- Email: [email protected]
Damage Identification and Mechanical Assessment of Impacted Sandwich Composites
Available online at www.sciencedirect.com
ScienceDirect Procedia Engineering 188 (2017) 178 – 185
6th Asia Pacific Workshop on Structural Health Monitoring, 6th APWSHM
Damage identification and mechanical assessment of impacted sandwich composites Nicolae Constantin*, Marin Sandu, Adriana Sandu, Mircea Găvan The University “Politehnica” of Bucharest, Laboratory for Integrity Evaluation of Composite Structures Splaiul Independenţei 303, Bucharest 060042, Romania
Abstract Sandwich composites are often subjected to low velocity impact, a random and thus feared type of loading. Vulnerability of sandwich materials under such an event depends on the skins and core structure and skin-core interfaces. The paper presents ways to identify the damage severity occurred during low velocity impact tests, with further assessment of the residual mechanical performance of some kind of sandwich composites. In this study, non-destructive inspections (NDI) have been mixed with mechanical destructive testing in a comprehensive structural health monitoring (SHM) approach, meant to evaluate the damage tolerance of such materials, intended to be used on some structural components of a green energy power unit. © 2016 2016The TheAuthors. Authors. Published by Elsevier Ltd. is an open access article under the CC BY-NC-ND license © Published by Elsevier Ltd. This (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of the 6th APWSHM. Peer-review under responsibility of the organizing committee of the 6th APWSHM Keywords: sandwich materials; low velocity impact; infrared thermography; bending after impact; residual mechanical performance.
1. Introduction The sandwich type materials are introducing an unrivalled variety of materials, with associated mechanical parameters, that are recommending them for many and diversified applications. Like all layered composites, sandwiches are prone to many structural threats during their service life, due to the particular interface between layers and the characteristics of each one. Difficulties in ensuring full reproducibility in the manufacturing technology can add extra-problems in offering reliable mechanical reference. Low velocity impact, followed by shear bending loading is the kind of in service event which can take advantage of these vulnerabilities and
* Corresponding author. Tel.: +40-21-402-9210; fax:+0-000-000-0000 . E-mail address: [email protected]
1877-7058 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of the 6th APWSHM
doi:10.1016/j.proeng.2017.04.472
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
compromise the safety of sandwich materials and associated structures. For this reason, extended tests have to be run in order to identify and characterize the damages inflicted during a variety of probable invasive events and evaluate the consequences on the structural performance, in a comprehensive SHM approach. A rich literature on sandwich materials reported results obtained worldwide during experimental work and modeling/simulation of low velocity impact [1]. It accounted the main aspects of this phenomenon: recording the main parameters during the impact, characterization of the incurred damages and the consequences they have on the residual mechanical performance of the material, damage detection, and damage initiation or progress prediction. Recent works focused in particular on the second [2,3,4] and third aspect [5,6]. This paper presents results obtained during low velocity impact tests, intermediate NDI using infrared thermography (IRT) and final assessment of the residual mechanical performance with bending after impact (BAI) tests, all performed on a category of sandwiches intended to be used as platforms of a wind turbine supporting tower, on which water tanks will be mounted. 2. Tested materials The materials used in this series of tests have been of two types: x Sandwich composites with honeycomb core x Sandwich composites with foam core Both sandwich types had glass fiber reinforced plastic (GFRP) skins. There were two thicknesses considered for the skin, corresponding to the number of layers. The thin skins, averaging 2.3 mm in thickness, involved four layers: two reinforced with felt type fabric at the outside and two reinforced with roving type fabric in the middle. The thick skins, averaging 4 mm in thickness, involved seven layers, with three more layers in the middle, reinforced with roving type fabric. In both cases, the matrix was a unique polyester resin. A gel coat layer was applied on the faces finally