AE monitoring of lightning-damaged CFRP laminates during compression-after-impact test.
Composite materials are increasingly used in aircraft structures because of their high specific stiffness and strength, good fatigue and corrosion resistances . Recently, the application of carbon-fiber reinforced plastics (CFRP) in commercial airframes becomes in reality even for primary structures such as fuselage as well as main wings. Excellent specific properties are the major driving forces to structural application of CFRP, but the sensitivity of laminated composites to impact loading in the thickness direction has been a serious obstacle to more widespread use of this class of materials. This is because the energy dissipated during impact is mainly absorbed by a combination of matrix damage, fiber fracture and fiber-matrix debonding, thus leading to significant reductions in the load-carrying capability of the material. In particular, compressive strength of brittle composite system is reduced remarkably after impact loading such as tool dropping and/or low-speed foreign object crashing . Therefore, the damage tolerance becomes a major concern of reliability assessment. Since these types of damage are virtually invisible, the evaluation of residual strength under the compression-after-impact (CAI) condition has been considered as a critical step in damage-tolerant design of numerous composite structures .
Another issue newly raised due to the use of composites in commercial airframe is their vulnerability to lightning-strike damage. Every commercial airplane can be exposed to lightning strikes on average about twice a year . With composites employed as the major structural materials for fuselage, a serious concern about the lightning-strike damage becomes in reality due to the poor electrical conductivity of composites comparing to that of aluminum alloys . When the fuselage is exposed to lightning strike, they shall suffer direct damages, such as puncture, spark propagation, local heating effect as well as indirect damages such as electromagnetic interference. If the external pressure acts on the surface of damaged area of fuselage, it will increase the severity of damages due to pressure from outside than the surface of the normal area. A study found that CFRP laminates made of carbon fibers coated with conductive nano-particles could reduce the lightning-strike damage by increased electrical conductivity .
In this study, CFRP coupons previously subjected to a simulated lightning strike were tested under the CAI condition and the damage progress due to compressive loading was monitored by acoustic emission (AE) activity.
Specimens and Experimental Setup
Six types of materials in total were tested; two types of prepregs, UD-CFRP and woven-CFRP were used and for each of them, three different laminates were prepared; the uncoated, the ITO 30% coated, and the ITO 40% coated. The uncoated UD-CFRP coupons were prepared as a reference from a unidirectional (UD) prepregs (USN125C, SK Chemicals). For the ITO-coated CFRP coupons, first the matrix resin was removed by soaking the prepregs in methyl-ether-ketone (MEK) and then indium-tin oxides (ITO) nano-particles of 30% and 40% concentrations in colloidal suspension were spray-coated directly on carbon fibers and dried. The ITO-coated carbon fiber bundles were placed between resin films on a hot press to reproduce a nanoparticles-coated CFRP prepregs. The stacking sequence of CAI test coupons was [[0/90].sub.8s], with the lateral dimension of 100 x 150 mm and the nominal thickness of 3.6 mm. The woven-CFRP coupons were prepared similarly from a plain weave, dry graphite fabric (BMS 9-8 3K-70-PW, Pacific Composites) except there was no need to remove the resin by soaking it in MEK. After the fabric was spray-coated with ITO nano-particles, CAI test coupons were prepared at hot press by using the same resin films as in the UD-CFRP coupons. The uncoated woven-CFRP coupons were also prepared as a reference.
Simulated Lightning Strikes and CAI Tests
The impact damage was induced by the simulated lightning strikes with an impulse current generator (ICG). The center of CFRP coupons was shot by the high voltage/high current ICG at 10, 20, 30 and 40 kA with a waveform of 8 [micro]s in rise time and 20 [micro]s in total duration. Damaged area of woven-CFRP coupons were virtually invisible even after the impulse current of 40 kA was applied. The CAI test fixture was prepared based on ASTM D7137 as shown in Fig. 1 , but was modified slightly from previous experiences of compression tests of CFRP. CAI tests were performed with a screw-driven mechanical test machine as shown in Fig. 2 where a CFRP coupon was in the CAI test jig. The crosshead speed was set to 1.25 mm/min for all tests under displacement control.
