METHOD FOR CHARACTERIZING AND MONITORING THE HOMOGENEITY OF METAL PARTS MANUFACTURED BY LASER INTERNING

Description

TECHNICAL FIELD

[0001] The invention is in the field of manufacturing metal parts by sintering and more specifically relates to a non-destructive method of characterizing and controlling the homogeneity of metal parts during their manufacture.

[0002] The invention also relates to a device for implementing the process according to the invention. PREVIOUS STATE OF THE ART

[0003] In many industrial sectors such as mechanical engineering, electronics, aerospace, metallurgy, and non-destructive testing, precise knowledge of a material's homogeneity and thermal properties is invaluable. This information is obtained through quality control, typically performed after parts have been manufactured.

[0004] We know of several techniques for carrying out this control, such as radiology, ultrasound control, and modulated photothermal radiometry.

[0005] Examining the structure or internal state of an object by radiography involves passing very short wavelength electromagnetic radiation (X-rays or γ-rays) through it and collecting the intensity modulations of the beam in the form of an image on a suitable receiver or, in most cases, a film.

[0006] Following the same principle, images can be obtained using particles other than photons, and thus, techniques such as neutronography can be implemented.

[0007] Radiography is a non-destructive testing method that produces an image of the material density of an object exposed to X-ray or gamma radiation. The principle of the method is based on the differential absorption of the medium according to the atomic number of its constituent atoms and its density. Any lack of material will induce lower absorption and therefore, locally, a higher optical density on the film or a higher gray level in the case of digital images. In industrial radiography, X-rays are produced, most often by an X-ray tube, or by a particle accelerator for high-energy applications. The gamma radiation sources used industrially are iridium-192, cobalt-60, and selenium-75. Radiography is a technique that allows visualization of material deficiencies in the volume of the object being inspected in a two-dimensional image.

[0008] Neutron imaging is a nuclear measurement technique (non-destructive testing) whose principle is similar to that of X-ray radiography, but which uses neutrons as the radiation source. A transmission image is produced by placing the object being inspected between a neutron source (often from a reactor) and a neutron detection system. The attenuation of the neutron flux is more or less pronounced depending on the nature of the material encountered, thus producing differences in contrast that allow for analysis of the object's contents.

[0009] The principle of ultrasonic testing involves emitting and propagating an ultrasonic wave within the part being inspected. The wave is then collected and analyzed after its interaction with the material. Based on this general principle, numerous specific techniques exist, depending on whether the inspection is performed by transmission or reflection, whether the transmitting and receiving devices are combined, and finally, the type and angle of the ultrasonic waves used, etc. The most common inspection method, known as reflection, is comparable to medical ultrasound. The transmitter and receiver (whether combined or not) are positioned on the same side of the part. The receiver collects the echoes generated by reflection or diffraction from obstacles encountered by the wave, such as defects, interfaces between materials, or the surface of the part.The transmitting and receiving devices, known as "ultrasonic transducers", are generally based on the piezoelectric effect. The main element, the transducer, consists of a piezoelectric pellet that converts an electrical signal into a mechanical vibration and vice versa.

[0010] We also know of methods for determining the thermophysical parameters of a body from the analysis of the thermal waves emitted by the body. These methods differ in that: The type of heating (temporal, spatial), the location of the heat source and the detector on the diagnostic object, the location of the heat source and the measurement point relative to each other (center of the beam or adjacent to it), and the number of measurement points are all factors to consider. Numerous theoretical methods already exist for different types of characterization.

[0011] By combining temperature measurement with heating modeling, it is possible to determine the thermal properties of a surface under certain conditions. This technique can be used more specifically to non-destructively and remotely measure the properties of a layer or coating on a known substrate. In the field of non-destructive active thermal monitoring, four different methods can be distinguished, which are listed and described below: Pulsed method; Continuous heating method; Periodic pulsed method; Synchronous detection modulated photothermal radiometry method, also called " "Lock-in."

