Manufacturing Fiberglass-Epoxy LSU-03 Aircraft Propeller Using Hand Lay-up and Vacuum Assisted Resin Transfer Moulding (VARTM) Methods

One of important aspect in designing a good quality product is on the method of how to manufacture it. In the world of aviation, products are required to meet several design criteria namely, strength, weight, and cost. Based on these design criteria, a study was carried out to manufacture LSU-03 propeller product sample using two manufacturing methods, i.e. hand lay-up and vacuum assisted resin transfer molding (VARTM). The objective is to design the manufacturing processes and to compare the quality of the propellers. The material used in this study are E-glass fiber cloth 135 and Epoxy “Lycal” resin. Prior to the propeller manufacturing process, several testing based on ASTM of manufactured composite plate was carried out to determine important parameters, i.e. thickness distribution, fiber volume fraction and voids content. A total of 3 propellers were manufactured. In order to determine the conformity of the product, propeller geometry is then measured and compared to the design mold. Finally, the comparison of the quality was made between the products of two methods. The quality of propeller products is measured based on calculated mass, thickness, void, fiber volume fraction, cost, surface quality and geometry conformity. It was concluded that VARTM is better than standard HLU and therefore more suitable for LSU-03 aircraft propeller production.


Introduction
In this study, the manufacturing process of propeller using Hand Lay-up (HLU) and Vacuum Assisted Resin Transfer Molding (VARTM) methods was carried out. The choice of this manufacturing method is due to its simplicity and cost competitiveness for manufacturing propeller parts.

Previous Work
Previous studies on composite manufacturing have examined variations in material properties, especially voids, fiber volume fractions, stiffness, and strength related to parameter of manufacturing process such as pressure, temperature, and other curing treatments [3][4][5][6]. Void is formed through several mechanisms. The resin flow advancing through woven or stitched fibrous preforms fills the empty regions between the fiber tows and within the fiber tows at different rates. This coupling between the two flow regimes is primarily responsible for incomplete filling of empty spaces within the fiber tows. These unfilled regions are what we term as microvoids. If the bulk flow front reaches the * Email: hendri.syamsudin@ae.itb.ac.id The International Conference on Aerospace and Aviation IOP Conf. Series: Materials Science and Engineering 645 (2019) 012018 IOP Publishing doi: 10.1088/1757-899X/645/1/012018 2 vent before covering a large region of the preform, a macrovoid or dry spot is created [7]. Unlike microvoids, dry spots can be visually detected. Microvoids can also be created due to entrapment of air during resin mixing operations or due to wrinkles or pockets that are created during the lay-up [8]. Voids can be entrapped in the inter-tow spaces and are called matrix voids, or in the intra-tow spaces and are called preform voids [9]. Void is formed due to either resin cannot enter the fiber spinning due to low fiber permeability and or high resin viscosity [2,9].

Hand Lay-up
HLU Manufacturing Process is the simplest manufacturing process. The resin is applied to the fiber layer until all the fiber parts are wetted. Then applying the second, third, and so on layers is carried out according to the desired number of layers. Finally, the curing process until the composite hardens. In some studies, parameter variations were carried out during the curing process. Antonio et al. [10] conducted a study of cure variations in the hand lay-up process. After the composite layer is impregnated, different treatments are carried out to achieve fully cured conditions. The results showed that curing by applying pressure resulted in a higher fiber volume fraction and a higher strength and smaller void volume. This is because by applying pressure it will push air bubbles (bubbles), which have a smaller pressure, out. Giving pressure will also suppress the composite layer so that the amount of resin will decrease so that the fiber volume fraction will rise.
In addition to the study parameters, a study of operator influence was carried out. This study was conducted in relation to the influence of the dominant operator during the impregnation process. A study by Kikuchi et al. [11] showed that the influence of the operator was unpredictable. Operators who have longer experience may not be able to perform the impregnation process faster and better than operators who do not have experience. This depends on the operator's consistency, including the direction of motion of the roll during the impregnation process, the pressure applied during the impregnation process, as well as the estimated total impregnation time by observing the gel time. With regard to strength and stiffness, the test results show that there are variations of up to 50%.

Vacuum Assisted Resin Transfer Moulding
VARTM Manufacturing Process utilizes the pressure difference in the impregnation process. With the difference in pressure, the resin can flow through the fiber layer that has been suspended. In the VARTM process, vacuum condition plays a vital role. The amount of vacuum pressure given will affect the flow rate of the resin which determines how quickly and effectively the impregnation process takes place. The flow rate which is too fast, resulting in the difficulty for the resin to wet the fiber. Vishwanath et al. [5] studied the effects of pressure, vacuum and temperature with polyester / Eglass material. It is stated that the impregnation time, which is influenced by the size of the vacuum, must be adjusted with the gel time polyester, so that the curing process does not occur when impregnation is still ongoing. Yenilmez et al. [13] recommend to keep vacuum condition during curing to reduce thickness gradients due to pressure differences. Modi [14] suggested to keep the resin flowing for some time after the impregnation finished. This allows fiber yarns, which have lower permeability than fiber webbing to continue the impregnation process. However, the study mentioned that how long or how much resin must be flowed is still difficult to determine. In addition to vacuum, there are several studies on the effect of pressure during curing in relation to layer thickness and fiber volume fraction. Pressure has a different effect when fiber is still dry, the impregnation process takes place, and the curing process. When the fiber is still dry, the pressure is limited to suppressing the fibers so that it is denser. This actually makes it difficult for the resin to flow through the fibers. During impregnation, the pressure could causes thickness variation in the composite layer, if there is a part that has not been moistened by resin or there are air bubbles that are trapped. At the curing process, the pressure will push the resin out so that the composite is densed and the fiber volume fraction rises.

