·Table of Contents
·Methods and Instrumentation
New Technique for the Evaluation of Delaminations on Low-Cost Natural Fibers Composite Laminates
G. BRIOTTI, C. SCARPONI , Dipartimento Aerospaziale
C. CANEVA AND M. VALENTE, Dipartimento Ingegneria Chimica, Materiali, Materie Prime, Metallurgia
University of Rome "La Sapienza" via Eudossiana 18 00184 Rome, Italy
The purpose of the present paper is to describe the experimental results obtained by the application of a new NDI technique, based on an ultrasonic system in the form of reflection, for the evaluation of delaminations, originated by a transversal impact load at low-velocity. Such technique has been pointed out for different base materials and fabrication technologies for aerospace applications. In the present paper the methodology has been extended to Juta/vynilester or hybrid Juta/E-glass/vynilester thicker laminates, obtained by vacuum bag molding.
Among the natural fibers to be used as reinforcement in composite laminates, such as cotton, coconut and banana, Juta is considered the most promising, du to its good thermal and electrical insulation characteristics, appreciable thoughness and, as a very important topic, wide availability at a very low-cost; on the contrary, the specific mechanical properties are lower than the syntetic fibers.
The NDI system is comprised of an ultrasonic signal generator and echo acquisition instrument, a probe arm with two position encoders, a probe of 5 MHz (for this application) and a PC with a suitable software for the signal analysis, the elaboration and post treatment of the ultrasonic maps (2D and 3D).
The application of the Multiple Echoes Procedure allows the sound velocity evaluation; in this case the knowledge of the specimens thickness is needed. Owing the high attenuation, due to the material characteristics and the poor production technology, the Depth Amplitude Compensation function have been utilized to obtain signal equalization along the specimens thickness. Each specimen has been subjected to ultrasonic scan before and after the impact tests in order to measure the amount of the delaminated areas. Moreover, each scansion was performed on both front and rear sides of the specimens.
Two different scansion modes have been utilized: the Maximum Amplitude Mode and the Relative Depth Mode. By the first mode, the system can record only the maximum response echo; by the second one the first echo is considered for the ultrasonic map, as a percentage of the specimen thickness.
Such a procedure allows to have more information about the delaminations inside the specimen and to eliminate the problem of the shadow zone, that usually covers the first ply-to-ply interface.
Ultrasonic colors maps and graphs describe the delaminations position along the thickness, extention and dependence from the impact energy. A peculiar aspect of such materials is the presence of several defects also before the impact, owing to the production process. For this reason high attention is necessary for both the scansions before and after impact.
KeyWords: Ultrasonic N.D.I., frequency, attenuation, delamination, juta, glass fiber, vynilester.
IN THE RECENT YEARS there has been mounting interest in the use of natural fibers as plastics reinforcements because of their intrinsic properties: biodegradability, renewable raw material, low density, high specific stiffness and strength and, more important, low cost per unit volume.
Already, the range of commerical applications for products reinforced with plant fibers is extensive: electrical insulation, semistructural apllication, wear parts, automotive industries [1-5].
Indipendently from the application, the knowledge of position and extension of delaminations on composite structures or, more in general, the complete determination of the internal damage status is a very important topic.
A NDI ultrasonic system, in form of reflection, has been successfully used for the delamination detection on different kinds of base materials and fabrication technologies. Typical materials for aerospace applications such as thermoset and thermoplastic CFRP fabric and tape with epoxy resins, in form of thin laminates and sandwich panels, have been studied [6-8]. Then GFRP/polyester and KFRP/polyester laminates, obtained by RTM technology, have been successfully tested . For this application an ultrasonic technique with 5 and 15 Mhz probes was used.
It is well known that the ultrasonic NDI effectiveness depends on several factors, such as materials (phisical and mechanical properties, as density and elastic moduli; fiber form, as tape, fabric or wowen roving), fabrication technology (that can cause voids  or discontinuities), geometry (essentially the thickness ), probe frequency and last, but not least, the operator skill.
