- Email: [email protected]
Experimental study on barely visible impact damage and visible impact damage for repair of small aircraft composite structure
Aerospace Science and Technology 29 (2013) 363–372
Contents lists available at SciVerse ScienceDirect
Aerospace Science and Technology www.elsevier.com/locate/aescte
Experimental study on barely visible impact damage and visible impact damage for repair of small aircraft composite structure Hyunbum Park a,∗ , Changduk Kong b a b
Department of Defense & Science Technology – Aeronautics, Howon University, 64 Howondae 3gil, Impi, 573-718 Gunsan, Republic of Korea Department of Aerospace Engineering, Chosun University, 375 Seosukdong, Donggu, 502-759 Gwangju, Republic of Korea
a r t i c l e
i n f o
Article history: Received 5 January 2012 Received in revised form 25 November 2012 Accepted 18 April 2013 Available online 23 April 2013 Keywords: Composite material Impact damage Structural design Finite element analysis
a b s t r a c t This work is focusing on the low velocity impact damage evaluation and the external patch repair techniques of carbon/epoxy UD and fabric laminate adopted developing aircraft. The impact damages of composite laminates of the carbon/epoxy UD and fabric are simulated by the drop-weight type impact test equipment. The damaged specimens are repaired using the external patch repair method after removing the damaged area. The compressive strength test and analysis results of the repaired impact damaged specimens are compared with the compressive strength test and analysis results of the undamaged specimens and the impact damaged specimens. Finally, through investigation of compressive strengths of the damaged specimens at different environmental conditions, the damage criteria for repairable design of both the impact damaged UD and fabric laminate structure are suggested. © 2013 Elsevier Masson SAS. All rights reserved.
1. Introduction Recently, as the utilization of small aircraft has been increasing, many countries have been developing various small aircraft. In Korea, the KC-100, which is a small scale piston propeller general aviation aircraft, has been developed to establish a domestic certificate infrastructure and system through the BASA (Bilateral Aviation Safety Agreement) program by KAI (Korea Aerospace Industries, Ltd.). Fig. 1 shows conceptual design result of KC-100 for BASA. The KC-100 aircraft was designed with a complete composite structure for structure weight reduction. However the carbon/epoxy composite structure, which is mainly used for this aircraft, is susceptible impact damage. Especially, the low velocity impact damage cannot be easily found during visual inspections. Thus the low velocity impact damage of the composite structure has become an important issue in composite structural design. Because the composite structure not only is weak to external impact damage but also has a different repair method from existing metallic materials, the repair method for the impact damage of composite structure is very important for certification of developing aircraft. According to the literature survey, several engineers have been performing various studies of low velocity impact damage and repair on the composite structure. In 1998, Eric J. Herup et al. performed the study on low-velocity impact damage initiation in graphite/epoxy/Nomex honeycomb-
*
Corresponding author. E-mail address: [email protected] (H. Park).
1270-9638/$ – see front matter © 2013 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.ast.2013.04.007
Fig. 1. Conceptual design configuration of Korean Civil Aircraft-100 (KC-100).
sandwich plates [5]. In this study, low velocity impact and static indentation tests on sandwich plates have been performed to characterize damage initiation as a function of face sheet thickness and loading rate. In 1999, Milan Mitrovic et al. studied the effect of loading parameters on the fatigue behavior of impact damaged composite laminates [10]. The long term mechanical fatigue of quasiisotropic graphite/epoxy laminates was investigated to determine the influence of loading parameters on impact induced delamination growth during constant amplitude and spectrum fatigue loading.
364
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
Fig. 2. Schematic diagram showing design load levels versus categories of damage severity.
J.J. Schubbe and S. Mall conducted the experimental investigation to characterize the post-repair fatigue crack growth behavior in 6.350 mm thick 2024-T3 aluminum panels repaired with the asymmetrically bonded full width boron/epoxy composite patch in 1999 [11]. In 2005, T. Mitrevski et al. studied the effect of impact shape on the impact response of composite laminates [9]. In this study, the effect of impactor shape on the impact response of thin woven carbon/epoxy laminates is investigated. In 2007, S. Fujimoto and H. Sekine conducted the study on method for identifying the locations and shapes of crack and disbond fronts in aircraft structural panels repaired with bonded FRP composite patches [6]. In this study, impact damage criterion was adopted for developing aircraft design. Therefore, this work is focusing on the impact damage evaluation. All structures of the wing were designed through investigation of impact damage. The spar and rib design of wing are firstly considered before skin design. This paper dealt with the study on the low velocity impact damage criterion for repair and repair method after impact damage on carbon/epoxy UD and fabric laminate adopted the four-seated small aircraft under development. Investigation of low velocity impact damage criteria for repair is firstly considered. In this study, impact damage was simulated on the specimen, and then the strength recovery rate was evaluated after comparing with the condition before impact damage through compressive strength test after repairing the composite by applying the repair method for impact damaged aircraft composite. The applied repair method was the external patch repair which is a method of removing damaged area and repairing the removed area with patches. 2. Investigation of low velocity impact damage criterion for repair According to a composite aircraft design handbook [1,7], the damage criteria are classified into five categories depending on the severity of damages for the damage tolerance design. The first category is the barely visible impact damage (BVID), and the second category is the visible impact damage (VID). Repair should be carried out from the second category. Thus the definition of the criteria for the VID category is important in structure design. In order to define the VID energy, this study produces impact dam-
