WO2016087719A1

WO2016087719A1 - Method and system for characterizing mechanical and thermal properties

Info

Publication number: WO2016087719A1
Application number: PCT/FR2014/053175
Authority: WO
Inventors: Jean-Pierre Nikolovski; Guillaume TRANNOY
Original assignee: Commissariat A L'energie Atomique Et Aux Energies Alternatives
Priority date: 2014-12-04
Filing date: 2014-12-04
Publication date: 2016-06-09

Abstract

The invention relates to a system (8, 9, 87, 88, 89, 97, 98, 99) and an ultrasound method of measuring, inspecting and classifying the mechanical and/or thermal properties of a medium (60, 61, 62, 63, 64, 65, 67), by means of at least one wideband source (20, 21, 22) of acoustic waves coupled to at least one dispersive tapered solid guide (10, 17, 18) utilizing elastic waves of known impedance and known effusivity, and which is housed and damped (43, 44, 55) in an acoustically and thermally insulating structure (41, 42, 52) and coupled with the medium on a determined surface (11, 14, 15, 16) with a bearing force (F) or a controlled pressure (80, 81), said coupling being able to result from a first coupling (92) of the solid guide with a liquid guide projected continuously with a constant or variable hydrostatic pressure, said liquid possibly also being heated and projected intermittently, in the form of a collimated jet (93) striking the surface of the medium. Applications relate to non-destructive testing, medical diagnosis, classification of the quality of foods, steelmaking, as well as directional velocimetry.

Description

METHOD AND SYSTEM FOR CHARACTERIZATION OF MECHANICAL PROPERTIES

AND THERMICS

DESCRIPTION

The present invention relates to high speed scanning non-destructive (NDT) control using a projected liquid coupling element, employing ultrasound elastography of elastic and viscoelastic samples and the identification or classification of the state of materials via their acoustic or thermal properties. It is particularly applicable to directional velocimetry.

The invention provides a device and method for measuring longitudinal and transverse characteristic acoustic impedances in organic materials or fabrics by simple contact, immersion or projection of a known impedance guide. The materials may be solids or fluids and in particular viscoelastic liquids in solidification phase or high temperature phase change.

The problem to be solved is to locally access the mechanical properties of elastic or viscoelastic heterogeneous materials or fabrics in space and / or time, by a local contact method, or even quasi-point, possibly accepting high temperatures. The parameters that are to be determined preferably simultaneously with controlled coupling conditions are in particular the longitudinal and transverse impedances of the materials. These parameters are used in many fields, in particular in the field of tactile acoustic learning interfaces described in patents FR0703651 of May 23, 2007, FR0955065 of July 21, 2009, FR1059657 of November 23, 2010, FR1156892 of July 28, 2011, FR1258664 of 14 September 2012, FR1258664 of September 14, 2012, exploiting the damping of resonance modes or the diffraction of ultrasonic waves based on the acoustic properties of the finger pulp which, when it comes into contact with the tactile surface, dampens, removes and diffracts the underlying vector vibration typically from the position of the touch or gesture to be recognized. The part of each of the phenomena is not yet known, and it is one of the advantages of the device proposed here to contribute to solving this problem. However, there are disparities from one person to another, as well as the age and activity of individuals, particularly in terms of callosity, contact area and the nature of the dermis. Another benefit of this device is to evaluate the acoustic properties of the finger pulp for a population of individuals, so that the locating method can be made functional for the entire population. Moreover, the industrial production of tactile surfaces for learning, for example in the form of tablets or tactile pads, requires the availability of artificial fingers made of silicone and having the same mechanical properties as human fingers, particularly from the point of view of the longitudinal and transverse impedance to reproduce. Indeed, it has been found for example that a silicone polymer material could be very close to a human finger with respect to the longitudinal characteristic impedance, but far away for the transverse characteristic impedance (or shear) so that with the same profile and contact area, the acoustic signatures of an artificial finger and a human finger could be quite different. By extension, the longitudinal and transversal impedances of materials subjected to high temperatures and in crystalline liquid-solid phase change situations are also variable in time and space, in particular in depth, and their monitoring over time (monitoring ) can provide a means of characterizing the thermal manufacturing process.

This same remark is valid for the control of the aging or the degradation of the mechanical properties of the dermis or of the underlying viscoelastic tissues, for example in the breast, because of a pathology or a cosmetic treatment aimed at improving mechanical flexibility. skin with creams. On the other hand, in the food field, the ripening of a fruit or vegetable often results in a variation of the mechanical properties of the superficial layer and the deeper layers. In this case, the simultaneous knowledge of the longitudinal and transversal impedances as well as thermal, combined with an indexation or a classification of these properties in an ordered database or associated with a scale of maturation, or even identification of the nature when it is for example to recognize animal meat, is a method for diagnosing the state of conservation or maturation of these foods. In addition, there are few elastographic and thermal measuring instruments that are affordable for the industry. Most of those who are developed exploit high-number of elements of transducers, typically a hundred, associated with significant memory size electronic means, in analog-digital conversion and digital-analog fast, typically 50 megahertz and in need fast processing and transfer of data, making these tools unaffordable for small businesses: independent doctors, farmers, farmers or even conservators.

To come back to the tactile surfaces, such as those mentioned in the above patents, using plastic, metal or glass materials, the contact with a human finger involves an interaction with a mechanical vibration that includes a longitudinal component and a shear component. With respect to the coupling with the pulp, there is an out-of-plane vibration penetrating perpendicularly to the pulp and a shearing component (ie located in the plane of the interface) from shearing the pulp. It is recalled that the effusivity of a material (in Jm ^ .k ^ .s ^"0 ^'5) is its ability to exchange thermal energy with the environment. The thermal effusivity of a material is given by formula = _A / i, where λ is its thermal conductivity, p its density and c its specific heat capacity The thermal effusivity makes it possible to determine the temperature of an interface when two semi-infinite objects, having different temperatures, are put in contact.

A "good" artificial finger is a finger that simultaneously has good similarity with a human finger on both the longitudinal and transverse impedances as well as on the viscous damping generated at the interface and backscattered by the composition of its volume. back, as well as on the thermal properties including the effusiveness that the artificial finger must reproduce in a manner comparable to that of the human finger which gives up part of its heat in a situation of non-thermal neutrality. In this way, the acoustic signature of an artificial finger is the same as the acoustic signature of a human finger and it is then possible to precisely calibrate or study the reliability using a robotic arm, a tactile surface comprising several hundred reference coordinates or various thermodynamic conditions.

Thus, when one seeks to develop an artificial finger and / or a tactile surface, one is finally interested in the same problems as those of the medical pulse elastography implementing low frequency shear waves and a high longitudinal acoustic illumination. frequency, for which it is necessary to have information such as density and Young's modulus, shear modulus and Poisson's ratio, or propagation velocities, or characteristic impedances, in order to predict the velocities of the waves in the material and their sensitivity, for example in temperature. However, these parameters are rarely explicitly provided by the manufacturers of materials, particularly for plastics and soft materials such as elastomers that propagate weakly shear waves. This type of information is even less available in manufacturing processes involving the coexistence of different crystalline phases. It is certainly possible to determine them from other information such as flexural and shear modulus or density, but this information is often approximate. In addition, the acoustic impedance parameter combines two values, velocity and density, which, when available, are often flawed or vary from one manufacturer to another, which reduces the accuracy available on the acoustic impedances. In addition, many materials undergo alteration of their physical properties depending on their aging or external agents, such as alteration by a chemical or mechanical wear. It may be necessary to check periodically or continuously the evolution of these properties over time.

In practice, it is therefore desirable firstly to be able to precisely and locally measure the characteristic acoustic impedances associated with the longitudinal and transverse modes during a measurement campaign on a population of individuals in order to study the dispersion of the mechanical properties. living tissues between individuals or on tactile materials whether they are plastic or flexible in order to control their similarity to human skin, or in food control, cases of high-temperature manufacturing processes, in medical diagnostics or in Non Destructive Testing (NDT). Here is provided a simple device, a local guided wave probe, associated with a process capable of rapidly characterizing the acoustic impedance of the materials without damaging them, by simple contact on a plane face of the order of one millimeter to one square centimeter and that possibly a variation of viscoelastic properties by scanning.

By extension, this device finds applications in the field of medical imaging by ultrasonic ultrasound or pulse elastography exploiting the combined use of longitudinal and transverse waves by means of one or more ultrasonic waveguides, possibly heated or cooled , propagating these two waves and brought into contact with the medium of interest in order to determine its mechanical and thermal properties. In the case of shear wave impulse elastography, the use of shear waves is limited to low frequencies, typically between 50 Hz and 5 kHz due to the strong damping of shear waves in diffusing viscoelastic media which propagate in the living tissues at a speed close to 3 m / s. A presentation of this method is for example given in the article by S. Catheline et al., "Diffraction Field of a Low Frequency Vibrator in Soft Tissues Using Transient Elastography," IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control 46, no. 4 (July 1999): 1013-1019.

Developments of this method are also published by L. Sandrin et al, "Shear Modulus Imaging with 2-D Transient Elastography," IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control 49, no. 4 (April 2002): 426-435, or by M. Tanter et al., "Ultrafast Compound Imaging for 2-D Motion Vector Estimation: Application to Transient Elastography," IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control, no. . 10 (October 2002): 1363-1374. Pulse elastography with low frequency shear waves has been the subject of numerous patent applications. One can quote the publication number 2 791 136 corresponding to the request FR9903157 of March 15, 1999 or the number 2 844 178 corresponding to the application FR0211074 of September 6, 2002. This type of elastography aims to detect cancerous nodules in the tissues humans, especially in the breast and liver and thus seeks to imaged these tissues to a depth of the order of 5 to 10 centimeters. The strong damping of shear waves in human tissues does not make it possible to operate at ultrasonic frequencies. Nevertheless, if one is restricted to short distances of the order of a millimeter, it is possible to exploit the shear waves at higher frequencies than those suggested in the state of the art, which are between 50 Hz and 5 kHz and reach ultrasonic frequencies. The objective is not only to exploit longitudinal waves to illuminate at an ultrafast rate of the order of 10 kHz the tissue being deformed under the effect of the passage of a pulse wave shear of high amplitude, tens of microns, generated by a vibratory pot, but also to exploit the shear wave as an additional wave, of different speed and polarization, generated by tapered waveguide transducers and exploited in traditional sonography for get a fine measurement. In addition, the present invention provides an additional means of characterizing the material by a thermal transfer process generating a thermal diffusion wave in the material. Finally, the present invention aims to propose a probe making it possible to generate the two types of waves at a distance and over a very limited lateral extent, by means of a projected collimated liquid coupler.

In the low-frequency shear wave impulse elastography, we seek to know the shear wave velocity in the tissue and in particular the variations of this velocity and the wavefront in the presence of a more rigid cancer nodule, which has the effect, on the one hand, of reducing the amplitude of the deformation of the fabric and, on the other hand, of accelerating the speed of propagation of the shear wave. The medical diagnosis then consists in solving the inverse problem to access the shear modulus, in particular its increase and its spatial extent in the tissue. The resolution of the process depends on the image acquisition frequency, ie the distance traveled by the bending wave between two longitudinal measurements, which is in practice of the order of a millimeter. Note that the isostatic compression modulus in human tissues is of the order of 1 GPa while the shear modulus is a thousand times smaller, in connection with the very low shear wave speeds observed. The device according to the invention also makes it possible to analyze or even image over a range of about one square centimeter by displacement and measurement of the position of the probe, the shear impedance in the vicinity of the dermis, or in general of the test surface, by means of a conical guide with dispersive bending waves, of quasi-point contact surface and therefore of lateral dimension comparable to or less than the working wavelength and therefore having a low radiation impedance typically 4 to 10 times smaller than the transverse impedance of the same volume material and thus more suitable for measuring small values of shear impedance or radiation. By way of example, if one wishes to characterize materials having a low transversal acoustic impedance, for example of the order of 50 kRayl, it will be necessary to reach a contact surface of less than 0.5 mm for 100 kHz, or even less than 0, 1mm for 600kHz.

It is thus possible to image by differential technique small variations of impedance on a surface of the order of one square centimeter. In addition, the tapered profile of the tip reduces the apparent thermal effusivity of the waveguide, generally metallic, which then approaches that of the skin, making it possible to minimize the heat exchange between the probe and the material to be characterized. The main benefit of a tip having an effusivity comparable to that of the test material is that the equilibrium temperature of the heated finger brought into contact with the test material is halfway to the differential temperature of the two objects before contact. For example, for a tip heated to 50 ° C and an ambient temperature of 25 ° C, the temperature of the tip will fall sharply, by at least 12.5 ° C, which will cause a change in transit time of the echo in the measurable tip with greater finesse. In the end, it is the measure of the effusivity of the material and its relative variations with respect to that of the finger that will be perceived with greater finesse. On the other hand, if the finger imposed its effusivity, if it was much higher than that of the materials which it touched, one would not see a big difference of temperature between the various materials. They would all be perceived as "hot" and can not be differentiated. Conversely, if it were the materials that had a much higher effusivity than the finger, the equilibrium temperature would remain in the vicinity of the ambient and it would be difficult to distinguish the different materials that would all be perceived as "Cold". Envariante, the same device will be able access by simple contact to the flow velocity of a fluid in a vein by ultrasonic differential transit time or by continuous or pulsed Doppler measurement.

From a physical point of view, the acoustic impedance Z is the ratio between the generated pressure P and the vibration velocity of the particles _v of the material in which the wave propagates:

When the lateral dimensions are sufficient to reason in plane waves, this term is the product of the density p by the speed of propagation _c in the medium of propagation:

Z = p .c The relation expressing the transmissions / reflections of the waves at the interfaces of different media can be expressed by the ratio of the acoustic impedances. A first approach to determine the impedance of a material is to evaluate separately the two parameters, density and propagation speed, to deduce the acoustic impedance of the material and to predict the transmission coefficient of a wave propagating from a first material to a second material sharing with the first a large lateral extent interface in front of the wavelength.

When the lateral dimension of the interface becomes comparable to the wavelength, we will no longer speak of characteristic impedance, but of radiation impedance. The radiation impedance involves the evolution of the contact section a little before and a little after the interface, over a distance at least equal to the half-wavelength and which corresponds to the volume zone of material coming from in response to the vibration at the interface.

The density can be known from the technical data of the material under consideration or evaluated by Archimedes' method, of which there are standard methods of immersion measurement, ISO 1183, DIN 53 479, ASTM D792. The propagation velocity can be measured according to the transit time of the wave through a known dimension of the material. Such a method is for example described in the article by AR Selfridge, "Approximate Material Properties in Isotropic Materials", Sonics and Ultrasonics, IEEE Transactions on, Vol. 32, no. 3, p. 381 -394, May 1985. When the damping is low, the speed can also be determined precisely by measuring the transit time of several echoes on the faces of the material. Several techniques for determining the speed of The propagation of the two fundamental modes and the characterization of the properties of materials from these data are thus presented in the chapter DK Pandey and S. Pandrey, "Acoustic Waves - Chapter 2: Ultrasonics: A technique of material characterization", in Acoustic waves, 2010. Another approach to accessing characteristic impedances is to start from the relationship Z = P / v. Numerous measurement methods based on different configurations of microphones and loudspeakers exploiting acoustic waveguides are found in the literature. The technique consists in emitting a wave, in monochromatic form or in a broad spectral band, emitted continuously or impulse which is reflected or transmitted by the test material. The wave is characterized in amplitude and / or in phase and the signal can undergo a mathematical treatment to go back to the intrinsic mechanical properties of the material such as the characteristic impedance.

For example, there is a method described in the patent application FR2547055B 1 exploiting sound waves in the air to evaluate the absorbent properties of materials. It operates a cylindrical waveguide, closed on one side by a sound source emitting a continuous wave and, on the other, by the test material whose impedance is to be determined. The waveguide is hermetically applied against a flat wall of the object to be analyzed. Two microphones are inserted into the waveguide and disposed at known distances from the interface so that the acoustic distribution of the resulting stationary wave can be measured and the impedance of the material can be deduced, with reference to the distribution of a standing wave for a standard material, the mechanical impedance of which is also known by another method. The applications of this method mainly concern civil engineering for building soundproofing.

Another method described in FR2806162B 1 is to emit a longitudinal wave beam through a fluid, for example water, to a plate of a material to be analyzed and for which it is desired to evaluate its homogeneity. A layer of impedance adapter varnish between the plate and the water allows maximum penetration of the wave into the material. Then, it is reflected on the opposite side until the transmitting transducer that has switched, meanwhile, in receiver mode. It is then possible by analysis of the different echoes to deduce the impedance of the material as well as its homogeneity, in particular when the material is a plate which one seeks to control the absence of defects by scanning over its entire area. . The use of a fluid as a medium ensures the transmission of a perfectly longitudinal wave. The process is applied to non destructive wing control of aircraft and aims to detect inhomogeneities and the possible presence of defects in the material.

