METHOD FOR NON-DESTRUCTIVE TESTING OF AN ELASTOMERIC CABLE INSULATION SHELL, DEVICE AND PROGRAM

Description

[0001] The invention relates to a method for inspecting the insulating sheaths of cables, particularly nuclear cables.

[0002] One area of ​​application of the invention relates to electrical cables in nuclear power plants.

[0003] In nuclear power plants, an average of 1500 km of cables (including 50 km in the reactor building) are installed per nuclear unit. During operation, these installed cables undergo aging, meaning they degrade.

[0004] Understanding and monitoring this degradation is essential to ensure the proper functioning of these cables.

[0005] The aging state of a nuclear cable is determined by the elongation at break of its insulation. This characteristic is measured in the laboratory using a uniaxial tensile test, a destructive test requiring the removal of at least 50 cm of the entire cable. Therefore, cable sampling campaigns are periodically scheduled on-site, and these cables are then analyzed in the laboratory to determine their end-of-life.

[0006] The main drawback of this type of inspection, which requires cable sampling, is its destructive nature. In other words, the removed cable is not put back into service, and its replacement incurs a significant cost.

[0007] The one-page summary document, published on November 28, 2017, of the thesis Pubellier et al.: "Influence of micrometric loads on the aging of polymer matrix composites", describes a method for controlling an insulating sheath made of a material of at least one elastomeric polymer of a cable, comprising a step of measurement by nuclear magnetic resonance on the insulating sheath to measure at least one first parameter characterizing the elastically active chains of the macromolecular network of the elastomeric polymer.

[0008] The document by Brice Gabrielle et al.: “Thermal Aging of Interfacial Polymer Chains in Ethylene-Propylene-Diene Terpolymer / Aluminum Hydroxide Composites: Solid-State NMR Study”, JOURNAL OF PHYSICAL CHEMISTRY PART B: CONDENSED MATTER, MATERIALS, SURFACES, INTERFACES & BIOPHYSICAL, September 26, 2011, vol. 115, no. 43, ISSN 1520-6106, pages 12392-12400, describes a non-destructive testing method for an EPDM cable sheath, with a decomposition of the transverse relaxation signal, measured by proton NMR, into two exponential components, namely: an exponential component AS with rapid decay, attributed to portions of chain constrained at their two ends by topological constraints, an exponential component AL with slower decay, attributed to portions of chains having only one free end.

[0009] The document BEN HASSINE ET AL: "Time to failure prediction in rubber components subjected to thermal ageing: A combined approach based upon the intrinsic defect concept and the fracture mechanics", MECHANICS OF MATERIALS, vol. 79, August 1, 2014, ISSN 0167-6636, pages 15 - 24, describes a measurement of the molar mass of elastically active chains for EPDM that can be used as electrical insulation.

[0010] The document BRICE GABRIELLE ET AL: "Probing Rubber Cross-Linking Generation of Industrial Polymer Networks at Nanometer Scale", JOURNAL OF PHYSICAL CHEMISTRY, part B, vol. 120, no. 24, June 2, 2016, ISSN 1520-6106, pages 5581 - 5589, describes a method for measuring double-quanta proton NMR.

[0011] The document GILLEN KT ET AL: "Condition monitoring methods applied to degradation of chlorosulfonated polyethylene cable jacketing materials", POLYMER DEGRADATION AND STABILITY, vol. 91, no. 6, 2006, ISSN 0141-3910, pages 1273 - 1288, describes a method for monitoring the CSPE sheath of an electrical cable.

[0012] The invention aims to obtain a non-destructive testing method (not requiring cable replacement) allowing conclusions to be drawn about the durability of cables.

[0013] To that end, a first object of the invention is a control method according to claim 1.

[0014] Claims 2 to 7 relate to embodiments of this process.

[0015] A second object of the invention is a computer program according to claim 8, comprising code instructions for implementing the process evaluation step as described above, when implemented on a computer.

[0016] A third object of the invention is a control device according to claim 9.

[0017] Claims 10 to 12 relate to embodiments of this device.

[0018] The invention will be better understood upon reading the following description, given solely by way of non-limiting example with reference to the accompanying drawings, in which: THE figures 1 and 2 schematically represent a molecular network of an elastomer in a cable sheath, the figure 3 schematically represents a transverse relaxation signal for an elastomeric network of a cable sheath, which can be used according to an embodiment of the invention, the figure 4 schematically represents a modular synoptic diagram of the non-destructive testing process according to an embodiment of the invention.