placed on the outside of the sandwich panel, while the inside layers in contact with the core material, reinforced with felt type fabric, ensured a stronger skin-core interface. The honeycomb core was made of impregnated paper, shaped in 9 mm side hexahedral cells. The foam core was made of medium density polyurethane foam. Both core materials were 30 mm thick. The tested materials have been single core or double core sandwiches, in the second case a single felt fabric reinforced layer separating the two cores in the sandwich mid-plane. The samples made from these materials had 100x450 mm in-plane dimensions. That dimensions complied with the rig used for mounting the standard low velocity impact samples [7] and with the requirements of the standard concerning the bending tests of sandwich composites [8]. 3. Low velocity impact tests The low velocity impact tests were performed with an instrumented impact hammer (Fig. 1), with the fixture used for standard laminated samples described in [7]. The DAQ NI USB 6009 collected data from a force transducer placed behind the hemispherical, 15 mm in diameter steel impact head, fitted with strain gauge sensors, with a 2,500 to 5,000 sampling rate. The impact tests have been performed at 10 J and 20 J energy levels. The impact force history had, in most of the cases, quite different shapes as compared to those obtained during impact events on laminates and thin sandwiches [9]. In Fig. 2 and Fig. 3 are presented the curves obtained for the samples proving the weakest post-impact behavior during BAI tests. Deep craters, close to penetration, were observed only on the samples having honeycomb cores and thin faces, at 20 J impact energy levels. In the other cases, the effect of the impact event was a small indentation or barely visible damages (BVD). Some honeycomb core sandwiches with thick faces experienced local buckling of the honeycomb paper wall under the impact area. The impact force history curves showed important distortions in the last situations or when a near penetration deep crater occurred, with consistent lowering of the peak force value. The variability of the impact force history curves obtained for sandwiches with honeycomb core depended also on the part of the cell opposed to the impact point: cell wall or empty mid-cell. The dips observed on many curves are supposed to denote the occurrence of the latter case or local skin-core de-bonding in the case of sandwiches with foam core.
179
180
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
Fig. 1. Instrumented impact hammer
a)
b)
Fig. 2. Impact force history curves obtained for sandwiches with honeycomb core and a) thin skins or b) thick skins.
a)
b)
Fig. 3. Impact force history curves obtained for sandwiches with foam core and a) thin skins or b) thick skins.
The link between the shape of the impact force history curves and the severity of the damage occurred in the sandwich material, with possible consequences on the residual mechanical performance, was cross-checked using IRT as an instrument for NDI of the impacted samples, in the way presented in the next section.
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
4. NDI through IRT The assessment of the damage state of the impacted samples was done using active IRT, with lockin technique. The setup is presented in Fig. 4.
Fig. 4. NDI setup for damage assessment using IRT.
The damage state was put in evidence quite differently, function of the different analysis variants of the lockin technique. The thermograms obtained for samples belonging to all four categories of assessed sandwich materials and exhibiting the weakest residual strength (see the next section) are presented in Fig. 5 to Fig. 8.
a)
b)
Fig. 5. Thermograms obtained with lockin IRT (a) on a 10 J impacted honeycomb core sandwich sample with thin skins, using the harmonic approximation frequency amplitude (HAFA) analysis and (b) on the same sample, using the harmonic approximation time amplitude (HATA) analysis.
a)
b)
Fig. 6. Thermograms obtained with lockin IRT (a) on a 20 J impacted honeycomb core sandwich sample with thick skins, using HAFA analysis and (b) on the same sample, using the HATA analysis.
181
182
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
a)
b)
Fig. 7. Thermograms obtained with lockin IRT (a) on a 20 J impacted foam core sandwich sample with thin skins, using HAFA analysis and (b) on the same sample, using the HATA analysis.
a)
b)
Fig. 8. Thermograms obtained with lockin IRT (a) on a 20 J impacted foam core sandwich sample with thick skins, using HAFA analysis and (b) on the same sample, using the HATA analysis.