ages with different impact energies in low velocity impact range, examines the visual finding possibility, and analyzes the decreasing tendency of compressive strength. Fig. 2 presents schematic diagram showing design load levels versus categories of damage severity. Impact damages depend on material properties, but also on the thickness of the laminate, the layup, its size, and boundary conditions. The stiffness of the thickness has a significant effect on the magnitude of the maximum contact force which, of course, will affect the extent of the damage induced. T. Mitrevski et al. studied the effect of impactor shape on the impact response of composite laminate. Using a drop weight test rig, specimens were impacted using steel hemispherical, ogival and conical impactors. The energy absorbed by the specimen was the highest for the conical impactor which also produced the largest indentation/penetration depth. The peak force was greatest for the hemispherical impactor which also produced the shortest contact duration. The damage threshold load was highest for the hemispherical impactor followed by the ogival and conical impactors, respectively [9]. Giangiacomo Minak et al. studied the influence of diameter and boundary conditions on low velocity impact response of CFRP circular laminated plates. Tests indicated that both dimensions and boundary conditions affect response and damage, higher target stiffness resulting in greater energy absorption and more extended delamination. An empirical relationship with absorbed energy and with maximum contact force, independent of the test configuration, was found for the delamination area [8]. In this study, boundary condition of the specimen is fixed by clamps and adopted on hemispherical impactor. The candidate composite materials for the KC-100 aircraft are considered among some AGATE (Advance General Aviation Transport Experimental) materials which were proposed for increasing design reliability as well as promoting the small aircraft industry of the USA. According to this consideration, some composite materials, which are produced by Toray Composites (America), Inc., are finally selected. Therefore, the main wing spar is designed with carbon/epoxy UD prepreg (P707AG-15), and the main wing rib is designed with carbon/epoxy fabric prepreg (F6273C-07M). The mechanical properties of adopted materials are shown in Table 1. In this study, specimens of spar and rib were manufactured by autoclave molding. The lay-up sequence of the spar is 32 plies with [45◦ /0◦ /−45◦ /90◦ ]4s and the lay-up sequence of the rib is 20 plies
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
Table 1 Mechanical properties of UD and fabric prepreg. Properties
UD prepreg
Longitudinal modulus [GPa] Transverse modulus [GPa] Shear modulus [GPa] Longitudinal compressive strength [MPa] Transverse compressive strength [MPa] Poisson’s ratio Ply thickness [mm]
124.82 8.40 4.22 1447 198 0.3 0.15
Properties
Fabric prepreg
Longitudinal modulus [GPa] Transverse modulus [GPa] Shear modulus [GPa] Longitudinal compressive strength [MPa] Transverse compressive strength [MPa] Poisson’s ratio Ply thickness [mm]
56.27 54.86 4.21 708 702 0.04 0.22
365
with [(45◦ /−45◦ )/(0◦ /90◦ )]5s . The manufacturing and experimental test process for impact damage is shown in Fig. 3. The impact damages on them are performed by drop-weight type impact test equipment. The specimen is fixed by clamps and located at the center of impactor. The impactor has a mass of 4.19 kg, a diameter of 12.7 mm and a hemispherical striker tip that follows ASTM D7136 [3]. In conjunction with the impact test machine, a model PQ43 data acquisition and analysis system was used to provide complete records of impact energy and force as functions of time. A model C3D load cell (Kolas Korea) was used to measure impact force and energy. A load cell located in the impactor registers the contact force between the test specimen and the impact up. The instrumented dropped-weight impactor with a hemispherical tip was dropped from different heights to generate different impact energy levels. Fig. 4 shows impact test machine. The dimension of specimen for impact test is 100 mm × 150 mm. The impact support fixture manufactured using a plate of 20 mm thick steel. Four clamps shall be used to restrain the specimen during impact. The
Fig. 3. Manufacturing and impact test process of specimen.
366
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
Fig. 4. Impact test machine.
clamp shall have a minimum holding capacity of 1100 N. Fig. 5 shows the configuration of impact tested specimen. Environmental conditions are firstly considered before producing impact damages simulated on the specimens. The room temperature dry condition and the elevated temperature wet condition should be taken into account because degradation of mechanical property of the composite material may take place frequently by environmental conditions. The specimens are immersed in a water tank with water temperature of 80 ◦ C which is to simulate the elevated temperature wet condition. When the specimens absorb moisture until saturation, their compressive strengths are measured at before and after impact damage. As the range of low velocity impact energy is less than 10 J, the impact damages are gradually carried within 10 J for the UD and fabric laminate according to ASTM D7136 [3], and their compressive strengths are evaluated. Finally, the damage criterion for repairable design of composite laminate is defined. In case of the UD laminate specimen, the compressive strength test results in the room temperature dry condition show that the compressive strength of the damaged specimen at impact energy of 5 J is reduced by 4% in comparison with the compressive strength of the undamaged specimen. The damage at this impact is started to find visually. It is reduced by 19% of the original strength at impact energy of 6 J and 32% of the original strength at impact energy of 7 J. However, the compressive strength test results of the UD specimen in the elevated temperature wet condition shows that the compressive strength of undamaged specimen is reduced by 5% in comparison with the strength of the undamaged specimen in the room temperature dry condition. Furthermore, the compressive strength is reduced by 11% of the original strength at impact en-
Fig. 5. Configuration of impact tested specimen.
Table 2 Strength reduction of UD laminates (RTD and ETW). RTD
No damage 4J 5J 6J 7J
ETW
Max. load [kN]
Strength reduction [%]
Max. load [kN]
Strength reduction [%]
215.80 211.50 208.73 176.73 146.58
– 1.99 3.27 18.10 32.07
205.01 192.06 179.11 148.90 129.48
5.00 11.00 17.00 31.00 40.00
ergy of 4 J, 17% of the original strength at impact energy of 5 J and 31% of the original strength at impact energy 6 J. Table 2 shows strength reduction ratio of UD laminate specimen considering environmental conditions. Barely Visible Impact Damage (BVID) are defined as those which are visible at a distance of less than 1.5 m and Visible Impact Damage (VID) defined as those which are visible at a distance of 1.5 m. Fig. 6 is a photo of checking the UD laminate specimen through visual inspection at a distance of 1.5 meters after damage
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
367
Fig. 6. Visual inspection at 1.5 m distance of UD laminate specimen.
