Arc tracking energy balance: application to copper and aluminium aeronautic wires

03/02/2015
OAI : oai:www.see.asso.fr:10638:20105
DOI :
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
 

Résumé

Arc tracking energy balance: application to copper and aluminium aeronautic wires

Collection

application/pdf Arc tracking energy balance: application to copper and aluminium aeronautic wires Th. André, Ph. Teulet, F. Valensi, A. Gleizes, Pascal Peyre, David Andissac
Détails de l'article
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

Arc tracking energy balance: application to copper and aluminium aeronautic wires

Métriques

11
7
308.05 Ko
 application/pdf
bitcache://40016dda998ace3be5fc3dcb3e3f2f34f556b911

Licence

Creative Commons Aucune (Tous droits réservés)

Sponsors

Organisateurs

logoaaaf-05_162x120.jpg
logo_see.gif

Sponsors

airbus.jpg
logo_safran_lps_grand.png
logo-aero-def-fd-blanc.jpg
logoirtsaintexupery.jpg
logo_onera.jpg
logotoulousemetropole.jpg
craquitaine.jpg
crmip.jpg
<resource  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
                xmlns="http://datacite.org/schema/kernel-4"
                xsi:schemaLocation="http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4/metadata.xsd">
        <identifier identifierType="DOI">10.23723/10638/20105</identifier><creators><creator><creatorName>Pascal Peyre</creatorName></creator><creator><creatorName>David Andissac</creatorName></creator><creator><creatorName>Th. André</creatorName></creator><creator><creatorName>Ph. Teulet</creatorName></creator><creator><creatorName>F. Valensi</creatorName></creator><creator><creatorName>A. Gleizes</creatorName></creator></creators><titles>
            <title>Arc tracking energy balance: application to copper and aluminium aeronautic wires</title></titles>
        <publisher>SEE</publisher>
        <publicationYear>2017</publicationYear>
        <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">Wed 19 Sep 2018</date>
	</dates>
        <alternateIdentifiers>
	    <alternateIdentifier alternateIdentifierType="bitstream">40016dda998ace3be5fc3dcb3e3f2f34f556b911</alternateIdentifier>
	</alternateIdentifiers>
        <formats>
	    <format>application/pdf</format>
	</formats>
	<version>34125</version>
        <descriptions>
            <description descriptionType="Abstract"></description>
        </descriptions>
    </resource>
.

