Acta Polytechnica https://doi.org/10.14311/AP.2025.65.0493 Acta Polytechnica 65(4):493–499, 2025 © 2025 The Author(s). Licensed under a CC-BY 4.0 licence Published by the Czech Technical University in Prague VERIFICATION OF MECHANICAL PROPERTIES OF FORGED MOULDING COMPOUND COMPOSITE FOR SMALL STRUCTURAL PARTS OF MICROSATELLITES Ondřej Uher∗, Josef Kruml, Michal Král Výzkumný a zkušební letecký ústav a.s, Beranových 130, 199 00 Prague, Czech Republic ∗ corresponding author: uher@vzlu.cz Abstract. The mechanical properties of a short fibre composite produced using the forged moulding compound technology were tested. Firstly, these properties were estimated using analytical calculations and the Tsai-Pagani equation. Next, technological tests were conducted, based on which testing samples were made for mechanical tests. The results of the mechanical tests were then compared to calculations. Finaly, a microsatellite part was produced using this technology and integrated into the satellite structure. Keywords: Composite, BMC, short fibre, space, satellite, epoxy, carbon fibre. 1. Introduction Composite materials have found their way into many industries, including space exploration. However, their use here is limited to large structures for launchers and satellites. In small CubeSat-type satellites (mi- crosats) and similar devices, the composites are mainly used as a base material for printed circuit boards (PCBs). This is because composite processing tech- nologies favour large parts. For smaller parts found in the construction of smaller satellites, aluminium alloy parts made by chip machining are preferred. In recent years, the development of composite process- ing and production technologies has made it possible to achieve competitive mechanical properties even for composites with short to medium lengths of non- oriented fibres. In general, composites can be divided according to the orientation of the reinforcement fibres into ori- ented and non-oriented reinforcement, and further according to the length of the reinforcement, into con- tinuous and discontinuous. When producing smaller parts, it is possible to achieve competitive parame- ters of mechanical properties by using discontinuous non-oriented reinforcements. This makes it possible to produce parts with complex shapes. In terms of mechanical properties, forged moulding compound (FMC) technology appears to be the most effective. “Forged carbon” (FMC) technology is related to Bulk Moulding Compound (BMC). Unlike BMC, where the reinforcement is mixed with the matrix, the resulting mass is placed in a mould, and then pressed [1], causing the material mixture to flow within the mould. This creep then straightens the fibres in the direction of the flow and the resulting compos- ite does not have isotropic or orthotropic properties. The aim of FMC is to reduce the flow of material in the mould to a minimum and thereby achieve nearly isotropic mechanical properties. In the case of both Material Tensile strength [MPa] Aluminium alloy 304 Forged carbon 192 Table 1. Maximum tensile strength according to [2]. technologies, the impregnated reinforcement is placed in the mould, which is then closed in the press un- der increased pressure until the matrix is cured. The mould is often heated to speed up the curing process. The closing of the mould and pressing takes place gradually, thereby achieving a gradual extrusion of the excess matrix from the mould, but not the rein- forcement. Knowing the volume of the manufactured part makes it possible to determine the amount of reinforcement and matrix to obtain the target volume fraction in advance, enabling a relatively accurate control. The technology described in [3] is a similar approach, but in the case of FMC, the reinforcement is discontinuous, randomly oriented, and manually placed. 