Four small AE sensors (B1025, Digital Wave) were mounted at each quadrant, with their location 36 mm away from the corner and toward the center of the CFRP coupon. They were fixed to the surface by using a spring-assisted attachment and grease couplant. Outputs of AE sensors were pre-amplified 40 dB and filtered by a band-pass filter of 100 kHz to 1.2 MHz, then processed recorded by using AE data acquisition boards (MISTRAS 2001, PAC). Threshold was set at 45 dB after pre-amplifier for the tests of UD-CFRP coupons, but was adjusted at 35 dB for woven-CFRP coupons since it appeared to be losing some low amplitude AE signals arrived before the maximum compressive load was reached.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
Degree of Damages and AE Activities
Figure 3 shows the photographs of UD-CFRP coupons after the completion of CAI tests. The left coupon was broken along the centerline containing the lightning-strike damaged area. The right coupon was broken along a line slightly above the damaged area. The latter is commonly observed as acceptable failure modes under ASTM 7137, however. Those of woven-CFRP coupons showed similar behavior but the broken surface was not as clear as shown in Fig. 3.
[FIGURE 3 OMITTED]
In Fig. 4, the primary data obtained during the CAI tests are shown as the relationship between compressive load and AE events with time for the uncoated, whereas the same data are shown in Fig. 5 for the ITO 40% coated. For both cases AE activity started earlier with the increasing level of impulse current applied, which corresponds to the severity of impact damage, but the onset of AE activity was significantly delayed with the ITO 40% coated in Fig. 5 regardless of the level of impulse current applied. In general, the maximum and residual compressive strengths were also lower for the uncoated and damaged than for the ITO 40% coated and damaged. This is because higher degree of damage was induced for the uncoated CFRP coupons with lower electrical conductivity at the same level of impulse current. On the other hand, more AE events were observed overall with increasing level of impulse current applied.
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
It appeared that the severe damage induced by stronger electrical shocks produced more sources of AE when the coupons were tested under CAI test mode. It becomes even clearer if one can combine the relationship between the maximum compressive strength and the level of impulse current given in Fig. 6, and the impact-damaged area measured by ultrasonic C-scan in Fig. 7. It could also be shown as the direct relationship between the damaged area and AE total event as given in Fig. 8.
In Fig. 6, the relationship between the maximum compressive strengths and the level of impulse current are shown for woven-CFRP. The maximum compressive strength appeared to be almost inversely proportional with the level of impulse current applied for all types of CFRP coupons; UD- and woven-CFRP, and the uncoated, the ITO 30% coated, and the ITO 40% coated. At the same level of impulse current applied, the uncoated showed always the lowest strength whereas the ITO 40% coated showed the highest strength. For UD-CFRP coupons, a similar behavior was observed although the maximum compressive strength was generally lower, and the detailed data was reported elsewhere .
[FIGURE 6 OMITTED]
The ultrasonic C-scan images of impact-damaged area are shown in Fig. 7 for UD-CFRP coupons. They were measured by using an ultrasonic scanning unit with 64-channel linear phased array transducer working at the center frequency of 5 MHz (Matrixeye EX, Toshiba) with a built-in function of area calculation. A visual image of damaged area for the uncoated (d) is also shown as a reference. The impact-damaged area was the largest for the uncoated, intermediate for the ITO 30% coated, and the smallest for the ITO 40% coated. The damaged area increased with increasing level of impulse current applied. For woven-CFRP coupons, a similar trend was observed although the impact-damaged area was virtually invisible and generally much smaller than those shown here for UD-CFRP.
[FIGURE 7 OMITTED]
The overall AE activity for each type of CFRP coupons are summarized in Fig. 08 (a) and (b) in terms of AE total events versus the impact-damaged area for three different conditions of CFRP's in terms electrical conductivity. AE event counts appeared to be almost proportional to the impact-damaged area, which is again proportional to the level of impulse current applied for both types of CFRP coupons. The good correlation between AE events and the impact-damaged area induced by the impulse current can be utilized as a useful indicator of the degree of lightning-strike damage in composite airframes.
[FIGURE 8 OMITTED]
The relationship between AE total events and the maximum compressive strength with the level of impulse current applied is shown in Fig. 9 for woven-CFRP coupons. With increasing level of impulse current, AE total events increased regardless of the condition of CFRP coupons, whereas the maximum compressive strength decreased. A similar behavior was observed for UD-CFRP coupons and reported elsewhere . Therefore, smaller counts of AE total events implied smaller impact-damaged area and the higher maximum compressive strength.
