Mekanika: Majalah Ilmiah Mekanika Surface Roughness and Fiber Angular Orientation Analysis Toward Laminated Composite Crack Propagation

the angular orientation of the carbon fiber and the surface roughness of the aluminum with the dependent variable response is the bridge crack rate. The manufacture of fiber metal laminates specimens uses the Vacuum Resin Infuse (VARI) method, which uses a vacuum pump as a means to flow the resin from the reservoir to the mold. This method is used to minimize the occurrence of air bubbles trapped on the specimen which causes porosity defects which will reduce the strength of the metal laminates specimen itself. Fatigue testing is performed using the stress amplitude method. is, the value of the load when the tensile test is one third of the tensile strength. After the fatigue test was carried out, the results were obtained on specimens with an angular orientation of 0/90 ° fibers, the crack propagation rate slowed down with a cycle value of 90000 in specimens with a surface roughness value of 2.128 µm then decreased cycles on specimens with a value of 2.887 µm, namely 11000 cycles.

Composite is a material that consisting of two or more materials, either micro or macro, where the properties of the material differ in shape and chemical composition from the original substance. In this study, fatigue testing of fiber metal composites was carried out to determine the rate of crack propagation so that the age of the fiber metal composite specimen was known. The independent variable in this research is the angular orientation of the carbon fiber and the surface roughness of the aluminum with the dependent variable response is the bridge crack rate. The manufacture of fiber metal laminates specimens uses the Vacuum Resin Infuse (VARI) method, which uses a vacuum pump as a means to flow the resin from the reservoir to the mold. This method is used to minimize the occurrence of air bubbles trapped on the specimen which causes porosity defects which will reduce the strength of the metal laminates specimen itself. Fatigue testing is performed using the stress amplitude method. That is, the value of the load when the tensile test is one third of the tensile strength. After the fatigue test was carried out, the results were obtained on specimens with an angular orientation of 0/90 ° fibers, the crack propagation rate slowed down with a cycle value of 90000 in specimens with a surface roughness value of 2.128 µm then decreased cycles on specimens with a value of 2.887 µm, namely 11000 cycles.

Introduction
In developments in the field of technical materials, most metals are processed into household utensils, tableware and drinking utensils, ornaments and weapons of war such as swords, armor and shields at the beginning of the discovery of metal by mankind, which was around 6000 BC [1]. However, with the development of science and technology, metals are then processed for industry. The automotive industry is one of the industrial fields that uses a lot of metal as raw material for its production. Such as in the manufacture of car frames, motorcycle frames, car bodies, auto and motorbike parts and components of motor vehicle engines. Besides being widely used in the automotive industry, metal is also widely used as a building construction material. Military defense is also an industry that uses metal as raw material for production, such as making tanks, cannons, rifles, and others.
The use of metals for engineering and industrial processes is indeed quite attractive as a raw material. Almost all industries use metal as raw material, but metal also has several drawbacks to be used as a material in engineering and industrial processes. The disadvantages of metal are that they are corrosive. To overcome the shortage of corrosive metals, composite has recently become a material that is widely used as raw material for production. Composites that have corrosion resistant properties and are generally lighter than metal are starting to be widely used as materials in industrial and engineering processes Composites are materials that are composed of a mixture of two or more materials that have different chemical and physical properties which will produce new materials that have different properties from the constituent materials [2]. The constituent material of the composite consists of four components, namely the matrix, the reinforcing material, the filler material and the additive material. Carbon fiber, carbon graphite or CF is a material which consists of very thin fibers about 0.005-0.010 mm and consists mostly of carbon atoms. The carbon atoms are bonded together with microscopic crystals that are parallel to the long axis of the fiber. This is what makes carbon fiber a very strong fiber for a fiber of its size. Several thousand carbon fibers are spun together to form a single thread, which can be used on its own or woven into cloth. [3] Carbon fiber has many different spun patterns and can be combined with plastic resin and molded to form composite materials such as carbon fiber reinforced plastic (CFRP) to make materials that have a high strength-to-weight ratio. The density of carbon fiber is also lower than that of steel, making it ideal for applications requiring lightweight materials. From this description, it shows that carbon fiber has the potential as a reinforcement or core in composites.
Fatigue or fatigue according to is defined as the process of permanent, progressive localized structural change in conditions that produce fluctuations in strain and stress below its tensile strength and at one point or many points that can peak into a crack or fracture as a whole after certain fluctuation [4]. There are three phases in the fatigue fracture, namely;