US6298726 discloses an acoustic impedance measuring device by analyzing reflected waves on several elements located at depth of the material. Particular parameters of the frequency response of the echo signal are extracted and compared with calibrated values to identify particular materials. An algorithm based on a polynomial equation, calibrated for the transducer, delay and sample set, enables the identification of the material. This device can adopt different configurations with a waveguide, focusing or not the longitudinal wave on a small surface. The excitation signal is a trapezoidal burst with rise and fall times adapted to cancel the echo in the material. Alternatively, the trapezoidal pulse may consist of two pulses shifted in time, the second pulse being shifted by an interval creating a wave in phase opposition with the wave from the first pulse and an amplitude to cancel the echo from the test material. The patents just cited are all intended to be used in a single mode of longitudinal propagation of the echo.

There are also publications dealing with acoustic impedance measurement in living materials or tissues. The publication "A contact method for the assessment of ultrasonic velocity and broadband attenuation in cortical and cancellous bone", Clin Phys Physiol Meas, vol. 11, no. 3, p. 243-249, August 1990, CM Langton et al, exploits a principle of emission of an acoustic wave in a living part of an animal or a human, here the horse paws and makes it possible to determine the speeds of propagation in different tissues and bone types, by analyzing transmitted waves or echoes of ultrasonic waves through the sample to be analyzed. A temporal representation makes it possible to determine the propagation speeds in the different materials. The goal is to diagnose fracture risks. But, it is not question of modes of propagation other than longitudinal, nor of characteristic impedance analysis.

The publication "Direct measurement of index finger mechanical impedance at low force", in Eurohaptics Conference, 2005 and Symposium on Haptic Interfaces for Virtual Environment and Teleoperator Systems, 2005. World Haptics 2005. First Joint, 2005, p. 657 - 659, C.-Y. Fu and M. Oliver, to evaluate the mechanical impedance of the whole finger at low frequency (<500 Hz) by the use of an electromagnetic excitation device (pot vibrating) and measuring movements with a laser vibrometer for different ranges of forces applied and measured using a force sensor.

The publication "An Ultrasonic Indentation System for Biomechanical Properties Assessment of Soft Tissues in-vivo," Biomedical Engineering, IEEE Transactions on, Vol. 43, no. 9, p. 912-918, Sept. 1996, Y.-P. Zheng and A. F. T. Mak, shows a device consisting of a cylindrical waveguide coupled to a force sensor. The purpose of this system is to apply the end of the waveguide to a biological tissue under the condition of a bearing force. The indentation of the probe in the tissue can be measured and an ultrasound wave emission system makes it possible to measure the thicknesses of the different layers of tissues such as the epidermis, dermis, muscle and bone. A burst of waves is sent, then the echoes on the different tissues are recorded. By knowing the speed of propagation and by measuring the flight time between the different echoes, it is possible to determine the thickness of each layer. Knowing the support force and the thickness of each layer of tissue, the device makes it possible to trace and determine a viscoelastic behavior profile of the organic element and the various layers constituting it. In this publication, the device exploits an ultrasonic emission through a waveguide in organic materials but does not aim to measure the characteristic impedance of the materials and does not distinguish between the type of waves propagating in the tissue (longitudinal or transverse).

With regard to ultrasonic waveguides with bending wave cone guides, the state of the art mentions the work of JP Nikolovski and D. Royer (1997) "Local and selective detection of acoustic waves at the surface of a material, 699-703 IEEE Ultrasonics Symposium Proceedings, Toronto, 1997, and Nikolovski's patent application, FR2755225 of 1996 concerning a non-contacting touch-sensitive tapered-wave ultrasonic probe, and a flowmeter with conical points with ultrasonic waves for gas flow measurement, as well as the patent FR0759603 of December 6, 2007 concerning the measurement of thermal effusivity of a sample at room temperature using a heated tip. These applications mention many means of transmitting / receiving bending waves in a solid or a fluid from a conical tip, possibly flattened at its end to reduce the radiation impedance. The use of this device is, however, controlled only in particular coupling situations where the coupling section does not need to be known. No work mentions the measurement of the transverse impedance of the test material, nor the refraction angle and wave detection in the test material when the wave velocity in the material is comparable to the phase velocity in the tips, nor the method of use of the tips when brought into contact with very hot materials, the temperature of which exceeds several hundred degrees Celsius Curie temperature of the piezoelectric ceramics used to emit and receive waves in the tips, or the coupling of three or more points for the simultaneous measurement of longitudinal and transverse velocities in the sample, or the mechanical integration of the points in a granular carrier structure simultaneously performing the damping, ultrasonic, acoustic and thermal insulation, heating or the measurement of the support force or the measurement of the hydrostatic pressure prevailing in the medium or the control of the radiation impedance by truncation of the tip tapered at a position where the phase velocity reaches a given value.

In FR2547055B 1 and FR2806162B 1, the use of a fluid medium (air and water) allows only a longitudinal propagation mode. However, in all the fields of application of a device according to the invention mentioned above, whether in metallurgy, cosmetics, dermatology, power supply, flow metering, or else tactile interfaces, the mechanical coupling implements shear waves with the test environment.

In addition, in non-destructive testing, or in medical ultrasound, the use of slow-spreading transverse waves can improve spatial resolution, and reveal small variations in transverse impedances (associated with malignant tissue). The sensitivity to small transverse impedance variations is also useful in C D, in zones of appearance of cracks, delamination or crystalline phase change.

In addition, the applicant has not found in the state of the art, particularly in the field of high-frequency elastography method of measuring shear impedance at the scale of one cubic millimeter. Finally, the Applicant has also not found in the literature information relating to the coupling between a dispersive waveguide and a collimated liquid guide carrying ultrasound and projected at high speed on a sample to be inspected, said sample being in very fast relative displacement relative to the liquid guide.

The device proposed here is composed of contact probes, solid or liquid wave or guided, contacted with the materials to be characterized. The contact surface is an essential component of the device because it defines the limit between fundamental speeds and propagation speeds in dispersive guided medium. It also defines the apparent thermal effusivity of the probe when there is also a heat transfer process between the probe and the sample. The method comprises generating longitudinal and transverse waves, or dispersive waves, at the desired frequencies, in a known impedance waveguide and profile, contacted with the test material. The material can have a high temperature because the piezoelectric transducers used to transmit and receive the ultrasonic waves are not directly in contact with the sample, a relatively long waveguide acting as a thermal buffer. Conversely the waveguide may be heated and yield some of its heat to the sample, while the sample is at room temperature prior to contact. It can also be cooled and have a lower temperature than the sample to absorb some of its heat. The objective is then a simultaneous measurement of the mechanical and thermal properties of the sample. An echo processing makes it possible to deduce the characteristic and radiation acoustic impedances of the sample, in the case where it is not plane, as well as the propagation velocities and the spatial and temporal variations of the propagation velocities in the sample. . A variant of the device makes it possible to determine local impedances by echographic method or by direct or differential coupling between several conical flexural waveguides.

In the field of time variations, the device can determine a flow rate of a fluid by contact with the solid or in the surrounding fluid. It can also be used to characterize a process of solidification by thermal cooling, with change of liquid-solid phase, for example in the gelation of food or polymerization of resin or in the case of a metal plate several centimeters thick coming out of a blast furnace, the core of the plate still being melted at 1200 ° C (one thousand two hundred degrees Celsius), while the outer surface of the plate begins to solidify. In this case, depending on the raw materials used, mixing proportions, we seek to increase expertise on the cooling rate and the temperature profile in the thickness of the material to better control the appearance of crystalline phases and in fine the mechanical properties of the material, for example cast irons or glass plates, or even steel plates, or other materials resulting from high temperature manufacturing processes by rolling or float bath.

The simultaneous use of several conical waveguide waveguides integrated in a barrel or a collimated liquid guide in laminar flow and conveying ultrasound according to embodiments proposed here, makes it possible to increase the measurement dynamics, the lateral and deep spatial resolution and obtain a better diagnosis, especially when scanning a multi-tip probe for two-dimensional imaging (curvilinear and depth) of cancerous nodules or atheroma plaque, by combining several types of measurements in a minimum volume, each type of measurement providing for example a B-scan image in a given color.

Preferred but nonlimiting aspects of the device proposed here will now be described. The knowledge of the characteristic impedances or propagation velocities of the two propagation modes, as well as the knowledge of the radiation impedance at the point of contact when the lateral extent of the guide is small in front of the wavelength, gives access to several mechanical properties of the material such as the modulus of traction / compression (Young's modulus), the modulus of flexion / shear, the Lamé coefficients, the Poisson's ratio, the radiation impedance of the material involving a zone of limited coupling, as well as thermal properties of the material in contact with the tip, when it is heated or subjected to a temperature different from that of the sample. The radiation impedance is defined here as the product of the phase velocity by the density of the material constituting the guide. A large-diameter guide with a flat end, for example a cylindrical guide with a diameter of 13 mm (0.5 inches), generates waves considered quasi-planar and the impedance of the guide is equal to the characteristic impedance of the material constituting it. . The volume guides are thus adapted to flat, soft and isotropic materials, while the tips are adapted to local measurements on curved interfaces and for materials with rapid spatial variation of mechanical properties or high surface temperatures.

The device proposed here makes it possible to identify in the same place, the longitudinal and transverse characteristic impedances and of radiation as well as the local variations of impedance and effusiveness simultaneously by simple contact, without deterioration of the medium to be analyzed. The system exploits the reflective properties at the interface of the waveguides and the material to be analyzed. The waveguides have dimensions adapted to the transit time specific to each type of wave so that a longitudinal wave, propagating faster than a transverse wave, does not disturb the transverse wave arriving earlier than this. last. The mechanical properties of the materials are directly calculated from a comparison between a reference measurement, obtained without contact or at a given point of the reference surface of the sample, and a measurement on the sample, without approximation by a calibrated model beforehand. In cases where the test medium has finite dimensions, whether laterally or at depth (stratified plate, dermis, food, ...), the guide waveform has a tapered geometry and is selected to propagate dispersive bending waves, the waveguide section at the interface being selected to work with a certain phase velocity of 1% and 100% of the speed of the transverse wave propagating in the guide and therefore a certain value of radiation impedance of between 1% and 100% of the transverse impedance of the material constituting the guide allowing an optimal transfer of the energy in the test environment. Alternatively, the tapered guide section decreases until the wave phase velocity in the guide is less than or equal to the velocity of pressure or longitudinal or transverse or guided waves generated in the medium. It is recalled here that the characteristic longitudinal longitudinal impedance, respectively transverse, is the product of the propagation velocity of the longitudinal wave, respectively transversal, by the density of the material. The waveguides are preferably made of materials having a characteristic impedance greater than or equal to that of the test material, for example low density metals such as aluminum or titanium when we want to characterize the dermis because the spike waveguides are very sharp with typical opening angles of 5 ° which provides on the one hand a sufficient bandwidth to work at a high frequency of the order of Megahertz and secondly a radiating impedance may be small, in practice close to the longitudinal characteristic impedance of the skin, which allows for contrasting measurements on the local impedance variation of the dermis. In particular, the tapered profile of a tip makes it possible to reduce its phase speed of the bending mode as it approaches its end, down to below the speed of propagation of the waves in the water . When the diameter of an aluminum tip typically reaches 150 μπι, the phase velocity at 600 kHz falls below 780 m / s (4 times lower than the transverse speed in a large aluminum block which is 3100 m / s) and twice as low as in water. In this configuration, the wave can no longer be refracted in water because the refraction angle is greater than 90 °. The tip must then be pushed deeper into the water until it reaches a depth where the phase velocity exceeds 1500 m / s and refraction is possible. The amplitude of the echo corresponding to the bending wave reflected inside the tip, at its end, then decreases very quickly. As an example of numerical values illustrating this variant, it is known that an aluminum block with volume has the following characteristics:

VT = 3100 m / s (transversal speed)

VL = 6400 m / s (longitudinal speed) p = 2700 kg / m ³ (density)

ZL = 17.3 MRayl (longitudinal impedance)

Ζτ = 8.4 MRayl (transverse impedance)

The water has the following characteristics:

V _L = 1500 m / s

p = 1000 Kg / m ³

Z _L = 1.5 MRayl

For example, a conical waveguide aluminum tip (vertex angle = 4.8 °), which is truncated when the section diameter reaches 0.15 mm, is preferably chosen to obtain:

V _t (d = 0, 15 mm) = 780 m / s

Zt (d = 0, 15 mm) = 2.1 MRayl

The use of a dispersive tip whose end is immersed or partially immersed in a fluid and whose phase velocity is comparable to the speed of propagation of the waves in the fluid makes it possible to obtain a waveform angle of pressure, substantially aligned with the central axis of the emitting tip with a phase inversion on either side of the axis of the tip. This property may be a disadvantage if it is desired to refract waves at 90 ° from the main axis of the tip.

This disadvantage can be overcome by the use of a dispersive tip whose end is curved in the form of a hook with a curvature which compensates for the increase of the refraction angle as the wave progresses towards the end of the tip so that the ultrasonic beam diffracts into the medium at a constant angle. This dispersive tip shape, the end of which is twisted or curved in the form of a fishhook, makes it possible to orient the acoustic beam at a refraction angle close to the perpendicular to the main axis of the tip, or even exceed the normal and produce a negative angle of refraction relative to the normal to the main axis of the tip, that is to say able to generate a wave in the middle, near the end, which turns back. As will be explained below in connection with the description of the figures, it is preferably provided, for refracting the beam at 90 ° from the main axis, that the curvature of the tip is greater than 60 ° at the tangent to the tip of the tip. The main interest of this geometry resides in the possibility that it gives to couple acoustically two points with very close bases and very distant ends. The interest is obvious in the case of a transit time flow sensor a two-point differential inserted in the pipe by a single small hole (for example less than 20 mm), this configuration makes it possible to insert the tips so that they move away as far as possible from one of the other to increase the resolution power flow while imposing that the beams refracted in the fluid are in the direction of maximum sensitivity. Note that to emit at least a portion of the beam in the main axis of a tip, it will suffice to bend the tip of about 30 ° at the tangent to the end.

In the case of a flow measurement, it is possible to use two pairs of probes inserted so that the ends of the probes of each pair are at different depths, one of the pairs having its ends opening in the vicinity of the wall, the other in the center of the pipe, so as to perform a bi-cord measurement in the middle of the pipe and at the wall. This makes it possible to establish the velocity profile of the fluid flowing in the pipe and thus correct the K factor of the flow meter and ultimately obtain a more precise measurement of the flow rate. The insertion of a metal tip into its single-walled or multiple-walled carrier structure can be accomplished by force insertion or more conveniently by inserting and crimping the carrier structure around the cylindrical base of the cone by means of a ferrule. In this case, the ultrasonic damping can be optimized by creating a heterogeneous granular structure, for example a sintered metal or plastic in the form of a cylindrical tube with deep but non-through notches so as to maintain the seal. For this, the notches can be staggered to facilitate the crushing of the metal or the polymer during the crimping operation.

It will also be noted that the end of the sleeve may have different geometries. Its primary function is to guide the tip and pinch a little upstream of its end so as to close the cavity filled with insulating air the tip of the sleeve and that cavity is sealed. This ensures that the acoustic waves propagate well to the end of the waveguide and do not begin to refract in the medium before reaching the end of the tip. This results in an intense acoustic source localized near the end, even when the sheath is immersed in a liquid or cast in a solid. The second function of the sheath near the end of the tip is to act on the refraction angle of the ultrasonic beam. For this purpose, it is possible for the sleeve to envelope the truncated conical tip in the nip region by presenting a surface that reflects the beam diffracted towards the front, ie parallel to the main axis of the conical tip. This is obtained in water when the pinch cone of the conical tip makes an opening angle between 135 and 145 °, for example about 140 °. The reflective surface can further extend to the level of the truncated tip so as to block any lateral radiation and impose a total radiation forward. In the opposite case, if the end of the sleeve stops very clearly at the nip, the angle of refraction with respect to the perpendicular to the main axis will extend from 30 ° to 90 ° in the water with a maximum observed around 60 °. It will be noted that the beam orientation function is valid if the mechanical impedance of the sheath is significantly greater than that of the ambient medium, for example air. Otherwise, if the impedance of the sleeve is comparable to that of the medium, for example water, then the orientation of the beam can advantageously be supported by a separate part by ensuring that the end of the sleeve stops net in the pinching area. The structure can then be completed with a metal collar formed around the sheath and having the appropriate reflective curvature, for example parabolic. The reflector can then be made of metallic material of mechanical impedance much greater than that of water to obtain an effective reflection. When the medium to be analyzed is a compressible gas, the pressure of the medium can be connected by the equation of state of the gas (virial equation) to the speed of the acoustic waves propagating in the gas. This speed is itself measurable by measuring the round-trip transit time of an ultrasound wave emitted forwards or laterally from a tip-mounted transducer in its sheath, itself provided with its metallic reflector collar for a transmission forward (or without collar for side emission). The metal reflector collar has in this case an additional interest, that of reinforcing the mechanical strength of the sheath and its end. The assembly then operates in transmission-reception via a reflector disposed in the middle, perpendicular to the main axis of the tapered waveguide. By knowing the length of the acoustic path between the tip and the plane reflector and from the measurement of the round trip transit time of the echo or successive echoes observed in time, we classically go back to the propagation velocity of ultrasound in the middle. This same speed can be obtained by measuring the frequency difference between two harmonics of the cavity formed by the tip with (or without) its reflective collar and the plane reflector disposed perpendicularly to the main axis of the tip.