[0019] Described below, with reference to the figure 4 The non-destructive testing method for an insulating sheath G of a CB cable, according to embodiments of the invention. The insulating sheath G of the CB cable is made of an electrically insulating material, comprising at least one elastomeric polymer. This material may be, for example, EPDM or other materials. The CB cable may be, for example, an electrical cable or other type of cable installed in a nuclear power plant. As illustrated in the figure 4In such a nuclear power plant, the CB cable and its insulating sheath G may be subjected to high stresses, including ionizing radiation of up to 0.1 Gy.h-1 (h denoting one hour) and temperatures of up to 50°C, which may lead to accelerated aging of the CB cable and its insulating sheath G. The invention is also applicable to any cable having an insulating sheath.

[0020] According to one embodiment of the invention, the process includes a first step E1 of taking a sample of the insulating sheath G of cable CB. This can be done by taking one or more micro-samples of the sheath material G (a few milligrams) which do not impair the functionality of the cables C and are therefore non-destructive. This sample is taken in small quantities. For example, the mass of the insulating sheath G sample taken from cable CB during the sampling step E1 is 20 to 40 mg. This small mass represents a sample of approximately (2 x 5 x 5) mm³. As an example, the maximum length of insulation that can be taken from a cable on-site, at the cable end, without impacting its functionality, is estimated at 1 cm. There are no constraints regarding the shape of the sample to be taken.The sample preferably always has the same shape, ideally symmetrical (like a disc, for example), so that the magnetic field of the magnet is always modified in the same way when inserted into the spectrometer of the measuring device during measurement step E2 described below. This step E1 is quick and relatively simple to implement. This ensures non-destructive monitoring of the CB cable's functionality, allowing it to remain in place to continue operating.

[0021] In one alternative, in addition to or instead of the first step E1 of taking a sample of the cable's insulating sheath, a second step E1bis may be provided, involving contacting a proton magnetic resonance (PMR) measuring device with the sheath G of the CB cable, for example, at the location where the CB cable is installed on-site, if the measuring device is portable and brought to that location. Here too, the E1bis step of contacting the PMR measuring device with the sheath G ensures non-destructive monitoring of the CB cable's functionality, allowing it to remain on-site and continue operating.

[0022] After the first step E1 and / or E1bis, a second step E2 involves solid-state proton magnetic resonance (NMR) measurement of the insulating sheath G. This measurement aims to determine at least one parameter P1 characterizing the elastically active chains A of the macromolecular network of the elastomeric polymer in the insulating sheath G of the CB cable. This second measurement step E2 can be performed using the proton magnetic resonance instrument that was brought to the cable's location in the case of step E1, and / or on the sample taken from the CB cable's sheath G by a proton magnetic resonance instrument located in a laboratory.

[0023] The first parameter P1 includes the average molar mass M c of the elastically active chains A of the macromolecular network of the elastomer polymer, which is calculated from the nuclear magnetic resonance measurement of protons.

[0024] The topology of the macromolecular network (chemistry) is correlated with the mechanical properties of the sheath polymer material G. Indeed, the relevant macromolecular material parameters reflect the state of the CB cable. These parameters include the molar mass distribution Mc between crosslinking nodes and, optionally, the FCEA fraction of elastically active chains A in the macromolecular network of the elastomer polymer, which is calculated from proton nuclear magnetic resonance measurements. Preferably, both of these analyses focus specifically on the elastically active chains A, to which a threshold S is applied. They are directly related to the mechanical strength of the insulating sheath G, impacting the lifespan of the cables C. The lifespan and functionality of the CB cable are linked to the integrity of the insulating sheath G. This integrity is described by the mechanical behavior of the material.The study of the macromolecular network of the polymer insulation in the G sheath requires only micro-sampling if step E1 described above is implemented, thus enabling non-destructive testing of the cable, or no sampling is required if step Elbis described above is implemented. It should be noted that the term "elastically active chains A" is specific to the elastomer family. For other polymer families, the term "macromolecular chains" is used.

[0025] Solid-state proton nuclear magnetic resonance (1H NMR) is a physicochemical analysis technique for polymers. It is a non-destructive technique that allows for the characterization of the macromolecular network. It can be used as a primary tool for monitoring material changes during aging.

