Methodology for predicting degradation shape and depth of a CFRP during short-circuit current injection

DOI : You do not have permission to access embedded form.
contenu protégé  Document accessible sous conditions - vous devez vous connecter ou vous enregistrer pour accéder à ou acquérir ce document.
- Accès libre pour les ayants-droit


Methodology for predicting degradation shape and depth of a CFRP during short-circuit current injection


2.99 Mo


Creative Commons Aucune (Tous droits réservés)





<resource  xmlns:xsi=""
        <identifier identifierType="DOI">10.23723/10638/20097</identifier><creators><creator><creatorName>Jean Rivenc</creatorName></creator><creator><creatorName>G. Peres</creatorName></creator><creator><creatorName>Richard Perraud</creatorName></creator><creator><creatorName>Pascal Peyre</creatorName></creator><creator><creatorName>David Andissac</creatorName></creator><creator><creatorName>Christopher Albrieux</creatorName></creator><creator><creatorName>Francois Pons</creatorName></creator><creator><creatorName>Pierre-Henri Cadaux</creatorName></creator></creators><titles>
            <title>Methodology for predicting degradation shape and depth of a CFRP during short-circuit current injection</title></titles>
        <resourceType resourceTypeGeneral="Text">Text</resourceType><dates>
	    <date dateType="Created">Sun 1 Oct 2017</date>
	    <date dateType="Updated">Sun 1 Oct 2017</date>
            <date dateType="Submitted">Sat 17 Feb 2018</date>
	    <alternateIdentifier alternateIdentifierType="bitstream">ad3ca2ca3f60c6f5b311e81e98c742faea0ea981</alternateIdentifier>
            <description descriptionType="Abstract"></description>