Comparative measurement [12,13] pointed out the importance of the probe frequency effects on the acoustic attenuation: the choice of the probe frequency is critical for the inspection and the criterium consists on a good compromise between the minimum size of the detectable defect and the maximum detectable thickness for the same material.
In the present paper the above said technique was applied to higher thickness laminates (3.5-8 mm), obtained by vacuum bag technology (with vynilester resin), using Juta fabric and E-glass fabric. The delaminations were produced by low-velocity impact tests performed by a spring gun impactor.
Due to the fabrication technology and material characteristics a wide range of internal defects were espected in all the specimens, thus a N.D.I. inspection was performed both before and after the impact tests as quality verification.
As a result, delamination areas evaluated by DAC function were determined for the selected materials.
MATERIALS AND SPECIMENS
The basic mechanical and physical properties for resin and fibers are reported in Tabb. 1,2. The ATCAC 850 is a vynilester resin produced by D.S.M. The fiber systems are all in form of woven fabric.
|Tab. 1. Fiber materials|
|Tab. 2. Resin materials|
Seven lay-up series were utilized: two based only on glass fiber, one based on juta fiber and four based on both juta and glass fiber. The mechanical and phisycal properties of the laminate series are reported in Tab. 3. All the specimen are square of 100 mm in side and depending on the lay-up, the thiknesses are from 3.5 mm to 8 mm.
|Tab. 3. Mechanical and physical properties of the laminates |
The specimens were marked with suitable codes: aaJjjNnns where aa are two characters for series identification, jj is a two digit integer for the impact energy value, nn is a two digits integer for the specimen progressive order and s in a one character code used in the scansion in order to evidentiate the scan kind (Depth or Amplitude Mode, top or bottom inspection side).
In Tab. 4 the specimens layup are reported.
||(E1)3 (J)2 E1 (J)2 (E1)6
||(E1)4 (J)2 (E1)2 (J)2(E1)4
||E2 (J)2 E2 (J)2 (E2)3
||(E2)2 (J)2 E2 (J)2 (E2)2
|Tab. 4. Specimens lay-up|
|Tab. 5. Sound velocity evaluation by USD-10 set up procedure|
1. IMPACT TEST PROCEDURE
The impact test was performed using an instrumented spring gun tester (Fig. 1). The projectile had an instrumented impactor with a hemi-spherical nose of 12.7 mm in diameter an it runs along two rails. The impact tester had a photo-electric sensor in order to measure the impactor velocity at the moment of contact beginning. The specimens were clamped by a square frame accordingly with the certification test [7, 8, 14-16].
All the tests were performed at nominal initial contact velocity of 2 m/s to 3.8 m/s and impactor mass of 2,5 Kg. As a consequence, the impactor energies observed are from 5 to 20 Joule.
Fig 1: Impact test apparatus
After the impact, due to the existing correlation between internal and barely visible damage, for each specimen the indentation was evaluated by a suitable gauge.
1. ULTRASONIC TEST APPARATUS
The NDI system is comprised of the following components:
- an ultrasonic signal generator and echo acquisition instrument, called USD 10, with a display and an user oriented menu for the I/O operations;
- a 5 MHz probe for both the emission and reception of the ultrasonic beam;
- a probe arm, called ANDSCAN, fixed on a reference surface by a simple suction-cup system. The arm has two position encoders, in order to obtain the probe position referred to the specimen surface;
- a PC386 with a dedicated software, that allows several operations such as scan map plotting in real--time and the image post treatment (e.g. change of colour scale, zooming, different angle point of view, sections, 3-D images, overpositioning of different scansion, etc.). The computer is connected with a colour printer.
Two data displays are available: the computer monitor (C-scan) and the USD-10 display (A-scan). Because of the data received from the probe position transducer and the USD-10, several complete scan maps are available on the computer monitor (e.g. 2-D, 3-D maps) for an immediate data evaluation.