[0026] After the second measurement step E2, a third step E3 is carried out to evaluate the aging state of the CB cable from the first parameter P1 and / or from a second parameter P2, which is determined from the first parameter characterizing the elastically active chains A. This third evaluation step E3 includes a comparison of the first parameter P1 and / or the second parameter to at least a prescribed evaluation threshold S and a determination that the CB cable is at the end of its life when the first parameter P1 and / or the second parameter P2 is below the prescribed evaluation threshold S. Use of the FCEA fraction of elastically active chains A:

[0027] A first family of embodiments of the process, using the first parameter P1, is described below.

[0028] In one embodiment, the first parameter P1 comprises the FCEA fraction of elastically active chains A of the macromolecular network of the elastomeric polymer. The evaluation step E3 involves comparing E31 the FCEA fraction of elastically active chains A of the macromolecular network of the elastomeric polymer to the prescribed evaluation threshold S, which is a threshold A of the elastically active chain fraction. This threshold A is identified according to the type of material. The CB cable is determined to be at the end of its service life when the FCEA fraction of elastically active chains A of the macromolecular network of the elastomeric polymer falls below the prescribed evaluation threshold S.

[0029] According to one embodiment, the measurement step E2 includes a measurement E21 of a transverse relaxation of protons by proton nuclear magnetic resonance on the insulating sheath G. The FCEA fraction of elastically active chains A of the macromolecular network of the elastomer polymer is determined at least from this measurement E21.

[0030] According to one embodiment, the E21 measurement of transverse proton relaxation yields a transverse relaxation signal 1 < H, which can be obtained using three complementary experiments: solid-state echo (at short times) and Hahn / CARR-Purcell-Meiboom-Gill (CPMG) echo (intermediate and long times). This E21 measurement allows the proportion of network components to be determined by measuring the transverse relaxation signal 1 < H. These experiments provide important information regarding the mobility of the network protons. Thus, three populations of macromolecular chains within the network can be identified due to very different dynamics within the polymer network, as shown by the figures 1 and 2 .

[0031] To figures 1 and 2Elastically active chains (A) are cross-linked or entangled at both ends. Hanging chains (B) are cross-linked or entangled at only one end, or chain ends. Extractables (C) are non-cross-linked / free chains within the macromolecular network. Neither hanging chains (B) nor extractables (C) contribute to elasticity; they are elastically inactive. Mechanical properties are therefore governed primarily by the elastically active chains (A). Analysis of the transverse relaxation signal (1 < H) allows for the determination of the fractions of the different chain segments (A, B, and C) within the network.

[0032] According to one embodiment, the mole fraction B of pendant chains and the mole fraction C of extractable material in the network are first determined via the transverse relaxation measurement 1< H. The fractions B and C are obtained from the transverse relaxation signal M(t) 1< H, measured by measurement E21, using the following analytical adjustment: M t = B exp − t / T 2 b + C exp − t / T 2 c where M(t) is the magnitude of the magnetization, T2b is the transverse relaxation time 1< H relative to the protons of the pendant chains B, which is obtained from M(t), T2c is the transverse relaxation time 1< H relative to the protons of the extractables C, which is obtained from M(t).

[0033] There figure 3schematically represents the transverse relaxation signal M(t) / M0 on the ordinate (logarithmic scale) as a function of time t on the abscissa, for an elastomeric network composed of the 3 relaxation components A, B, and C, where M0 is a defined constant. As shown in the figure 3 , these extractables C (represented in dashed lines) correspond to the points of M(t), located after extinction of the component relating to the protons of the elastically active chains A (which is the first relaxation component).

[0034] The simple subtraction FCEA = 1 - B - C gives the FCEA fraction of elastically active chains A in the network.

[0035] Determining the FCEA fraction of elastically active chains A by NMR allows us to know the position relative to the threshold A THRESHOLD: if FCEA > A THRESHOLD, the mechanical strength of the material is considered satisfactory, if FCEA < A THRESHOLD, the material is considered degraded and no longer able to ensure good strength.

[0036] Furthermore, if we have the evolution of the kinetics of the FCEA fraction of elastically active chains A (with accelerated aging tests for example), we can then determine the residual life of the material.

[0037] Thus, a measurement of the FCEA fraction of elastically active chains A below the threshold A THRESHOLD indicates that the CB cable is at the end of its service life, and therefore the CB cable is in too advanced a state of aging and must be replaced with a new CB cable on site.