2. Prediction of mechanical properties Based on the data from the literature, we expect the modulus of elasticity to be about 10–15 % lower than that of aluminium alloys. The expected tensile strength is presented below in Table 1. The estimation of mechanical properties can be performed using many analytical formulas such as Tsai-Pagani equation, see below. Although these the- oretical calculations can provide a good initial estima- tion for engineering works, material testing is always decisive. Using the Tsai-Pagani equation [4], we can predict tensile modulus E using the following process: E = 3 8E11 + 5 8E22, (1) 493 https://doi.org/10.14311/AP.2025.65.0493 https://creativecommons.org/licenses/by/4.0/ https://www.cvut.cz/en O. Uher, J. Kruml, M. Král Acta Polytechnica where parameters E11 and E22 are obtained from Equations (2) and (3), respectively, as shown below: E11 = 1 + 2 lf df ηLVf 1 − ηLVf Em, (2) E22 = 1 + 2ηT Vf 1 − ηT Vf Em, (3) where parameters ηL and ηT are obtained from Equa- tions (4) and (5), respectively: ηL = Ef Em − 1 Ef Em + 2 lf df , (4) ηT = Ef Em − 1 Ef Em + 2 , (5) where: Em is the tensile modulus of matrix, Ef is the tensile modulus of the fibres, lf is the fibre length, df is the fibre diameter, Vf is the volume fraction of fibres. Calculation example for a reinforcement length of 6 mm: Matrix: Hexion MGS285 (Em = 3.2 GPa), Car- bon fibre: Toray T300 (Ef = 230 GPa), lf = 6 mm, df = 7 µm a Vf = 0.6 [4]: ηT = 230 3.2 − 1 230 3.2 + 2 = 0.959, [4] (6) ηL = 230 3.2 − 1 230 3.2 + 2 6 0.007 = 0.040, [4] (7) E22 = 1 + 2 × 0.959 × 0.6 1 − 0.959 × 0.6 × 3.2 = 16.222 GPa, (8) E11 = 1 + 2 ( 6 0.007 ) 0.040 × 0.6 1 − 0.040 × 0.6 × 3.2 = 137.067 GPa, (9) E = 3 8137.067 + 5 816.222 = 61.539 GPa. (10) The calculation is analogous for a reinforcement length of 12 mm. See Table 2 for calculated tensile modules. 3. Technological tests The technological tests began with the design and 3D printing of a closed mould made from PETG material. The mould consisted of four parts: the lower part, two side parts and the upper part (see Figure 1). The upper part, which has a bevel, was closed. After 3D printing, a wax-based release agent was applied to the mould in several layers, then the mould was assembled and ready for the production of the first part. The first parts were made from Hexion MGS 285 epoxy resin and 3 mm chopped glass reinforcement. A target fibre volume fraction (FVC) of 60 % was Reinforcement Tensile length [mm] modulus [GPa] 6 61.539 12 61.950 Table 2. Calculated tensile modules for different lengths of reinforcement fibres. Figure 1. Disassembled mould. chosen, for which the exact volume and weight of the required reinforcement portion were then deter- mined, and subsequently the amount of matrix was determined as 125 % of the reinforcement weight. The reason was to achieve the requested shape accuracy and surface quality of the part. Thanks to the gradual increase in pressure when closing the mould and the bevel on the closing part, the excess matrix is allowed to flow out of the mould. First, a thin layer of matrix was first applied to the bottom of the mould to wet its surface. Then about one-fifth of the dry reinforcement was inserted into the mould, which was then covered with a propor- tional amount of the matrix. After several repetitions, all the reinforcement and the matrix were uniformly distributed in the mould. This was followed by closing the upper part of the mould and gradually increas- ing the closing pressure. The pressure was gradually increased in 5–10 minute steps until the mould was completely closed, which took about 30 minutes. Af- ter 24 hours, the matrix had cured sufficiently (i.e. it had achieved handling strength) and the mould was disassembled and the part cleaned, see Figure 2. 