Methodology for predicting degradation shape and depth of a CFRP during short-circuit current injection Jean Rivenc * (1), G Peres (1), Richard Perraud (2), Pascal Peyre (3), David Andissac (3), Christopher Albrieux (4), Francois Pons (4), Pierre-Henri Cadaux (4) * Corresponding author 1 : Airbus Group Innovations, Department TX5EE, building D42, 18 rue Marius Terce, 31025 Toulouse Cedex 2 : Airbus Group Innovations, Department TX5EE, 12 rue Pasteur, 92150, Suresnes, France 3 : Airbus SAS, Department EYNI1, building M01, 316 route de Bayonne, 31300 Toulouse, France 4 : Airbus SAS, Department ESMC3, 18 rue Marius Terce, building D41, 31025, Toulouse Cedex Email: (with – for composed names, e.g. pierre-henri) Abstract - A methodology is presented in order to predict the degradation shape and depth of a CFRP (Carbon Fiber Reinforced Plastic) coupon when an ab-normal short-circuit current is injected. The study context and stakes are explained. The conditions under which degradation initiates, propagates and aborts, are exposed, depending on short-circuit (voltage, current) and aircraft breakers characteristics. The general methodology is exposed: the relationship between dissipated power, temperature rise and degradation are given. It is necessary to know how material characteristics are modified during degradation. Hence, measurements of material dynamic resistance and thermal characteristics during short-circuit from one part, and binocular observation after degradation from another part, are presented on latest generation of CFRPs. Parameters like resin sublimation temperature and state change enthalpy, are discussed. Analysis of correlations between electrical and physical material behaviours, is made. At last, figures regarding numerical solver choice are discussed. Introduction Important technical evolutions have been taking place in aeronautics for the last decade. For instance, A350 fuselage is entirely made of CFRP (Carbon Fiber Reinforced Plastic). From another part, more and more functions are being electrified: on-board power has been strongly increasing, from 400 kVA on A330 aircraft to 800 kVA on A380 1 . Wider use of high voltage is expected in future power architectures (“More Electrical Aircraft”), with a HVDC (High Voltage Direct Current) network of voltages +270/- 270V. These new constraints require developing skills and tools in order to anticipate the consequence of electrical faults in early phases of aircraft development programme. One of these faults is the case of a harness which chafes the CFRP fuselage: when harness insulation and CFRP surface are damaged, a short-circuit occurs and current is injected into CFRP structure. It is essential to predict the shape and the depth of CFRP degradation, as a function of injected current, voltage and duration, in order to anticipate the expected consequences on the aircraft installation, their criticality, and whether specific countermeasures shall be anticipated. This is the aim of the presented work. In this paper, firstly the physical phenomenon of material degradation, and the way to calculate it, is presented. A description of CFRP structure with plies and interplies, is proposed. The relationship between injected power and temperature rise is given, and a criterion for material degradation is proposed. Some parameters have to be determined, like the material dynamic resistance during degradation: they are measured and presented. A physical analysis, based on binocular observations, is made on tested materials, and a correlation with measured electrical behaviour is proposed. A numerical model which describes the degradation initiation, propagation and abortion, is presented and discussed. Finally, conclusions are given, and next steps of study are discussed. Physics of degradation mechanism and modelling For a better understanding, a simple case is considered: a 2D material model, made of 3 plies composed with carbon fibers (grey cells), separated by 2 inter-plies (white cells), is depicted in Fig 1. Each interply is described by three rows of cells. Current is injected from an electrode on top side (red), and leaves the sample through grounded right electrode (black). Plies conductivity (grey cells) mainly depends on carbon fibers conductivity. Previous measurements showed that its order of magnitude was 40 000 S/m in the direction of fibers 2,3 . Interplies conductivity (white cells) is less trivial to understand; indeed, for the studied material, resin was insulating, however measured transverse conductivity (bulk) was around 1 S/m 2,3 . This indicates that there was quite a high density of connecting points between plies, distributed inside interplies, ensuring path for electrical current in transverse direction 4 . The conductivity ratio between interplies and carbon fibers was on the order of 1/40000, although it may vary a lot from one material kind to another. The methodology which follows is valid for any kind of material, no matter of number of plies and of fibers orientation. For modelling purposes, the material is divided in elementary cells, in which currents, impedances, electrical power and temperature rise, are computed. A) Degradation initiation Material at initial state is depicted in Fig 1, when power Pk is computed on each resin cell (white). When current injection starts, temperature rise of a k- th cell is given by:         − × × + = − τ t k thk a k e P R T t T 1 ) ( , (1) where Ta is the ambient temperature, Rthk is the thermal resistance of the k-th cell, Pk is dissipated power (i.e. 2 k k k i r P × = ), and τ is the thermal time constant. The final temperature of the k-th cell at steady-state is given by: k thk a k P R T T × + = ∞ . (2) This expression is valid if the following conditions are fulfilled: • Effect of heat diffusion within the material is negligible, i.e. speed of temperature rise is homogeneous within the material; • Convection is negligible; • Heat radiation is negligible. This assumption is reasonable when considering phenomena which occur inside the material. The two first assumptions have been verified by experiments (not presented here). From power calculation and thermal resistance measurements, it is easy to compute with Equ (2) the expected final temperature ∞ k T of every cell, and to identify the hottest j-th cell in interplies, where