Frequency analysis is performed in the frequency space by DFT (Discrete Fourier Transform). The advantage of the device proposed here lies in the fact that the lowering of the radiation impedance of the tip by exploitation of dispersive bending waves increases the efficiency of the solid tapered-gas coupling and makes it possible to measure low pressures, typically to pressures below atmospheric pressure. This is not always the case when it is necessary to couple an acoustic wave through a wall to measure the composition of a gas and the pressure that prevails therein. The device proposed here thus finds a direct application in the nuclear field, in particular for the measurement of the release of gases under pressure, in particular Krypton (Kr) and Xenon (Xe) during the fission process of uranium oxide pellets. contained in the nuclear fuel rods and as indicated in the patent No. 2 934 409 of Jean-Yves Ferrandis et al of 24 July 2008 entitled "Acoustic sensor for measuring a gas in a chamber and assembly comprising a speaker and such a gas ". The internal pressure of the fuel rods ranges from a few tens of bars at the beginning of activity to more than 300 bars at the end of activity and the temperature can exceed 300 ° C. The solution disclosed in the Ferrandis et al. Patent is well suited to a certain pressure and temperature range. Nevertheless, it faces two limits: the first relates to low pressures where the planar or radial configuration of the longitudinal waveguide limits the effectiveness of the acoustic-gas solid coupling so that the signal / noise ratio is unusable below a few tens of bars and does not measure the speed of sound nor the pressure or the composition of the release gas. But this first limitation is not a problem because it only exists at the beginning of the fuel life. The second limitation is more troublesome because it concerns the evaluation of the end of life of the fuel, while the pressure and the temperature increase and that the end-of-life diagnosis is accompanied by costly maintenance operations in fuel change. During its activity, the medium can also reach a working temperature exceeding 300 ° C which is not compatible with the use of piezoelectric ceramics having the best piezoelectric coefficients, limited to maximum working temperatures close to 250 ° C. If one wants to be able to work at 500 ° C, it is necessary ceramics PZT whose piezoelectric coefficients are 10 to 20 times weaker than that operating at 250 ° C. In this context, the signals are too weak to make accurate measurements of sound velocity, gas composition and media pressure.

However, by exploiting a configuration with a metal sleeve crimped and welded to the tip at the base of the conical waveguide and associating a conical bending wave tip with its forward transmitting collar (or without but in a cylindrical cavity (see FIG. 26) and a plane reflector disposed on the main axis, it is possible to obtain a narrow sensor of diameter typically less than 10 mm, that is to say comparable to the diameter of a pencil of fuel and can therefore be mounted at the end of a pencil thus typically improving by a factor 30 relative to a planar configuration, the effectiveness of the acoustic coupling between the solid guide and the surrounding gas, which ultimately allows to work with PZT ceramics which perform poorly piezoelectrically but are suitable for high working temperatures up to 550 ° C such as PZ46 (Ferroperm, Denmark). The effectiveness of the tapered solid coupling with bending-fluid wave thus constitutes an appreciable advantage when the measurement of pressure and in particular the resolution on the pressure depend on the measurement of the amplitude of the echo and thus the signal / noise ratio. . It will be noted that in a compressible fluid, the acoustic impedance of the fluid increases with the pressure, which reduces the solid-gas guide acoustic mismatch and increases de facto the amplitude of the echo signal or transmitted in the medium. In conclusion, the measurement of the amplitude of the echo (or of a signal transmitted in the case of a system with several points) is directly exploitable for the measurement of a pressure. Nevertheless, it is advisable to combine a pressure measurement by capacitive method of the deformation of the carrier structure or the sheath that will be reserved at low pressures typically around a few bars, to have a high resolution, to a measurement of pressure from the measurement of the amplitude of the reflected or transmitted signal that will be reserved for higher pressures, the low-pressure calibration by capacitive method can also be used for high-pressure calibration using the amplitude of the signal . It should also be noted that for large pressure variations, the measurement signal must undergo an automatic gain control.

When the sample is the dermis or food, the tip is preferably aluminum alloy or titanium. The radiation impedance of the tip is lowered by refining its end by polishing so that an incident wave is fully transmitted (or the reflection coefficient is lowered to a threshold value) in the skin or food when the end of the tip is in contact with the dermis or food.

The section of the tips being very small, around 0.1 mm ² (tenth of a square millimeter), the impedance measurements can be carried out finely and on the scale of the square millimeter which is of interest in cosmetics and dermatology. In particular, the speed of shear waves in the viscoelastic media such as the dermis can decrease very strongly to reach values of the order of one meter per second. In this case, the velocities measured will be those of Rayleigh waves or horizontal transversals close to each other within 5%. Where possible, and the radiation impedance of the tip is comparable to that of the medium, the tip can be driven into the test medium (silicone, viscous liquids, vegetables, fruits, etc.) to determine the height necessary to cancel any echo inside the tip. This height is characteristic of the longitudinal impedance of the medium. In practice, rather than canceling the echo, a depth of penetration will be determined where the echo is decreased by 50% or 75%, which is easier to measure.

A bipod may consist of two pointed waveguides in contact with the test surface. The axes of the points are in the same plane perpendicular to the plane tangent to the surface at one of the points of contact. Deep drilling is carried out in the osculating plane with a lateral resolution determined in particular by the contact diameter of the tips, which is in practice less than one millimeter. The lateral extent of the beam is also conditioned by pulsed or continuous operation. In pulsed mode, the configuration is adapted to measurements by differential transit time or pulsed Doppler velocimetry. A bipod with superalloy metal conical guides, for example Inconel 600 or 625 can be used to probe materials carried at high temperatures of several hundred degrees Celsius. The contact surface between the test material and the conical guide is reduced, which slows down the heat transfer mechanism and creates an apparent thermal effusivity well below the thermal effusivity of the same volume material. The height and the opening of the conical guide in superalloy, is at least 25 mm and is preferably chosen between 50 mm and 150 mm, with a good compromise between sensitivity and bandwidth obtained with a tip height of 100 mm for an angle at top of 4.8 °. The waveguide then has an acoustic transfer function between its base and its end which is similar to a low-pass filtering of cutoff frequency greater than 700 kHz. These dimensions make it possible to work at a center frequency between 600 and 700 kHz for an aluminum alloy. The height and the angle at the apex also make it possible to lower the temperature between the zone of contact, which can be carried for example at a temperature greater than 1000 ° C. and the base of the cone where the piezoelectric elements are bonded, generally ferroelectric, for which it is not necessary to exceed the temperature of Curie. It is also a significant advantage of the device presented here to provide a device for insertion of longitudinal acoustic waves and shear by having a radiation impedance adapted to the test medium. Finally, the base of the cone where the transducer is glued PZT is exposed to a temperature below the Curie temperature of the piezoelectric element used to generate and receive the ultrasonic waves.

The base can also be actively cooled via a heat transfer circuit, so that its temperature does not exceed 300 to 500 ° C, while the end is heated to a temperature exceeding 1000 ° C. Reducing the section of the waveguide here brings a considerable benefit since it amounts to lowering the apparent thermal effusivity of the material constituting the tip. As a reminder, the thermal effusivity is the square root of the product of the density of the material by the specific heat and its thermal conductivity. It characterizes the process of heat transfer between two materials brought to different temperatures. It makes it possible to define the interface temperature as the centroid of temperatures weighted by the effusivity of the materials. But this definition is valid only under conditions where the interface is flat. When the lateral extent of one of the materials is very small in front of the other, as in the case of a point in contact with a plane, the apparent thermal effusivity of the tip is weak and it is therefore the plane which imposes its temperature on the tip. Nevertheless, as one goes up towards the base of the point, the thermal inertia increases as much as the apparent thermal effusivity which approaches that of the voluminal material, so that it is the base of the cone that imposes the temperature. The conical guide therefore functions as a thermal buffer. This configuration makes it possible to mechanically couple the guide to a medium heated to a temperature well above the Curie temperature of the piezoelectric element. In addition, the piezoelectric element is generally bonded to the guide with an epoxy adhesive. The latter can hardly withstand temperatures above 300 ° C. Beyond that, it is necessary to consider, for example, a solid couplant, such as a gold layer. Therefore, in order to operate an epoxy adhesive with a sample heated to 1000 ° C, it is desirable, on the one hand, that the section or volume of the guide exposed to a high temperature sample represents a smaller surface area or volume, less than 100 times smaller, than the section or the volume of the guide coupled to the ultrasonic source and exposed to the colder temperature of the heat transfer fluid, without limiting the amplitude of the transmitted signal. This is the advantage of conical bending waveguides which simultaneously have a high bandwidth of between 0.3 and 3 MHz, depending on the shape of the waveguide, a reduced apparent thermal effusivity at the end and a radiation impedance. typically 4 times weaker than transverse impedance.

A bipode according to one embodiment comprises a transmitter and a receiver distant from each other by a value adapted to the laws of refraction in the sample and depending on the depth probed. The ends of the tips are carefully truncated so that the phase velocity at the ends is just greater than or equal to the velocity of the longitudinal waves generated in the medium.

A preferred arrangement in so-called tripod transmission consists in arranging two receiving points at equal distance from and in alignment with a bending wave emitter tip. The receiving tips are arranged angularly with respect to their central axis so that the received signals are in phase opposition. The excitation is impulse or continuous. The summation of these signals provides a spatial differential measurement allowing small variations in the local mechanical properties to be observed between the receiving points. A tripod arrangement can advantageously be exploited by reversing the transceiver roles, with the transmitters continuously vibrating in phase opposition, resulting from a phase inversion between the two electrical excitation signals, either because of the angular orientation points. The tips are typically spaced 1 mm apart. More generally, spike transducers can operate in bending mode at frequencies between 20 kHz and 2 MHz. But they may preferentially be excited by a sinusoidal pulse at the center frequency of the tip or modulated in frequency by performing a digital frequency scan for example between 300 kHz and 900 kHz (called "chirp" in English). The frequency range is for example divided into 200 frequencies in the 3 kHz step, uniformly distributed between 300 kHz and 900 kHz. The burst consists of a complete period of each of the 200 frequencies distributed. In this case, the signal is preferably a square signal. There is then only one receiver associated with a single variable gain electronic amplifier, followed by a bandpass filter between 300 kHz and 900 kHz, followed by a single analog-to-digital converter, digitizing the signal on 12 bits, at a preferred rate of 5 to 12 MHz (or mega samples per second) and a single random access memory (RAM) of typical size 2 megabytes.

The rate of fire is preferably between 20 and 1000 shots per second. Each shot is digitized over typically 5,000 to 20,000 samples, representing an acquisition time of 1 to 4 ms and 10 to 40 kbytes of memory. RAM can store 50 to 200 successive shots. The reception signal is digitized at the same time as the transmission signal. This configuration makes it possible to probe the space underlying the central receiver by masking the direct signal between transmitters and central receiver to measure disturbances backscattered by the medium. The received signal can then be further amplified without saturating the electronics. The relative variations of the received signal are calculated by slowly sliding the tripod along a curve on the surface of the sample, the relative variation is calculated at the end of the sweep or when the memory is full by taking as reference one of the acquisitions made during the contact. Depending on the rate of fire, the acquisition lasts between 0.05 seconds and 10 seconds. The relative variation of the signal with respect to a reference signal is performed in the time domain or in the frequency space (in module and in phase) after calculation of a Fast Fourier Transform (FFT). The relative variations of the signal are representative of variations in the mechanical properties of the backscattering medium during displacement of the tripod. In other words, the configuration with two spike transducers generating a phase-opposite acoustic field at the receiver, makes it possible to perform a measurement of relative multi-frequency perturbation. When working in the frequency space, the frequency components whose amplitude does not exceed 3% of the maximum amplitude observed are eliminated. Then the Euclidean norm of the relative perturbation vector in modulus is calculated. This vector is representative of the diffusing power of the medium over a set of frequency components. Any finding of a significant variation in the standard of this signal during manual or automatic mechanical scanning may result in a new scan possibly changing the scan rate or measurement rate to better visualize a change of ownership zone. backscattering.

The slow displacement of a tripod assembly on the surface of the medium, in the direction perpendicular to the segment connecting the three points, provides information on the spatial variations of impedance. It gives information on inhomogeneities of the environment. The temporal variations of a motionless tripod provide information on variations in the environment of the tripod, for example linked to mechanical deformations caused by the passage of blood or by swelling of the medium associated with breathing or muscle contraction or by thermal diffusion. When the point of contact of the device is skin, the tip may be truncated so as to avoid damaging the dermis, for example by keeping a diameter of at least 0.4 mm. The tip is brought into contact with the skin (or any other flexible material) with a constant force (for example equal to its weight when no dynamometer is available) or slightly greater by a few grams than its practical weight. at 15 grams) using a precision digital balance (having a resolution of at least 0.1 g and in practice 0.01 to 0.001 g and a 4-20 mA or digital output for controlling the position from the peak to a target force). As a variant, the force exerted on the tip is directly measured by a capacitive method based on the deformation of the bearing structure of the tip, as will be described later (see Figure 22).

Deformation of the supporting structure under the action of a force or pressure can cause the sliding of one wall relative to the other (in the case of a double-walled probe) or the slump of a wall (for example in the case of a tube probe with internal cavity and maintaining the tip at its base and its end). There is thus an air cavity around the cone that can sag or swell under the action of a hydrostatic pressure. Assuming that the metal tip is at the electrical ground and constitutes a first electrode and that the outer surface of the sleeve is coated with a conductive layer of revolution (or partial for example on an angular sector with a conductive strip from the base towards the end of the sheath) forming another electrode, materializes a capacitor whose variation in thickness under the effect of the hydrostatic pressure is representative of the external pressure exerted by the medium. This configuration lends itself well to the realization of flowmeter probes with manometric function with minimum bulk.

The sheath with its tip tip is advantageously covered with an insulating varnish avoiding oxidation of the metal tip and / or a protective layer for example a Teflon-based varnish (PTFE or FEP or PFA) or silicone fungicide for avoid the accumulation of organic matter on its surface. In case of use of a metal reflector mounted on the end of the sleeve, the cavity or flag surrounding the end of the tip is preferably filled with a soft silicone, transparent for ultrasound and preventing scale buildup.

In the case of a two-point flowmeter with insertion perpendicular to the wall of the pipe, it is possible to provide shoulders either on the sheaths or on the bearing structure, from which support arms carrying a reflector plate placed under the spikes, perpendicular to their main axis at a distance comparable to the interpoined distance, themselves aligned with the direction of flow, the assembly forming a transmission-reception system with direct and reflected acoustic paths forming an equilateral triangle and characterizing a refraction angle in water close to 60 °. At least one of the sleeves, downstream, is provided with a collar for screening the direct beam.

The use of two tripodes oriented 90 ° from each other, and sharing the same central point, constitutes a pentapode, that is to say a system with 5 points. The pentapode is the configuration the most complex, but allows the measurement of different parameters during a single contact with the material to be characterized. In such a system, it is advantageous for the peripheral peaks to operate as a transmitter so that there is only one electronic amplifier connected to the single central receiver. The two additional points can thus operate at the same frequencies or at different frequencies of the two transmitting points of the first tripod. In particular, they can vibrate in polarization oriented differently from the other two.

Each of the tips of a pentapode has a four quadrant ceramic functioning as a linear combination of two dipoles oriented at 90 ° to each other and can be combined in amplitude and / or phase. This gives a total control of the acoustic vibration in the contact plane. The tips may in particular be programmed to emit shear waves or longitudinal waves towards the central tip depending on whether their direction of vibration is collinear or parallel to the main sensitivity axis of the central tip. When the central tip is operating in reception, the pentapode is in echographic configuration. In this configuration, it is sought to bring the points as close as possible so as to be able to detect rapid spatial variations and to exploit high frequency shear waves, preferably with a frequency greater than 20 kHz.