494 vol. 65 no. 4/2025 Verification of mechanical properties of forged moulding compound . . . Figure 2. Process samples with fiberglass reinforcement, 70 % FVC (top) and 60 % FVC (bottom). Sample No. 1 2 Theoretical FVC [%] 60 70 Actual FVC [%] 56.3 62.8 Density [kg m−3] 1 875.6 1 961.2 Void content [%] 6.5 5.3 Table 3. Results of technological tests. The actual achieved fibre volume content (FVC) was measured. For the first samples with a theoretical FVC of 60 %, the actual fibre volume content was found to be 56.3 %. The second attempt followed with a theoretical fibre volume fraction of 70 %, where the actual fibre volume content was found to be 62.8 %. For detailed results see Table 3. The next step were technological tests with carbon fibres measuring 6 and 12 mm in length. Unfortu- nately, insufficient amount of the matrix caused cavi- ties in these samples. For that reason, the amount of the matrix was increased to 130 % for the next attempt. However, even the increased amount of matrix did not help. The manufacturing process was thus modified, and the matrix was mixed with the reinforcement prior to being placed into the mould. This modifi- cation improved the result, the surface quality, and decreased the void content. After adjusting the man- ufacturing process, the samples for mechanical tests were produced (see Figure 3 for freshly demoulded sample as well as cleaned sample): 5 samples with a reinforcement length of 6 mm and 5 samples with a reinforcement length of 12 mm. In the Figures 4 and 5 respectively are shown samples with 6 mm and 12 mm length of reinforcement before and after the testing was performed. 4. Mechanical testing The mechanical tests were performed using the In- stron 55R1185 universal tensile machine equipped with a 10 kN load cell, and strain was measured us- ing the Instron 3560 extensometer. The tests were performed according to ISO 527-2 standard, i.e. with a loading speed of 1 mm s−1, test temperature of 22.4 °C, and a humidity level of 31.1 % RH. 4.1. Test results Individual test results for both types of the test sam- ples are listed in the following Tables 4 and 5. 4.2. Evaluation of test results The results show a considerable variance. The signifi- cant variation in the measured mechanical properties is probably due to the large differences in fibre ori- entation distribution between the test samples. It is given by the manufacturing process used, which does not allow it to be controlled. The greater vari- ation coefficient (C.V.) for the 12 mm reinforcement (30.4 %) compared to the 6 mm reinforcement (24.4 %) also supports this hypothesis. The distribution of the orientations of the shorter fibres appears to be closer to that of an isotropic material than the longer ones. The other important factor is also the variation in FVC between the individual test samples. In order to identify the volume fraction and void content of the test samples, corresponding tests were carried out on two samples from each set. The target FVC of 60 % was not achieved, with a considerable variation in the values obtained for each set. The results are summarised in the following tables. They also contain a comparison of the measured and calcu- lated E-moduli of the individual test samples. The E-moduli were calculated using the approach shown in Section 2 for the measured FVC. For detailed results of test samples see Tables 6 and 7. The following conclusions were drawn from the anal- ysis of the obtained data: • It was expected that a higher modulus would corre- spond to a higher strength. However, this relation- ship was not demonstrated. Although the detailed analysis of the test samples showed the expected tendency, some of the other samples for a given length of reinforcement exhibited the reverse phe- nomenon. This is probably due to the nature of the randomly distributed reinforcement, which is of finite length and the random overlap of fibres. The modulus measurement was performed using an extensometer with a base of 50 mm, which was placed approximately in the middle of the sample. If sample failure occurs outside this area, the mod- ulus and strength may be completely independent, 495 O. Uher, J. Kruml, M. Král Acta Polytechnica Figure 3. Carbon reinforced samples, 6 mm long fibres (top), 12 mm long fibres (bottom), not yet cleaned after removal from the mould. (a). Before test. (b). After test. Figure 4. Specimens with 6 mm reinforcement before test and after test. (a). Before test. (b). After test. Figure 5. Specimens with 12 mm reinforcement before test and after test. 496 vol. 65 no. 4/2025 Verification of mechanical properties of forged moulding compound . . . Sample Failure Width