Arc tracking energy balance: application to copper and aluminium aeronautic wires Th. André (1)(a), Ph. Teulet (1)(b), F. Valensi (1)(c), A. Gleizes (2)(d), P. Peyre(3)(e), D. Andissac(3)(f) 1: Université de Toulouse, UPS, LAPLACE, 118 route de Narbonne, F-31062 Toulouse cedex 9, France 2: CNRS, LAPLACE, F-31062 Toulouse, France 3: Airbus Operations S.A.S., Site de Saint Martin du Touch, 316 route de Bayonne, F-31060 Toulouse Cedex 9, France a: andre@laplace.univ-tlse.fr, b: teulet@laplace.univ-tlse.fr, c: valensi@laplace.univ-tlse.fr, d: gleizes@laplace.univ-tlse.fr, e: pascal.peyre@airbus.com, f: david.andissac@airbus.com Abstract Arc tracking tests have been carried out between two voluntarily damaged aeronautic wires. Copper or aluminium cables have been submitted to short circuits under direct current. Electrical data (arc voltage and current) enabled to establish the total average power. The average power transferred to the electrodes was determined by two ways, an electrical one (by means of the electrode voltage fall), and a thermodynamic one (with the cable ablated mass). Heat flux sensors have made possible to estimate the radiated power, taking the absorption of VUV-radiation into account. It is observed that the vaporization is more significant in the case of copper than for aluminium, since more fumes are released, while aluminium rather provides metal droplets. Introduction Fault arcs occurring in aeronautic electrical circuits may have a catastrophic effect and should be eliminated or at least controlled to limit their consequences. In particular, short circuits occurring between two adjacent cables may lead to a particular phenomenon called “arc tracking” [1] corresponding to the propagation of the arc along the cables due to the cable metallic core ablation and to the surrounding dielectric material degradation. New airplane concept introduces two strong constraints: lowering of weight and increase of the electrical power. The consequences of these constraints are the partial or total replacement of copper with aluminium in the cables, the increase of alternating current (AC), the strong increase of direct current (DC) parts and DC buses in the electrical circuit, and the replacement of metal by composite materials in the structure. These changes require the problem of electric arcs and arc tracking to be carefully considered again. The present study aims to quantify the energy transfers from the tracking arc to materials in DC circuit in conditions of atmospheric pressure. The main objective is to evaluate the various terms of energy losses based on several experimental measurements and on some theoretical analysis. We present thus an experimental part based on electrical measurements, on fast imaging and on radiation measurements. Material An experimental setup was developed to create a fault arc between two cables for an adjustable period of time. The generator delivers continuous power with current regulation adjustable up to 100 A with a voltage of 110 V. A commutation box, involving two Insulated Gate Bipolar Transistors (IGBT) driven by a computer, was operated via an interface system. This device allowed us to shift from a first loop including only the generator and a ballast resistance to the working circuit including the cables. The procedure consists first in waiting for current stabilization in the first loop, and then in triggering the shift to the working circuit. The control unit allows duration to be modified with steps of 10 ms, up to 9.9 s. The tested cables were mainly made whether of copper (then they were called DR) or of aluminium (then they were called AD). Both had the same insulating material, consisting in a polyimide layer wrapped with a sheath of PTFE. Two diameters were tested for each type, named after the American Wire Gauge (AWG) system. Each cable is presented in the following table, giving its nominal current and section. Type Main material Nominal current Section DR-18 Copper 12.5 A 1.02 mm DR-20 10 A 0.81 mm AD-16 Aluminium 12.5 A 1.29 mm AD-18 10 A 1.02 mm Table 1 – Characteristics of the various tested cable types. For each test, the two cables were from the same type. The tests have been performed with various values of direct current, from around 45 A to around 100 A, depending on the type of cable. The open loop voltage was fixed at 110 V. The cables were weighed before and after each test, in order to determine the mass loss. Three kinds of experimental diagnostics could be done simultaneously: electrical measurements, high speed imaging and radiative heat flux measurements. The starting signal from the computer was used as a synchronization trigger for all measurement devices, allowing a comparison of the various results. Total average power Electrical data were acquired using current and voltage probes connected to a computer, with a time resolution of 50 μs. The electrical measurements give information not only on voltage and intensity, but also on the occurrence of discharge. In Fig. 1, a favourable case (i.e. arc duration longer than 80 ms) is presented. We may note that the behaviour of the arc remains erratic. Besides, due to the arc variable impedance, the actual current value may differ from the input value. The most typical behaviour is a first step with current flowing through the wires without any arc, but with material temperature increase. This step can extend up to a few seconds, and then the arc ignites. Concerning the voltage, the value of 110 V corresponds to the case of operating in open loop: once the arc is established, the voltage drops to values depending mainly on the distance between the electrodes and their nature. Fig. 1: Electrical data from a test on a DR-18 cable with a 86A current input. For the same gauge, the total average power is higher with aluminium than for copper (see Fig. 2), since the nominal current is smaller. For the same metal and current, we observe that the total average power is higher when the gauge is smaller. Fig. 2: Total average power obtained at various