When the central tip is transmitting and the four peripheral points are receiving, the pentapode is in directional velocimetric configuration. In this case, the central tip operates in precession mode, while the 4 peripheral points operate in pairs of receivers diametrically opposed to the transmitter and detect the direct signal emitted by the central transmitter. The measurement is then a measure of differential transit time per pair of receivers. It provides information on temporal variations, for example a flow in the plane of the ends. In this configuration we try to move the peaks as far as possible in order to increase the resolution on the speed measurement. This configuration is more expensive in electronic amplification and digitization / memory components.

The invention thus makes it possible to simultaneously evaluate several parameters of a medium without damaging it, in particular: · The speed of the longitudinal waves propagating in a sample (and thus indirectly the longitudinal impedance by knowledge of the density). • The velocity of transverse horizontal or Rayleigh waves propagating in a sample (and thus indirectly the transverse impedance by knowledge of the density).

• Variations in the homogeneity of the medium by differential echography of the interface medium, and the associated two-dimensional imaging (B-Scan) in the temporal space or frequencies, that is to say a grayscale or color (curvilinear position, amplitude of the temporal signal) or (curvilinear position, modulus or phase of a frequency component obtained by Fourier transform).

• The speed and direction of a directional flow in the plane or in a plane underlying the directional velocimetry point plane.

• The simultaneous measurement in a volume less than 1 cm ³ of characteristic properties of a thermal effusivity or a support force.

The waveguides are chosen according to the desired working frequency which can vary for elastic materials from 300 kHz to 900 kHz and for viscoelastic media from 20 kHz to 4 MHz.

Thus, an embodiment provides a system for characterizing the mechanical and thermal properties of a solid or fluid medium, comprising at least one broadband ultrasonic transducer coupled to a first end of at least one solid tapered waveguide. to generate in the solid tapered waveguide at least one continuous or pulsed dispersive bending wave which propagates, the tapered solid waveguide being in contact, on the side of a second end, of section smaller than that from the first end, with the solid or fluid medium, wherein: the dispersive bending wave transmitted in the waveguide is refracted into an ultrasonic elastic wave in the medium, the maximum lateral extent of the contact surface between the tapered solid waveguide and the medium is less than the wavelength of the ultrasonic elastic waves generated in the medium, the tapered solid waveguide is t housed in a supporting structure of which at least one wall allows the acoustic insulation and the maintenance of the tapered guide as well as the damping of the ultrasonic waves, the guide being maintained by the supporting structure on the one hand on its periphery at its base of the broadband ultrasonic transducer side and, secondly, upstream of its tapered end, the system further comprising a device for measuring the amplitude or phase variations of the reflected waves and / or transmitted in the medium and a device for calculating mechanical and / or thermal properties of the medium from the measurements provided by the measuring device.

According to one embodiment, the tapered solid waveguide has a cylindrical axis of symmetry, has a solid or hollow conical profile having a section that is refined until the phase velocity in the refraction zone is lower than or equal to the speed of the longitudinal or transverse or guided waves generated in the medium or that its radiation impedance at the point of contact is between 1% and 100% of the transverse impedance of the material constituting the guide.

According to one embodiment, the computing device determines the mechanical impedance Z2 of the medium determined by the formula: Z-, = Z · ^ ⁺ with a = · -, where Z _a i _r is the mechanical impedance of the air, A _r2 is the reflection coefficient at the tapered waveguide interface - medium to be characterized, A _r i is the reflection coefficient at the tapered waveguide - air interface, and is the impedance of contact of the tapered waveguide.

According to one embodiment, the waveguide is made of a material having a characteristic impedance greater than or equal to that of the medium, for example a low density metal such as aluminum or titanium, or a hard plastic such as plexiglass or polycarbonate.

According to one embodiment, the tapered solid waveguide is a solid cone whose end is truncated at the point where the phase velocity is equal to the speed of the longitudinal acoustic waves generated in the medium. According to one embodiment, the system also has the following one or more features: the first end of the tapered waveguide is exposed at a first temperature below the Curie temperature Te of the ultrasonic transducer; the tapered waveguide is a superalloy, for example in inconel; the second end of the tapered waveguide is in contact with a medium of a second temperature greater than the first and has an area at least 100 times smaller than that of the first end of the tapered waveguide; the tapered waveguide has a length which ensures the maintenance of a temperature difference between the first and second ends of the guide, the temperature difference resulting from a phenomenon of convection or free or forced conduction by a cooling system heat transfer fluid, and regulates the temperature of the transducer at the first temperature. According to one embodiment, the system further comprises second and third tapered waveguides shaped conical points aligned to form a differential tripod, and two receiving points arranged at equal distances from a bending wave emitter tip, and aligned therewith, the receiving tips being angularly disposed with respect to their central axis so that the received signals are in phase opposition.

According to one embodiment, the system further comprises four tapered tapered waveguides to form a pentapode or two tripods oriented at 90 ° to each other and sharing the same central tip, the central point of the system operating as a receiver and peripheral points transmitter, the system further comprising a single electronic amplifier connected to the single central receiver.

According to one embodiment, the acoustic transducers associated with the tapered waveguides are four-quadrant PZT ceramic discs operating in a linear combination of two dipoles oriented at 90 ° to each other to be combined in amplitude and / or in phase so as to impose an acoustic vibration direction in the contact plane.

According to one embodiment, the system further comprises a fourth and a fifth tapered waveguides to form a pentapode organized in two tripodes oriented at 90 ° to each other, for the depth measurement of the speed variations. longitudinal or transverse.

According to one embodiment, the system further comprises four additional taper-shaped guides to form a pentapode, a central tip operating in transmission mode while the other four peripheral peaks operate in reception mode, the central tip operating in precession mode, while that the four peripheral peaks operate by pair of receivers diametrically opposite to the transmitter, the measuring device detecting the differential transit time per pair of receivers.

According to one embodiment, the system further comprises four conical-point guides forming a pentapode in the transit time measurement mode between a central emitter tip and the other four peripheral tips, the emitter tip producing a horizontal transverse wave in vibration. impulse precession, the two pairs of peripheral tips being oriented to detect the horizontal transverse wave, the measuring device giving a speed of propagation of this wave in two perpendicular directions. According to one embodiment, the system is adapted to contact with the epidermis and the conical tip transducers are truncated and slightly pressed against the epidermis so that the entire section of the guide is in contact with the epidermis and that in the contact zone, the section of the guide is such that the phase velocity in the tapered solid waveguide is equal to that of the longitudinal waves of the medium, ie about 1500 m / s.

According to one embodiment, the system is adapted to contact with an air environment and further comprises four additional tips for forming a pentapode with a central emitting tip E vibrating in precession mode and four peripheral receiving points responsive to a flexion mode , the central tip having a round end, preferably of section diameter at its end less than 0.1 mm, and vibrating in a combination of two bending modes oriented at 90 ° to each other.

According to one embodiment, the carrier structure is sealed and the tapered solid waveguide is pointed and partially immersed in a fluid medium to where its phase velocity is equal to the speed of propagation of the pressure waves. in the middle.

According to one embodiment, the carrier structure comprises a U-shaped double wall ensuring: the control of a set temperature of an internal wall of the supporting structure, for example by means of a resistive coil for heating the conical tip by Joule effect; damping successive echoes passing through the mechanical coupling zone of the waveguide to the inner wall of the carrier structure; the thermal insulation of the inner wall by the outer wall of the supporting structure, separated from the inner wall by a thermal insulator; the relative displacement of the inner wall of the supporting structure relative to the outer wall of the supporting structure under the effect of a contact force of the tip on the sample or a deforming pressure exerted by the medium on a part of the supporting structure or the holding sleeve of the point; intermittently measuring the value of a capacitance of at least one capacitor that accounts for the relative position of the inner wall relative to the outer wall of the supporting structure or the inner wall of the supporting structure relative to the guide tapered solid wave, and one of the electrodes is integral with the waveguide and the other electrode is integral with the inner wall or the outer wall, the value of the capacitance being deduced from a resonance frequency or oscillations of a relaxation oscillator, or a divider-bridge measurement of a reference capacitance and the law of variation of the capacitance being connected by a correspondence table, or a mathematical formula such as a law of polynomial interpolation, at the force of support of the tip on the sample, or at the pressure of the medium; or a safety function for the tip and the sample, imposing a limit on the maximum penetration at zero bearing force of the tip in the sample by a maximum value of between 0 and 10 mm, and preferably less than 0 , 5 mm. According to one embodiment, the support structure is composed of a part ensuring the positioning, the orientation, the insulation and the damping of the waveguides and a bayonet cowling consisting of elastic deformable parts ensuring a force holding the probes in the inner wall and locking the flexible elements on the guide element.

According to one embodiment, the system further comprises: a conduit for feeding a pressurized liquid waveguide at the end of the tapered solid waveguide; an end of the solid waveguide immersed partially in the liquid waveguide, in a connection chamber, on a limited extent, of the order of 1 mm and a volume of the order of 1 mm 3; a liquid waveguide ejection nozzle of straight conical or curved hook shape for inspecting external, lateral, or internal surfaces of a sample, such as a railroad rail; a protector skirt of the liquid waveguide against dislocations or fragmentation by a stream of air over at least a part of its length to the surface of the sample or containment and suction splashing of the jet liquid sprayed onto the surface of the sample, wherein: the hydrostatic fluid ejection pressure varies over time to provide low frequency shear acoustic wave generation; the acoustic waves are controlled in amplitude and / or in phase, in particular between the jets of a configuration with several liquid waveguides; the hydrostatic pressure is modulated to reach an intermittent ejection regime with an ejection rate of between 1Hz and 10kHz.

According to one embodiment, the end of the tapered solid waveguide is curved.

According to one embodiment, the second end of the solid tapered waveguide is forcefully inserted into a sheath formed in the supporting structure, the assembly being rigidly mechanically coupled to be sealed by crimping the sleeve or by metal brazing between the base of the tip and the sheath if it is metallic.

According to one embodiment, the tapered solid waveguide is inserted into a sheath formed in the supporting structure provided at its end with a reflective collar intended to orient the emitted ultrasonic beam or to shield the received ultrasonic beam from a predefined direction . According to one embodiment, the system further comprises a second solid waveguide tapered to form a bipode and a reflector integral with the support structure disposed under the points, perpendicular to their main axis, at a distance between 75 and 125% of the inter-point distance.

According to one embodiment, the system further comprises three additional tapered solid waveguides to form two bipodes inserted perpendicular to a pipe at two different depths, each bipode being aligned with the direction of the flow to be characterized, one bipods having their ends arriving at a first depth in the pipeline so as to access the flow velocity at the center of the pipeline, while the other bipod has its ends arriving at a second depth of the pipeline, distinct from the first depth, in the vicinity of the inner wall of the pipe so as to access the flow rate near the wall.

According to one embodiment, the liquid ejection tip is separate from the carrier structure and is bonded to the carrier structure in a position where at least one of the ultrasonic beams refracted into the liquid from the end of the tip is reflected on at least one of the faces of the tip and aligns after reflection with the main axis of the nozzle ejection cannula. According to one embodiment, the tapered waveguide is inserted into a sheath formed in the supporting structure, the end of the sheath with its open tapered solid waveguide being covered or filled with a protective anti-adherence varnish and or fungicide, for example based on Teflon (PTFE or FEP or PFA) or silicone.

According to one embodiment, the computing device deduces the pressure of a medium directly from the amplitude of an echo or a signal transmitted in the medium in combination with a capacitive measurement of the deformation of the carrier structure or a sheath formed in the supporting structure.

An embodiment further provides a method of characterizing the mechanical and thermal properties of a solid or fluid medium using a system as proposed above, comprising the following steps: generating in the waveguide solid tapered at least one dispersive bending wave in continuous or pulsed regime that propagates, the dispersive bending wave transmitted in the waveguide being refracted into an ultrasonic elastic wave in the medium; measuring amplitude or phase variations of the reflected and / or transmitted waves in the medium, and calculating mechanical and / or thermal properties of the medium from measurements provided by the measuring device.

According to one embodiment, the method further comprises the following steps: generating and detecting a reference echo in the medium by bipode (89), tripod or pentapode type probe (98) successively contacting the medium at a set of reference positions along a curvilinear reference path; storing the reference echoes associated with each position of the probe on the medium, along the curvilinear reference path; recognizing contact positions of the probe on the medium by calculating the minimum distance or by calculating the maximum of an inter-correlation function between a measurement echo obtained during a second passage on the curvilinear path and the one of the echoes corresponding to the learned reference positions, and to provide amplitude and phase information on the spatial variations of impedance along the curvilinear reference path.

According to one embodiment, the method further comprises a step of embedding the waveguide in the middle of a characteristic depth for which the amplitude of the internal echo in the waveguide is decreased by 50% at 75% of its value with respect to conditions where the end of the waveguide is free.

According to one embodiment, the method further comprises the following steps: moving a differential tip tripod to the surface of the medium, in the direction perpendicular to the segment connecting the three points; measure amplitude and phase on spatial variations of impedance.

According to one embodiment, the method further comprises: a step of constituting a reference database characterizing different media or the different states of a medium and associating with each medium or with each reference state the measurement of at least one of its effusivity, longitudinal and transverse impedance as well as angular anisotropy impedance values in at least two different directions, for example orthogonal directions, from bipod, tripod or pentapode type probes; a step of recognizing a medium or the state of a medium by measuring its effusivity values, longitudinal ZL and transverse ZT impedances as well as angular anisotropy of the impedances in at least two different directions, using a bipod, tripod or pentapode type probe by searching for a corresponding element closest to a reference database on this medium or on the states of this medium.

DETAILED DESCRIPTION OF EMBODIMENTS

In its simplest expression, exploiting wave-wave guides, therefore of lateral extent at least equal to three wavelengths, the velocity determination method proposed here exploits a comparative measurement between a reference measurement, for example without contact and a measurement during a contact on the sample to be analyzed. Then we use formulas giving the reflection coefficients at the interfaces as a function of the impedances of the media. It is a deterministic process.

• Among the applications envisaged are tactile hulls with artificial finger calibration. The prior knowledge of the impedance of the human finger from the device described here makes it possible to then make artificial fingers for different user profiles (children, adults), to optimize the thicknesses and coatings of the touch hulls by choosing materials whose longitudinal and transverse radiation impedances optimize the energy transfer with the finger.

• The device also finds applications in the analysis and monitoring of the viscoelastic property variation of materials, especially when the tip transducers are oriented to transmit and detect horizontal shear waves. A tripod or a pentapode in echographic mode in shear configuration also makes it possible to check a possible angular anisotropy in the viscosity of a fluid, this thanks to its ability to perform the same measurement in two perpendicular directions.

The device proposed here is also a powerful means for measuring spatial and temporal viscosity, particularly in the field of magnetorheological fluids or elastomers, which can rapidly change state under the action of a magnetic field with a response time of less than the millisecond. Indeed, even in pulsed mode, the tip transducers can be excited at a high rate of up to 1 to 2 kHz (or PRF for Pulse Repetition Frequency in English) and therefore likely to display rapid changes in state.

• In contrast, the device can find applications in slow processes of change of state, for example in the control of the maturity of food held at constant temperature in a cold room, the hardening of concrete, or lava, the polymerization of glues, sintering of powders, changes of crystalline phase, manufacture of gels.

• The device can give an estimate of the spatial variations of longitudinal or transverse acoustic impedance by a mechanical scanning in the plane of the surface of the sample (C-scan) and thus allow a mapping of a human tissue or a any material. The device according to the present invention can learn to recognize the successive contact positions with the medium during a mechanical scanning phase of a bipod, tripod or pentapode consisting in a first step to memorize the echoes associated with each position of the probe then to recognize the positions by distance calculation or inter-correlation during a second pass, following the first constituting a learning phase.

• The device proposed here can allow a classification of materials by the characterization of the impedances of the two fundamental modes and according to two directions of propagation, thus providing information on the anisotropy of the material as well as on the efficiency of the thermal transfer of a tip heated or cooled, the lowering or elevation of temperature after a contact of determined duration with the sample at room temperature being measured. In conjunction with the use of a database of references that may be constituted, the device can be a "classifier" of materials, applicable to elastomers, human tissues, food and meat in particular for the control of their state of conservation or ripening, as well as various metal and plastic materials.