Crack Initiation
The fatigue mechanism generally starts from the crack initiation that occurs at weak material surfaces or areas where concentration occurs surface tension (such as scratches, notches, hole-pits etc.) due to presence repeated loading.

Crack Propagation
This crack initiation develops into microcracks. Propagation or the combination of these microcracks then forms the macrocracks that will leading to failure.

Fracture
Fracture occurs when the material has undergone a stress cycle and strain that produces permanent damage To find out the process of the three phases above, it is necessary to do a fatigue test. Fatigue testing can be used to see the resistance of a material to dynamic or repetitive loading where the load is below its yield strength. The fatigue strength is influenced by many variables, including: specimen size, specimen shape, final work, type of load / stress. By using a variable load or voltage, it produces different cycles, so that an S-N graph can be made. Fatigue testing is generally carried out by applying a uniaxial stress or dynamic load. The given dynamic level varies, it can be pull-pull, pull-compress and compress-compress.  Figure 3 shows a repeated tensile stress cycle with maximum stress (S max) and minimum stress (S min).
To produce the composites in this research uses the vacuum assist resin infusion method (VARI). Vacuum assisted resin infusion (VARI) is a method of making composite materials that uses low pressure applications to regulate the passage of resin into lamina. The material that becomes the matrix is placed in a mold, then a vacuum process is carried out to pull the resin flow into the mold.
The VARI method produces composite materials that have a high fiber-resin ratio compared to the hand lay-up method. The hand lay-up method uses the manual method to flow the resin, whereas in the VARI method the resin flow is carried out by a constant vacuum pressure. The use of this constant vacuum pressure which regulates the distribution of resin to remain in a certain amount. This results in a high fiberresin ratio resulting in a strong and lightweight composite material.
Some of the basic steps in the VARI process are as follows: 1. The fiber which functions as a filler is placed in a mold covered with a vacuum bag. 2. Liquid resin functions as a matrix poured in a container connected to the mold and vacuum machine 3. The air pressure inside the mold is lowered by the vacuum machine 4. The resin is dispensed under low pressure. 5. The curing process is carried out after the resin forms the lamina.

Volume 20 (1) 2021
The VARI method is classified into two types, namely the Surface Infusion method and the Interlaminar Infusion method [5]. In the surface infusion method, the resin is passed over the surface of the lamina, with the greatest disadvantage being the cost due to the preparation of machine operation, and the increasing complexity if this method is applied on a large scale. Whereas in the interlaminar infusion method, the resin is flowed through the space between the lamina. The Interlaminar Infusion method has many advantages when applied on a large scale. The resin flows between the laminae so that the thickness of the resin is maintained in the space between the laminae. In addition, the resin flow process is faster because it passes through a space that has been maintained in thickness. This more sustainable process also causes less waste material to be wasted.

Materials a. Carbon Fiber
The fiber used in this research is carbon fiber with twill woven type with the following specifications: For the matrix in this research using epoxy resin 174 / Eposchon-A and Versamide 140 / Eposchon-B (hardener) c. Aluminium for the outer structure of this material using aluminum. With a thickness of 0.5 mm and the type of aluminum is Al 1100.