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
In Fig. 10, the stress level at which AE activity started were calculated from the data shown in Figs. 4 and 5 and correlated with the maximum compressive strength for each condition of woven-CFRP coupons. The onset of AE activity was based on the detected of AE events with the pre-determined threshold level. Since the maximum compressive strength appeared almost inversely proportional to the levels of impulse current as shown in Fig. 6, the lower maximum compressive strength means higher degree of the impact-damage. The lower left part of Fig. 10 represents this condition. The stress level at the onset of AE activity appeared much lower for the uncoated than for the ITO 40% coated. On the other hand, the upper right part of Fig. 10 represents the higher maximum compressive strength, hence lower degree of the impact-damage. In this case, the difference between different conditions of woven-CFRP was smaller. A similar behavior was observed for UD-CFRP coupons . Therefore, this could be another useful indicator of the degree of damage induced by lightning strikes in composite airframes.
AE activity of CFRP coupons made of nano-particles-coated carbon fibers and then subjected to a simulated lightning strike by using a high voltage/high current impulse current generator was monitored under CAI test mode. AE data were evaluated in terms of the level of impulse current applied to simulate the lightning strike, hence the degree of impact-damage and the maximum compressive strength measured by the CAI tests. Coating of ITO nano-particles up to 40% in colloidal suspension appeared to be an effective means to prevent from lightning-strike damage in composite airframes. Compressive strength after the impact has close correlation with the level of impulse current applied and the lightning-strike damaged area. A good correlation between AE total events and the impact-damaged area induced by the impulse current applied can be utilized as a useful indicator of the degree of lightning-strike damage in composite airframes. The correlation between the maximum compressive strength and the stress level at the onset of AE activity could be another useful indicator of the degree of lightning-strike damage.
This work was supported by Inha University Research Grant. References
 J.R. Kollgaard and S.G. LaRiviere, "NDE Challenges with Future Commercial Aircraft--A Boeing Perspective", Review of Progress in Quantitative NDE, 27, 2008, 23-28.
 F. Aymerich and P. Priolo, "Characterization of fracture modes in stitched and unstitched cross-ply laminates subjected to low-velocity impact and compression after impact loading", International J. of Impact Engineering, 35, 2008, 591-608.
 I. Ohsawa, I. Kimpara, K. Kageyama, T. Suzuki, and A. Yamashita, "AE Characterization of Compressive Residual Strength of Impact-Damaged CFRP Laminates", Progress in Acoustic Emission VII, JSNDI, 1994, pp. 511-516.
 F.A. Fisher, J.A. Plumer and R.A. Perala, Aircraft Lightning Protection Handbook, Federal Aviation Administration, DOT/FAA/CT-89/22, September 1989.
 J. Guo, Y. Tang, F. Liang, Z. Zhao, D. Firsich and J. Fielding, "Carbon nanofiber paper for lightning strike protection of composite materials", Composites Part B: Engineering, 41, (2), 2010, 192-198.
 S.-W. Seo, M.-S. Ha, O.-Y. Kwon and H.-S. Choi, "Improvement of Electrical Conductivity of Carbon-Fiber Reinforced Plastics by Nano-Particles Coating", J. of the Korean Soc. for Composite Materials, 23, (6), 2010, 1-6.
 ASTM D7137/D7137M-07 "Standard Test Method for Compressive Residual Strength Properties of Damaged Polymer Matrix Composite Plates", American Soc. for Testing and Materials, 2007.
 J.-H. Shin, O.-Y. Kwon and S.-W. Seo, "Acoustic Emission Monitoring of Compression-After-Impact Test of CFRP Laminates Damaged by Simulated Lightning Strikes", Progress in Acoustic Emission XV, JSNDI, 2010, pp. 115-120.
JAE-HA SHIN, OH-YANG KWON
Department of Mechanical Engineering, Inha University, Incheon 402-751, Korea
|Printer friendly Cite/link Email Feedback|
|Author:||Shin, Jae-Ha; Kwon, Oh-Yang|
|Publication:||Journal of Acoustic Emission|
|Date:||Jan 1, 2011|
|Previous Article:||Detection of transient zones during water boiling by acoustic emission.|
|Next Article:||Neural network techniques for burst pressure prediction in Kevlar/epoxy pressure vessels using acoustic emission data.|