Fig. 7. View observed in ultrasonic inspection of UD laminate specimen. Table 3 Strength reduction of fabric laminates (RTD and ETW). RTD
No damage 4J 5J 6J 7J
ETW
Max. load [kN]
Strength reduction [%]
Max. load [kN]
Strength reduction [%]
169.20 163.84 151.18 126.37 118.05
–
133.66 130.12 125.83 107.61 91.23
21.00 23.09 25.63 36.40 46.08
3.16 10.64 25.31 30.23
has occurred by impact energy of 4 J, 5 J and 6 J. Also as internal damage is more serious than surface damage in case of a composite material, the validity of visual inspection has been reexamined by analyzing it using ultrasonic inspection which is a non-destructive inspection method in order to check the degree of internal damage. Fig. 7 shows the results of ultrasonic detection at 5 J and 6 J impact energy. No. 1 is configuration of the surface of damaged specimen. From 1 to 3 are photographs showing detection of the damaged areas in direction of depth of the specimen. The configuration of No. 2 is middle plane in direction of depth of damaged specimen. After 5 J impact energy is applied, it was
found that slight damage has occurred inside the specimen and after application of 6 J impact energy, the internal damage was so serious that the area was checked to require repair. Accordingly, as a result of checking, visual inspection was found to be valid. In case of the fabric laminate specimen, the impact damages are produced within impact energy of 10 J at the same environmental conditions as the previous case. In the room temperature dry condition, its compressive strength is reduced by 10% of the original strength at impact energy of 5 J and 25% of the original strength at impact energy of 6 J which is possible to find visually the damage. As the previous UD laminate case, the strength test of the fabric laminate specimen also is carried out in the elevated temperature wet condition. The test results show that the compressive strength is reduced by 25% of the original strength at impact energy of 5 J and 36% of the original strength at impact energy of 6 J. Table 3 shows strength reduction ratio of fabric laminate specimen considering environmental conditions. Fig. 8 is a view of fabric laminate specimen through visual inspection at a distance of 1.5 meters after damage has occurred by impact energy of 5 J and 6 J. The validity of visual inspection of fabric laminate specimen also has been reexamined using ultra-
368
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
Fig. 8. Visual inspection at 1.5 m distance of fabric laminate specimen.
Fig. 9. View observed in ultrasonic inspection of fabric laminate specimen.
sonic inspection in order to check the degree of internal damage. Fig. 9 shows views of ultrasonic detection of 5 J and 6 J impact energy. Through the investigation above, the damage criteria for repairable design of composite laminate are suggested. Because the safety factor of 1.5 is used in most aircraft design, the repair of the damage tolerance structure can be performed at 33% reduction of its original strength. If this criterion is applied, the UD spar and fabric rib structure must be repaired if the structure has the impact damage equivalent to having impact energy of 6 J in the elevated temperature wet condition. Figs. 10 and 11 show configuration before and after compression test of impact damaged UD laminate specimen. Figs. 12 and 13 show configuration before and after compression test of impact damaged fabric laminate specimen. 3. Repair process using external patch method The external patch repair method, which is first removing damaged area and then repairing the removed area with external patches using adhesive, is chosen as the repair for both the UD composite spar structure and the fabric composite rib structure [2].
Fig. 10. Configuration before compression test of impact damaged UD laminate specimen.
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
369
Fig. 11. Configuration after compression test of impact damaged UD laminate specimen.
Fig. 14. External patch repair process of impact damaged composite laminate. Fig. 12. Configuration before compression test of impact damaged fabric laminate specimen.
Fig. 13. Configuration after compression test of impact damaged fabric laminate specimen.
Fig. 14 shows external patch repair process to composite laminate structure. The patch was designed for a single sided repair of a 15 mm damaged area on 100 mm × 150 mm of composite panel. In the repair patch design approach, the lay-up sequence of impact damaged area was determined considering compressive strength. The dimension of the patch is 23 mm × 23 mm and the patches having 4 plies were designed. The peach a prefabricated piece was adopted. This work shows how to apply the proposed external patch repair method with the following procedure. Firstly the impact damage on the specimen is made with impact energy of 6 J by the drop-weight type impact test equipment, and then the impact damaged area is removed by a special tool. After that the adhesive is filled in the removed damaged area. The adhesive was used Hyson EA 9360 given by Henkel Corporation. Then the patch is applied on the neighboring surface around the adhesive filled area, and both vacuum and heat are applied. Finally the patch-applied specimen is cured using the autoclave. This autoclave curing is to enhance the repair performance. The use of special hot bonder is probably the most popular method for performing hot-cured repair. This study used the autoclave in the same manner and simulated the similar conditions. Fig. 15 shows the configuration of patched composite plate for curing using autoclave. The real process view of applied external patch on damaged area is shown as Figs. 16 and 17.
370
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
Fig. 15. Curing of patched composite plate using autoclave.
Fig. 16. Repaired specimen after impact damage (UD laminate specimen).
Fig. 17. Repaired specimen after impact damage (fabric laminate specimen).
In order to investigate the strength recovery effect, the compressive strength of the repaired specimen is compared with the original specimen strength before impact damage. The compressive strength test of the UD laminate specimen is performed to investigate the compressive strength recovery effect of the repaired specimen by the ASTM D7137 [4] standard procedure. The specimen before damage is fractured at an average compressive load of 215.80 kN, but the specimen after repair is fractured at an aver-
age compressive load of 197.22 kN. The test result shows that the compressive strength of the UD laminate specimen after repair is recovered to 91.39% of the compressive strength before damage. The investigation of structural strength recovery of the fabric laminate specimen is carried out by the same ASTM D7137 [4] standard procedure. In case of fabric laminate, the specimen before damage is fractured at an average compressive load of 168.96 kN, but the specimen after repair is fractured at an average
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
371
Table 4 Strength recovery of repaired UD laminate.