Thickness E modulus Poisson Rm ID type [mm] [mm] [GPa] ratio [-] [MPa] 1 L2 11.52 3.98 41.14 0.178 98.8 2 L0 11.85 3.95 37.54 0.249 94.7 3 Grip 11.79 3.95 42.26 0.448 158.9 5 L2 11.88 4.06 42.58 0.247 106.2 6 L0 11.72 4.03 35.59 0.378 146.7 Avg. 11.75 3.99 39.82 0.300 121.1 S.D. 0.143 0.05 3.097 0.110 29.6 C.V. 1.22 1.28 7.78 36.73 24.4 Min. 11.52 3.95 35.59 0.178 94.7 Max. 11.88 4.06 42.58 0.448 158.9 Table 4. Tensile test result of the samples with 6 mm reinforcement. Sample Failure Width Thickness E modulus Poisson Rm ID type [mm] [mm] [GPa] ratio [-] [MPa] 1 L0 12.55 3.88 31.89 0.335 72.0 2 L0 11.87 3.95 64.72 0.117 65.0 3 L0 11.74 4.07 42.28 0.420 134.1 4 L0 11.87 4.13 64.32 0.019 87.4 5 L0 11.99 3.99 57.55 0.426 85.1 Avg. 12.01 4.00 52.15 0.264 88.7 S.D. 0.318 0.10 14.52 0.185 27.0 C.V. 2.65 2.43 27.84 70.20 30.44 Min. 11.74 3.88 31.89 0.019 65.0 Max. 12.55 4.13 64.72 0.426 134.1 Table 5. Tensile test result of the samples with 12 mm reinforcement. Sample Density FVC Void E modulus [GPa] Rm [kg m−3] [%] content [%] Measured Calculated [MPa] 6 mm / 1 1 501.5 57.02 3.05 41.14 58.10 98.8 6 mm / 2 1 469.1 51.31 2.85 37.54 51.80 94.7 Table 6. Parameters of 6 mm samples. Sample Density FVC Void E modulus [GPa] Rm [kg m−3] [%] content [%] Measured Calculated [MPa] 12 mm / 4 1 456.6 51.24 3.86 64.32 52.15 87.4 12 mm / 5 1 515.6 58.76 2.76 57.55 60.51 85.1 Table 7. Parameters of 12 mm samples. however, this is also the case when failure occurs in the measured area. As the measured modulus is calculated from the strain of the given section, it is not possible to distinguish local weak points related to reduction in strength. • An interesting phenomenon is also the different average moduli for both lengths of the reinforcement. The 6 mm reinforcement shows a significantly lower modulus (about 40 GPa) compared to the 12 mm reinforcement (about 50 GPa) at the same FVC. The modulus of the longer fibres is also closer to the predicted modulus values. It is probably due to the orientation of fibres, where the longer fibres were oriented along the longer side of the test sample, while the 6 mm were oriented in multiple directions. • It was also expected that the effect of porosity on strength would also have a major impact, as the presence of air bubbles in the material can reduce the load transfer from fibre to fibre. However, for the measured range of porosity, it did not have a major impact on the strength value. 497 O. Uher, J. Kruml, M. Král Acta Polytechnica 5. Verification of the production of the satellite structural part The next step was to verify the feasibility of manu- facturing more complex parts. A structural partition of microsatellite construction was chosen, which is usually milled from an aluminium alloy. Similarly complex parts, produced using the same FMC tech- nology, are used, for example, in sports cars [5]. The design of the mould consists of 6 components due to the complexity of the part. The mould is divided to the base, 4 side components, and the cover. Figure 6 shows the mould already filled with a mixture of re- inforcement and matrix. Nine threaded inserts were placed around the perimeter of the part during the fill- ing of the mould. The same type of reinforcing fibres as in previous tests were used and a length of 12 mm was chosen. The matrix system was also the same as that used in previous test, i.e. Hexion MGS 285 epoxy matrix. The target fibre volume content was 55 %. Once the mould had been filled, the cover was in- stalled, the entire assembly was placed in a press, and pressed until the matrix had achieved handling strength, i.e. for about 24 hours. Subsequently, the part was post-cured at an elevated temperature of 60 °C for 2 hours. Finally, the part was deburred and cleaned, as can be seen in Figure 7. The finished parts suffered from surface defects, some of these can be seen in Figure 7. However, these can be fixed after demoulding and are not of criti- cal nature. The fibre volume content was not tested for these parts. A mass saving of around 46 % was achieved at around 30 % of stiffness loss compared to the aluminium variant, assuming that target value of 55 % was achieved. These benefits/losses can be fur- ther optimised by changes in the FMC part geometry. The two manufactured parts were implemented in the prototype of a semi-composite microsatellite struc- ture, which is being developed as part of VZLU re- search activities as shown in the Figure 8. 