• This device can finally allow the inspection of moving surfaces by combining a solid guide coupled to a liquid guide projected towards the sample. This type of probe finds applications in nondestructive, high speed control. The probe is constituted by a double wall bearing structure possibly heated to improve the ultrasonic damping and the viscosity of the fluid in circulation and comprising two waveguides arranged in series, the first being solid and consisting of a guide to conical dispersive tip for lowering the radiation impedance and optimizing the transfer to the second liquid guide, coupled to the first via a connection chamber and opening onto an ejection nozzle of the liquid coupling in the form of a collimated jet directional high pressure. The liquid jet is protected from drafts by a skirt, for example in the form of a cylindrical multi-strand brush. The jet may be intermittent to strike the surface of the sample regularly and generate low frequency shear waves. The latter point is of interest in pulse elastography and is an alternative to percussion methods vibrating pot or other electromagnetic or static contact usually juxtaposing the ultrasound transducer. In the case of deformation of the skin during a medical examination, the jet in intermittent mode must allow a significant deflection of the skin, typically of the order of 1 mm. The viscosity and the surface tension of the projected fluid make it possible to create a waveguide having a spatial continuity and for which a laminar flow can be imposed if the Reynolds number associated with the diameter of the jet, the density of the liquid, its speed ejection and its viscosity, remains below 2300. Thus, the fluid may advantageously be more viscous than water in order to maintain a laminar regime and therefore a collimated projection over a greater distance or to limit the spread of fluid on the inspection surface, for example for the sake of cleanliness. By way of example, a jet of water with a diameter of 0.8 mm, with a viscosity of 1 mPa.s and a density of 1000 kg / m ³ , expelled at a speed of 1 m / s, is associated with a Reynolds number. value 800. The turbulent regime is then reached for an expulsion speed of about 3 m / s.

DESCRIPTION OF THE FIGURES

Other aspects, objects and advantages of the present invention will appear better on reading the following detailed description of preferred embodiments thereof, given by way of nonlimiting example and with reference to the appended drawings, in which: :

FIG. 1 is a schematic view of a transduction assembly with a waveguide 150 forming an elementary probe for a longitudinal or transverse propagation mode, coupled to a material to be analyzed 100,

FIG. 2 is a schematic view of the complete device 400 according to one embodiment allowing the characterization of the two basic modes of propagation simultaneously, thanks to two separate probes 125 and 225 in contact with the material 100.

FIG. 3 is a schematic view defining the incident waves 600, reflected 620 and transmitted 610 at the interface with the sample 100.

• Figure 4 defines the incident waves 600, reflected (620, 640 and 650) and transmitted 610 and 630 in the case where the sample is composed of two layers 100 and 101.

• Figure 5 shows the interface echo in the case of a plexiglass guide with longitudinal wave, bathed in the air.

• Figure 6 shows the interface echo in the case of a transverse wave plexiglass guide, bathed in the air.

• Figure 7 shows an example of peak-to-peak amplitude variation of the longitudinal interface echo, an attenuation of about 50% between ambient air and a sample of human skin on a subject of study, highlighting the effect of a strong variation of load longitudinal impedance and making it possible to determine the longitudinal impedance of this sample (approximately 1 MRayl) which is approximately 32% Plexiglas (Z _L = 3, 1 MRayl).

Figure 8 shows an example of a peak-to-peak amplitude variation of the interface transverse echo, or 9%, between the ambient air (high value) and the same human skin sample (low value) highlighting a transverse load impedance (0.06 MRayl) much lower than that of Plexiglas (Z _T = 1.26 MRayl).

FIGS. 9A and 9B schematically illustrate two variants of the device according to an embodiment associating transducers 260, 270, 280 and 290 for the two fundamental modes of propagation on a single waveguide 150 according to, respectively, a concentric configuration.

FIG. 10 is a block diagram of the mounting of the transducer 540 according to one embodiment with the transmission and reception electronics using a limiter 510 avoiding to saturate the amplification elements 520.

FIG. 11 is a block diagram of the mounting of the transducer 540 according to one embodiment with the transmission and reception electronics using a fast switching 515 between the transmitted signal and the measured echoes making it possible to avoid saturation of the electronics amplification 520.

Fig. 12 illustrates a polished and truncated end dispersive flexural waveguide and its operation.

Figure 13 illustrates a modeling of the dispersive propagation of a bending wave in a cone.

Figure 14 illustrates a calculation of the refraction angle of the longitudinal waves in the sample as a function of the phase velocity in the tip.

FIG. 15 illustrates a bending wave bipod 89 integrated in a supporting structure. Figure 16 shows probe variants for impedance measurement depending on the desired application.

Fig. 17 illustrates a pentapode in ultrasound mode according to one embodiment. FIG. 18 illustrates an application of an embodiment of the bipod impedance meter in a velocimetric probe 89. • Figure 19 illustrates the use of a pulse transducer Doppler velocimetry pulsed according to one embodiment.

• Figure 20 illustrates the use of a directional velocimeter pentapode according to another embodiment.

Figure 21 is a sectional view of a probe according to one embodiment.

Fig. 22 is a sectional view of a conical tip probe according to one embodiment.

Fig. 23 shows a waveguide probe according to one embodiment.

FIG. 24 is a schematic sectional view of an ultrasonic probe 99 for non-destructive testing according to another embodiment.

• Figure 25 illustrates a dispersive tip whose end is bent hook-shaped according to one embodiment.

Figure 26 illustrates a pressure sensor according to one embodiment.

A. VOLUME WAVE PROBES

FIG. 1 represents a first device using volume waveguides making it possible to determine the characteristic impedances of the skin on a panel of individuals, as well as to identify the characteristics of different materials. The device allows by simple contact to evaluate the acoustic impedance parameter of a homogeneous material, by a non-destructive method and for the two fundamental modes of propagation, transverse and longitudinal. By contacting two cylindrical probes for each fundamental mode with the sample, the two longitudinal and transverse characteristic impedances are determined.

The complete system proposed here consists of a structure on which are fixed the two probes forming waveguides for sending and receiving acoustic pulses on the surface or in the material. The maintenance of the probes on the structure is done in such a way that there is an acoustic insulation, by an impedance break, between the structure and the waveguides. Insulation is necessary to prevent the interface echo from being disturbed by an acoustic short circuit via the waveguide holding structure.

A transducer is attached to each waveguide to transmit / receive the acoustic pulses. This transduction system, which exploits elements that can be piezoelectric, can generate pulses under the best conditions. In the In the case of a piezoelectric active element, an impedance matching with the waveguide allows a better insertion of the modes at the desired frequency by the addition of a so-called "quarter-wave plate" whose acoustic impedance is preferably intermediate between that of the active element z _a and that of the waveguide Z _g : jZ _a .Z _g . Finally, a "backing" wave damping system arranged on the rear face or on the side of the piezoelectric element allows a better temporal resolution of the pulse by limiting the intrinsic resonances of the piezoelectric element.

The system is simultaneously connected to a high voltage signal generator for transmission and an amplifier for reception. The receiving electronics tolerate the high excitation voltage of the transducer. Typically, a 20 volt pulse is applied to the transducer, and then the interface echo is magnified by a factor of 200 to produce an output signal of a few hundred millivolts. The constraint is that the amplitude excitation signal 20 Volts is also amplified by a factor of 200, which saturates the reception electronics, which can not deliver a signal of amplitude greater than +/- 10 V. For to overcome this case, we will use a clipper to limit the input voltage of the amplifier to a maximum allowable not causing too much latencies in the measurement or a fast switching between transmission and reception on the same transducer so that according to the desired action on the active element, the system operates sometimes in transmission and sometimes in reception. The output of the amplifier allows the measurement of the echo (s) of the acoustic pulse, coming from the interface and having passed through the waveguide.

Figure 1 schematically illustrates a transducer assembly 300 consisting of a non-dispersive planar waveguide measurement probe. This device comprises a waveguide 150 also serving as a probe, which is kept in contact with a sample 100. The waveguide 150 is made of a material whose known acoustic impedance Z _g is close to and greater than that of of the material that is to be determined, preferably less than 2 times the impedance of the material. Typically it may consist of low density metals such as aluminum or titanium, or more preferably plastics such as polycarbonate or plexiglass. On the waveguide 150, in contrast to the material to be analyzed, there is provided a piezoelectric component 250 having a sufficient bandwidth, for example 100% around the central frequency, which is equivalent to saying that its response impulse is short. The component works both as a transmitter and a receiver. An impedance matching blade 200 is fixed on the active face of the component 250, between this component and the guide 150, ideally sized to have an impedance Z _L = and a

dL thickness corresponding to a quarter of the wavelength of interest, typically of the order of a millimeter, and a damping block 350 on the opposite side (referenced by the term "backing" in English) so as to improve the resolution temporal by limiting the wave reflections on this face of the transducer, which limits the impulse response of the active element 250 to a bipolar pulse. It is recalled here that the working frequency will preferably be between 300 kHz and 900 kHz for elastic materials and between 20 kHz and 4 MHz for viscoelastic media. By way of example, for plexiglass, the longitudinal speed is of the order of VL = 2697 m / s and the transverse velocity is of the order of VT = 1335 m / s, which leads to a length of wave of interest between 1.1 mm and 2.2 mm at 300 kHz, between 0.37 mm and 0.75 mm at 900 kHz, between 0.08 mm and 0.16 mm at 4 MHz, etc.

The transducer assembly 300 may be a ready-to-use commercial component, for example the Paneameter transducers (Waltham, Massachussetts, USA) Ml 10 longitudinal wave and the transducer Panametrics VI 53 (Waltham, Massachussetts, USA) to transverse waves. These transducers have a respective center frequency of 5 MHz and 1 MHz. To optimize the acoustic transmission between the probe 150 and the sample 100, a viscous paste can be used as couplant.

A coupling adapted to the propagation mode of the wave is essential in the case of the transverse mode. Coupling referenced SWC (shear wave coupling in English) Panametrics company allows the transmission of the transverse mode. Its thickness is reduced as much as possible. Its use is not necessary with dispersive transducers spikes or when the nature of the material studied allows a satisfactory coupling, such as with fluids, gels, elastomers.

FIG. 2 diagrammatically illustrates a complete device according to an embodiment 400 with two non-dispersive probes such as that illustrated in FIG. 1. A first longitudinal wave probe 125 has its non-dispersive plane waveguide having a length 1 10, the other transverse wave probe 225 has its non-dispersive planar waveguide of length 120, distinct from the length 110. The guides are integrated in a support, typically made of sintered plastic, providing good acoustic insulation between the waveguides. wave. The probes are in contact with the sample 100 having a lateral extent and a depth of at least five wavelengths and preferably at least ten wavelengths or longer. Means for regulating the bearing force may be envisaged such as springs at the rear of the probes. Fasteners to isolate "acoustically" probes from the structure. They can be foam, having a very low acoustic impedance compared to waveguides. The lengths of the guides 110 and 120 are dimensioned so that the pulses do not interfere with each other. For this, it suffices that the acoustic pulses traveling along the length 110 at the longitudinal speed and the length 120 at the transversal speed reach the sample at the same time. In addition, the lengths are calculated so that the incident wave and the reflected wave are well separated temporally. Concerning the last condition, the lengths 110 and 120 must have a minimum satisfactory value

2 /

the equation τ = - <-, in which T represents the duration of the pulse, which is the ratio

/ ^c

the number _n of periods desired by the central frequency / pulse. This period must be less than one round trip of the wave, ie twice the distance /, reported on the speed of propagation of the wave _c generated by the transducer. We will therefore have the following dimensional condition on the length 1 of each probe:

2 - f With regard to the simultaneous arrival of the two waves on the material, it is necessary to check the equality

- = -, in which the lengths of the probes 110 and 120 (respectively lno and 1 ₁₂₀ ) c _L c _T

depend on the longitudinal velocities _{c L} and transverse c _r . It is also possible to include a delay of excitation of the fastest wave equal to the difference of the transit times to reach the interface with respect to the slow wave, which makes it possible to preserve equal lengths of guides 110 and 120. Considering the fast propagation speed in the guide 110 and the slow propagation speed in the guide 120, we obtain the equation of the arrival time delay of the wave fronts on the sample in both probes: delay = ---.

c _T c _L

Figure 3 shows the wave evolution at the time of measurement. An incident wave 600 propagates in the guide 150 towards the sample 100. At the interface, a part of the energy is reflected, which leads to a reflected wave 620 and another part 610 is transmitted in the sample. The precise knowledge of the material constituting the impedance guide 150 Z intervenes in the reflection coefficients r and transmission _t with the impedance z _e of the sample 100 according to the following formulas: AZ - ZA 2-Z

r ₌ L ₌ ^ - __. _e tt = ^ = ^e - (i)

4 _{e +} z _g 4

In which, ^ denotes the amplitude of the reflected wave 620 and A _t , the amplitude of the incident wave 600, while A _t denotes the amplitude of the transmitted wave 610.

The relations (1) constitute a system of two equations with 2 unknowns connecting the ratio of the amplitudes of the reflected and transmitted waves to the characteristic impedances of the materials. Its resolution provides the impedance of the sample z _e .

Furthermore, the amplitude of the incident wave A _t being capable of changing according to the temperature conditions and since it is not desired to use additional equipment for calibrating the probe, a measurement relating to a reference makes it possible to deduce the impedance z _e by avoiding the variations of attenuation properties, temperature and size of the waveguides.

The measurement used as a reference is the empty measurement, that is to say without contact. Indeed, for known propagation conditions in the probe under given temperature and humidity conditions, the air impedance z _air is known and very low. In this case, the probe provides a reflected wave of given amplitude A _rl . The guide 150 is then placed in contact with the sample 100 and a new measurement is made. An amplitude A _r2 corresponding to the reflected wave 620 is obtained on the unknown impedance material z _e . We obtain the two equations:

This system of equation leads to the relationship

\ + A ₂ Z _mr Z

with a = - (2)

1 - oc A _rl Z "," + Z Thus knowing the impedance Z _g of the guide, the _air impedance z _air and from the two successive measurements of amplitude of the received signal, one determines the characteristic contact impedance of the material to be analyzed z _e .

For this method to work, it is necessary that the material is at least locally homogeneous and isotropic. The use of a system such as that of Figure 2 makes it possible to perform the acquisition of the two fundamental modes simultaneously. The use of these probes associated with a mechanical scanning on the surface, of the C-scan type, makes it possible to map the impedance and determine its homogeneity. The isotropy of a material can be deduced if the sample has a suitable geometry, for example cubic and has 3 orthogonal faces and for each of them, 3 measurements of impedances are realized: impedance of the longitudinal mode and transversal mode in two directions, typically orthogonal.

If we know the impedance of the air z _atr , the impedance of the waveguide Z the amplitude of the first empty echo, and the contact echo 620 we deduce the impedance of the sample z _e .

On the other hand, one can establish the relation describing the progressing transmitted wave art in the sample 610:, which leads

expression of the wave portion transmitted on the reflected wave

In the case of a stratified sample (see Figure 4), which there is a second layer 101 of _an impedance, one can establish the relationship of the reflected wave on the second interface of the wave retransmitted in relation the first layer towards the waveguide

We deduce z _e ,, the impedance of the second layer of material, from the established relationships

As ₀ _ As ₀ Λ40 A ₁₀ _ ^{2 - Z} _e Z <-Z. ² - ^Z e { ^Z e + ^Z _S ) 4 4 · ¾ Z, - Z.

A20 A40 AlO A20 ^Z _R + ^Z e ^Z e! + ^Z e ^Z e + ^Z _R ^Z e + ^Z _R ^Z e + ^Z e At _7n 4 · Ζ ² Ζ, + Ζ

By posing β -, there comes the relation:

50 Z e Ί "Ζ g Z e Z e

FIG. 4 defines the incident waves 600, reflected (620, 640 and 650) and transmitted 610 and 630 in the case where the sample is composed of two layers 100 and 101 stacked, the layer 10 being in contact with the guide of In this case, in addition to the waves previously mentioned in relation with FIG. 3, a wave 630 is transmitted to the interface between the layers 100 and 101 in the layer 101, and a wave 640 is reflected at this interface in FIG. the layer 100. A second reflected wave 650, resulting from the wave 640, is thus reflected in the waveguide 150.

Figure 5 shows an example of a curve recorded for a longitudinal propagation mode in a plexiglass guide in the open air (voltage measured as a function of time). The signal near the instant t = 15 μβ, corresponds to the guide-air interface echo. The incident wave 600 passes through the probe to the interface where a portion corresponding to the characteristic impedance ratio is reflected under the identity 620 to the transducer. The wave is reflected again in the guide and makes a second round trip, visible with a strong attenuation at time t = 30 μβ.

FIG. 6 shows an echo acquisition curve 620 for the transverse propagation mode, always with a Plexiglas guide in the air (voltage measured as a function of time). The amplitude peak at t = 31 corresponding to the guide / air interface echo of the transverse mode. It is temporally more spread because of a lower working frequency, 1 MHz, and comes late compared to the previous assembly because of a lower propagation speed than the longitudinal mode. The curve also shows the interference by the longitudinal mode with the arrival of a portion of the energy of the longitudinal wave at t = 15 and a longitudinal / shear conversion at the guide / air interface arriving at t = 23 μβ .

