎 ±!" !! = !!" ! = 177.2  𝑀𝑃𝑎   [0  90  +/-­‐45]2s:   𝑤 = 9.98𝑚𝑚;  𝑡 = 2.15𝑚𝑚;  𝐴 = 𝑤𝑡 = 21.45𝑚𝑚! ;  𝐿!" ≈ 13𝑘𝑁   ∴ 𝜎 !  !"±!" !! = !!" ! = 605.8  𝑀𝑃𝑎   [0]8s:   𝑤 = 10.15𝑚𝑚;  𝑡 = 2.27𝑚𝑚;  𝐴 = 𝑤𝑡 = 23.04𝑚𝑚! ;  𝐿!" ≈ 57𝑘𝑁   ∴ 𝜎 ! !! = !!" ! = 2473.9  𝑀𝑃𝑎   Analysis:  From the results:   Discussion   Experimental  Errors:  From the testing, it is conclusive that the Rule of Mixtures when used against test data is accurate and is generally conservative. Errors for the two layup’s and pull directions shown are minimal and is conclusively a fast approach to calculate alternative layups/loading directions based on application requirements. Errors such as the 8.9% found in the [+/-45]4s can be due to experimental errors in setting up the apparatus/sample, overshooting the deflection selection, sliding of the measurement device within the tensile test apparatus. An alternative may be to use strain gauges on a repeated investigation, which would provide a change in length at the extreme fibres, thus the stress and modulus can be calculated.   Layup  Inconsistency:  The composite layup plays a significant factor for inconsistencies, however repeated tests would outline these errors. Manufacturing of composites involves layering of laminates in various fibre forms, which could leave pockets of air and impurities hence causing inter-laminar failure during loading.   Conclusions:  Using ABD matrices one can also see the same results as the test and 10% rule, however is a time consuming task and requires information on detailed material properties of the matrix and its fibre interaction. Overall the Rule of mixtures allows the composite designer to predict and size the composite based on the load requirements.   Comparison  to  other  materials:  In comparison to other materials, composite solutions play an advantage in weight saving, as the strength can be directionally oriented to that of the principle stress axis. Fatigue resistance is also much greater and allows for manufacturing of components that usually could not be manufactured in single pieces.   References   [1] Messiry, M.E., Theoretical analysis of natural fiber volume fraction of reinforced composites. Alexandria Engineering Journal, 2013. 52(3): p. 301-306. [2] McGrath, J.J. and J.M. Wille, Determination of 3D fiber orientation distribution in thermoplastic injection molding. Composites Science and Technology, 1995. 53(2): p. 133-143. [3] Kim, H.S., On the rule of mixtures for the hardness of particle reinforced composites. Materials Science and Engineering: A, 2000. 289(1–2): p. 30-33. Figure 4 - Failure Types from Tensile tests
  • 4.
    Amit Ramji –A4 – University of Hertfordshire