Load [kN] Test 1 Test 2 Test 3 Average Strength reduction [%] Strength recovery [%]
No damage
After impact (6 J)
After repair
215.91 217.46 214.04 215.80 – –
176.73 178.09 171.20 175.34 18.74 –
198.53 203.63 189.52 197.22 8.60 91.39
No damage
After impact (6 J)
After repair
169.20 170.67 167.02 168.96 – –
126.37 127.47 124.74 126.19 25.31 –
154.08 155.78 150.78 153.54 9.12 90.88
Table 5 Strength recovery of repaired fabric laminate.
Load [kN] Test 1 Test 2 Test 3 Average Strength reduction [%] Strength recovery [%]
compressive load of 153.54 kN. The compressive strength of the fabric specimen after repair is recovered to 90.88% of the strength before damage. Tables 4 and 5 show the compressive strength test results of the UD and fabric specimen before damage, after impact and after repair.
Fig. 18. The location of strain gage for stress measurement test. Table 6 Comparison between test and analysis results of repaired specimens. UD laminate
Stress
Fabric laminate
Test
FEM analysis
Test
FEM analysis
84.7 MPa
97.8 MPa
131.8 MPa
139.0 MPa
4. Finite element analysis The finite element analysis is performed to compare with the test results mentioned above. The finite element analysis FEA is used to analyze the strength recovery before the specimen test so that the repair method can be applicable to various structures. The analysis models simulate almost the same shape of the repaired specimen using external patch. The modeling was performed by increasing the number of element and applying more elements in the damaged area for accuracy of analysis using the shell element. As for the boundary condition, the lower part fixed boundary condition of the specimen was applied equally to the specimen test condition and the load was applied in the longitudinal direction of the specimen in the upper part. Finite element analysis is performed using the MSC Nastran solver and finally the results are obtained in the form of stresses, strains. Total number of elements for FEM mesh generation were 3677 including 2637 for the patched region mesh. The lay-up sequence of the UD laminate for finite element analysis is 32 plies with [45◦ /0◦ /−45◦ /90◦ ]4s and the lay-up sequence of the fabric laminate for finite element analysis is 20 plies with [(45◦ /−45◦ )/(0◦ /90◦ )]5s . The FEM analysis results are compared with the test results measured by the strain gage. In this paper, the strain when the specimen was fractured by compressive load was measured by the strain gage and then we compared the result of the strain measured by the strain gage with the result of the analysis. Finally recalculated stress considering elastic modulus from strain was compared the result of test with the result of the finite element analysis. The location of strain gage is a distance of 25 mm from the top and side of specimen. Two strain gage locations of the specimen are shown in Fig. 18. Table 6 shows the comparison of stress between the analysis results and the test results of repaired specimens. Through this comparison, it was found that the finite element stress analysis results are well agreed with the test results in both specimen cases of UD laminate and fabric laminate. Fig. 19 shows stress analysis result of repaired UD laminate and Fig. 20 shows stress analysis result of repaired fabric laminate.
Fig. 19. Stress analysis result of repaired UD laminate.
5. Conclusion This research is focusing on the low velocity impact damage evaluation and external patch repair techniques of carbon/epoxy UD and fabric laminate adopted small aircraft composite structure. Through application of gradual impact energies and investigation
372
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
specimen and the original specimen strength before impact damage, the strength recovery effect is investigated. The validity of the finite element analysis model suggested in this study was verified through the comparative analysis of the stress between the specimen test results and the finite element analysis results. Acknowledgement This study was supported by research funds from Howon University and Chosun University. References
Fig. 20. Stress analysis result of repaired fabric laminate.
of compressive strengths of the damaged specimens at different environmental conditions, the damage criteria for repairable design of both the impact damaged UD laminate and fabric laminate are suggested. The repair method using external patch is proposed to recover the reduced strength of the impact damaged specimens. Through comparison between the compressive strength of the repaired
[1] AC20-107B, Composite Aircraft Structures, FAA, 1984. [2] Keith B. Armstron, et al., Care and Repair of Advanced Composite, SAE International, 2005. [3] ASTM D7136, Standard test method for measuring the damage resistance of a fiber-reinforced polymer matrix composite to a drop-weight impact event, 2005. [4] ASTM D7137, Standard test method for compressive residual strength properties of damaged polymer matrix composite plates, 2005. [5] Eric J. Herup, Anthony N. Palazotto, Low-velocity impact damage initiation in graphite/epoxy/Nomex honeycomb-sandwich plates, Composites Science and Technology 57 (1998) 1581–1598. [6] Shin-etsu Fujimoto, Hideki Sekine, Identification of crack and disband fronts in repaired aircraft structural panels with bonded FRP composite patches, Composite Structures 77 (2007) 533–545. [7] MIL-HDBK-17-3F, Composite Materials Handbook, 2002. [8] Giangiacomo Minak, Daniele Ghelli, Influence of diameter and boundary conditions on low velocity impact response of CFRP circular laminated plates, Composites Part B: Engineering 39 (2008) 962–972. [9] T. Mitrevski, I.H. Marchall, R. Thomson, R. Jones, B. Whittingham, The effect of impactor shape on the impact response of composite laminates, Composite Structures 67 (2005) 139–148. [10] Milan Mitrovic, H. Thomas Hahn, Greg P. Carman, Peter Shyprykevich, Effect of loading parameters on the fatigue behavior of impact damaged composite laminates, Composites Science and Technology 59 (1999) 2059–2078. [11] J.J. Schubbe, S. Mall, Investigation of a cracked thick aluminum panel repaired with a bonded composite patch, Engineering Fracture Mechanics 63 (1999) 305–323.