6. Conclusion This research involved estimating the mechanical prop- erties of the FMC composite using analytical and experimental methods. Initial screening of the me- chanical properties was necessary to verify if the imple- mentation of FMC composite into the microsatellite design is feasible and has a positive contribution. The production of 3D printed moulds was verified during the work on producing test samples and prototype parts. The results of the mechanical tests showed con- siderable variance and deviations from the theoretical values. The large variance of the measured mechanical properties is probably due to the large differences in the fibre orientation distribution between the test sam- ples. It is given by the manufacturing process used, which does not allow it to be controlled. Nevertheless, the manufacturing of the microsatellite prototype part Figure 6. The mould filled with a mixture of rein- forcement and matrix. Figure 7. The mould filled with a mixture of rein- forcement and matrix. Figure 8. Parts in the assembly of semi-composite microsat. 498 vol. 65 no. 4/2025 Verification of mechanical properties of forged moulding compound . . . has shown significant potential for weight and cost savings for complex shaped parts manufactured using the FMC technology. There are two possible ways to enable the use of this manufacturing process for microsatellites. The first option is to considerably increase the number of test specimens (more than 15) in order to obtain statistically reliable design allowances. The second option is to improve manufacturing the process to be more predictable. One of the possibilities is to combine the continuous reinforcement and short fibres. The continuous reinforcement (fabric or UD) could be used in flat or simply curved sections and serve as the load bearing member. The short fibres can then be used to fill complex shaped sections, providing bending stiffness and enable the use of inserts for mechanical joints with a surrounding structure. References [1] P. Malnati. Forged molding compound: Extending SMC capabilities, 2021. [2022-12-05]. https: //www.compositesworld.com/articles/forged- molding-compound-extending-smc-capabilities [2] Easy Composites. Comparing the mechanical properties of forged carbon fibre, 2022. [2022-05-15]. https://www.easycomposites.eu/learning/ mechanical-properties-of-forged-carbon-fibre [3] V. Haguenauer, E. Becker, R. Bigot, et al. Forging C/Thermoplastic printed composite, shaping parameters impact. Procedia Manufacturing 47:169–173, 2020. https://doi.org/10.1016/j.promfg.2020.04.165 [4] J. Biagiotti, S. Fiori, L. Torre, et al. Mechanical properties of polypropylene matrix composites reinforced with natural fibers: A statistical approach. Polymer Composites 25(1):26–36, 2004. https://doi.org/10.1002/pc.20002 [5] P. Feraboli, F. Gasco, B. Wade, et al. Lamborghini “Forged Composite®” technology for the suspension arms of the Sesto Elemento, 2011. 499 https://www.compositesworld.com/articles/forged-molding-compound-extending-smc-capabilities https://www.compositesworld.com/articles/forged-molding-compound-extending-smc-capabilities https://www.compositesworld.com/articles/forged-molding-compound-extending-smc-capabilities https://www.easycomposites.eu/learning/mechanical-properties-of-forged-carbon-fibre https://www.easycomposites.eu/learning/mechanical-properties-of-forged-carbon-fibre https://doi.org/10.1016/j.promfg.2020.04.165 https://doi.org/10.1002/pc.20002 Acta Polytechnica 65(4):493–499, 2025 1 Introduction 2 Prediction of mechanical properties 3 Technological tests 4 Mechanical testing 4.1 Test results 4.2 Evaluation of test results 5 Verification of the production of the satellite structural part 6 Conclusion References