[0012] In pulsed heating, the material is subjected to a single pulse from the heat source (e.g., a pulsed laser) with known energy parameters. Using the material's cooling curve, the desired thermophysical properties are then obtained by inversely solving the heat transfer equation. This technique allows a solution to be derived from the data of a relatively short measurement.

[0013] Numerous pulsed methods have been developed to characterize homogeneous materials and those coated with one or more layers. The most widely used method involves measuring thermal phenomena on the front face of the sample, i.e., the heated surface. Most studies focus on measurements at the center of the laser beam, but temperature variations can also be monitored in a lateral direction. One application is the characterization of defects. The thermal properties to be determined are diffusivity, effusivity, thermal conductivity, or combinations of these properties.

[0014] The second category of active pyrometry methods is based on continuous surface heating. As with the first method, the main parameter is the surface temperature and its deviation from the reference materials, known as thermal contrast. The advantage of this method, as with the first, is its speed. However, unlike the weak cooling temperature signal of pulsed heating, continuous heating allows for measurements at higher, and therefore more precise, temperatures. Nevertheless, heating carries the risk of overheating the material and, consequently, altering its properties.

[0015] As previously mentioned, step heating allows, by comparing the estimated model with the experiment, to determine the characteristics of the layer.

[0016] Repetitive pulsed heating overcomes the drawbacks of the two previous methods. The temperature is maintained at a level sufficient for measurements, while reducing the risk of overheating the surface.

[0017] Tolev, Jordan & Mandelis, Andreas. (2010). "Laser photothermal non-destructive inspection method for hairline crack detection in unsintered automotive parts: A statistical approach." NDT & E International. 43. 283-296. discloses the photothermal inspection of green parts. Another example of photothermal characterization of a metal part is disclosed by FR 3 007 523 A1.

[0018] One problem with the methods described above is that they cannot be implemented during the manufacturing process because the material's emissivity is generally unknown, leading to inaccurate results or even incorrect temperature measurements. Furthermore, determining thermal properties from temperature measurements requires knowing the heat flux absorbed by the surface and, therefore, the laser power and absorption coefficient.

[0019] The aim of the invention is to achieve real-time defect detection on a part manufactured by sintering without prior knowledge of the heating flux absorbed by the surface and, therefore, the laser power and the absorption coefficient. DESCRIPTION OF THE INVENTION

[0020] This goal is achieved by means of a non-destructive process for characterizing and controlling the homogeneity of metallic parts comprising several distinct zones manufactured by sintering, in which laser radiation is applied successively to each zone and an analysis of each sintered zone is carried out simultaneously in real time by synchronous detection laser active radiometry.

[0021] Preferably, laser radiation is applied successively to successive sub-zones of each zone so as to use low-power laser radiation, the size of said sub-zones having a regular shape on which the laser radiation is applied line by line so as to form in real time an image of the sintered zone.

[0022] According to another feature of the invention, each zone is heated with frequency modulation, generating a single thermal wave at each frequency, and for each frequency, the phase shift between the light signal from the laser applied to the part and the thermal signal emitted by the part is measured in real time.

[0023] According to the invention, the thickness L (µm) and the thermal diffusivity D (m² / s) of each zone are determined by the following formulas: L = r 0 ζ φ ln 90 φ min D = 1 ζ f f min L r 0 Or r 0 is the laser beam radius at 1 / e in intensity; φ min And f min represent respectively the minimum phase shift and the laser beam rate; ζ φ And ζ f represent known coefficients that depend respectively on the minimum phase shift φ min and the report r 0 / L.

[0024] The process according to the invention is implemented by a device comprising a laser source adapted to successively apply laser radiation to each area of ​​the controlled part, an infrared radiation detector adapted to capture and measure in real time the thermal radiation emitted by each area of ​​said part, a synchronous detector intended to detect the phase shift between the light signal of the laser radiation and the thermal signal of the infrared radiation emitted by each area of ​​said part.

