Keywords

1 Introduction

FRP is referred to Fiber Reinforced Plastics by using resin and reinforcing fiber in order to improve the strength of the product. The important characteristics of the FRP are good thermal insulation, resistance to heat, weather and chemical, electrical insulating and radio wave transmitting. FRP can be manufactured to various shapes, free color decorating, light weight and excellent strength. Therefore, it is widely used in variety applications such as aerospace industry, railway, automobile, construction industry sports and some medical fields [1].

The fabrication methods of the FRP are including hand lay-up method, spray-up method and press method. The most common and traditional method is the hand lay-up method [2]. A general process of this method is pre-placed reinforced fiber in a mold and then impregnated resin to the fiber using a brash or a roller by a hand. The hand lay-up method is used for preparing the high or low volume productions, which is presented an advantage on method of complex shape products. However, a disadvantage of this process is also found. That is the quality of the products will be affected by the skill of operators [3]. In our series of study, glass fiber reinforced plastic (GFRP) plate was created by the hand lay-up method. We examined the difference in skills of individual operators in this method [4].

In this study, we focused on preparing the GFRP plate by the hand lay-up method from the difference of an individual workmanship. The targets of this research are five of inexperience operators of hand lay-up molding [5]. The working procedure and the working time of each operator from the hand lay-up method were recorded using a video camera. Mechanical properties of GFRP plates were carried out by tensile testing. The relationships of the total working times and other parametents of fabrication and the strength of each specimen from each operator were elucidated.

2 Experimental Methodology

2.1 Subject

The targets of this research are five of inexperience operators of hand lay-up molding. One operator made three GFRP plates, which each specimen was made by duration of three days.

2.2 Hand Lay-up Method

The procedures for creating the GFRP plate are listed in the following.

  1. Step 1.

    Place reinforced fiber (glass fiber) in the mold plate.

  2. Step 2.

    Pour the resin into the mold plate.

  3. Step 3.

    Natural impregnated between the resin and the fiber.

  4. Step 4.

    Use a roller for impregnated of the resin and the fiber.

  5. Step 5.

    Degassing by using the iron bar.

  6. Step 6.

    Free the mold plate by putting the weight.

  7. Step 7.

    Wait for complete cure.

Figure 1 shows the hand lay-up mold plated used in this study. The hand lay-up method was used a roller in order to fabricated the resin on the smooth metal molded plate of 300 mm × 300 mm. After that 40 kg of the weight was applied in order to adjust the thickness and then the molded was left to cure for 24 h. In addition, the molded was put for 2 h in an oven at temperature of 100°C for completely curing of the resin. The targets of this research are five of inexperience operators of hand lay-up molding. One operator made three of GFRP plates, which each specimen was made by duration of three days. It measured the working time during the molding of each operator.

Fig. 1.
figure 1

Photograph of the hand lay-up mold plate

2.3 Measurements of Working Time

A video camera was used for recording the working of each operator. The camera was taken for the entire of making the GFRP plate until the laminating process was finished. The working time of the molding in each step was also recorded.

We focused on the impregnation of the resin and the fiber from the procedure of steps 3–5, which would impact on the properties of the molding. The working process significantly influenced on the properties of the GFRP molded article. The observations were included as the following.

  1. 1.

    Natural impregnation (sheet1).

  2. 2.

    Use the roller of the resin impregnation (sheet1).

  3. 3.

    Natural impregnation (sheet2).

  4. 4.

    Use the roller of the resin impregnation (sheet1).

  5. 5.

    Extrusion of resin.

2.4 Test Materials and Test Pieces

The resin was unsaturated polyester resin (RIGOLAC150HRBQTNW). The chopped glass fiber strand mat (M137, Owens Corning® Inc.) was used as the reinforced fiber. Kayamec M was used as curing agent. It was mixed at a ratio of 0.07 phr of the mass of the resin. The amount of resin was designated as 300 ml.

The size of the GFRP plate was 200 mm in length and 20 mm in width. Test pieces were prepared for three specimens.

2.5 Mechanical Properties

The tensile test was carried out by Intsron universal testing machine (INSTRON type 4206) at speed of 1 mm/min. The load cell was used at 10 tons. The gauge length was 100 mm. The load-displacement curve was illustrated, which was used for characterizing the molding technique of various operators. *1.

3 Experimental Results

3.1 Load-Displacement Diagrams

Figure 2 shows the load-displacement diagram of the tensile test of the GFRP specimens prepared at the first round of molding by each operator. Figure 3 presents the third round of molding from each operator. It can be seen that the load-displacement diagram of the first round molding showed the variation of each sample and the graphs were different from each operator as shown in Fig. 2. On the other hand, the load displacement curves from the third molding round by each operator presented some of an evenness of the displacement curves as illustrated in Fig. 3. These results would be used for indicating an experience and a skill of each operator in the hand lay-up method. Thus, the experience in the hand lay-up molding of the inexperienced person would be focused, especially on the technique for impregnating the resin. It can be said that it is difficult to improve their molding skill within three times of the molding.