The USD-10 display shows several echo peaks due to the interlaminar surfaces. When a discontinuity increases (i.e. a delamination occours) the corresponding peak rises, due to the high reflection of the ultra-sonic wave. If the peak reaches the instrument gate (horizontal marked line), its position in depth is shown on the USD-10 display. The first and last peaks correspond to the top and bottom surfaces of the specimen. More details on ultrasonic system are given in [6,9].
2. INSPECTION PROCEDURE
The investigation methodology is widely described in [6,9]. For a better understanding of the paper the main items are here described. This tester allowed two working modes: the Maximum Amplitude Mode and the Relative Depth Mode.
By the Amplitude Mode the maximum ultrasonic echo amplitudes (i.e. the highest peak in the gate range) is measured, thus a precise damage localization was possible, because of the relationship between the damage extent and the echo amplitude. The USD-10 is arranged in order to measure and record the highest peak level. Because of the first and bottom peak high level, the operator must exclude them from the gate. As a result of this inspection, a colours map is obtained. The colours scale indicates the reflection energy level.
By the Relative Depth Mode the ultrasonic echo delay is measured, that is, the damage localization in terms of relative depth, as a percentage of the specimen thickness. In this case, the USD-10 is arranged in order to measure and record the level of the peak above to the threshold limit and nearest to the specimen top surface (flank mode). Because of this working mode, the operator can retain the last echo inside the gate, but always the first echo must be kept out of the gate. In the Depth Map the colours scale indicates the relative depth, that is the damage position as a percentage of the specimen thickness.
Each specimen was subjected to a double inspection: from the impact side and the opposite side with both the Amplitude and Depth Mode.
3. DAC FUNCTION
Due to the high scattering and general high attenuation, the DAC (Depth Amplitude Compensation) function (Fig. 2) that allows the signal equalization, has been applied for all the tests .
Fig 2: Scheme of signal equalization by DAC function.
The indentation evaluation is shown in Fig. 3. It is possible to see the higher (about ten times) indentation for juta fiber based laminates in respect to the other laminates. This behaviour seems related to the higher thickness and poor mechanical properties of the J9 specimen
As espected, due to the importance of first ply composition for the indentation phenomena, the other laminates show a very similar behaviour.
Fig 3.a: Indentations
Fig 3.b: Indentations
Fig 3.c: Indentations
The delamination areas are reported in Tab. 6. In this table, for each laminate, the original defect area is reported (i.e. the defects due to the fabrication process and material characteristics); then for each impact energy level in the others columns the increase of delamination due to the impact phenomena are reported. The delamination area for the J9 specimens impacted at 20 Joule are not reported because the measure uncertainty due to the high indentation caused by the impact test.
|Delam. increases due to impact|
|Tab. 6. Delamination areas|
In Fig. 4 the relationship between delamination areas and impact energy is shown for glass (Fig. 3.a.), hybrid (Fig. 3.b.) and juta (Fig. 3.c.) laminates.
It is interesting to observe that the juta fiber based specimens (i.e. J9 series) seem to absorb the impact energy by indentation more than by delamination.
Fig 4.a: Delaminations
Fig 4.b: Delaminations
Fig 4.c: Delaminations
The test performed by ultrasonic N.D.I. system show good accuracy even for these materials with high attenuation and scattering. Of course, the probe and frequency choices become mandatory for the correct internal damage evaluation. Moreover, a good set up of the ultrasonic system in terms of sound velocity, probe delay and D.A.C. function is required.
The authors are gratefully to the Aerospace Engineering Dept. staff at Politecnico di Milano for the impact test.
- Bolton A.J., 1994 "Natural fibers for plastic reinforcement" Mat Tech vol. 9 pp. 12-20.
- Piercy R., Kelbert E., 1999 "Natural fibres reinforced polymer matrix composites" p. 653
- Gowda T.M., A.C.B. Naidu, R. Chaya, 1999 "Some mechanical properties of jute fabric reinforced polyester composites", Composites Part A, n. 30, pp. 277-284.