[0025] The laser source of the device according to the invention is configured to apply modulated heating to each zone, generating a single thermal wave at each frequency, and the synchronous detector is configured to measure, for each frequency, the phase shift between the light signal from the laser applied to the part and the thermal signal emitted by the part. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Other features and advantages of the invention will become apparent from the following description, taken by way of non-limiting example, with reference to the accompanying figures in which: there figure 1 schematically illustrates a sample of a metal part obtained by the additive manufacturing method using the process according to the invention; the figure 2 illustrates the dimensions of the different areas of the sample of the figure 1 ; there figure 3 schematically illustrates a device for characterizing and controlling the homogeneity of metal parts according to the invention; figure 4 schematically illustrates one embodiment of the optical system used in the device for characterizing and controlling the homogeneity of metallic parts according to the invention; figure 5 is a curve illustrating the variations of the measured phase shifts as a function of frequency for the sample of the figure 1for two different thicknesses; the figure 6 schematically illustrates a first setup for measuring the phase shift as a function of frequency for the metallic part of the figure 4 in the direction of the x-axis; the figure 7 is a curve illustrating the phase shifts measured by the setup of the figure 6 ; there figure 8 schematically illustrates a second setup for measuring the phase shift as a function of frequency for the metallic part of the figure 4 in the direction of the z-axis; the figure 9 is a curve illustrating the phase shifts measured by the setup of the figure 8 . DETAILED DESCRIPTION OF SPECIFIC METHODS OF IMPLEMENTATION

[0027] The invention will be described in an application for the real-time manufacturing and characterization of metal parts used in the nuclear industry.

[0028] As an illustrative example, the method will be applied to a metallic sample 2 with dimensions of 20 mm × 20 mm (laterals) and thicknesses of 180 µm to 2430 µm, obtained by additive manufacturing using 316L stainless steel powder, as illustrated by the figures 1 and 2 .

[0029] Sample 2 comprises four distinct zones A, B, C, and D with thicknesses of 180 ± 20 µm (zone A), 400 ± 20 µm (zone B), 1170 ± 20 µm (zone C), and 2430 ± 20 µm (zone D), respectively. The surface roughness is ± 20 µm. The thermal diffusivity of this 316L stainless steel sample is D = 0.039 ± 0.004 cm² / s [4] at ambient temperature.

[0030] Zones A and B were examined by the method according to the invention during manufacturing using device 3. This device comprises a fiber laser source 4 with a wavelength of 1080 nm for heating sample 2 with a tunable average power of 0 to 50 W. This average power can be modulated. The beam r The wavelength of the collimated beam at sample 2 is equal to 1740 µm (radius at 1 / e intensity). The laser power (amplitude and rate) is managed by a function generator 6. The output of generator 6 (output impedance: 50 Ω) is split into two channels, one for laser operation (input impedance: 50 Ω) and the other for use as the reference signal for a synchronous detection amplifier (input impedance: 1 MΩ).

[0031] Function generator 6 can produce a continuous signal or a sinusoidal, rectangular, or triangular modulated signal of variable amplitude and frequency. Laser source 4 delivers up to 50 W of continuous output power and up to 25 W (average power) in sinusoidal mode, with a beam quality close to the diffraction limit (M2 < 1.1). The laser characteristics are shown below. Characteristics of the fiber laser source 4.

[0032] Operating method Continuous, modulated Central wavelength nm 1080 Beam shape Gaussian, collimated Beam quality (M2<) 1,1 Beam diameter at 1 / e µm 1740 ± 30 Maximum power (peak or cw) W 50 Modulation rate Hz ≤ 5 kHz Maximum generator voltage V 5 Dimensions mm 448 x 451 x 132

[0033] Laser source 4 features a low-power infrared beam to indicate beam direction and facilitate optical alignment of samples prior to experimental testing. The average power of the sinusoidal laser signal is only slightly dependent on the applied frequency. The variation in laser signal power remains small, so the average heating temperature can be considered constant for frequency ranges from 1 Hz to 1 kHz. This ensures consistent thermal properties throughout the measurements at different frequency ranges, resulting in improved measurement accuracy.