current values with AD18 and DR18 cables. Average power transferred to the electrodes The average power transferred to the electrodes was estimated by two different ways: by an electrical approach (by the means of the electrode voltage fall), and with a thermodynamic approach (with the ablated mass) [2,3]. The electrical approach consists in multiplying the electrode voltage fall by the average value of the current. The electrode voltage fall is obtained assuming that the total voltage is a linear function of the length of the arc column, such as , where is the total voltage, is the electrode voltage fall (considered not to be independent of the current), is the average electric field within the plasma column, and is the length of the arc column. High speed imaging was performed, with a recording speed of 3000 frame/s, allowing to measure (along with ), and extrapolate the value of . The obtained values (18 V for copper and 19 V for aluminium) correspond to other data found in literature, considering the incertitude. For the thermodynamic approach, we consider that the whole ablated metal mass is melted, but only a small part of it is vaporized. For those calculations involving the ablated mass, the proportion of metal and insulating material, whether for copper or aluminium wires, were taken into account. The exact vaporized proportion is unknown, and we consider a range of 1%-10% of the ablated mass. We observe that a considerable amount of fumes is emitted in the case of copper, resulting from recondensed vaporized metal, while aluminium rather projects numerous metal droplets, directly produced by the fusion of metal. Thus, the proportion of vaporized matter must be smaller for aluminium than for copper. Radiated power Four heat flux sensors were placed around the cable samples (at the level of the unsheathed core), at a distance of 4.5 cm from them, in a plane perpendicular to the cable axis. These sensors integrated all radiation from Vacuum Ultra Violet (VUV) (0.1 μm) up to long wavelength infrared (12 μm). As their response time was 80 ms, the measurements performed were considered only if the arc duration was longer than 100 ms. Assuming that the arc is punctual and that the radiation is isotropic, the measured radiated energy is deduced from the expression ∫ , where is the distance between the arc and the sensors (4.5 cm), and is the measured radiated flux. This energy is then transformed into a power , which then corresponds to the radiated power over the surface of a sphere (centred on the arc, and with a radius equal to ). The assumption of punctual source is rather good considering the size of the plasma (around 1-3 mm). As the experiments have been performed at atmospheric pressure, all the radiation with a wavelength lower than 200 nm (called Vacuum Ultra Violet or VUV radiation) is absorbed in the ambient air within a few microns, because of photodissociation and photoionization of oxygen molecules mainly. Due to this phenomenon, the radiation measured by our sensors (located a few cm from the arc) does not include the VUV part. It is possible to determine the total radiated power by the means of the Net Emission Coefficient (NEC). The NEC is computed as the divergence of the total radiative flux at the centre of an isothermal sphere of radius Rp (power per volume unit and per solid angle unit). Radiation power is integrated over a large range of photon energy corresponding to the wavelength range from 30 nm to 5 μm. In a previous work, our team calculated the NEC and partial NEC (i.e. for λ>200 nm, which does not include the VUV part) of air-Cu and air-Al plasmas. An example is presented in Fig. 3, presenting the variation of the total NEC and the partial NEC (without VUV) for a large temperature range, in a plasma containing 99% air and 1% Cu. Fig. 3: Comparison between the NEC and the partial NEC (without VUV) for a 99% air - 1%Cu mixture (mass proportions) and for 2 values of Rp (plasma radius). The value Rp=0 corresponds to a fictitious case without absorption, while the value Rp=2 mm corresponds to our experimental conditions. The composition of the plasma (proportion metal-air) is unknown and we have considered two cases: 1% or 10% of metal. In our case, the temperature of the plasma is around 13kK. Thus for the 90% air - 10%Cu mixture, there are 50% VUV radiation and 50% non- VUV radiation, so the power has to be multiplied by 2 in order to take the whole radiation into account. For the same proportion with aluminium (90% air – 10% Al), the ratio is approximatively 80% VUV and 20% non-VUV, so the power has to be multiplied by 5. As the measurements provided by the sensors are quite comparable between copper and aluminium, we obtain that the radiated power is significantly higher for aluminium that for copper. This seems to be in contradiction with the observation that copper releases a lot of fumes, while aluminium produces metal droplets. Conclusions Since they are based on the electrical measurements, which are accurate and instantaneous, the total average power and the average power transferred to the electrodes seem to be the most reliable calculated results. A spectroscopy study should be interesting, in order to determine the concentration of each species in the plasma. This would allow us to have a better understanding of the respective behaviour of copper and aluminium in arc tracking, notably concerning the vaporized part, and its relationship with the radiated power. We envisage carrying out this study in a near future. References 1 F. Dricot et al, Survey of arc tracking on aerospace cables and wires, IEEE Transactions on Dielectrics and Electrical Insulation, 1994, Vol.1 n°5, pp.896-903 2 H. El Bayda et al, Energy losses from an arc tracking in aeronautic cables in DC circuits, IEEE Transactions on Dielectrics and Electrical Insulation, 2013, Vol.20 n°1, pp.19-27 3 H. El Bayda, Étude du transfert d’énergie entre un arc de court-circuit et son environnement : application à l’ « arc tracking », PhD thesis, Université de Toulouse, 2012