[0038] Thus, the method according to the invention makes it possible to obtain reliable information on the wear of the CB cable, based on which a maintenance operation on the cable can be decided. Using the average molar mass M c of elastically active chains A:

[0039] A second family of embodiments of the process, using the second parameter P2, is described below.

[0040] According to one embodiment, the second parameter P2 includes an elongation ε r at break, having been determined from the average molar mass M c of the elastically active chains A of the first parameter P1.

[0041] In one embodiment, the evaluation step E3 involves comparing the elongation εr at break to the prescribed evaluation threshold S, which is an elongation at break threshold S2. The CB cable is determined to be at the end of its service life when the elongation εr at break falls below the prescribed evaluation threshold S2.

[0042] The measurement step E2 includes a two-quanta proton coherence rise measurement E22 by solid-state proton nuclear magnetic resonance on the insulating sheath G. The average molar mass M c of the elastically active chains A of the macromolecular network of the elastomer polymer is determined at least from this measurement E22.

[0043] According to one embodiment, the E22 measurement of coherence rise to two quanta 1< H is obtained based on the Baum & Pines sequence, improved by Saalwächter.

[0044] The analysis allows us to obtain the experimental curve I DQ showing the evolution of the amplitude of the two-quanta coherences 1< H as a function of the excitation time t DQ.

[0045] The analytical description of the I DQ curves of coherence rise 1< H gives the value of the residual coupling distribution D res of the network, according to the following equation: I DQ t DQ = ∫ 0 D Stat P D res f t DQ D res dD res where D stat is a defined constant distribution, P(D res ) is a defined density distribution.

[0046] D res in Hz corresponds to the residual dipole coupling, measured using NMR experiments. It is the average dipole coupling of a chain segment between topological constraints in the lattice. The further apart the topological constraints are, the lower D res is because it is averaged by more molecular motions.

[0047] Δstat (in Hz) corresponds to the static dipole coupling, which is measured for protons in the absence of molecular motion (obtaining it would require a measurement at very low temperature for a polymer). Δstat is the value of the static dipole coupling 1< H- 1< H. Δstat can be calculated from the distances 1< H- 1< H at a repeating unit, disregarding any molecular motion.

[0048] k is dimensionless; it is a parameter that allows for an additional averaging effect arising from movements within a Kuhn segment (movements on the order of a few hundred picoseconds, librations, and conformational jumps). k is specific to each polymer.

[0049] P(Dres) is the distribution function of the residual dipolar coupling for the material under study. For some materials, P(Dres) can be a Gaussian probability density function. For EPDM, in order to describe heterogeneities in P(Dres), P(Dres) is calculated from an asymmetric probability density function known as the "log-normal" distribution, which is as follows: where µ and σ are two adjustable parameters.

[0050] From the P(Dres) distribution parameters determined for crosslinked, unfilled EPDM (µ = 0.752; σ = 0.506), it is possible to plot the distribution of residual dipolar couplings within the network, before aging.

[0051] f(t DQ, D res) is a core function that describes the rise of the curve I DQ(t DQ), obtained by the double quanta measurement DQ 1 < H, if only one residual dipole coupling were present in the network. In other words, this equation allows us to describe the rise of coherences for a perfect unimodal network, without the aid of any distribution.

[0052] For the elastomeric polymer of the material, the function f(t DQ ,D res ) can be of the form: f t DQ D res = 0.5 1 − e − Qt DQ 1.5 cos UD res t DQ where Q and U are predetermined first and second coefficients.

[0053] For example, for the aforementioned EPDM material, Q=0.378 and U =0.583 and the function f(t DQ ,D res ) is: f t DQ D res = 0.5 1 − e − 0.378 D Res t DQ 1.5 cos 0.583 D res t DQ

[0054] Of course, the function f(t DQ ,D res ) may be different for another elastomeric polymer or for another polymer.