[0034] Device 3 further includes an IR detector 10 for measuring the thermal radiation of the sample 2 when it is illuminated by the laser source 4. The useful spectral range of the IR detector 10 extends from 1.5 µm to 11 µm. A convex ZnSe lens 12 is used to image the heated area retained on a detector surface with a radius of 800 µm. A germanium filter 14, transmitting only wavelengths between 2 µm and 14 µm, has been placed in front of the IR detector 10 to prevent diffuse reflection of the laser signal from interfering with detection (since the detector has residual sensitivity at a wavelength of 1 µm).

[0035] A synchronous detection amplifier 16 is used to determine the phase shift between the laser light signal and the thermal signal as a function of the laser signal frequency.

[0036] The synchronous detection amplifier 16 is connected to an electronic oscilloscope 17. The power of the laser signal is controlled by a laser power detector 18 also connected to the electronic oscilloscope 17.

[0037] Analysis of the phase shift curve allows us to determine certain properties of sample 2.

[0038] To avoid beam reflections into the laser, the sample surface is positioned at a small angle θ relative to the plane perpendicular to the incident normal axis. The beam size is then multiplied by cos θ in one direction. A small angle θ is chosen, i.e., less than 10°, to remain within the approximation cos θ ~ 1.

[0039] There figure 4schematically represents the optical system used in device 1. This optical system includes a ZnSe 20 lens (transmission spectral range from 0.6 µm to 15 µm) used to focus the thermal flux onto the sensitive area of ​​the IR detector 10 and a Germanium 22 filter (transmission spectral range from 1.8 µm to 23 µm) to cut all wavelengths up to 1.8 µm, thus filtering the wavelength of the laser.

[0040] The convex ZnSe 20 lens, with a focal length of 50 mm and a diameter of 25 mm, images the center of the area heated by the laser onto the IR detector 10. The IR detector's capture area has a radius of 800 µm. To image only the central part of the area heated by the laser beam, limited to half its radius (i.e., rheating / 2 = 870 µm), the magnification (rsensor / rcollected) is greater than or equal to 1, and the lens-sample distance is 2f = 100 mm.

[0041] As an example of implementing the method according to the invention, the phase shift curves as a function of the laser signal frequency will be determined for zones A and B of sample 2. The phase shift curves obtained make it possible to determine the optimal rate for better distinction (maximum Δφ) between zones A and B. All measurements were carried out on the homogeneous side where the plateaus of the sample are not visible.

[0042] Once the optimal rate is determined, the laser beam is moved along the x and z axes to study the homogeneity of the sample's properties along these axes. The axes were chosen so that it is possible to change the thickness along the x-axis without changing the thickness along the z-axis.

[0043] During operation, an area is heated with the Laser 4 source, varying the laser signal frequency from 1 to 200 Hz. The average power is 11 W (generator amplitude 3 V). Since frequency variation has very little effect on the laser output power, temperature variation must be minimal to avoid affecting measurements at different rates.

[0044] The measurements were carried out on two areas of different thicknesses, one at 180 µm and the other at 400 µm.

[0045] There figure 5 represents the phase shifts obtained. This curve allows us to determine the minimum phase shift φ min and the corresponding frequency f min and, by the following formulas (1) and (2), the thickness L (µm) and thermal diffusivity D (m² / s) in the tested areas: L = r 0 ζ φ ln 90 φ min D = 1 ζ f f min L r 0 Or r 0 is the radius of the laser beam at 1 / e in intensity; ζ φ And ζ fare known coefficients that depend respectively on φ min and the report r 0 / L.

[0046] The measurement results are presented in Table 1 below: Measurement results with sample 2.

[0047] φ min f min Thickness L = 180 µm -77,1 ± 0,1° 9 Hz Thickness L = 400 µm -61,7 ± 0,1° 3 Hz

[0048] Based on these results and the coefficients ζ φ and ζ f As presented below, it is possible to determine the thickness and diffusivity in zones A and B, which are shown in the table below. ζ φ = 1.535 ± 0.005 (180 µm) and 1.53 ± 0.005 (400 µm) ζ f = 0.534 ± 0.005 (180 µm) and 0.472 ± 0.005 (400 µm)