Fig. 2.
figure 2

The load displacement curve from tensile testing of the GFRP at the first round of molding by each operator.

Fig. 3.
figure 3

The load displacement curve from tensile testing of the GFRP at the third round of molding by each operator.

3.2 Tensile Properties

The tensile strength of the GFRP as a function of the total working times is shown in Fig. 4. The total working times were focused on the step of natural impregnation, roller impregnation and degassing by the iron bar (extrude of resin). The working time of each research step obtained the variation of the tensile strength of the GFRP plates. From Fig. 4, we considered separately in the first half and the second half from the middle of the molding as illustrated in Figs. 5, 6, 7, 8, 9 and 10. Figure 5 shows the relationship of the tensile strength as a function of the degassing step. From this figure, the first half study (red line) showed a positive correlation while the second half study (blue line) showed a negative correlation between the tensile strength and the molding (working) time of the degassing step. It can be seen that the correlation coefficient of the second half was low. It might be attributed to unevenness of the molding from the operator.

Fig. 4.
figure 4

The tensile strength as a function of the working time for natural impregnation (sheet1)

Fig. 5.
figure 5

The tensile strength as a function of the working time for impregnation with roller (sheet1).

Figures 6 and 7 show the results from the step of natural impregnation in the sheet 1 (first layer) and the sheet 2 (second layer), respectively. On the contrary, the results from the step of roller impregnation of the sheet 1 and sheet 2 are presented in Figs. 8 and 9. In addition, we illustrated the relationship of the total working times and the tensile strength is presented in Fig. 10. From these results, it could be observed the irregularity of the tensile strength at various molding times of the natural as well as the roller impregnation and the total working times from the inexperience operators (Figs. 11, 12, 13, 14 and 15).

Fig. 6.
figure 6

The tensile strength as a function of the working time for natural impregnation (sheet2)

Fig. 7.
figure 7

The tensile strength as a function of the working time for impregnation with roller (sheet2).

Fig. 8.
figure 8

The tensile strength as a function of the working time for the degassing by iron bar

Fig. 9.
figure 9

The tensile strength as a function of the total working time

Fig. 10.
figure 10

The tensile strength as a function of the working time for the degassing by iron bar

Fig. 11.
figure 11

The tensile strength as a function of the working time for natural impregnation (sheet1).

Fig. 12.
figure 12

The tensile strength as a function of the working time for impregnation with roller (sheet1).

Fig. 13.
figure 13

The tensile strength as a function of the working time for natural impregnation (sheet2)

Fig. 14.
figure 14

The tensile strength as a function of the working time for impregnation with roller (sheet2).

Fig. 15.
figure 15

The tensile strength as a function of the total working time

3.3 Correlation Table of the Working Time and the Tensile Strength

Table 1 summarizes the correlation coefficient from the results of the working time (time of each process, the first half and the second half) and the tensile strength of each analysis step. The correlation coefficient of the natural impregnation and the roller impregnation was low. On the other hand, the step of the degassing process showed the higher values of the correlation coefficient. From the results in Table 1, it can be considered that the experience of the operator is important to obtain the regularity of the strength of the GFRP plate specimens. It can be noted that the evenness of the tensile strength was found from the step of the degassing, especially at the first half of the molding process. This degassing step was significantly influenced the mechanical strength of the GFRP molding products. It can be considered that air bubble would remain in the plate when the degassing time was low, which effected on the declination of the mechanical properties of the GFRP plate. Therefore, the degassing step with the iron bar was the most affected on the mechanical properties of the GFRP plate making by the inexperience operator. It can be noted that the fully degassing out of the molding product strongly suggested for the hand lay-up method in order to maintain the high strength of the GFRP products (Tables 2, 3 and 4).

Table 1. E,σ,δ First plate
Table 2. E,σ,δ Second plate
Table 3. E,σ,δ Third plate
Table 4. The correlation coefficient of the relationship between the tensile strength and the Conclusion.

4 Conclusion

The relationship between mechanical properties and the working time in the hand lay-up method was presented in this study. The inexperience operators in the GFRP plate by the hand lay-up method were selected. The tensile strength as the function of the working time of the GFRP plate molding was different in each analysis step of the natural impregnation, the roller impregnation and the degassing process. The correlation coefficient from the relationship between the tensile strength and the working time of each analysis step indicated that the degassing process was significantly influenced on the mechanical strength of the GFRP molding products. Therefore, the degassing step with the iron bar was the most affected on the mechanical properties of the GFRP plate making by the inexperience operator. It can be noted that the fully degassing out of the molding product strongly suggested for the hand lay-up method in order to maintain the high strength of the GFRP products.