From Fig. 5 and Fig. 6 one can see that, in the case of sandwiches with honeycomb core, the difference between the HAFA and HATA analysis is significant: the first put in evidence only the local damage, while the second, damages extended in the surrounding area. On the other hand, in the case of sandwiches with foam core, Fig. 7 and Fig. 8 show that there is no important difference between the thermograms obtained with HAFA and HATA analysis, both retaining only local damages occurred during the impact event. The dark spot on the upper left sample surface visible in Fig. 8 is due to some stain on the sample surface. These observations will prove very much in line with the results presented in the next section. 5. Damage tolerance assessment The final evaluation of the sandwich materials was done through the BAI tests. As mentioned before, the samples have been tailored to dimensions meant to comply with the standards concerning the impact tests [7] and bending tests [8]. In this way, the width of the samples was big enough to make relevant the bending tests of samples made of sandwiches with honeycomb core. It was proved that small width for such samples, in comparison with the cell dimensions, can be source for additional scattering of results [10]. The averaged data obtained from BAI tests for each category of sandwich materials previously subjected to 10 J and 20 J low velocity impact tests, together with results obtained from bending tests applied to pristine samples, are presented in Fig. 9 to Fig. 12. The results have to be correlated with previous results to have sound conclusions
183
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
concerning the damage tolerance of the materials studied in this research work. It is obvious that the sandwiches having honeycomb core have a weak damage tolerance to low velocity impact events. This intolerance is almost dramatic in the case of sandwiches with thick skins. This happens in spite of BVD class impact consequences observed for such sandwiches, face to deep craters appeared in the case of honeycomb sandwiches with thin skins. The delaminations on large surfaces put in evidence by IRT scans for that kind of sandwiches following impact events (see Fig. 5,b and Fig. 6,b) explain that high sensitivity to such aggressive loading. Also, the flat impact force history curves, at quite low impact force levels (see Fig. 2,b) can predict the extended internal damages, like delaminations and cell buckling, paving the way towards consistently diminished residual strength of honeycomb sandwiches with thick skins.
a)
b)
Fig. 9. Averaged results obtained during bending tests made on pristine and impacted sandwich materials with honeycomb core and (a) thin skins or (b) thick skins.
a)
b)
Fig. 10. Averaged results obtained during bending tests made on pristine and impacted sandwich materials with foam core and (a) thin skins or (b) thick skins.
184
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
On the other hand, the damage tolerance of the sandwiches with foam core appears to be very good, proved by almost unchanged bending behavior of impacted samples versus that of pristine samples (see figures above). The encouraging forecast made by the increased impact force level (see Fig. 3,b) is backed by the thermograms, which are indicating only local damages, visible through both used analysis (see Fig. 7 and Fig. 8). Otherwise, the maximum force during the bending tests of such materials follows almost a plateau till the displacement reaches at least four times the values at which the maximum force for the sandwiches with honeycomb core is reached. This behavior can be explained by the very much delayed core-skins delamination (Fig. 11,a) or core fracture. The considerably lower maximum force is due to the continuously diminishing overall thickness under the loading force, in conditions of reduced compression resistance of the foam core. Due to the much weaker core-skin interface, the delaminations in sandwiches with honeycomb core appears shortly after the maximum force is reached, being the main cause of the abrupt fall of that force (Fig. 11,b).
a)
b)
Fig. 11. Critical bending loading with (a) large displacement with no delamination in the case of honeycomb core sandwich material and with (b) generalized upper skin-core delamination in the case of foam core sandwich material.
6. Final conclusions and further research work The results of the combined evaluation performed on the sandwich materials described in this study prove that their behavior is quite different in terms of strength, stiffness and damage tolerance. The adequacy of one or the other depends on the conditions of exploitation during their service life: sandwich material with honeycomb core for structures with no excessive deformations and prevented from low velocity impact events and sandwich material with foam core for structures exposed at large deformations and probable low velocity impact events. The much higher strength in pristine conditions makes the former very attractive in usual in service conditions concerning the loads and the environment. Nevertheless, in order to increase the damage tolerance of the former, solutions for reducing the effect of