[0049] The thicknesses and thermal diffusivities determined by the modulated photothermal radiometry method for sample 2 are given in the following table 2: Measured by Lock-in Reference for T = 400 K Thickness L (zone A) 175,3 ± 26 µm 180 ± 20 µm Thickness L (zone B) 429,3 ± 30 µm 400 ± 20 µm D and zone A 0.051 ± 0.005 cm² / s 0.047 ± 0.003 cm² / s D and zone B 0.047 ± 0.005 cm² / s 0.047 ± 0.003 cm² / s

[0050] These thicknesses, determined by the modulated radiometry method, correspond well with the thicknesses measured by calipers. The thermal diffusivities measured in two different areas of sample 2 are the same.

[0051] The measurements carried out allow, firstly, for the analysis of sample 2 in the direction where there is a change in thickness, and secondly, in the direction where there is no visible change in either thickness or thermophysical properties.

[0052] For this purpose, the frequency of the laser beam is fixed, and is equal to that which gives the greatest phase shift when passing from one zone to another, i.e. about 20 Hz, which makes it possible to clearly highlight the different levels.

[0053] There figure 6illustrates the first case of analysis of sample 2 in the direction where there is a change in thickness.

[0054] In this case, sample 2 is heated in zone A, which has a thickness of 180 µm, and then moved along the x-axis (towards zone B) over a distance of 15 mm with a step size of 0.5 mm. The phase shifts measured for each position are shown by the curve of the figure 7 .

[0055] The two phase shift levels represent two thicknesses with two steps located at -80° and -45°. A final step is located at -40°. The chosen frequency is 20 Hz to clearly differentiate the two steps.

[0056] We note that the frequency is not suitable to show the difference between the last two levels of 400 and 1170 µm ( Figure 2 ).

[0057] The first part of the curve located for x between -4 and -2 mm shows a defect due to the manufacturing of the part.

[0058] It is observed that the method can easily reveal non-homogeneity (a plateau) in an inspected part. The spatial resolution is approximately one millimeter and is determined by the width of the area tested by the VIGO detector (1.6 mm). This resolution can be improved by using a higher-resolution optical path.

[0059] There figure 8 illustrates the second case of analysis of sample 2 in the direction where there is no change in thickness.

[0060] In this case, the sample is heated in zone A, which has a thickness of 180 µm, and then moved along the z-axis over a distance of 10 mm with a step size of 0.5 mm. The phase shifts measured for each position are shown by the curve of the figure 9 .

[0061] The sample is heated in zone A with a thickness of 180 µm, then moved along the z-axis over a distance of 10 mm with a step of 0.5 mm.

[0062] The phase shift is uniform. Indeed, the curve is flat and located at -80°. This explains why the part has no internal or external defects.

[0063] The measurements are taken in real time, that is, during the manufacturing of the part. This makes it possible to correct the sintering manufacturing process in real time.

Claims

1. A Non-destructive method for characterising and monitoring the homogeneity of a metal part (2) manufactured by sintering including several distinct zones, wherein, upon manufacturing said part, the method is characterised by the following steps: - a laser radiation is successively applied on each zone and an analysis of each sintered zone is simultaneously made in real time by synchronous detection active laser radiometry, - a frequency modulated heating generating at each frequency a single thermal wave is applied on each zone, and for each frequency, the phase shift between the luminous signal of the laser applied to the part and the thermal signal emitted by the part is measured in real time, and, - the thickness L (µm) and thermal diffusivity D (m2 / s) of each zone are determined by the following formulae: L = r 0 ζ φ ln 90 φ min D = 1 ζ f f min L r 0 where r0 is the laser beam radius at 1 / e intensity; φmin and fmin represent the minimum phase shift and the laser beam rate respectively; ζφ and ζf represent known coefficients which depend on the minimum phase shift φmin and the ratio r0 / L respectively.

2. The method according to claim 1, wherein the laser radiation is successively applied on sub-zones of each zone so as to use a low power laser radiation.

3. The method according to claim 2, wherein the size of said sub-zones has a regular shape on which the laser radiation is applied row by row so as to form an image of the sintered zone in real time.