Figures 7 and 8 show the variations of the respectively longitudinal (48%) and transverse (9%) reflection coefficient, with respect to air, when the sample (100) is an exposed part of human skin (amplitude in mV referenced by sample number). Slight amplitude variations are observed during successive free / contact measurements. They can be caused by temperature variations during contact or variations in couplant thickness due to crushing or thermal equilibrium if temperatures are not not exactly identical. It is for these reasons that the coupler will preferably have a very small thickness, typically less than 50 μπι, and the temperature equals between the probe and the test material in order to avoid any phenomenon of thermal diffusion or disturbance. the transverse load impedance by the impedance of the couplant. In addition, neglecting the thermal expansion in the case of small temperature variations, ie constant density and known dimensions of the guide, the arrival of the first echoes is the round trip in the guide of known length. Therefore, the precise knowledge of the propagation speed _c for a known density p guide, allows to refine its impedance Z _g = pc. The impedance of the material to be analyzed will therefore be characterized for the ambient temperature. We thus find longitudinal impedances of 1 MRayl and transverse impedances of 0.06 MRayl for this measurement on human skin. Variations of this method are possible, such as that of establishing a reference table of the reference echo in the open air as a function of the temperature. The measurements can also be made in a climatic chamber in order to impose the temperature and relative humidity, homogeneous between the probe and the material to be analyzed.

FIGS. 9A and 9B show two variants of the arrangement making it possible to use the same waveguide 150 and therefore the same probe to generate the two fundamental modes of propagation.

FIG. 9A shows a first concentric mounting of the transducers 260 and 270, disposed on impedance adapter blades 210 and 220, themselves bonded to the waveguide 150. A backing 350 is disposed on the rear face of the transducers 260 and 270 to improve the temporal resolution of the pulse. The impedance measurements according to each mode will be carried out successively so as not to interfere with each other. Given the high propagation speeds compared to a manual movement by a user, both acquisitions can be considered simultaneous. The transducer 260 will preferably be a longitudinal wave transducer and the horizontal transverse wave transducer 270.

Figure 9B illustrates a serial assembly of the transducers for each fundamental mode. The backing 350 is fixed to the back of a first transducer 280 resting on an impedance matching blade 230 which rests on a second transducer 290 and on a second impedance matching blade 240 and finally on the guide of The two impedances of the transducers being equivalent, the impedance matching blade 230 can be neglected. For a shear mode transducer 290, the adapter layer 240 will be optimal for transmitting a transverse wave in the waveguide 150. The transducer 280 will be in extension mode to emit a longitudinal wave in the guide of the waveguide. wave through the transducer 290 and the adapter layer 240. In the case of the longitudinal wave, the transfer will not be optimal, but the assembly allows to have a probe to emit the two fundamental modes in the sample 100. The impedance matching blade 240 is provided for the transducer 290 and a transverse mode

Ύ ²⁴⁰ / ^ 280 ^ 150

, but whose longitudinal mode impedance will be similar to that desired ^¾ LL because the transducers 280 and 290 have identical properties and if the ratio of longitudinal to transverse impedances of the adapter layer 240 is equivalent to that

7240 7150

^≡ of the wave-guide · The size of the thickness of the adapter plate is adjusted to match a thickness of a quarter of the wavelength of the transverse mode ²⁴⁰ = λγ /. In solids, the propagation velocity of the longitudinal mode is greater and at least one times the propagation velocity of the transverse mode, which implies a smaller wavelength. By using a higher excitation frequency for the longitudinal wave emitting transducer 280, a frequency can be chosen such that λ ⁸⁰ = λ ⁹⁰ and thus the impedance matching blade 240 will allow correct transmission of both transverse waves as longitudinal waves of the transducer 280.

Figure 10 shows a block diagram of the acquisition chain. This device is implemented for each transducer, each propagating one of the two basic modes. A signal generator 500 sends an electrical pulse into a transducer 540. This signal is also present at the input of a limiter 510 whose role is to limit the maximum voltage in order to protect the amplification electronics which carries out a measurement under high impedance. The clipper 510 passes it to an amplifier 520 providing a signal in the time domain, the extraction of the useful parameters being obtained by means of a measuring circuit 530 (for example an oscilloscope).

Figure 11 is an alternative configuration of the measurement chain. In this configuration, a switch 515 is used to firstly connect the transducer 540 to the pulse generator 500 and in a second step to the amplification circuit 520. In this way, the high amplitude excitation signal does not disturb the electronics amplifier and the low amplitude reception signal. As before, a measurement circuit makes it possible to extract the parameters that are useful for evaluating the characteristic impedance (for example an oscilloscope).

The method thus makes it possible to measure with a good resolution, in a short time, the characteristic impedances of any material, by simple contact on its flat surface and without degrading it.

B. DISPERSIVE GUIDED WAVE PROBES

The waveguide method is a fairly simple and intuitive approach to measurement of longitudinal and transverse impedance. But, it is limited to locally homogeneous and isotropic surfaces with a sufficient measurement volume.

However, there are in practice many situations where the impedance to be characterized is poorly suited to the use of plane waves and can also be very low compared to the impedance of the waveguide. This is already the case with the human skin of Figure 8 which represents only 4.7% of the transverse impedance of Plexiglas. As can be seen, the acoustic load drops the signal around 140 mVcc, with an uncertainty of 2 mV. The measurement impedance is thus between 0.06 MRayl and 0.07 MRayl. The effect of the couplant is difficult to evaluate. With a plexiglass guide of 12 mm diameter, the same process applied to the pulp of the index on a panel of 10 people provides an average longitudinal impedance of 767 kRayl with a dispersion of between 480 and 1100 kRayl and a mean transverse impedance of 100 kRayl with a dispersion of between 52 and 220 kRayl.

The waveguide waveguides may be used for the impedance measurement in particular so as to reduce the impedance of the waveguide to bring it closer to that of the medium to be analyzed. The guides may be made of metals or plastics, for example aluminum, titanium, Plexiglass (Poly Methyl Methacrylate) or Polycarbonate. By introducing a dispersive effect, the probe has a reduced impedance, smaller than its characteristic value by a factor of 4 to 10. The impedance of the guide is then determined by the phase velocity of the wave in the vicinity of the interface.

To understand the phenomena involved, we can model the decay of the phase velocity of a bending wave in a conical tip. Figure 12- (a) shows the truncated and polished machining of the tip 10 near its end 11 while Figure 12- (b) gives the field radiated by the far field tip. In far field, the vibration at the end of the cone is a dipolar source. The conical profile and the polishing of the end reduce the mechanical impedance of radiation by a factor of 4 to 10 and bring it closer to the longitudinal impedance of the finger pulp, which allows the transfer of most of the the energy in the medium and therefore subsequently to have a better signal-to-noise ratio for the fine analysis of the impedance variations of the medium. Figure 12- (c) schematically shows a tip in sectional view with a damping material filling a side tank 13. Figure 12- (d) schematically shows two methods of electrical connection for generating waveforms. bending, one operating a uniform polarization PZT disk 21 and series connection of the two piezoelectric capacitors formed by the two upper electrodes in half-disk and the common lower electrode, the other operating an alternating polarization 20 per half-disk and an electrical excitation E in parallel of the two associated piezoelectric capacitors.

Near the base of the cone, the velocity of the wave is close to the velocity VT, transverse waves. Then, during its propagation, the speed decreases in first approximation like the speed of the first AoS asymmetric Lamb mode in an isotropic thin plate. The dispersion equation of this mode is used to describe the phase velocity of a dispersive wave A in a conical tip: if we call k the wave number, the working frequency, the radius of the section, VT the velocity of the transverse waves, VL the velocity of the longitudinal waves, and σ the Poisson's ratio, and if kh "l, it comes:

Example: Let's take a conical tip of half-angle at the top 2.4 ° and whose end is polished so as to further reduce its thickness to 0, 1 mm, then with f = 600 kHz, VT = 3100 m / s, VL = 6420 m / s, σ = 0.345, we obtain, V _p = 5355 m / s, Vd = 763 m / s and λ = 1.27 mm. This example and in particular formula (6) is illustrated in FIG. 13 (b).

Having specified the law of variation of the phase velocity near the end of a conical point, we must now specify its behavior as an acoustic antenna.

1. Field radiated by a conical waveguide Let u be a solution of the sinusoidal propagation equation in the medium in which the tip is immersed:

At + k ² u = 0 (8) where k is fixed and denotes the wave vector: k = co / V _a (V _a = air velocity if the medium is for example a fluid and in the occurrence of air) Let G be the Green solution function of the propagation equation in the case of a point source localized in:

AG (r ₀ - r ') + k ² G (r ₀ - r') = -δ (r ₀ - r ') (9)

In a problem with spherical symmetry, and in the case where the source point is coincident with the origin of the coordinates, the solution of the system of equations (8) and (9) is:

1 c ^ikr

( _{r) =} J_fl (10)

4π r

a mathematical treatment of the system of equation (8) and (9) leads to the Kirchhoff-Sommerfeld formulation of the Huyghens-Fresnel principle stating that the diffracted field at r ₀ = (x ₀ , y ₀ ) is obtained by summation of Spherical waves G (r ₀ - r ') centered in r' = (x ', 0): u (x ₀ , y ₀ ) =' u (x, 0) ^ l ^ dx (H)

x = '0 or r For simplicity, the mechanical displacement at the base of the cone is chosen unitary u (/, 0) = 1. The progressive nature of the wave is taken into account and therefore the sources are not synchronous by introducing a phase shift e ^Vj associated with the source of position (, 0).

We also take into account the reduction of the section of the cone and therefore of the increase of the amplitude of the mechanical vibration by conservation of the momentum by writing that the vibration undergoes a mechanical gain g (x, 0): Where u (x, 0) is the vibration at the end of the cone, u _ref the vibration at position l ^, a factor describing the progression of gain near the end; / is the length of the cone.

If in addition, we transform the three-dimensional problem into a two-dimensional problem (for example by taking a slice of the middle, the points being placed on the edge of the slice). Equations (8) to (11) then lead to the following diffraction integral:

with _r = J (x - x ₀ ) ² +

In pulsed excitation, the integral (12) must be further transformed. It is first written that the time signal u (t) has a frequency spectrum U (ω) = Ε (ω) Ηβ (ω) given by its Fourier transform. Ε [ω) is the Fourier transform of the electrical excitation and He (co), the Fourier transform of the electroacoustic impulse response of the piezoelectric element (assumed wider band than Ε [ω)). Conversely, when we know U (a>) we can go back to the temporal response of the signal. We then have:

Taking into account the diffraction, we obtain the general spatio-temporal shape of the radiated field:

where Re designates the real part of the double integral.

In burst mode, the spectrum U (co) is very selective around the excitation frequency. We

'approximates by a Dirac distribution centered at the excitation frequency ω,

Taking into account the wave reflected at the end of the emitting cone

An important phenomenon in the impedance measurement method is taking into account the reflection of the incident wave at the end of the guide. When the tip is free and bathes in the ambient air the mechanism is simple because the reflection takes place without phase change and is worth practically +1.

The field refracted in the medium taking into account the wave reflected in the cone and the phase velocity v in the duralumin emitter cone is:

The phase velocity of the formula (16) is modified with respect to the formula (6) to take account of the fact that the cone is truncated and that the phase velocity is not zero when one is at the end of the cone (x = 0). Moreover, a cone can be locally flattened at its end by polishing. Its opening angle is then locally modified in the polishing zone. This treatment is only interesting for radiating energy in gaseous media characterized by a low speed of wave propagation. It is then necessary to polish the point so that the phase velocity approaches that of the gas.

When one is interested in the dermis 64 or the water 66, it may happen that the phase velocity in the tip is lower than that of the refracted waves. The bending wave A can then no longer radiate in the medium and it is necessary to drive the tip until its phase velocity again exceeds that of the medium.

Figure 13 illustrates a modeling of the dispersive propagation of a bending wave in a duralumin cone at 600 kHz. The modeling assumes the emission of a reflection wave packet at the end 11 of the cone and interference between the incident wave packet and the reflected wave packet.

More particularly, this figure illustrates: Fig 13 (a): experimental phase velocity deduced from the distances between maxima = half-wavelength;

- Fig 13 (b): theoretical phase speed of Lamb A0 mode in a Duralumin plate whose section decreases linearly at the same angle Θ as the cone. FIG. 13 (c): exemplary embodiment of a broadband flexural wave transducer incorporating a conical waveguide 10 and constructed from a PZT disk with alternating polarization 20. The impulse response of the disk is damped via an annular reservoir 13 filled with saturated tungsten powder saturated polymer 30.

FIG. 13 (d): Time signals of the vibration velocities Vibl and Vib2 measured perpendicular to the flat end of the tip 11 and 4.5 mm upstream of the end, for a 20 Vcc electrical pulse consisting of a burst of 5 square periods at 700 kHz.

FIG. 13 (e): Peak-to-peak amplitudes of the vibration velocities by scanning using a laser vibrometer along the tip, perpendicular to the flat end under the same conditions as above.

- Fig 13 (f): Modeling of the radiation in the air, very close to the generator (at 0, 1 mm), without reflected wave (dashed line) and with reflected wave (solid line).

As shown in graph (b) of FIG. 13, the phase velocity in the tip is 1500 m / s for a section diameter of 0.4 mm which is reached at 4.8 mm distance from the end of a perfect conical tip (angle at the top 4.8 °). It is therefore necessary to push a point of several millimeters in the water to totally cancel the echo inside the tip.

Taking into account the reflected wave is illustrated on the graph (f) of Figure 13. On the same graph, the absence of reflected wave gives the dashed curve. In addition, in the presence of reflected waves, there appears an interference phenomenon well made by the experimental measurement (graph (e) of Figure 13). According to formula (17), the distance between the minima is given by the condition: It corresponds to a local half-wavelength in the

cone.

Figure 14 illustrates a calculation of the refraction angle of the longitudinal waves in the sample as a function of the phase velocity in the tip. The medium is here directly the air 67 that is to say the weakest impedance. The angle of refraction is then closest to the perpendicular to the generator Ng of the cone, it is 14 ° with respect to the generatrix of the cone for a duralumin cone of half-angle at the top 2.4 ° having a speed of dispersion theoretical according to graph (b) of Figure 13 which gives a phase velocity close to 800 m / s at its end 11.

In conclusion, the angle of refraction is all the greater (ie directed towards the axis of the point) that the impedance of the medium will be higher or that the section of the point is smaller (see Fig. 14 -(at)). Thus, for water, the emission angle can reach 90 ° when the phase velocity is less than or equal to the speed of wave propagation in water, while it is about 14 °, when a duralumin point has an angle at the top of 4.8 °, that the excitation frequency is 600 kHz, that its diameter at its truncated end is 0.4 mm and that it is immersed in the air. This is shown in Figure 14- (b): the progressive wave in the cone loses its energy due to radiation.

FIG. 15 illustrates a flexural wave bipod 89, integrated in a sintered plastic support structure providing, on the one hand, an acoustic insulation between the guides, on the other hand, making it possible to manually enter the assembly without disturbing the propagation. in the guides, finally to simply manipulate the assembly in the manner of a large pen that is brought into contact with the sample, here water 66. The metal waveguides are fixed (43, 44 ) on the very damping carrier structure for bending waves. The structure is made by rapid prototyping sintered polymer type PolyAmide PA12. The conical tips have a T-shaped geometry (10, 12) and are held around their periphery by force insertion into the support structure providing a double hold on the periphery of the tip at the level of the base of the cone and typically at 20 mm. upstream of the end 11, so as not to dampen the vibration by holding the tips too close to their end. The acoustic path in the carrier structure requires a transit time greater than that of passing through the conical guides and the sample. The distance e between the ends of the tips is adjusted according to the material to be probed. It is typically 5 mm for a pentapode in mixed use (ultrasound or velocimeter) and for an elastic material having a transverse velocity of between 500 m / s and 3000 m / s, but much less, about 0.5 mm for materials. thin or diffusing viscoelastic, small localized variations of impedance to be detected. The distance can be increased to 10 mm, if one tries to analyze relatively elastic materials, homogeneous on the surface, but having variations of impedance in depth or also if one wishes to have a good resolution in velocimeter mode . The phase velocity of the bending waves near their end is close to 800 m / s for aluminum, which allows a very good insertion of the waves into the dermis and into the silicones and a radiation towards the front of a longitudinal beam, close 90 ° (ie almost parallel to the main axis of the tip). The spacing and inclination of the tips is therefore optimal for a certain sample thickness as shown in Figures 15 (a) and (c). The tips open out from the supporting structure, projecting only a very small amount, approximately 0.5 mm, so that the supporting structure also serves as support and avoids damaging or injuring the probed medium, particularly if is the dermis. The probe can then be dragged or moved point by point (pixel by pixel) on the surface of the test material. In this configuration, the latter must be dry and clean so as not to pollute the tips, in particular fill the gaps between the tips and the carrier structure which would lead to measurement errors. The images thus constituted are C-Scans, that is to say that it is possible to look deeper and deeper in the study material according to whether one looks at the head or more inside the wave packet. The tips have a typical length of 85 mm, for a base diameter of 6.6 mm (half-angle at the top of 2.4 °) which corresponds to a working frequency of 600 kHz. The images can be amplitude or phase images (in particular, a cartography is taken by taking a particular zero crossing of the wave packet and we look at how the zero crossing changes according to the medium sampled during a displacement. lateral bipod). There is therefore interest in bringing the points closer to each other when doing a phase mapping.