Contents lists available at SciVerse ScienceDirect
Aerospace Science and Technology www.elsevier.com/locate/aescte
Experimental study on barely visible impact damage and visible impact damage for repair of small aircraft composite structure Hyunbum Park a,∗ , Changduk Kong b a b
Department of Defense & Science Technology – Aeronautics, Howon University, 64 Howondae 3gil, Impi, 573-718 Gunsan, Republic of Korea Department of Aerospace Engineering, Chosun University, 375 Seosukdong, Donggu, 502-759 Gwangju, Republic of Korea
a r t i c l e
i n f o
Article history: Received 5 January 2012 Received in revised form 25 November 2012 Accepted 18 April 2013 Available online 23 April 2013 Keywords: Composite material Impact damage Structural design Finite element analysis
a b s t r a c t This work is focusing on the low velocity impact damage evaluation and the external patch repair techniques of carbon/epoxy UD and fabric laminate adopted developing aircraft. The impact damages of composite laminates of the carbon/epoxy UD and fabric are simulated by the drop-weight type impact test equipment. The damaged specimens are repaired using the external patch repair method after removing the damaged area. The compressive strength test and analysis results of the repaired impact damaged specimens are compared with the compressive strength test and analysis results of the undamaged specimens and the impact damaged specimens. Finally, through investigation of compressive strengths of the damaged specimens at different environmental conditions, the damage criteria for repairable design of both the impact damaged UD and fabric laminate structure are suggested. © 2013 Elsevier Masson SAS. All rights reserved.
1. Introduction Recently, as the utilization of small aircraft has been increasing, many countries have been developing various small aircraft. In Korea, the KC-100, which is a small scale piston propeller general aviation aircraft, has been developed to establish a domestic certificate infrastructure and system through the BASA (Bilateral Aviation Safety Agreement) program by KAI (Korea Aerospace Industries, Ltd.). Fig. 1 shows conceptual design result of KC-100 for BASA. The KC-100 aircraft was designed with a complete composite structure for structure weight reduction. However the carbon/epoxy composite structure, which is mainly used for this aircraft, is susceptible impact damage. Especially, the low velocity impact damage cannot be easily found during visual inspections. Thus the low velocity impact damage of the composite structure has become an important issue in composite structural design. Because the composite structure not only is weak to external impact damage but also has a different repair method from existing metallic materials, the repair method for the impact damage of composite structure is very important for certification of developing aircraft. According to the literature survey, several engineers have been performing various studies of low velocity impact damage and repair on the composite structure. In 1998, Eric J. Herup et al. performed the study on low-velocity impact damage initiation in graphite/epoxy/Nomex honeycomb-
*
Corresponding author. E-mail address: [email protected] (H. Park).
1270-9638/$ – see front matter © 2013 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.ast.2013.04.007
Fig. 1. Conceptual design configuration of Korean Civil Aircraft-100 (KC-100).
sandwich plates [5]. In this study, low velocity impact and static indentation tests on sandwich plates have been performed to characterize damage initiation as a function of face sheet thickness and loading rate. In 1999, Milan Mitrovic et al. studied the effect of loading parameters on the fatigue behavior of impact damaged composite laminates [10]. The long term mechanical fatigue of quasiisotropic graphite/epoxy laminates was investigated to determine the influence of loading parameters on impact induced delamination growth during constant amplitude and spectrum fatigue loading.
364
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
Fig. 2. Schematic diagram showing design load levels versus categories of damage severity.
J.J. Schubbe and S. Mall conducted the experimental investigation to characterize the post-repair fatigue crack growth behavior in 6.350 mm thick 2024-T3 aluminum panels repaired with the asymmetrically bonded full width boron/epoxy composite patch in 1999 [11]. In 2005, T. Mitrevski et al. studied the effect of impact shape on the impact response of composite laminates [9]. In this study, the effect of impactor shape on the impact response of thin woven carbon/epoxy laminates is investigated. In 2007, S. Fujimoto and H. Sekine conducted the study on method for identifying the locations and shapes of crack and disbond fronts in aircraft structural panels repaired with bonded FRP composite patches [6]. In this study, impact damage criterion was adopted for developing aircraft design. Therefore, this work is focusing on the impact damage evaluation. All structures of the wing were designed through investigation of impact damage. The spar and rib design of wing are firstly considered before skin design. This paper dealt with the study on the low velocity impact damage criterion for repair and repair method after impact damage on carbon/epoxy UD and fabric laminate adopted the four-seated small aircraft under development. Investigation of low velocity impact damage criteria for repair is firstly considered. In this study, impact damage was simulated on the specimen, and then the strength recovery rate was evaluated after comparing with the condition before impact damage through compressive strength test after repairing the composite by applying the repair method for impact damaged aircraft composite. The applied repair method was the external patch repair which is a method of removing damaged area and repairing the removed area with patches. 2. Investigation of low velocity impact damage criterion for repair According to a composite aircraft design handbook [1,7], the damage criteria are classified into five categories depending on the severity of damages for the damage tolerance design. The first category is the barely visible impact damage (BVID), and the second category is the visible impact damage (VID). Repair should be carried out from the second category. Thus the definition of the criteria for the VID category is important in structure design. In order to define the VID energy, this study produces impact dam-