degradation will initiate. In the following, index j will refer to this cell, while index k will refer to other cells (white). Let’s assume, as an example, that initation starts in the white cell below injecting electrode, represented in Fig 1. Consider now the resin temperature sublimation Ts. If injected current and dissipated power are sufficient to reach a temperature s j T T > ∞ , then resin sublimation starts to occur from this cell. According to Equ (1), this occurs at time tj such that: s t k thj a T e P R T j =         − × × + − τ 1 . (3) P1 P2 P3 P4 Pj P5 P6 P7 P8 P9 … … … … … … … … … … Fig. 1 – Numerical description of a 2D CFRP material made of 3 plies (grey) and 2 interplies (three white rows). B) Degradation propagation After resin sublimation starts, the material physical and electrical properties locally change. It is assumed that resin exits the material and is “replaced” by entangled carbon fibers which touch each other. From an electrical point of view, it is assumed that material local conductivity changes from interplies conductivity to fibers conductivity. These assumptions will be justified in the next Sections. New electro-thermal initial conditions are: 1. The conductivity of j-th cell (i.e. the first cell where Ts is reached), turns from interplies conductivity (white) to fibers conductivity (grey). Conductivity of all other cells, remains unchanged; 2. Temperature of each k-th cell is given by Equ (1), for the time t = tj. In particular, the temperature of the j-th cell is Tj = Ts. Since conductivity has locally changed, currents and power distributions also change, and have to be computed again. Pk changed for all cells, therefore temperature rise also changes and has to be computed again according to Equ (1). The next cell where degradation will occur, is then identified. It is expected to be located in the vicinity of the cell in the previous step. In that way, an iterative numerical process is put in place. It keeps going as long as current levels and dissipated power are such that there is an cell for which final temperature ∞ k T is above Ts. C) Degradation abortion When tripping time of aircraft breaker device is reached, electrical current is shut down. Due to material inertia, degradation may keep continuing for a while, but quickly material cools down and temperature becomes lower than Ts everywhere in the material. At this point, degradation mechanism aborts. In some cases, for impedant short-circuits (typically In400s), this area enlarged and sample resistance decreased even more. Fig. 4 – Sample aspect after test (top electrode); applied electrical stress is shown in Fig. 3. Fig. 5 – Binocular observations of sample thickness (4mm) according to cross-section along dash-dot line in Fig 4. Thermal measurements Four thermocouples were introduced on a sample and a controlled electrical power of 8.5W was injected in transverse direction (current of 5.6A). Measurements gave: thermal resistance Rth = 7.5° C/W, and time constant τ = 90 s. Given values are at sample center and, for both parameters, relative deviation was less than 10% depending on the 4 thermocouples positions inside the sample. These parameters were used in the calculation of Equ (1). At last, temperature Ts has been estimated by detecting apparition of first smokes during material degradation; estimation gave Ts = 180° C. Conclusions and future work The methodology for predicting a CFRP degradation shape and depth when a short-circuit current is injected, has been presented in details. Electrical, thermal and physical characterizations of the material were conducted in order to obtain input parameters for accurate description of material behavior. Experimental observations, correlated to binocular analysis, showed that, during current injection, resin sublimation occurred; in the area where degradation occurred, the initial structure of carbon plies, separated by resin interplies, was replaced by a more conductive area made of entangled carbon fibers, touching each other; this area enlarged as degradation propagation occurred, and material resistance strongly decreased. Thermal measurements showed that, for injection durations lower than material time constant (90s), temperature rise and thermal resistance deviations were less than 10%, and effect of heat propagation was not of first order: hence when considering current injections of < 10 s (linked to breakers tripping times), heat propagation effect may be neglected. The next step is to implement a complete electro- thermal simulation. Two numerical strategies may be used: one is to use an electromagnetic code (e.g. numerical tool ASERIS-BE based on Boundary Cell Method in frequency domain) and to integrate a sub- routine for thermal calculations. The other one is to use a multiphysics code (e.g. Comsol) and to integrate a sub-routine to integrate resistance change during degradation. The latter seems to be promising and is currently under construction. The latest parameter, which has not been taken into account so far, is the resin massic sublimation enthalpy. It will be measured in the next future by Differential Scanning Calorimetry. Knowing the mass of a k-th cell and the dissipated power in this cell, it will be possible to take into account, in a dynamic way, the requested duration for resin sublimation. Once the model will be implemented, a test under real conditions (airplane voltages and typical short-circuit currents) will be performed on CFRP; the degradation obtained during experiments will be compared to the degradation predicted by the tool, in order to confirm results validity, and to check self-consistency of the global approach. References 1 Classe Affaire n° 31, 4 th quarter 2005 and internal Airbus data. 2 A. Piche, D. Andissac, I. Revel, B. Lepetit, Dynamic electrical behavior for a composite material during a short-circuit, EMC Europe, Sept 26-30 th , 2011. 3 A. Piche, I. Revel, G. Peres, F. Pons, B Lazorthes, B. Gauthier, P-H Cadaux, Numerical modeling support for the understanding of current distribution in carbon fibers composites, EMC Europe, June 11- 12 th , 2009. 4 JP Park, TK Hwang, HG Kim and YD Doh, Experimental and numerical study of the electrical anisotropy in unidirectional CFRP composites, Smart Materials and structures, 2007, Vol. 16, n° 1, pp57-6 6.