One measurement technique is to use a second tip disposed next to the first at the distance e as illustrated in Figure 15 (c). The sample is characterized by its thickness x and the wave refracted by the tip is reflected on the back side of the sample. The thicker the sample, the longer the tips need to be spaced to find the maximum refracted signal as shown in Fig. 15 (a). FIG. 15- (b) shows the phase velocity in the vicinity of the end of an unpolished, duralumin conical tip for an apex angle of 4.8 ° and a truncated end 14 when the diameter is 0, 4 mm. When performing amplitude mapping, it may be advantageous to use a tripod or probe with three conical points, comprising a transmitting tip in the center and two receiving points aligned with the transmitting tip, at equal distances from it and all oriented so that their axes of vibration are collinear. This configuration is called longitudinal or Rayleigh because the emitting tip pushes the material towards the receiving points. If the medium is not homogeneous, variations in homogeneity can be measured via the two receiving tips which normally provide opposite signals in phase. It is then sufficient to add these signals to obtain a minimum reception signal whose amplitude will vary substantially with the local variations of impedance of the medium. echo inside the emitter tip provides information on the average load impedance, while differential measurement provides information on longitudinal impedance fluctuations in the material (because the wave must propagate parallel to the surface of the proof material ).

To measure a shear rate, simply turn the tips 90 °. Horizontal transverse vibrations are then detected (Fig. 16 (e)). These speeds are particularly low in diffusing viscoelastic materials such as organic tissues and transit times between two points spaced 1 mm apart can reach 0.3 milliseconds. The tips are therefore as close together as possible, for example at 0.5 mm from each other, and the working frequency is lowered to the minimum of the transducer bandwidth, a compromise can be found between 20 kHz and 600 kHz.

Different configurations of spike transducers are therefore possible depending on whether it is desired to measure a longitudinal or shear impedance or to carry out a surface or in-depth shear impedance mapping.

FIG. 16 shows probe variants for impedance measurement depending on the desired application: a) Penetration into a soft, molten sample 63 of the tip having a very fine end 16 until the reduction of Ar echo of 75% and possibly for the plates, refraction and adjustment of the position of the truncated receiving tip 14 until detection of the maximum signal. Then, deduction of the angle of maximum refraction and / or depth of penetration into the middle of the longitudinal impedance of the medium.

Radiation in the medium by depressing a transmitting tip 16, until cancellation or attenuation of the reflection coefficient A _r inside the emitting tip to a set value, for example A _r = 25%. The tips are provided with PZT disk transistors 20 alternating polarization per half-disk, denoted (P, -P), glued on the base. The faces of a PZT disk are covered with a weldable paste. The base of the cone of typical height 80 mm and angle at the top 4.8 ° is surrounded by a tank 13 machined in the cone, with thin walls, typically 0.2 to 0.5 mm, filled with a damper 30, for example sintered metal powder. In the case where it would be necessary to work at high temperature, the powder is for example a refractory metal, for example tungsten, and the cone Inconel. For uses at ambient temperature and in contact with food or the dermis, the cone is made of aluminum alloy or stainless steel or platinum and the powder is aggregated by a saturated polymer binder to a density of at least 5. The filling height of the reservoir 13 is typically 2 mm. PZT ceramics are typically PZ29 or PZ27 (references from the company Ferroperm located in Denmark). At high operating temperature PZ46 will be chosen for operation up to 550 ° C. The reservoir can also be traversed on its inner wall by a winding consisting of a conduit in which circulates a coolant 70 for example water having the effect of maintaining the ceramic PZT at a temperature below its Curie temperature. The end 16 of an Inconel tip can then be brought to a temperature of around 1000 ° C in contact with a cast iron plate during cooling, while the base of the cones is maintained at a temperature below 550 ° C. The receiving tip is remote from the transmitting tip, until reaching a spacing e (max) corresponding to a signal of maximum amplitude, refracted and reflected on the rear face of the semi-solid multiphase medium. The amplitude curve of the signal received as a function of the inter-peak spacing corresponds to a certain associated velocity profile and therefore of crystalline phase in the thickness of the material. After complete cooling and posterior crystallographic analysis of the material in its thickness, abacuses are obtained making it possible to connect, for a given mixture, a phase or impedance profile to an inter-peak spacing-amplitude curve. Process based on the modification of the reflection conditions at the end 14 of a probe 88 with a truncated conical point, not immersed but in contact and loaded by the longitudinal impedance of the hard sample 60 and thus on the displacement of the nodes of interference as a function of load impedance Z2. It can implement a laser vibrometer generator scan to measure the position of the nodes or a measurement of the displacement of the zero crossing of an echo inside the transmitting tip.

At ambient temperature, the impedance of the sample can be determined from a given radiation impedance corresponding to a truncated tip 14 and the position of the interference nodes in the vicinity of the end. When the tip is free, the end vibrates according to a belly of vibration, whereas when the point is blocked, charged by a much higher impedance, there is a node of vibrations. Between these two extremes, the tip can be charged first by a reference impedance Zref and produce a first vibration node at a distance from the end. When the reference impedance is replaced by an unknown impedance Z ₂ , the node moves. This displacement can be physically observed using a laser vibrometer performing a scan of the generator of the cone. It can also be deduced from the phase shift of the electrical signal of the peak echo. A correspondence is then established between the impedance of the medium and an offset table of one or more minima or phase shift of one or more reference points of the echo electrical signal, for example of a shift to a zero crossing. c) Probe 87 with a thin end 15 to obtain a mapping based on echo analysis with constant support force (F) on a 3D hard surface 61.

When a sample has a very rugged curved profile, it can be associated with a lower radiation impedance that the surface is more convex (pointed). The use of a very tapered point 15 with low radiation impedance makes it possible to transfer the energy in the object more or less according to the local radius of curvature of the object (concave or convex). An amplitude mapping of the first echo carried out with constant force F provides a surface radiation impedance image of the material. In such a case, the impedance variations are low, control of the support force is critical. d) Longitudinal velocity measurement method VL2 or Rayleigh (VR) in a viscolelastic medium 61 and mapping based on longitudinal or Rayleigh radiation disturbance.

The previous assembly 87 can be used for deep analysis with a bipod. If the vibration axis of the emitting tip 10 is collinear with the sensitivity axis of the receiving tip (Fig. 16-d), the device uses Rayleigh or longitudinal waves in the tip plane, and now the two points at a constant distance, for example 1 mm, and a constant support force F, for example 1 Newton, a scan of the surface is carried out with a spatial sampling period of less than or equal to half a length in the middle, typically 1 to 2 mm. Impedance variations of the medium at different depths can then be visualized by C-Scan (visualization of a map showing the amplitude of the signal at different time slots in the wave packet). The tips are oriented so that the vibration axes according to their bending mode are collinear. e) Method of measuring transverse velocity of a sample 62 able to propagate them, from the generation and detection of horizontal transverse waves.

The previous assembly can be reproduced with one difference, the tips are rotated 90 ° with respect to their main axis, so that the direction of vibration / maximum sensitivity is perpendicular to the segment which connects the ends 14. The assembly is then fit to generate and detect horizontal transverse waves if the medium 62 allows it. The coupling between the tips gives access to a horizontal transverse propagation velocity in the medium. The spacing between the tips is constant as well as the support forces F. Such an assembly is also interesting for detecting contiguous cracks. This configuration can furthermore be used to characterize diffusing viscoelastic media such as living organic tissues. The measurement consists of contiguous points and move them in small steps, typically 10 μπι. We then measure the phase shift and damping constant of the shear wave that can be compared to the decay curve of a shear wave in an elastic solid. When starting from contiguous points, the speed of the shear waves can be calculated by phase shift measurement over a very short distance of the order of 100 μπι. In such a case, the approximation or calibrated distance points can be achieved simply by a clamping screw through the carrier structure shown in Figure 15 making a quarter turn.

Fig. 17 illustrates a pentapode in ultrasound mode according to one embodiment. Depending on the medium, the distance between the central tip and the peripheral tips varies from 0.5 mm to 5 mm. This configuration is particularly interesting in practice because, the excitation of the emitting points (Ex-, Ex +, Ey-, Ey +) are excited so as to cancel the signal at the center for the longitudinal 1 and transverse 2 configurations, or to double the signal in the center for the longitudinal 3 and transversal configurations 4. A pentapode can be closed and sealed by a thick silicone membrane of thickness 0.5 mm to 1 mm with total transfer of the waves from the tips to the membrane and then the membrane to the sample (not shown). The emitting tips immersed in the silicone membrane refract in the medium at a very penetrating angle. In this configuration, the pentapod is not likely to be contaminated and can be brought into contact with the skin, covered with a gel that provides sliding and a good acoustic coupling of the probe. The four transmitting probes all operate a piezoelectric disc with four quadrants 22; they can switch electronically from a shear configuration to a configuration longitudinal. A mixed configuration (a) can also be envisaged combining longitudinal 3 and shear 4 emissions simultaneously.

Figure 17 shows the configuration of a 5-point device combining a central tip operating in reception mode R and whose directivity is programmable by means of the four-quadrant PZT 22 disk equipping its base. Two other pairs of points are aligned with the central point and are arranged in quadrature so as to occupy the vertices of a square. These tips operate as transmitters and are also equipped with four quadrant PZT 22 discs able to vibrate the tips in an electronically programmable main direction. There are three particular configurations. In the configuration (e), the vibrations of the emitting points are tangent to the circle passing through the vertices of the square and all oriented in the direct (or indirect) direction. The signals (Ex-, Ex +) are 90 ° out of phase with the signals (Ey-, Ey +). This configuration is optimized for transmitting / receiving shear waves. The central tip exploits its four electrodes to form two signals RI and R2 corresponding to squares of sensitivity in quadrature. In the configuration (b), the vibrations of the transmitting tips are perpendicular to the circle and antisymmetric with respect to the center of the circle. The longitudinal or Rayleigh signals are emitted in the direction of the center and cancel in the center if they are emitted in opposition of phase 3. Otherwise they are added if they are in phase 1. The pairs are also excited in quadrature (one pair compared to the other). The central receiver remains in the same configuration with its signals RI and R2 that we sum. The configuration (a) is called mixed. Two emitters of one pair vibrate in directions tangent to the circle symmetrically 4 with respect to the central point and generate a shear wave, while the other two emitters of another pair vibrate in a direction perpendicular to the circle, so that antisymmetric 3 with respect to the central point. This configuration approaches in his mind the low frequency (50 to 200 Hz) bending wave impulse elastography propagating in a viscoelastic tissue, revealed by high frequency longitudinal wave waves (3 to 5 MHz) illuminating the medium in which propagates the shear wave. Nevertheless, in the case of the present invention, the two peaks responsible for the generation of the shear wave operate at an ultrasonic frequency between 20 kHz and 600 kHz. The configuration (a) thus makes it possible to access longitudinal and transverse speeds and impedances as well as to illuminate the sample in different ways. This process can be followed in time, whereas the tips are preheated, for example at a temperature of 50 ° C and that they are brought into contact with the medium. The diffusion of heat in the medium then changes the elasticity of the medium in a manner that differs according to the fundamental wave considered.

C. LOCAL VELOCIMETRIC PROBE. 1. Differential transit time

A bipod 89 may also operate as a local velocimetric probe as shown in FIG.

This FIG. 18 illustrates an application of the bipod impedance meter in an alternating measurement velocity probe 89: the two waveguides 10 function alternately as transmitters then as receivers. The transit time to raise the blood flow is greater than the transit time to lower the blood flow. The differential transit time is proportional to the speed of the blood. Contact measurement requires a very precise time measurement, with a resolution less than one-tenth of a nanosecond. The use of high frequency point probes (2 to 3 MHz) consisting of conical tips with a base diameter of 2 mm and a height of 25 mm in combination with time-digital components (or TDC: Time to Digital Converter) is here advantageous for performing local controls, especially on the venous network. In addition, the measurement rate must be high around 1000 measurements per second in order to be able to average the measurements and also to avoid that the dimensions change between an upstream measurement and a downstream measurement. It may be advantageous to operate a tripod, the three probes being aligned with the vein and the emitter probe disposed in the center. The vibration axes are collinear and aligned with the direction of the vein. The upstream-downstream differential measurement is thus simultaneous, which makes it possible to avoid errors related to dimensional variations generated by local variations in pressure and swelling of the vein.

Both transducers implement the principle of differential transit time measurement. They function alternately as transmitters and receivers. A longitudinal acoustic wave refracts in the dermis, enters a vein and descends or rises blood flow. It undergoes a positive or negative phase shift depending on the path taken, the amplitude of which will depend on the path traveled and the speed of the blood.

Application: Let Vf be the speed of blood flow and Ve = the speed of propagation of ultrasonic waves in the blood (which is taken as equal to that of water).

Spacing of the tips and length of the path traveled in the vein: e = 5 mm Upstream transit time: It consists of the transit time in the dermis to reach the Tderme vein, cumulated transit time in the vein T ™ _P in the upstream direction, cumulated transit time in the dermis Tderme to go back to the receiving tip.

Downstream transit time:

It consists of the transit time in the dermis Tderme and in the vein T _v down in the downstream direction.

The upstream transit time in the vein intervenes in the relation: e = (V-Vf) Tvu _P

The downstream transit time in the vein occurs in the relation: e = (V + Vf) T _v down. By combining the two times we obtain:

ΔΤ = T.,

(V _e - V _f ) (V _{e +} V _f ) v _f ^■ At (17)

2nd

Digital application: Ve = 1500 m / s; e = 5 mm, Δΐ = 0, 1 ns; Vf = 2 cm / s

By way of comparison, the average speed of blood in the arteries is:

Aorta: 40 cm / s

Arteries: 10 - 40 cm / s

Arterioles: 0, 1 - 10 cm / s

Capillary: <0, 1 cm / s

Veinules: <0.3 cm / s

Veins: 0.3 - 5 cm / s

Cellar vein: 5 - 20 cm / s

As can be seen, the differential transit time velocimetry does not allow to reach a resolution sufficient to detect the blood flow in the venules. In addition, dimensional changes due to deformation of the skin may taint the measurement of errors, especially when performed upstream-downstream alternation and the measurement rate is too low compared to dimensional variations (a variation dimension of 1 mm per second results in a transit time variation of about 0.6 ns between a shot upstream and downstream fire (for a rate of 1000 shots / sec). It is then advantageous to use a tripod, ie three probes aligned with the vein, the emitting probe being at the center and equidistant from the receiving probes. The vibration axes are collinear and aligned with the direction of the vein. The upstream measurement is thus carried out simultaneously from the downstream measurement, which makes it possible to avoid errors related to dimensional variations related to local variations in pressure. Moreover, with a probe permanently operating in transmitter mode and two probes continuously operating in receiver mode, there is no longer any switching noise related to the control electronics. Finally, if the medium is diffusing, which is the case of red blood cells contained in the blood, the time signal can account for a velocity profile according to the zero crossing considered (speed near the surface for the head of the packet of waves and deeper and deeper for the tail of the wave packet). Figure 18 further provides an exemplary embodiment in a sectional view, the support conical tips. This support is preferably made of metal or plastic firpt. The interest is a partial damping of the waves in the sintered structure no longer requiring the use of a reservoir 13 lateral damping of the resonance of ceramics. This structure is also designed to avoid an acoustic short circuit between the transmitting tip and the receiving tip. Indeed, the acoustic path to go from the base of the emitter cone to the base of the receiving cone requires a much longer transit time through the structure that by the tips and the sample. Such a structure is feasible by 3D rapid prototyping, by stereo lithography of powder, for example PA12 polyamide material ("DuraForm ® PA Plastic," 2013, "Ensinger," 2013) having an average grain size of 58 μπι with a dispersion 25 μπι-92 μπι representing 90% of the particles. The tips are T-shaped and are held laterally in two places, one 43 on an extension z1 situated at the base on a length of 5 to 50 mm and the other 44 about 20 mm before the end on a very short length of the order of a millimeter. They are inserted in force in the supporting structure. In the area of contact with the dermis 64, the supporting structure serves as a counter-support and normalizes the contact between the tips and the dermis. The supporting structure separating the two points has a withdrawal of 1 to 2 mm to avoid contact with the dermis. The tips thus protrude from a short end, about 0.5 mm which is called the support depth PA and are spaced about 5 mm. The base diameter of the tips is 6.6 mm and their height is 85 mm. The flanges 12 at the base have a small thickness, close to 0.5 mm. The collar has an outer diameter of about 10 mm. Its external face does not come flush with the base of the cone, but has a shrinkage of about 1 mm, which allows the ceramic PZT alternating polarization 20 to build a main resonance radially at the frequency of 600 kHz. Once the conical tips are inserted in force, the external faces of the ceramics are connected to the central part of the coaxial connectors 46 and 47 of subvisive or subclic type mounted on a cap 45, while the braid of the coaxial cables is connected via these connectors to the flanges 12 and constitutes the electric mass. The inner face of the cap and the carrier structure may optionally be metallized to form an electrical shield.