low velocity impact events are considered, by applying damping layers on the exposed faces [10]. The damage assessment and the evaluation of the residual mechanical performance in such conditions will be made in the near future. Evaluations of double core sandwiches, in terms of sensitivity to low velocity impact events and subsequent residual mechanical performance is also underway. The middle layer is expected to have important effect in preventing the buckling of the cell wall for sandwiches with honeycomb core. The differences in the thermograms obtained with different analysis options clearly show that NDI through IRT must take into consideration and take advantage of the quite large number of techniques used by this fast NDI method, which have to be carefully tuned on the particular inspected material. The review of the impact force history curves shows that their shape is generally quite different from that recorded for laminates and thin sandwiches. Behind that difference, there are some common features that can predict damages which can consistently affect the residual mechanical performance: (i) the high irregularities on the
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
ascending or descending parts and (ii) the unusual low values in comparison with curves recorded at lower energy levels. In predictive or on condition maintenance management, cross-checking is highly necessary for ensuring high in service reliability, having in view the important number of NDI methods and progress in the performance of related equipment. Acknowledgement The research work was performed under project 258/2014, financed by the Romanian Ministry of Education and Scientific Research, through its dedicated body (MEN-UEFISCDI), in the “Partnership in priority domains – PN II” programme. References [1] G.B. Chai, S. Zhu, A review of low velocity impact on sandwich structures, J. Mat.: Design and Appl. 225(4) (2011) 207–230. [2] A. Shipsha, D.Zenkert, Compression-after-impact strength of sandwich panels with core crushing damage, Appl. Composite Mat. 12(3) (2005) 149-164. [3] K.S. Raju, B.L. Smith, Impact damage resistance and tolerance of honeycomb core sandwich panels, J. Composite Mat. 42(4) (2008) 385-413. [4] I.M. Daniel, Impact response and damage tolerance of composite sandwich structures, in Dynamic Failure of Materials and Structures, Springer, London, 2010, pp. 191-233. [5] K.Diamanti, C. Soutis, J.B. Hodgkinson, Non-destructive inspection of sandwich and repaired composite laminated structures, Composites Sci. Tech. 65 (2005) 2059–2067. [6] J. Wang, A.M. Waas, H. Wang, Experimental study on the low-velocity impact behavior of foam-core sandwich panels, 53rd AIAA/ASME/ASCE/ASC Structures, Structural Dynamics and Materials Conference, Honolulu, Hawaii (2012), to be found on http://arc.aiaa.org/DOI:10.2514/6.2012-1701. [7] ASTM D 7136 – Standard test method for measuring the damage resistance of a fiber reinforced polymer matrix composite to a drop impact event, in Annual Book of ASTM Standards 2005, pp. 430-445. [8] ASTM Standard C 293-11 (2011) “Flexural properties of flat sandwich constructions” American Society for Testing and Materials, West Conshohocken, Pennsylvania (first issued in 1957). [9] N. Constantin et all., Links between low velocity impact results and the residual mechanical performance of sandwich composite materials, Advanced Manufacturing Technologies 834 (2016) 167-172. [10] N. Constantin et all., Some aspects concerning the mechanical characterization versus manufacturing technology of sandwich composite materials, Advanced Manufacturing Technologies 834 (2016) 161-166..
185
ScienceDirect Procedia Engineering 188 (2017) 178 – 185
6th Asia Pacific Workshop on Structural Health Monitoring, 6th APWSHM
Damage identification and mechanical assessment of impacted sandwich composites Nicolae Constantin*, Marin Sandu, Adriana Sandu, Mircea Găvan The University “Politehnica” of Bucharest, Laboratory for Integrity Evaluation of Composite Structures Splaiul Independenţei 303, Bucharest 060042, Romania
Abstract Sandwich composites are often subjected to low velocity impact, a random and thus feared type of loading. Vulnerability of sandwich materials under such an event depends on the skins and core structure and skin-core interfaces. The paper presents ways to identify the damage severity occurred during low velocity impact tests, with further assessment of the residual mechanical performance of some kind of sandwich composites. In this study, non-destructive inspections (NDI) have been mixed with mechanical destructive testing in a comprehensive structural health monitoring (SHM) approach, meant to evaluate the damage tolerance of such materials, intended to be used on some structural components of a green energy power unit. © 2016 2016The TheAuthors. Authors. Published by Elsevier Ltd. is an open access article under the CC BY-NC-ND license © Published by Elsevier Ltd. This (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of the 6th APWSHM. Peer-review under responsibility of the organizing committee of the 6th APWSHM Keywords: sandwich materials; low velocity impact; infrared thermography; bending after impact; residual mechanical performance.