- Van Dam E.G., 1993 "Environmental advantages of application of juta fibers in innovative industrial products", FAO-FIAT ESC: JU/IC 93/12 August.
- Rowell R.M., 1993 "Opportunities for composite materials from jute and Kenaf ", FAO-FIAT ESC: JU/IC 93/15 September.
- Scarponi C., G. Briotti, 1997 "Ultrasonic detection of delaminations on composite materials" Jrnl. of Reinforced Plastics and Composites vol. 16, n. 9, pp. 768-790.
- C. Scarponi, G. Briotti, R. Barboni, A. Marcone "Reduction of Tensile Strength in Angle-ply Composite Laminates due to Low-Velocity Impact" Jrnl. of Reinforced Plastics vol. 18, n. 1/1999, pp. 63-85.
- Scarponi C., G. Briotti, R. Barboni, A. Marcone, M. Iannone, 1996 "Impact testing on composites laminates and sandwich panels" Jrnl. of Composite Materials vol. 30, n. 17, pp. 1873-1911.
- Scaroni C., G. Briotti, 1999 "Ultrasonic techniques for the evaluation of delaminations on CFRP, KFRP and GFRP composite laminates", Proceedings of ICCE/6 Orlando (USA), June, pp. 727-728.
- H. Jeong, 1997 "Effect of void on the mechanical strength and ultrasonic attenuation of laminated composites" Jrnl. of Composite Materials, 31:277-292.
- D.K. Hsu, M.S. Hughes, 1992 "Simultaneous ultrasonic velocity and sample thickness measurement and applications in composites" Jrnl. of Acoustical Society of America Vol. 92, n. 2 August pp. 669.
- Briotti G., C. Scarponi, 1999 "The importance of the acoustic attenuation for the ultrasonic N.D.I. detection of delaminations on composite laminates" Jrnl. of Reinforced Plastics and Composites [in press].
- T.D. Lhermitte, S.M. Handley, M.R. Holland, J.G. Miller, 1992 "Anisotropy of the Frequency-Dependent Ultrasonic Attenuation in Unidirectional Graphite/Epoxy Composite Material" Proc. IEEE Ultrasonics Symposium, Orlando, pp 819-823.
- Iannone M., A. Marcone, M. de Leo, P. Ferraro, 1993 "Impact damage in composites - controlling factors" Ninth Intrnl. Conference on Composite Materials, 12-16 July, Madrid, Spain.
- P. Kumar, B. Rai, 1993 "Delaminations of barely visible impact damage in CFRP laminates", Composite Structures vol. 23, pp. 313-318.
- NTA 62463. Alenia internal technical note.
- H. Kaczmarek, 1995 "Ultrasonic detection of damage in CFRPs" Jrnl. of Composite Materials, 29:59-95.
- J. Krautkrämer, H. Krautkrämer, 1990 Ultrasonic Testing of Materials, Ed. Springer Verlag.
- A.E. Brown, "Rationale and summary of methods for determining ultrasonic properties of materials at Lawrence Livermore National Laboratory", UCRL-ID-119958 L. Livermore Nat. Lab.
- A. R. Selfridge "Approximate Material Properties in Isotropic Materials", IEEE Transactions on Sonics and Ultrasonics, vol SU-32, N. 3, May 1985
- A. P. Christoforou, A. S. Yigit, 1994 "Normalized impact response and damage in a thin composite laminate supported by a rigid substrate" Jrnl. of Composite Materials, 28:1553-1573
- Sjöblom, P. O., J. T. Hartness and T.M. Cordell. 1988. "On low-velocity impact testing of composite materials", Jrnl. of Composite Materials, 22:30-52.
- Wang, T. Vu-Khanh, 1994 "Damage extension in carbon fiber/PEEK crossply laminates under low-velocity impact", Jrnl. of Composite Materials, 28:684-707.