[0055] The distribution of D res is transformed into a distribution of the average molar mass M c between macromolecular chains (topological constraints that encompass entanglements and crosslinking nodes), according to the following equation: k D res D stat = 3 5 r 2 N where N is the number of Khun segments between crosslinking nodes, which is proportional to the average molar mass Mc of the polymer, and r and k are material parameters that can be determined by molecular dynamics. r2 corresponds to the square of the ratio between the end-to-end vector (or the end-to-end chain length) for a portion of chains under topological stress (R2) and the average value of this vector (or the end-to-end chain length) at equilibrium (R2 average). r2 often follows a Gaussian distribution. Dres depends on the inverse of the average molar mass Mc and the inverse of the average molar mass Me of the entangled chains. D res ∝ 1 / M c + 1 / M e

[0056] This Dres distribution can be assimilated to the distribution of molar mass between topological constraints (M).

[0057] We assume that 1 / M e is much less than 1 / M c (we neglect entanglements).

[0058] Mc is calculated using the equation above, which links Dres to the molar mass Mc via the factor G = (Dstat / k) * 3 / 5 * r2. We can thus calculate: D res = D stat / k .3 / 5 . r 2 / M c

[0059] Thus we have: M c = D stat / k .3 / 5 . r 2 / D res Thus, Mc = G / Dres

[0060] The predominant parameter governing mechanical properties and their degradation being elastically active chains, the evolution of their molar mass (M c ) is a good indicator and is directly correlated with the evolution of the elongation at break ε r .

[0061] Knowing the average molar mass M c of the elastically active chains A, a multi-scale relationship linking M c to the elongation ε r at break allows us to predict the value of the elongation ε r at break of the analyzed material and therefore to conclude on its state of degradation with respect to the critical threshold value S2.

[0062] Thus, measuring the average molar mass (Mc: average molar mass of the chains between crosslinking nodes and Me: average molar mass of the entangled chains) of the material will give an indication of the elongation εr value at break. For example, an elongation εr value at break less than 50% of the S2 elongation at break threshold in absolute value indicates that the cable has exceeded its service life limit.

[0063] If we have the evolution of the material's kinetics (Mc) (using accelerated aging tests, for example), we can then determine the evolution of the elongation kinetics εr at break using the multiscale relationship. This approach allows us to determine the remaining lifetime of the material.

[0064] An example of the relationship between the average molar mass Mc of elastically active chains A and the elongation ε r at break is described below.

[0065] An example of the relationship between the macromolecular scale (M c ) and the macroscopic scale (ε r ) uses a methodology based on fracture mechanics and the concept of intrinsic defect to predict the fracture of the elastomer (for a material in which post-crosslinking is the predominant process during aging).

[0066] The input data is: The value of the tear energy Jc0 of the unaged material, obtained on a cracked specimen: SENT specimen (single notch). The elongation at break of a specimen in tension, as well as the stress-strain curve on unaged material. (□ c0, W0 = f(□)), the subscript 0 referring to the unaged material.

[0067] From these input data, we can calculate the size of the intrinsic defect a th<: a th = J c 0 4 . k λ c 0 . W c 0 with k λ = π λ where a th< is the theoretical size of the intrinsic defect, J c0 is the tear energy of the unaged material, k (λ) is a proportionality factor in the case of a DENT specimen.

[0068] In the case of EPDM, we found an intrinsic defect length (2a th< ) equal to 0.5 mm (diameter of a fictitious circular defect in a uniaxial tensile specimen) and which will be accepted for any other EPDM.

[0069] M c = f (t×a T ) : The evolution of the molar mass M c as a function of reduced time t×a T is: M c = M c 1 + a 1 e a 2 ta T obtained by time-temperature equivalence. t×a T denotes t multiplied by a T. a T is a slip factor of the time-temperature equivalence curve. a 1 , a 2 , are constants obtained by least squares regressions, M c1 parameter obtained by least squares regression, t is the time in hours.

[0070] The output data are J c and W c.

[0071] J is the critical tearing energy of rupture.

[0072] W c is the strain energy density at break.

[0073] Jc = f(Mc): The value of Jc for a given value of the molar mass Mc is: I c = A M c 0 M c − M c 0

[0074] For EPDM: A = 125 kJ / m 2< and M c0 = 678 g / mol (represents the threshold value below which the material has zero tearing energy).

[0075] W c = f(M c ) : the value of the corresponding strain energy density W c.

[0076] As a first approximation, we assume a Gaussian model which allows us to write that: W = ρRT 2 M c 1 − 2 f λ 1 2 + 2 2 + λ 3 2 − 3 where R and T are the gas constant and absolute temperature respectively, ρ the density, f a functionality factor depending on the number of segments leaving a crosslinking node, W is the strain energy density, λ1, λ2, λ3 are the eigenvalues ​​of the strain gradient tensor.