ages with different impact energies in low velocity impact range, examines the visual finding possibility, and analyzes the decreasing tendency of compressive strength. Fig. 2 presents schematic diagram showing design load levels versus categories of damage severity. Impact damages depend on material properties, but also on the thickness of the laminate, the layup, its size, and boundary conditions. The stiffness of the thickness has a significant effect on the magnitude of the maximum contact force which, of course, will affect the extent of the damage induced. T. Mitrevski et al. studied the effect of impactor shape on the impact response of composite laminate. Using a drop weight test rig, specimens were impacted using steel hemispherical, ogival and conical impactors. The energy absorbed by the specimen was the highest for the conical impactor which also produced the largest indentation/penetration depth. The peak force was greatest for the hemispherical impactor which also produced the shortest contact duration. The damage threshold load was highest for the hemispherical impactor followed by the ogival and conical impactors, respectively [9]. Giangiacomo Minak et al. studied the influence of diameter and boundary conditions on low velocity impact response of CFRP circular laminated plates. Tests indicated that both dimensions and boundary conditions affect response and damage, higher target stiffness resulting in greater energy absorption and more extended delamination. An empirical relationship with absorbed energy and with maximum contact force, independent of the test configuration, was found for the delamination area [8]. In this study, boundary condition of the specimen is fixed by clamps and adopted on hemispherical impactor. The candidate composite materials for the KC-100 aircraft are considered among some AGATE (Advance General Aviation Transport Experimental) materials which were proposed for increasing design reliability as well as promoting the small aircraft industry of the USA. According to this consideration, some composite materials, which are produced by Toray Composites (America), Inc., are finally selected. Therefore, the main wing spar is designed with carbon/epoxy UD prepreg (P707AG-15), and the main wing rib is designed with carbon/epoxy fabric prepreg (F6273C-07M). The mechanical properties of adopted materials are shown in Table 1. In this study, specimens of spar and rib were manufactured by autoclave molding. The lay-up sequence of the spar is 32 plies with [45◦ /0◦ /−45◦ /90◦ ]4s and the lay-up sequence of the rib is 20 plies
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
Table 1 Mechanical properties of UD and fabric prepreg. Properties
UD prepreg
Longitudinal modulus [GPa] Transverse modulus [GPa] Shear modulus [GPa] Longitudinal compressive strength [MPa] Transverse compressive strength [MPa] Poisson’s ratio Ply thickness [mm]
124.82 8.40 4.22 1447 198 0.3 0.15
Properties
Fabric prepreg
Longitudinal modulus [GPa] Transverse modulus [GPa] Shear modulus [GPa] Longitudinal compressive strength [MPa] Transverse compressive strength [MPa] Poisson’s ratio Ply thickness [mm]
56.27 54.86 4.21 708 702 0.04 0.22
365
with [(45◦ /−45◦ )/(0◦ /90◦ )]5s . The manufacturing and experimental test process for impact damage is shown in Fig. 3. The impact damages on them are performed by drop-weight type impact test equipment. The specimen is fixed by clamps and located at the center of impactor. The impactor has a mass of 4.19 kg, a diameter of 12.7 mm and a hemispherical striker tip that follows ASTM D7136 [3]. In conjunction with the impact test machine, a model PQ43 data acquisition and analysis system was used to provide complete records of impact energy and force as functions of time. A model C3D load cell (Kolas Korea) was used to measure impact force and energy. A load cell located in the impactor registers the contact force between the test specimen and the impact up. The instrumented dropped-weight impactor with a hemispherical tip was dropped from different heights to generate different impact energy levels. Fig. 4 shows impact test machine. The dimension of specimen for impact test is 100 mm × 150 mm. The impact support fixture manufactured using a plate of 20 mm thick steel. Four clamps shall be used to restrain the specimen during impact. The
Fig. 3. Manufacturing and impact test process of specimen.
366
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
Fig. 4. Impact test machine.
clamp shall have a minimum holding capacity of 1100 N. Fig. 5 shows the configuration of impact tested specimen. Environmental conditions are firstly considered before producing impact damages simulated on the specimens. The room temperature dry condition and the elevated temperature wet condition should be taken into account because degradation of mechanical property of the composite material may take place frequently by environmental conditions. The specimens are immersed in a water tank with water temperature of 80 ◦ C which is to simulate the elevated temperature wet condition. When the specimens absorb moisture until saturation, their compressive strengths are measured at before and after impact damage. As the range of low velocity impact energy is less than 10 J, the impact damages are gradually carried within 10 J for the UD and fabric laminate according to ASTM D7136 [3], and their compressive strengths are evaluated. Finally, the damage criterion for repairable design of composite laminate is defined. In case of the UD laminate specimen, the compressive strength test results in the room temperature dry condition show that the compressive strength of the damaged specimen at impact energy of 5 J is reduced by 4% in comparison with the compressive strength of the undamaged specimen. The damage at this impact is started to find visually. It is reduced by 19% of the original strength at impact energy of 6 J and 32% of the original strength at impact energy of 7 J. However, the compressive strength test results of the UD specimen in the elevated temperature wet condition shows that the compressive strength of undamaged specimen is reduced by 5% in comparison with the strength of the undamaged specimen in the room temperature dry condition. Furthermore, the compressive strength is reduced by 11% of the original strength at impact en-
Fig. 5. Configuration of impact tested specimen.
Table 2 Strength reduction of UD laminates (RTD and ETW). RTD
No damage 4J 5J 6J 7J
ETW
Max. load [kN]
Strength reduction [%]
Max. load [kN]
Strength reduction [%]
215.80 211.50 208.73 176.73 146.58
– 1.99 3.27 18.10 32.07
205.01 192.06 179.11 148.90 129.48
5.00 11.00 17.00 31.00 40.00
ergy of 4 J, 17% of the original strength at impact energy of 5 J and 31% of the original strength at impact energy 6 J. Table 2 shows strength reduction ratio of UD laminate specimen considering environmental conditions. Barely Visible Impact Damage (BVID) are defined as those which are visible at a distance of less than 1.5 m and Visible Impact Damage (VID) defined as those which are visible at a distance of 1.5 m. Fig. 6 is a photo of checking the UD laminate specimen through visual inspection at a distance of 1.5 meters after damage
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
367
Fig. 6. Visual inspection at 1.5 m distance of UD laminate specimen.