• Independent Variable
The independent variable in this study is the angular orientation of the carbon fiber with the magnitude of the fiber angle as follows : 1. 0/90° 2. 45/45° Then the second independent variable is surface roughness ( ) with the magnitude of the surface roughness as follows : 1. The dependent variable in this study is crack propagation on fiber metal laminates (FML) composite.
• The manufacturing process of the composite specimen The steps for making the test specimen are following the standard D638-03 tensile test specimen and the ASTM D3479 fatigue test using laminate composites with the vacuum method as follows: 1. Prepare materials. (an aluminum plate that has been given surface roughness, carbon fiber that has been cut, epoxy, hardener, and composite making equipment using the vacuum method). 2. Lubricate the base/mold with grease to prevent the specimens from sticking when unloading. 3. Construct a composite FML structure with aluminum as the outer layer and as the core is a carbon fiber composite with a variety of fiber angles. 4. Prepare flow media and mesh with predetermined dimensions. 5. Arrange the FML structure on the base and wrap it with a vacuum bag. 6. Turn on the vacuum pump as a trial to make sure there are no leaks in the vacuum bag, if a leak occurs then the pressure on the pressure gauge will drop 7. Prepare epoxy resin and hardener with a ratio of 2: 1. 8. If there are no leaks, then start the process of suction the resin dough into the vacuum bag that has been prepared previously. 9. Drying the composite at a temperature that has been isolated in a vacuum bag. 10. After drying, the specimen can be unloaded. Figure 5 The manufacturing process of the composite using the VARI method.

• The Tensile Test Specimen
Tensile specimens according to ASTM D638 were made using the VARI process with the following dimensions : [8]  • The Fatigue Test Specimen Tensile specimens according to ASTM D3479 were made using the VARI process with the following dimensions : [9] Volume 20 (1) 2021 The crack propagation measure steps are as follows : 1. Apply load the FML specimen using a fatigue testing machine with a specified cycle. 2. After the cycle is reached in the first step, measure the cracks that occur using the dinocapture software 3. Perform the same steps for each subsequent cycle

Fatigue Test of Fiber Metal Laminates (FML)
The fatigue test in this study was carried out to observe the crack propagation rate of metal laminates fiber specimens for each variation of surface roughness and angular orientation of the fibers. By carrying out this fatigue test, the number of cycles data of the metal laminates specimen for each surface roughness variation and fiber angular orientation variation is shown in the following table : The results of fatigue testing of fiber metal laminates specimens with an angular orientation of 45 / 45ᵒ fibers are shown in the following table: In addition to knowing the data on the number of cycles of each fiber metal laminates specimen, this test also aims to determine the crack propagation of the specimen for each variation of surface roughness and fiber angular orientation. To find out the rate of crack propagation of fiber metal laminates specimens, it can be seen in the following sample data: Volume 20 (1) 2021 In the fatigue test, the point is to determine the rate of crack propagation of the specimen with a small but repetitive load, or material is subjected to cyclic loading. Then the two sides of the notch are measured and the average crack length is calculated (a).
To find the average crack length (a) for each cycle when testing is carried out, it can be calculated using the following formula.
So for the crack propagation rate (da / dN) in the 10000 th cycle of fiber metal laminates specimens under fatigue test is 0.0001897 mm / cycle. Likewise, several other cycles can be calculated using the same method to determine the rate of crack propagation.