[0077] For a given value of Mc, we can deduce that of J ccorresponding (equation 3), as well as the corresponding strain energy density W c calculated by a numerical method from equation 4 with λ 2 =λ 3 =λ 1 -1 / 2< in the case of uniaxial tension.

[0078] Knowing the size of the intrinsic defect a th< , we can obtain the elongation λ c at break by solving the following equation: J c = 4 k c a th W c

[0079] With : k c = π λ c

[0080] The elongation εr at break is expressed by: ε r = λ c − 1

[0081] The molar mass between topological stresses M obtained here involves the contribution of both entanglements (Me) and crosslinking nodes (Mc). That is to say, M (obtained by NMR via the previous equation) can be decomposed: M = M c + M e where M c is the molar mass between crosslinking nodes and M e is the molar mass between entanglements and Me is usually given in textbooks. Mc = M − Me

[0082] The end-of-life criterion is set at a breakage elongation threshold of 50% in absolute terms.

[0083] The approach described can be enhanced by two distinct applications: Use of the tool for evaluating the state of the material at time t knowing its M c (measured by NMR), Use as a predictive tool: knowing the evolution of the molar mass M c of the material, we can obtain the evolution of the elongation ε r at break and thus predict the end of life of the cable.

[0084] In these first and second families of embodiments, the third evaluation step E3 can implement the comparison E31 and / or the comparison E32.

[0085] The process described above can be implemented on a computer (which may be a computer, processor(s), microprocessor(s), or other) for evaluation step E3. This computer may have been programmed by a computer program containing code instructions for implementing the process when it is implemented on that computer.

[0086] Of course, the above embodiments, features, possibilities and examples can be combined with each other or selected independently of each other, but the invention is defined only by the attached claims.

Claims

1. A method for non-destructive testing of an insulating sheath (G) made of a material of at least one elastomeric polymer of a cable (CB), including a step (E2) of measurement using proton nuclear magnetic resonance on the insulating sheath (G) in order to measure at least a first parameter (P1) characterizing the elastically active chains (A) of the macromolecular network of the elastomeric polymer, the measurement step (E2) including a measurement (E22) of the increase in proton two-quantum coherences using proton nuclear magnetic resonance on the insulating sheath (G) in order to determine at least from this measurement (E22) the average molar mass (Mc) of the elastically active chains (A) of the macromolecular network of the elastomeric polymer, the first parameter (P1) comprising the average molar mass (Mc) of the elastically active chains (A) of the macromolecular network of the elastomeric polymer, wherein the average molar mass Mc of the elastically active chains (A) is calculated from the equation M c = D stat / k .3 / 5 . r 2 / D res where Dstat is a predetermined static dipolar coupling, Dres is the residual dipolar coupling having been measured by the measurement (E22) of the increase in proton two-quantum coherences using proton nuclear magnetic resonance, k is a predetermined parameter of local coupling topology and of motions between segments of the material, r is the predetermined ratio of the end-to-end chain length for a portion of chains between topological constraints and the average value of this end-to-end chain length at equilibrium for the material, and an assessment step (E3) including a comparison of the first parameter (P1) characterizing the elastically active chains (A) and / or of a second parameter (P2), having been determined from the first parameter (P1) characterizing the elastically active chains (A), with at least one prescribed assessment threshold (S) in order to determine that the cable (CB) is at the end of its life when the first parameter (P1) and / or the second parameter (P2) is below the prescribed assessment threshold, wherein the second parameter (P2) includes an elongation (εr) at break having been determined from the average molar mass (Mc) of the elastically active chains (A).

2. The method according to claim 1, characterized in that the first parameter (P1) further comprises the fraction (FCEA) of elastically active chains (A) of the macromolecular network of the elastomeric polymer, the assessment step (E3) including the comparison (E31) of the fraction (FCEA) of elastically active chains (A) of the macromolecular network of the elastomeric polymer with the prescribed assessment threshold (S), in order to determine that the cable (CB) is at the end of its life when the fraction (FCEA) of elastically active chains (A) of the macromolecular network of the elastomeric polymer is below the prescribed assessment threshold (S).