Fig. 7. View observed in ultrasonic inspection of UD laminate specimen. Table 3 Strength reduction of fabric laminates (RTD and ETW). RTD
No damage 4J 5J 6J 7J
ETW
Max. load [kN]
Strength reduction [%]
Max. load [kN]
Strength reduction [%]
169.20 163.84 151.18 126.37 118.05
–
133.66 130.12 125.83 107.61 91.23
21.00 23.09 25.63 36.40 46.08
3.16 10.64 25.31 30.23
has occurred by impact energy of 4 J, 5 J and 6 J. Also as internal damage is more serious than surface damage in case of a composite material, the validity of visual inspection has been reexamined by analyzing it using ultrasonic inspection which is a non-destructive inspection method in order to check the degree of internal damage. Fig. 7 shows the results of ultrasonic detection at 5 J and 6 J impact energy. No. 1 is configuration of the surface of damaged specimen. From 1 to 3 are photographs showing detection of the damaged areas in direction of depth of the specimen. The configuration of No. 2 is middle plane in direction of depth of damaged specimen. After 5 J impact energy is applied, it was
found that slight damage has occurred inside the specimen and after application of 6 J impact energy, the internal damage was so serious that the area was checked to require repair. Accordingly, as a result of checking, visual inspection was found to be valid. In case of the fabric laminate specimen, the impact damages are produced within impact energy of 10 J at the same environmental conditions as the previous case. In the room temperature dry condition, its compressive strength is reduced by 10% of the original strength at impact energy of 5 J and 25% of the original strength at impact energy of 6 J which is possible to find visually the damage. As the previous UD laminate case, the strength test of the fabric laminate specimen also is carried out in the elevated temperature wet condition. The test results show that the compressive strength is reduced by 25% of the original strength at impact energy of 5 J and 36% of the original strength at impact energy of 6 J. Table 3 shows strength reduction ratio of fabric laminate specimen considering environmental conditions. Fig. 8 is a view of fabric laminate specimen through visual inspection at a distance of 1.5 meters after damage has occurred by impact energy of 5 J and 6 J. The validity of visual inspection of fabric laminate specimen also has been reexamined using ultra-
368
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
Fig. 8. Visual inspection at 1.5 m distance of fabric laminate specimen.
Fig. 9. View observed in ultrasonic inspection of fabric laminate specimen.
sonic inspection in order to check the degree of internal damage. Fig. 9 shows views of ultrasonic detection of 5 J and 6 J impact energy. Through the investigation above, the damage criteria for repairable design of composite laminate are suggested. Because the safety factor of 1.5 is used in most aircraft design, the repair of the damage tolerance structure can be performed at 33% reduction of its original strength. If this criterion is applied, the UD spar and fabric rib structure must be repaired if the structure has the impact damage equivalent to having impact energy of 6 J in the elevated temperature wet condition. Figs. 10 and 11 show configuration before and after compression test of impact damaged UD laminate specimen. Figs. 12 and 13 show configuration before and after compression test of impact damaged fabric laminate specimen. 3. Repair process using external patch method The external patch repair method, which is first removing damaged area and then repairing the removed area with external patches using adhesive, is chosen as the repair for both the UD composite spar structure and the fabric composite rib structure [2].
Fig. 10. Configuration before compression test of impact damaged UD laminate specimen.
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
369
Fig. 11. Configuration after compression test of impact damaged UD laminate specimen.
Fig. 14. External patch repair process of impact damaged composite laminate. Fig. 12. Configuration before compression test of impact damaged fabric laminate specimen.
Fig. 13. Configuration after compression test of impact damaged fabric laminate specimen.
Fig. 14 shows external patch repair process to composite laminate structure. The patch was designed for a single sided repair of a 15 mm damaged area on 100 mm × 150 mm of composite panel. In the repair patch design approach, the lay-up sequence of impact damaged area was determined considering compressive strength. The dimension of the patch is 23 mm × 23 mm and the patches having 4 plies were designed. The peach a prefabricated piece was adopted. This work shows how to apply the proposed external patch repair method with the following procedure. Firstly the impact damage on the specimen is made with impact energy of 6 J by the drop-weight type impact test equipment, and then the impact damaged area is removed by a special tool. After that the adhesive is filled in the removed damaged area. The adhesive was used Hyson EA 9360 given by Henkel Corporation. Then the patch is applied on the neighboring surface around the adhesive filled area, and both vacuum and heat are applied. Finally the patch-applied specimen is cured using the autoclave. This autoclave curing is to enhance the repair performance. The use of special hot bonder is probably the most popular method for performing hot-cured repair. This study used the autoclave in the same manner and simulated the similar conditions. Fig. 15 shows the configuration of patched composite plate for curing using autoclave. The real process view of applied external patch on damaged area is shown as Figs. 16 and 17.
370
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
Fig. 15. Curing of patched composite plate using autoclave.
Fig. 16. Repaired specimen after impact damage (UD laminate specimen).
Fig. 17. Repaired specimen after impact damage (fabric laminate specimen).
In order to investigate the strength recovery effect, the compressive strength of the repaired specimen is compared with the original specimen strength before impact damage. The compressive strength test of the UD laminate specimen is performed to investigate the compressive strength recovery effect of the repaired specimen by the ASTM D7137 [4] standard procedure. The specimen before damage is fractured at an average compressive load of 215.80 kN, but the specimen after repair is fractured at an aver-
age compressive load of 197.22 kN. The test result shows that the compressive strength of the UD laminate specimen after repair is recovered to 91.39% of the compressive strength before damage. The investigation of structural strength recovery of the fabric laminate specimen is carried out by the same ASTM D7137 [4] standard procedure. In case of fabric laminate, the specimen before damage is fractured at an average compressive load of 168.96 kN, but the specimen after repair is fractured at an average
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
371
Table 4 Strength recovery of repaired UD laminate.