1. Introduction The sandwich type materials are introducing an unrivalled variety of materials, with associated mechanical parameters, that are recommending them for many and diversified applications. Like all layered composites, sandwiches are prone to many structural threats during their service life, due to the particular interface between layers and the characteristics of each one. Difficulties in ensuring full reproducibility in the manufacturing technology can add extra-problems in offering reliable mechanical reference. Low velocity impact, followed by shear bending loading is the kind of in service event which can take advantage of these vulnerabilities and
* Corresponding author. Tel.: +40-21-402-9210; fax:+0-000-000-0000 . E-mail address: [email protected]
1877-7058 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of the 6th APWSHM
doi:10.1016/j.proeng.2017.04.472
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
compromise the safety of sandwich materials and associated structures. For this reason, extended tests have to be run in order to identify and characterize the damages inflicted during a variety of probable invasive events and evaluate the consequences on the structural performance, in a comprehensive SHM approach. A rich literature on sandwich materials reported results obtained worldwide during experimental work and modeling/simulation of low velocity impact [1]. It accounted the main aspects of this phenomenon: recording the main parameters during the impact, characterization of the incurred damages and the consequences they have on the residual mechanical performance of the material, damage detection, and damage initiation or progress prediction. Recent works focused in particular on the second [2,3,4] and third aspect [5,6]. This paper presents results obtained during low velocity impact tests, intermediate NDI using infrared thermography (IRT) and final assessment of the residual mechanical performance with bending after impact (BAI) tests, all performed on a category of sandwiches intended to be used as platforms of a wind turbine supporting tower, on which water tanks will be mounted. 2. Tested materials The materials used in this series of tests have been of two types: x Sandwich composites with honeycomb core x Sandwich composites with foam core Both sandwich types had glass fiber reinforced plastic (GFRP) skins. There were two thicknesses considered for the skin, corresponding to the number of layers. The thin skins, averaging 2.3 mm in thickness, involved four layers: two reinforced with felt type fabric at the outside and two reinforced with roving type fabric in the middle. The thick skins, averaging 4 mm in thickness, involved seven layers, with three more layers in the middle, reinforced with roving type fabric. In both cases, the matrix was a unique polyester resin. A gel coat layer was applied on the faces finally placed on the outside of the sandwich panel, while the inside layers in contact with the core material, reinforced with felt type fabric, ensured a stronger skin-core interface. The honeycomb core was made of impregnated paper, shaped in 9 mm side hexahedral cells. The foam core was made of medium density polyurethane foam. Both core materials were 30 mm thick. The tested materials have been single core or double core sandwiches, in the second case a single felt fabric reinforced layer separating the two cores in the sandwich mid-plane. The samples made from these materials had 100x450 mm in-plane dimensions. That dimensions complied with the rig used for mounting the standard low velocity impact samples [7] and with the requirements of the standard concerning the bending tests of sandwich composites [8]. 3. Low velocity impact tests The low velocity impact tests were performed with an instrumented impact hammer (Fig. 1), with the fixture used for standard laminated samples described in [7]. The DAQ NI USB 6009 collected data from a force transducer placed behind the hemispherical, 15 mm in diameter steel impact head, fitted with strain gauge sensors, with a 2,500 to 5,000 sampling rate. The impact tests have been performed at 10 J and 20 J energy levels. The impact force history had, in most of the cases, quite different shapes as compared to those obtained during impact events on laminates and thin sandwiches [9]. In Fig. 2 and Fig. 3 are presented the curves obtained for the samples proving the weakest post-impact behavior during BAI tests. Deep craters, close to penetration, were observed only on the samples having honeycomb cores and thin faces, at 20 J impact energy levels. In the other cases, the effect of the impact event was a small indentation or barely visible damages (BVD). Some honeycomb core sandwiches with thick faces experienced local buckling of the honeycomb paper wall under the impact area. The impact force history curves showed important distortions in the last situations or when a near penetration deep crater occurred, with consistent lowering of the peak force value. The variability of the impact force history curves obtained for sandwiches with honeycomb core depended also on the part of the cell opposed to the impact point: cell wall or empty mid-cell. The dips observed on many curves are supposed to denote the occurrence of the latter case or local skin-core de-bonding in the case of sandwiches with foam core.
179
180
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
Fig. 1. Instrumented impact hammer
a)
b)
Fig. 2. Impact force history curves obtained for sandwiches with honeycomb core and a) thin skins or b) thick skins.
a)
b)
Fig. 3. Impact force history curves obtained for sandwiches with foam core and a) thin skins or b) thick skins.
The link between the shape of the impact force history curves and the severity of the damage occurred in the sandwich material, with possible consequences on the residual mechanical performance, was cross-checked using IRT as an instrument for NDI of the impacted samples, in the way presented in the next section.
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
4. NDI through IRT The assessment of the damage state of the impacted samples was done using active IRT, with lockin technique. The setup is presented in Fig. 4.
Fig. 4. NDI setup for damage assessment using IRT.