3. The method according to claim 2, characterized in that the measurement step (E2) includes a measurement (E21) of a transverse proton relaxation using proton nuclear magnetic resonance on the insulating sheath (G) in order to determine at least from this measurement (E21) the fraction (FCEA) of elastically active chains (A) of the macromolecular network of the elastomeric polymer.

4. The method according to any one of the preceding claims, characterized in that the assessment step (E3) includes the comparison (E32) of the elongation (εr) at break with the prescribed assessment threshold (S) in order to determine that the cable (CB) is at the end of its life when the elongation (εr) at break is below the prescribed assessment threshold (S).

5. The method according to any one of the preceding claims, characterized in that it includes a step (E1) of taking a sample of the insulating cable sheath, the step (E2) of measurement using proton nuclear magnetic resonance being carried out on the sample.

6. The method according to claim 5, characterized in that the taken mass of the sample of the insulating sheath (G) of the cable (CB) during the sampling step (E1) is from 20 to 40 mg.

7. The method according to any one of the preceding claims, including a step (Elbis) of bringing a device for measurement using proton nuclear magnetic resonance into contact with the insulating sheath (G) of the cable (CB), the measurement step (E2) being carried out by the device for measurement using proton nuclear magnetic resonance.

8. A computer program, including code instructions for implementing the assessment step (E3) of the method according to any one of the preceding claims, when it is implemented on a calculator.

9. A device for non-destructive testing of an insulating sheath (G) made of a material of at least one elastomeric polymer of a cable (CB), wherein the device includes a device (E1, E1bis) for measurement using proton nuclear magnetic resonance on the insulating sheath (G) in order to measure at least a first parameter (P1) characterizing the elastically active chains (A) of the macromolecular network of the elastomeric polymer, the first parameter (P1) comprising the average molar mass (Mc) of the elastically active chains (A) of the macromolecular network of the elastomeric polymer, the device being able to carry out a measurement (E22) of the increase in proton two-quantum coherences using proton nuclear magnetic resonance on the insulating sheath (G) in order to determine at least from this measurement (E22) the average molar mass (Mc) of the elastically active chains (A) of the macromolecular network of the elastomeric polymer, wherein the average molar mass Mc of the elastically active chains (A) is calculated from the equation M c = D stat / k .3 / 5 . r 2 / D res where Dstat is a predetermined static dipolar coupling, Dres is the residual dipolar coupling having been measured by the measurement (E22) of the increase in proton two-quantum coherences using proton nuclear magnetic resonance, k is a predetermined parameter of local coupling topology and of motions between segments of the material, r is the predetermined ratio of the end-to-end chain length for a portion of chains between topological constraints and the average value of this end-to-end chain length at equilibrium for the material, and a calculator enabling to carry out an assessment step (E3) including a comparison of the first parameter (P1) characterizing the elastically active chains (A) and / or of a second parameter (P2), having been determined from the first parameter (P1) characterizing the elastically active chains (A), with at least one prescribed assessment threshold (S) in order to determine that the cable (CB) is at the end of its life when the first parameter (P1) and / or the second parameter (P2) is below the prescribed assessment threshold, wherein the second parameter (P2) includes an elongation (εr) at break having been determined from the average molar mass (Mc) of the elastically active chains (A).

10. The device according to claim 9, in which the first parameter (P1) further comprises the fraction (FCEA) of elastically active chains (A) of the macromolecular network of the elastomeric polymer, the calculator enabling to carry out the assessment step (E3) including the comparison (E31) of the fraction (FCEA) of elastically active chains (A) of the macromolecular network of the elastomeric polymer with the prescribed assessment threshold (S), in order to determine that the cable (CB) is at the end of its life when the fraction (FCEA) of elastically active chains (A) of the macromolecular network of the elastomeric polymer is below the prescribed assessment threshold (S).

11. The device according to claim 9 or 10, in which the device (E1, Elbis) for measurement is able to carry out a measurement (E21) of a transverse proton relaxation using proton nuclear magnetic resonance on the insulating sheath (G) in order to determine at least from this measurement (E21) the fraction (FCEA) of elastically active chains (A) of the macromolecular network of the elastomeric polymer.

12. The device according to any one of the preceding claims, in which the calculator enables to carry out the assessment step (E3) including the comparison (E32) of the elongation (εr) at break with the prescribed assessment threshold (S) in order to determine that the cable (CB) is at the end of its life when the elongation (εr) at break is below the prescribed assessment threshold (S).