Load [kN] Test 1 Test 2 Test 3 Average Strength reduction [%] Strength recovery [%]
No damage
After impact (6 J)
After repair
215.91 217.46 214.04 215.80 – –
176.73 178.09 171.20 175.34 18.74 –
198.53 203.63 189.52 197.22 8.60 91.39
No damage
After impact (6 J)
After repair
169.20 170.67 167.02 168.96 – –
126.37 127.47 124.74 126.19 25.31 –
154.08 155.78 150.78 153.54 9.12 90.88
Table 5 Strength recovery of repaired fabric laminate.
Load [kN] Test 1 Test 2 Test 3 Average Strength reduction [%] Strength recovery [%]
compressive load of 153.54 kN. The compressive strength of the fabric specimen after repair is recovered to 90.88% of the strength before damage. Tables 4 and 5 show the compressive strength test results of the UD and fabric specimen before damage, after impact and after repair.
Fig. 18. The location of strain gage for stress measurement test. Table 6 Comparison between test and analysis results of repaired specimens. UD laminate
Stress
Fabric laminate
Test
FEM analysis
Test
FEM analysis
84.7 MPa
97.8 MPa
131.8 MPa
139.0 MPa
4. Finite element analysis The finite element analysis is performed to compare with the test results mentioned above. The finite element analysis FEA is used to analyze the strength recovery before the specimen test so that the repair method can be applicable to various structures. The analysis models simulate almost the same shape of the repaired specimen using external patch. The modeling was performed by increasing the number of element and applying more elements in the damaged area for accuracy of analysis using the shell element. As for the boundary condition, the lower part fixed boundary condition of the specimen was applied equally to the specimen test condition and the load was applied in the longitudinal direction of the specimen in the upper part. Finite element analysis is performed using the MSC Nastran solver and finally the results are obtained in the form of stresses, strains. Total number of elements for FEM mesh generation were 3677 including 2637 for the patched region mesh. The lay-up sequence of the UD laminate for finite element analysis is 32 plies with [45◦ /0◦ /−45◦ /90◦ ]4s and the lay-up sequence of the fabric laminate for finite element analysis is 20 plies with [(45◦ /−45◦ )/(0◦ /90◦ )]5s . The FEM analysis results are compared with the test results measured by the strain gage. In this paper, the strain when the specimen was fractured by compressive load was measured by the strain gage and then we compared the result of the strain measured by the strain gage with the result of the analysis. Finally recalculated stress considering elastic modulus from strain was compared the result of test with the result of the finite element analysis. The location of strain gage is a distance of 25 mm from the top and side of specimen. Two strain gage locations of the specimen are shown in Fig. 18. Table 6 shows the comparison of stress between the analysis results and the test results of repaired specimens. Through this comparison, it was found that the finite element stress analysis results are well agreed with the test results in both specimen cases of UD laminate and fabric laminate. Fig. 19 shows stress analysis result of repaired UD laminate and Fig. 20 shows stress analysis result of repaired fabric laminate.
Fig. 19. Stress analysis result of repaired UD laminate.
5. Conclusion This research is focusing on the low velocity impact damage evaluation and external patch repair techniques of carbon/epoxy UD and fabric laminate adopted small aircraft composite structure. Through application of gradual impact energies and investigation
372
H. Park, C. Kong / Aerospace Science and Technology 29 (2013) 363–372
specimen and the original specimen strength before impact damage, the strength recovery effect is investigated. The validity of the finite element analysis model suggested in this study was verified through the comparative analysis of the stress between the specimen test results and the finite element analysis results. Acknowledgement This study was supported by research funds from Howon University and Chosun University. References
Fig. 20. Stress analysis result of repaired fabric laminate.
of compressive strengths of the damaged specimens at different environmental conditions, the damage criteria for repairable design of both the impact damaged UD laminate and fabric laminate are suggested. The repair method using external patch is proposed to recover the reduced strength of the impact damaged specimens. Through comparison between the compressive strength of the repaired
[1] AC20-107B, Composite Aircraft Structures, FAA, 1984. [2] Keith B. Armstron, et al., Care and Repair of Advanced Composite, SAE International, 2005. [3] ASTM D7136, Standard test method for measuring the damage resistance of a fiber-reinforced polymer matrix composite to a drop-weight impact event, 2005. [4] ASTM D7137, Standard test method for compressive residual strength properties of damaged polymer matrix composite plates, 2005. [5] Eric J. Herup, Anthony N. Palazotto, Low-velocity impact damage initiation in graphite/epoxy/Nomex honeycomb-sandwich plates, Composites Science and Technology 57 (1998) 1581–1598. [6] Shin-etsu Fujimoto, Hideki Sekine, Identification of crack and disband fronts in repaired aircraft structural panels with bonded FRP composite patches, Composite Structures 77 (2007) 533–545. [7] MIL-HDBK-17-3F, Composite Materials Handbook, 2002. [8] Giangiacomo Minak, Daniele Ghelli, Influence of diameter and boundary conditions on low velocity impact response of CFRP circular laminated plates, Composites Part B: Engineering 39 (2008) 962–972. [9] T. Mitrevski, I.H. Marchall, R. Thomson, R. Jones, B. Whittingham, The effect of impactor shape on the impact response of composite laminates, Composite Structures 67 (2005) 139–148. [10] Milan Mitrovic, H. Thomas Hahn, Greg P. Carman, Peter Shyprykevich, Effect of loading parameters on the fatigue behavior of impact damaged composite laminates, Composites Science and Technology 59 (1999) 2059–2078. [11] J.J. Schubbe, S. Mall, Investigation of a cracked thick aluminum panel repaired with a bonded composite patch, Engineering Fracture Mechanics 63 (1999) 305–323.