The damage state was put in evidence quite differently, function of the different analysis variants of the lockin technique. The thermograms obtained for samples belonging to all four categories of assessed sandwich materials and exhibiting the weakest residual strength (see the next section) are presented in Fig. 5 to Fig. 8.
a)
b)
Fig. 5. Thermograms obtained with lockin IRT (a) on a 10 J impacted honeycomb core sandwich sample with thin skins, using the harmonic approximation frequency amplitude (HAFA) analysis and (b) on the same sample, using the harmonic approximation time amplitude (HATA) analysis.
a)
b)
Fig. 6. Thermograms obtained with lockin IRT (a) on a 20 J impacted honeycomb core sandwich sample with thick skins, using HAFA analysis and (b) on the same sample, using the HATA analysis.
181
182
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
a)
b)
Fig. 7. Thermograms obtained with lockin IRT (a) on a 20 J impacted foam core sandwich sample with thin skins, using HAFA analysis and (b) on the same sample, using the HATA analysis.
a)
b)
Fig. 8. Thermograms obtained with lockin IRT (a) on a 20 J impacted foam core sandwich sample with thick skins, using HAFA analysis and (b) on the same sample, using the HATA analysis.
From Fig. 5 and Fig. 6 one can see that, in the case of sandwiches with honeycomb core, the difference between the HAFA and HATA analysis is significant: the first put in evidence only the local damage, while the second, damages extended in the surrounding area. On the other hand, in the case of sandwiches with foam core, Fig. 7 and Fig. 8 show that there is no important difference between the thermograms obtained with HAFA and HATA analysis, both retaining only local damages occurred during the impact event. The dark spot on the upper left sample surface visible in Fig. 8 is due to some stain on the sample surface. These observations will prove very much in line with the results presented in the next section. 5. Damage tolerance assessment The final evaluation of the sandwich materials was done through the BAI tests. As mentioned before, the samples have been tailored to dimensions meant to comply with the standards concerning the impact tests [7] and bending tests [8]. In this way, the width of the samples was big enough to make relevant the bending tests of samples made of sandwiches with honeycomb core. It was proved that small width for such samples, in comparison with the cell dimensions, can be source for additional scattering of results [10]. The averaged data obtained from BAI tests for each category of sandwich materials previously subjected to 10 J and 20 J low velocity impact tests, together with results obtained from bending tests applied to pristine samples, are presented in Fig. 9 to Fig. 12. The results have to be correlated with previous results to have sound conclusions
183
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
concerning the damage tolerance of the materials studied in this research work. It is obvious that the sandwiches having honeycomb core have a weak damage tolerance to low velocity impact events. This intolerance is almost dramatic in the case of sandwiches with thick skins. This happens in spite of BVD class impact consequences observed for such sandwiches, face to deep craters appeared in the case of honeycomb sandwiches with thin skins. The delaminations on large surfaces put in evidence by IRT scans for that kind of sandwiches following impact events (see Fig. 5,b and Fig. 6,b) explain that high sensitivity to such aggressive loading. Also, the flat impact force history curves, at quite low impact force levels (see Fig. 2,b) can predict the extended internal damages, like delaminations and cell buckling, paving the way towards consistently diminished residual strength of honeycomb sandwiches with thick skins.
a)
b)
Fig. 9. Averaged results obtained during bending tests made on pristine and impacted sandwich materials with honeycomb core and (a) thin skins or (b) thick skins.
a)
b)
Fig. 10. Averaged results obtained during bending tests made on pristine and impacted sandwich materials with foam core and (a) thin skins or (b) thick skins.
184
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
On the other hand, the damage tolerance of the sandwiches with foam core appears to be very good, proved by almost unchanged bending behavior of impacted samples versus that of pristine samples (see figures above). The encouraging forecast made by the increased impact force level (see Fig. 3,b) is backed by the thermograms, which are indicating only local damages, visible through both used analysis (see Fig. 7 and Fig. 8). Otherwise, the maximum force during the bending tests of such materials follows almost a plateau till the displacement reaches at least four times the values at which the maximum force for the sandwiches with honeycomb core is reached. This behavior can be explained by the very much delayed core-skins delamination (Fig. 11,a) or core fracture. The considerably lower maximum force is due to the continuously diminishing overall thickness under the loading force, in conditions of reduced compression resistance of the foam core. Due to the much weaker core-skin interface, the delaminations in sandwiches with honeycomb core appears shortly after the maximum force is reached, being the main cause of the abrupt fall of that force (Fig. 11,b).
a)
b)
Fig. 11. Critical bending loading with (a) large displacement with no delamination in the case of honeycomb core sandwich material and with (b) generalized upper skin-core delamination in the case of foam core sandwich material.