Effect of Surface Roughness and Fiber Angle Orientation 0/90°
The discussion in this study is focused on the effect of aluminum surface roughness and carbon fiber angular orientation on the rate of crack propagation in fiber metal laminates specimens. To determine the effect of surface roughness and fiber angle orientation of 0/90° on the rate of crack propagation, it can be seen in the following figure: The graph in Figure 6 explains the relationship between the number of cycles to the crack length of the metal laminates specimen for each surface roughness variation (Ra) with the fiber angular orientation 0/90°. In the graph, it can be observed that specimens with a value of Ra 0.33 µm are fractured in the 25000 In Figure 6, the graph of the relationship between the number of cycles and the crack length of the metal laminates specimen is explained. In this study, in addition to knowing the relationship between the number of cycles and the length of the crack, it was also observed how the effect of surface roughness (Ra) and fiber angle orientation 0/90 ° on the rate of crack propagation. The fiber metal laminates specimen is shown in the following figure. Figure 9 Graph of the relationship between the average crack length (a) to the crack propagation rate (da / dN) of the fiber metal laminates specimens for each surface roughness variation with an angular orientation of 0/90° Figure 9 explained the relationship between the average crack length (a) and the crack propagation rate (da / dN) of the metal laminates specimens for each surface roughness variation (Ra) with fiber angular orientation 0/90 °. Apart from decreasing specimen age, surface roughness also affects the rate of crack propagation of fiber metal laminates specimens. Fiber metal laminates specimens with a value of Ra 0.33 µm, Ra 1.68 µm, Ra 1.78 µm, the higher the surface roughness value, the crack propagation rate decreased, but on the fiber metal laminates specimens with a value of Ra 1.93 µm to Ra 2,887 µm crack propagation rate was even faster By using the stress amplitude method in the fatigue test, the load received by the specimen depends on the tensile strength value for each variation of surface roughness. This is the basis for determining the loading value when fatigue testing is carried out. The value of fatigue loading can be determined by the following formula.   Table 3.3 shows that the higher the Ra value, the higher the tensile strength value as well as the higher the loading when fatigue testing is carried out. The number of cycles when testing increased for specimens with a value of Ra 0.33 µm, Ra 1.68 µm, and Ra 1.78 µm with loading values of 1.3 kN, 1.36 kN, and 1.7 kN, respectively. Then for specimens with a value of Ra 1.93 µm, Ra 2.128 µm, and Ra 2.887 µm with loading values above 1.7 kN, the number of cycles decreased. The loading value that is too high affects the fiber metal laminates specimen. The loading value that is too high makes the age of the material lower, which is indicated by the decreasing number of cycles during fatigue testing and the high rate of crack propagation.

Effect of Surface Roughness and Fiber Angle Orientation 45/45°
Two variables of fiber angular orientation were used in this study, namely the angular orientation of the fiber 0/90 ° and 45/45 °. To determine the effect of surface roughness and angle orientation of the 45/45 ° fibers on the crack growth of fiber metal laminates specimens, it is shown in Figure 8 below. To determine the effect of surface roughness (Ra) and fiber angle orientation 45/45° on the crack propagation rate of fiber metal laminates specimens can be seen in Figure 9 below. Volume 20 (1) 2021 Figure 11 Graph of the relationship between the average crack length (a) to the crack propagation rate (da / dN) of fiber metal laminates specimens for each surface roughness variation with fiber angular orientation 45/45° Figure 9 explained the relationship between the average crack length (a) and the crack propagation rate (da / dN) of the metal laminates specimens for each surface roughness variation (Ra) with the fiber angular orientation 45/45°. Variations in surface roughness in this study affect the cracking rate of fiber metal laminates specimens. In fiber metal laminates specimens with a fiber angle of 45/45°, the higher the surface roughness value or the coarser the aluminum surface, the bridge crack rate tends to increase. Then in specimens with a surface roughness value of 2.887 µm, the crack propagation rate was below the specimen with a surface roughness value of 1.93 µm and Ra 2.128 µm. This happens because the loading value during the fatigue test equates to the value of the specimen loading with a surface roughness of 2.887 µm with an angular orientation of 0 / 90ᵒ fibers. This is done because if the loading value is adjusted to the actual loading value of the specimen with a surface roughness of 2.887 µm and angular orientation of 4 5/45°, the crack propagation rate will be difficult to observe. After all, the data recorded is small. After all, the loading value is too large.
The loading value during the fatigue test of fiber metal laminates specimens with an angular orientation of 45/45° fibers based on the formula for calculating the loading values discussed previously can be seen in the following table. In table 3.4 it can be observed that the higher the surface roughness value, namely the specimens with a Ra value of 0.33 µm to Ra 2.128 µm, the higher the tensile strength value which causes the higher the loading value when the fatigue test is carried out. The higher the Ra value in the specimen with a fiber angle orientation of 45/45 °, the age of the specimen is the lower, this is because the load given to the specimen is also greater. Loads that are too large can affect the fiber metal laminates specimens which affects the lower the age of the material, which is indicated by the decreasing cycle value during fatigue testing and the high rate of crack propagation that occurs. However, the specimen with a Ra value of 2.887 µm, the cycle increased with a value of 60000 cycles because during this test the loading was adjusted to the specimen loading value of Ra 2.887 µm with a fiber angle orientation of 0/90 °. Because the load given is adjusted to the actual loading value of the specimen Ra 2.887 µm with an angular orientation of 45/45 ° fibers, which is 2.5 kN, the crack propagation rate cannot be observed because of the few data recorded.
The application of stress amplitude based on the results of this study is less suitable when applied to composites. It is possible that stress amplitude can be applied to materials that have high fatigue cycles or high fatigue cycles due to the ductility of these materials so that they are suitable for applying stress amplitudes. In crack propagation, the initial crack or crack initiation occurs in a material that is easily deformed plastic or material with low strength, after which the crack propagates until the material breaks or fails. In this study, the use of stress amplitude may be appropriate if the loading value is adjusted to the material that has the lowest tensile strength, namely aluminum alone without carbon reinforcement.