Material dan Design
During manufacturing process, the type of fiber and resin used are the same for HLU and VARTM. Fiber type is E-glass fiber cloth 135, while the resin type is Lycal 1011. The dimensions of the propellers are following the same geometry as the original conventional wood propellers.

Hand Lay-up Process
The HLU process uses a brush to flatten the resin. After the resin has wetted the fiber, an emphasis is carried out to increase the fiber volume fraction.

Vacuum Assisted Resin Transfer Moulding Process
The pressure of vacuum infusion process is maintained at 70 mm Hg by adjusting the valve of the pump. During the infusion process, the following resin conditions are maintained. Firstly, the resin stock is enough to avoid incoming air bubbles. Secondly, the resin that comes out of the outlet point is kept to be free of air bubbles. This can be done by continuing to run the pump and opening the resin inlet for a while after all the fiber parts have been moistened.

Assembly
The assembly process is designed to join the upper and lower parts of the propeller. It is done by combining the upper and lower parts with the connection at the Trailling Edge (TE) and Leading Edge (LE), using cold bonding technique. The material used as the join is a woven E-glass placed at the orientation of ± 45.

Mass comparison
Using Krisbow KW0600377 Precision Scale, the results show that the propellers manufactured by the VARTM method are 24% lighter than HLU. Mass measurement results are in line with the amount of pressure given during the curing process. At VARTM the pressure given is about 2 atm while HLU is 1 atm. With higher pressure, the amount of resin will decrease and automatically the total mass will decrease.

Fibre Volume Fraction and Void Comparison
Good mechanical properties of composite structure can generally be known from low percentage of voids and high percentage of fiber volume fractions [4]. Measurement of void and fiber volume fraction refers to ASTM 2734-09 with the void measurement referring to ASTM 3171-99. Measurement of fiber volume fraction shows that VARTM has a fraction of 20% higher. While VARTM voids are 34% lower.

Geometry Conformity
Conformity to design geometry at the various position of the propeller is shown in the following figure. It can be seen that the two manufacturing methods achieve around 90% of design geometry.

Cost Comparison
Costs can be divided into investment costs and production costs. Investment costs are costs for equipment that can be used for a period of time, such as pump, jig, fixture etc . While production costs are the costs needed for production, such as consumable material, fibre, resin, salary, electricity, etc. During this study, for manufacturing three propellers, the cost of using VARTM method is double of  the HLU. However, it was found that the cost of production using VARTM is reduced with higher number of propellers.

Assembly
Generally good, there are only a few miss geometries in the LE and TE sections because the number of layers is too thick. With this condition, it is certain that there will be a miss aerodynamic calculation. One way to overcome this is by sanding.

Balancing
A good propeller must have a center of mass right on the shaft (center). This is to avoid mass imbalances which result in vibration when the propeller rotates. Continuous vibration can cause the loss of propeller power to cause the propeller to break. Also to maintain the balance of the lift force associated with symmetry between the sides of the propeller. In the LSU-03 propeller manufacturing, the product that has been completed at the assembly is carried out in a simple equilibrium analysis [15]. By using an iron plate to get the center of mass. And the results show that the center point of the propeller mass is around the shaft. This indicates that the propeller is balanced and can be used for further testing purposes (rotary / vibration test, aerodynamic test, and structural test).

Discussions
In the aerospace world, two most important aspects are quality and cost. The quality of VARTM products is superior. This can be seen from a lighter mass of around 24%, a lower void of about 34%, and a higher fiber volume fraction of around 20%. This is the main point in the assessment of composite products. VARTM products are lighter due to greater pressure during curing than the HLU process. Thus the amount of resin contained in the final product will be less. With the same amount of fiber, the VARTM product produces lighter mass. While voids are related to the presence of vacuum during the curing process. In the VARTM process, the pressure is maintained at a condition of around 0 atm so that the possibility of air being trapped in the composite layer is smaller. Which also reduces the possibility of voids. The higher fiber volume fraction in VARTM products is again due to the higher pressure on the curing process so that it reduces the amount of resin. With a fixed amount of fiber, the fiber volume ratio to the total composite volume will be higher.
In terms of production, namely costs, VARTM requires approximately 2 times the cost of HLU both for awall investment and the cost of producing datu. This is quite reasonable because VARTM method uses several materials that cannot be used repeatedly such as bagging films, breather, and peel ply if production is only done once. With mass production, there is a possibility that the VARTM method produces products with a cost that may be only 20% more expensive than HLU products. While in terms of geometry, both HLU and VARTM have a geometry match of about 90-95%. This indicates that the level of trust, especially the aerodynamic aspects of the propeller reaches 90-95%. In terms of mass balance, both VARTM and HLU have produced good products.
To determine the method suitable for the LSU-03 propeller manufacturing process, quality parameters (mass, mechanical properties, and geometry), production (cost), and ease of manufacturing processes are used [16].

Conclusions
In the composite manufacturing process, there are parameters that affect both the manufacturing process and the final product. In general these parameters are pressure and vacuum. Pressure has a role to drive the resin so that the fiber volume fraction rises. While vacuuming is related to the number of voids. The lower the vacuum, the lower the voids. In this study, the manufacturing process of fiber cloth 135 E-glass / Epoxy using material HLU and VARTM was carried out. The study results show that VARTM products are superior in terms of quality while HLU is superior in terms of costs and manufacturing processes. The suitability of HLU and VARTM products is quite good, in the range of 90-95%. The center point of both HLU and VARTM propeller mass is good. By conducting comparative and assessment studies covering the quality, cost, and other aspects of production with the point-point assessment method, it can be concluded that the VARTM method is a more suitable method for manufacturing aircraft propellers LSU-03.