6. Final conclusions and further research work The results of the combined evaluation performed on the sandwich materials described in this study prove that their behavior is quite different in terms of strength, stiffness and damage tolerance. The adequacy of one or the other depends on the conditions of exploitation during their service life: sandwich material with honeycomb core for structures with no excessive deformations and prevented from low velocity impact events and sandwich material with foam core for structures exposed at large deformations and probable low velocity impact events. The much higher strength in pristine conditions makes the former very attractive in usual in service conditions concerning the loads and the environment. Nevertheless, in order to increase the damage tolerance of the former, solutions for reducing the effect of low velocity impact events are considered, by applying damping layers on the exposed faces [10]. The damage assessment and the evaluation of the residual mechanical performance in such conditions will be made in the near future. Evaluations of double core sandwiches, in terms of sensitivity to low velocity impact events and subsequent residual mechanical performance is also underway. The middle layer is expected to have important effect in preventing the buckling of the cell wall for sandwiches with honeycomb core. The differences in the thermograms obtained with different analysis options clearly show that NDI through IRT must take into consideration and take advantage of the quite large number of techniques used by this fast NDI method, which have to be carefully tuned on the particular inspected material. The review of the impact force history curves shows that their shape is generally quite different from that recorded for laminates and thin sandwiches. Behind that difference, there are some common features that can predict damages which can consistently affect the residual mechanical performance: (i) the high irregularities on the
Nicolae Constantin et al. / Procedia Engineering 188 (2017) 178 – 185
ascending or descending parts and (ii) the unusual low values in comparison with curves recorded at lower energy levels. In predictive or on condition maintenance management, cross-checking is highly necessary for ensuring high in service reliability, having in view the important number of NDI methods and progress in the performance of related equipment. Acknowledgement The research work was performed under project 258/2014, financed by the Romanian Ministry of Education and Scientific Research, through its dedicated body (MEN-UEFISCDI), in the “Partnership in priority domains – PN II” programme. References [1] G.B. Chai, S. Zhu, A review of low velocity impact on sandwich structures, J. Mat.: Design and Appl. 225(4) (2011) 207–230. [2] A. Shipsha, D.Zenkert, Compression-after-impact strength of sandwich panels with core crushing damage, Appl. Composite Mat. 12(3) (2005) 149-164. [3] K.S. Raju, B.L. Smith, Impact damage resistance and tolerance of honeycomb core sandwich panels, J. Composite Mat. 42(4) (2008) 385-413. [4] I.M. Daniel, Impact response and damage tolerance of composite sandwich structures, in Dynamic Failure of Materials and Structures, Springer, London, 2010, pp. 191-233. [5] K.Diamanti, C. Soutis, J.B. Hodgkinson, Non-destructive inspection of sandwich and repaired composite laminated structures, Composites Sci. Tech. 65 (2005) 2059–2067. [6] J. Wang, A.M. Waas, H. Wang, Experimental study on the low-velocity impact behavior of foam-core sandwich panels, 53rd AIAA/ASME/ASCE/ASC Structures, Structural Dynamics and Materials Conference, Honolulu, Hawaii (2012), to be found on http://arc.aiaa.org/DOI:10.2514/6.2012-1701. [7] ASTM D 7136 – Standard test method for measuring the damage resistance of a fiber reinforced polymer matrix composite to a drop impact event, in Annual Book of ASTM Standards 2005, pp. 430-445. [8] ASTM Standard C 293-11 (2011) “Flexural properties of flat sandwich constructions” American Society for Testing and Materials, West Conshohocken, Pennsylvania (first issued in 1957). [9] N. Constantin et all., Links between low velocity impact results and the residual mechanical performance of sandwich composite materials, Advanced Manufacturing Technologies 834 (2016) 167-172. [10] N. Constantin et all., Some aspects concerning the mechanical characterization versus manufacturing technology of sandwich composite materials, Advanced Manufacturing Technologies 834 (2016) 161-166..
185