Delamination on Fatigue Test Fiber Metal Laminates Specimen
Delamination is a failure model in steel or composite materials with a lamina or layer structure. This failure is caused by a variety of things, for example, repeated cyclic loads, collisions (impact), or other influences that cause the layers to separate [6]. The easy-to-separate layer can significantly reduce the toughness of a material. Delamination can also occur because of the weak bond between the fibers and the matrix. Besides, the ability of the matrix to fill the space between the fibers also affects the delamination temperature. In this study, delamination affects the bond between layers of aluminum and carbon fiber composites in the fiber metal laminates specimens. The bond between these layers also affects the mechanical properties of the fiber metal laminates specimen. To detect delamination in a specimen, an observation is needed regarding the area of delamination that occurs in the specimen. The wider the delamination area, the lower the specimen strength and vice versa. [7] The area of delamination that occurs in the fatigue metal laminates test specimen was also observed in this study. To determine the delamination that occurs in the fatigue test specimen, the specimens that have been tested for fatigue are immersed in ink, then the area is observed with the help of ImageJ software. The image of the delamination area that occurs in the fatigue test specimen can be seen in the following figure. (sample delamination images measured using ImageJ software on FML specimens with angular orientation 0/90°) Volume 20 (1) 2021 Figure 13 Fiber metal laminates specimen with the value of surface roughness is 1,68 µm To determine the area of delamination that occurred in the specimen, it was measured using ImageJ software. The results of measuring the area of delamination on fatigue test specimens using ImageJ software can be seen in the following table. To observe the trend of how surface roughness affects the area of delamination it can be seen in the following graph. In addition to observing the delamination area of the specimen with an angle of 0/90° orientation, this study also observed the area of the specimen delamination with a 45/45° fiber angle orientation. The following is a picture of the delamination area that occurs in the specimen with an angular orientation of 45/45° fibers. (sample delamination images measured using ImageJ software) To determine the area of delamination that occurred in the specimen, it was measured using ImageJ software. The results of measuring the area of delamination on fatigue test specimens using ImageJ software can be seen in the following table. To observe the trend of how surface roughness affects the area of delamination it can be seen in the following graph The data that has been discussed shows the phenomenon that the surface roughness value of 0.33 µm in specimens with 45/45 fiber angular orientation has the highest cycle with a value of 440000.This shows that crack propagation at FML is affected not only because of surface roughness but also due to surface roughness. the stress concentration at the crack tip at the time of notching prior to the fatigue test. The rougher the surface, the greater the stress concentration value at the end of the notch, which is indicated by the larger area of delamination [10].