Velocimeter with pulsed Doppler effect tip

The conical waveguide probes 10 may also be used in continuous or pulsed Doppler velocimetry.

Figure 19 illustrates the use of a spike transducer in pulsed Doppler velocimetry. The end 14 of the conical Duralumin tip is truncated at 0.4 mm in diameter and lightly pressed against the dermis 65 so that the dermis surrounds the tip to a depth of at least 1 millimeter. The dispersive wave refracts the fields 6 and 7 in the dermis with a maximum amplitude according to the angles α1 and α2 with respect to the axis of the tip which disperses the Doppler angle between the extremes Θ 1 and Θ2. By choosing a phase velocity equal to that of the longitudinal waves of the medium, the dispersion of the Doppler angle is reduced.

The formula giving the speed Vf of the liquid is:

For the same Doppler frequency fd, a slight dispersion of measured velocities between two extremes Vn and \ ¾ appears. The dispersion is greater as the refractive angles i and a ₂ of the acoustic fields 6 and 7 are higher. In principle, the angles ai and a ₂ should be identical, but the reflection at the epidermis / air interface increases ai (and therefore decreases θι). It is therefore appropriate to choose a tip whose phase velocity Vd at the end is slightly greater and very close to the speed of propagation in the dermis and in general in the diffusing fluid inspected. The angles ai and a ₂ are then close and weak, which reduces the dispersion of the Doppler angle. A single tip 10 is used with its peripheral reservoir 13 and a PZT ceramic (PZ27 or PZ29) 20 or 21 with uniform polarization, but the external electrode of which is divided into two half-disks respectively borne at one phase. Burst electric (t) and are logical opposite. The burst burst consists of a burst of 5 slots of amplitude 0 to 100 V. The excitation E (t) with two opposite phases thus generates the bending wave. After transmission of the burst, the transducer is switched to receiver mode R (t). One of the outer electrodes shaped half-disk is then grounded while the other comes into input of a transimpedance amplifier.

The remainder is in accordance with a conventional Doppler measurement (flat transducer without dispersive waveguide). In particular, in Doppler Velocimetry, the duration of a burst burst is rather long, especially when it comes to measuring the low speeds, which goes well with the length of the waveguide and its broadband response. . A measurement rate (or PRF = Pulse Repetition Frequency) of 1 kHz is a good compromise between resolution and measurement range. The lateral resolution is very good and comparable to the size of the tip, which is the main advantage of the device with its ability to operate at high interface temperature.

3. Angle Angular Probe

The granular plastic carrier structure 89 of FIG. 18 can be modified to form a 5-point barrel pentapode.

FIG. 20 illustrates this exemplary use of a directional velocimeter pentapode in a configuration with a central emitting tip E vibrating in precession mode and four receiving points Rx-, Rx +, Ry-, Ry + peripherals sensitive to the modes of bending. The precession tip has a round end 16, the finest possible (diameter close to 0.1 mm refined by polishing) and propagates a combination of two modes of bending at 90 °. The vibration is obtained using a uniformly polarized PZT ceramic 22, divided on its upper face into 4 electrodes each covering 1 quadrant and an electrode forming a solid disc on its underside. The upper electrodes operate in pairs opposite from the center. Each electrode of a pair receives an electrical phase consisting of a burst called Burst (cot) for one electrode and its logical inverse for the other. The other pair is out of phase by π / 2 and comprises an electrode brought to the Burst voltage (cot-7t / 2) where ω is the angular pulse and another electrode brought to its logical inverse. The burst function is a burst a number of slots, 5 in a privileged way. The Burst switch between two voltages 0-100 VDC and are characterized by an extremely stable rise and fall time of the order of a few nanoseconds. The ceramic typically has a diameter of 6.6 mm, a thickness of 0.2 mm, a center frequency of 600 kHz and is made for example PZ27 or PZ29 (Ferroperm, Denmark). The tips are made of a low density metal material (for example aluminum alloy). The central tip has a length of 85 mm, a base diameter of 6.6 mm, a tank ring 13 with thin walls, thickness 0.4 mm, height 5 mm and diameter 20 mm, filled with epoxy resin saturated tungsten powder 13 to a density of at least 5 g / cm3. The ends 11 of the peripheral tips are flattened by polishing, the normal to the polishing plane being directed towards the central tip. This has the effect of increasing the air coupling between the emitter tip and the receiving tips and the directionality of the receiving tips. The receiving tips do not need to be damped like the emitting tip, they are simply inserted into force in the sintered plastic carrier structure 40 which partially damps their resonance and this especially as it is achieved by sintering of plastic powder and that it is heated for example at a temperature of 50 ° C. They have a simple T-shaped geometry. Their ceramic 21, of the PZT type, has an upper electrode divided into two half-disks and a differential amplification of 90 dB is performed between these two half-electrodes, around the central frequency of 600. kHz. The lower electrode is solid and constitutes a virtual mass. Alternatively, with a ceramic 20, the upper electrodes are alternating polarization and then electrically connected, while the lower electrode is the electrical mass. This increases the available electrical charges and therefore the signal-to-noise ratio in the case of a transimpedance amplification. A zero crossing detection using an ultrafast comparator (the switching time 10% -90% is of the order of one nanosecond) is performed on the amplified electrical signal. For the most stable time measurement, we choose the zero crossing corresponding to the maximum amplitude. This configuration makes it possible to reach a measurement rate of 1000 shots per second and a temporal resolution of the order of one nanosecond (in the air) which sets the detection threshold at approximately 1 cm / s per axis in the air at 340 m / s (25 ° C). The maximum speed exceeds several tens of meters / second. The tips comprise a central tip 16 vibrating according to a precession mode and 4 peripheral dipole points 11 for measuring a component of the air velocity vector, in the plane of the ends. The tips are inserted into the U-shaped double wall U-shaped sintered structure 40 and the measurement is similar to that of the "wet finger" in which the finger is raised towards the sky, and feels cold. because of the calories absorbed by the part dried by the wind to determine its direction.

Here, the five points open slightly (about 2 mm) of the rounded carrier structure, to facilitate the flow of wind 67 on the generatrix of the carrier structure. Peaks out sufficiently to reach the maximum coupling between them in the air, obtained for an adjacent area of about 2 mm.

D. VOLUMIC WAVE MIXED PROBES AND DISPERSIVE GUIDED WAVES Figure 21 shows a mixed configuration of the devices detailed in Parts A, "Wave Probes" and B, "Dispersed Guided Wave Probes".

More particularly, it illustrates a sectional view of a longitudinal waveguide probe 17 and shear 17 comprising a solid cylindrical guide 18 propagating the longitudinal waves and a concentric tube 17 propagating flexural waves. The figure illustrates the location and resonance mode of the transducers as well as the location of the dampers 31 and 32

In this figure, a cylindrical waveguide 18 in the center is used to convey longitudinal waves of a longitudinal wave transducer 25 damped by a "backing" 31 and having a coaxial connector 48. The longitudinal waves are routed to the material to be analyzed with sufficient delay. Knowing the impedance of the waveguide makes it possible to deduce the longitudinal characteristic impedance of the sample 61. A tube 17 with a tapered section towards the sample makes it possible to guide a dispersive wave. This wave is generated by an annular transducer operating in radial mode and concentric with the rod 18. The tube is secured to the rod via the porous structure 49, but the two guides are considered to be acoustically isolated. Again, the annular transducer can be divided into two half-rings vibrating in phase opposition to change the radiation pattern of the transverse waves in line with the central guide 18. In the same way that the propagation in the tips to the party B, the wave propagating towards the sample 61 has a decrease in its phase velocity Vd. By adjusting the thickness of the hollow conical waveguide in contact with the sample, it is possible to lower its impedance of radiation in the sample at a value close to the transverse characteristic impedance of the latter. This optimizes the reflection coefficient on the sample for the annular transducer. The wave reaches the sample by shearing the material out of the waveguide tube and generates a shear wave in sample 7. As in Part A, the difference analysis echoes received for a reference measurement, for example in air, and when the probe is in contact with the sample, makes it possible to determine the longitudinal and transverse characteristic impedances of the studied material. E. DUAL WALL GUIDED GUIDED WAVE PROBE

FIG. 22 is a sectional view of a Joule-effect tapered tip probe 97 integrated into an acoustically and thermally insulating bearing structure with integrated measurement of the bearing force. The double U-shaped wall of the supporting structure simultaneously distributes the heat in the inner wall in contact with the waveguide, the thermal insulation of the cone vis-à-vis the air currents by means of the outer wall separated from the inner wall by a thermal insulator, for example a layer of air. The outer wall also performs the acoustic isolation of the cone vis-à-vis the manipulations, while the inner wall carries the maintenance and also the ultrasonic damping of the pulse bending mode, partially dissipated in the sintered granular structure.

In the absence of heating, the advantage of the double wall lies first of all in the possibility of avoiding thermal disturbances on the measurement of the transit time by thermally isolating the waveguide. In addition, the heating makes it possible to impose a temperature profile by means of a heating device 72 applied to the inner wall, as well as to increase the damping of the acoustic waves. Heating also makes it possible to simultaneously perform an effusiveness measurement, for example according to the transit time method explained in French patent FR0759603 of December 6, 2007, by Nikolovski et al. The double wall U, 41 and 42 can thus provide several features:

- The heating of the inner wall for example joule effect by means of an electric current flowing in a wire 82 wound around the inner wall.

the damping of the ultrasounds when the successive echoes pass in the mechanical coupling zone of the waveguide to the sintered inner wall so that in the space of four to five successive echoes in the guide, the wave is damped at least 99% of its initial amplitude, which makes it possible to have a high measurement rate. - The thermal insulation of the inner wall by the outer wall, separated from the inner wall by a thermal insulator 73, for example a layer of air.

- The relative displacement of the inner wall relative to the outer wall under the effect of a contact force of the tip on the sample or a pressure on a part of the structure causing a deformation of the carrier structure in its lower part 50. - intermittently measuring the value of a capacitance of at least one capacitor 81 accounting for the relative position of the inner wall with respect to the outer wall and of which one of the electrodes is integral with the inner wall and the other electrode is integral with the outer wall, the value of the capacitance being for example deduced from a resonance frequency or oscillations of a relaxation oscillator, or of a capacitance measurement technique by a divider bridge from a reference capacitor and the law of variation of the capacitance being connected by a correspondence table, or a mathematical formula such as a polynomial interpolation law, to the force support of the tip on the sample. Three capacitors can be used (but not shown), one to account for the sliding of the two inner and outer walls relative to each other under the effect of a force component located in the axis of the tip, while two other capacitors arranged between the inner and outer walls account for the force components perpendicular to the main axis of the conical guide and in two perpendicular directions having the effect of bending at least one of the walls under the effect of an oblique force,

- Finally, the double wall provides a safety function for the tip and / or the sample, imposing a limit to the maximum penetration at zero support force of the tip in the sample by a maximum value corresponding to the depth of PA support between 0 and 10 mm, and preferably between 0 and 1 mm. Thus, when the tip is supported on a hard surface, the relative sliding of the inner wall relative to the outer wall allows the tip to fully return to the outer wall. For this, the end of the U-shaped double wall is thin and deformable in the region 50 of the folding, under the action of a bearing force.

Furthermore, the inner wall is heated by joule effect obtained by circulating a strong current, of the order of 0.3 to 1 ampere at 3 to 10 V, in a thin wire covered with an insulating varnish, wound around the wall internal 41 on a height corresponding to the contact area between the inner wall and the waveguide. The latter is cylindrical on a typical height of 5 to 10 mm at its base. The winding length produces a total resistance close to 5 to 10 ohms. The inner wall is thermally insulated from the outside air currents by means of the outer wall 42, separated from the previous by an air layer 73 with a thickness of 0.5 to 5 mm. The ohmic winding is housed between the two walls. The spacing is sufficient so that the winding does not touch the outer wall. The outer and inner walls are made of sintered plastic materials and covered with electrodes on at least a part of their surface and are electrically ground on at least another part of their surface. One can also imagine a variant where the inner wall is sintered metal and the outer wall sintered plastic. The inner wall thus provides thermal conduction, while the outer wall provides insulation thermal. The heating circuit is also connected to an annular electrode fixed to a printed circuit 80, said printed circuit being mechanically secured to the base of the outer wall and forming with the upper part of the inner wall a capacitor 81 whose capacitance depends on the bearing force on the tip in the direction of the main axis of the tip. The distance between the top of the inner wall and the ring integral with the outer wall can thus vary as a function of the force exerted on the tip. For this purpose it is ensured that the double wall is deformable by thinning the wall in the zone 50 of folding of the U. Thus the coaxial connector connected to the heating circuit by the wire 82 can also be used to measure a resonant or resonant frequency. oscillations of a relaxation oscillator, determined by the average distance between the ring and the top of the inner wall via the wire 83. The winding of the ohmic heating circuit is also characterized by an inductance L and the whole heating circuit and outer wall is an LC circuit which can be measured resonant frequency that can be correlated to the support force. Furthermore, the outer wall provides acoustic insulation of the cone vis-à-vis the manipulations, while the inner wall ensures the maintenance of the conical tip in two areas, one located near the base, the other 44 located closer to the end. The first zone is used for the ultrasonic damping of the pulsed bending mode, partially dissipated in the sintered granular structure, while the second zone extends over a very limited portion and serves only to guide the tip for mechanical retention. The bottom of the U-shaped double wall bears on the sample. It makes it possible to standardize the depth of support PA and to avoid damaging the sample or the tip. The bottom 50 of the double wall is deformable on its periphery thus allowing the tip to return into its housing under the effect of the bearing force F. The relative displacement of the tip relative to the ring attached to the base of the outer wall is thus measurable from the measurement of a capacitance. This law of variation is easily connected by a correspondence table or a mathematical formula translating the elasticity of the wall in the form of, for example, a polynomial interpolation law, to the support force. It is assumed of course that the driving stroke of the tip is limited, and at most equal to the support depth PA, preferably less than 0.5 mm and in all cases the maximum between 0 and 10 mm. The nominal value of the measuring capacity of the relative position of the two walls is of the order of 1 pF and the resolution of the order of 1 fF (femtoFarad).

Alternatively, during the production of the double-walled heated probe, the heating can be achieved by winding a copper enamelled wire 150μπι thick around a metal tube section. This part is inserted in force on the cylindrical part of the base of the sheath. If necessary to facilitate insertion, heating is started until the softening or melting of the outer surface of the thermoplastic sheath is achieved. Then, a reference electrode consisting of a conductive half-ring is created on a piece bearing on a shoulder of the outer wall. It is possible to provide three coaxial connection cables, one for the capacitive measurement (C), the other for sending / receiving the ultrasound (U), the last for the resistive heating (R). The relative displacement of the outer wall relative to the inner wall (sheath) is made possible by a limited number of beams for holding the sleeve to the outer wall. There are thus 4 beams near the end of the probe and 4 beams just below the heating cylinder. The 8 beams are angularly spaced 45 °. The metal tip is electric ground. The capacitive measurement is obtained by emitting a sinusoidal or square electric burst centered on 1 MHz (or in the range 0.5-2MHz) on the upper electrode of the PZT disk glued at the base of the cone. This pulse simultaneously generates an ultrasonic wave of flexion in the cone while it is transmitted by capacitive coupling to the reference electrode. This configuration has the advantage of simplifying the measurement process because the same ultrasonic excitation signal is used for capacitive measurement. The reference electrode, fixed with respect to the outer wall, is suspended above the upper electrode of the ceramic PZT. The measurement of force or pressure therefore consists in measuring the capacity of the capacitor constituted by the reference electrode and the upper electrode of the ceramic PZT. When the tip is pressed, the sleeve in which the probe is inserted rises and the inter-electrode distance decreases so that the value of the capacitance increases. The nominal distance between the two electrodes of the capacitor for measuring the force or the pressure at rest is 1 to 2 mm. The force measurement then consists in detecting the peak or effective amplitude of the excitation signal transmitted by capacitive coupling. For this, the capacitive signal comes for example at the input of an operational amplifier with a high input impedance and a very low input capacitance which will constitute, with the shielding of the coaxial cable, a reference capacitor with respect to which the measuring capacitance is measured. . The measurement signal Vcap is amplified and then filtered by a bandpass filter before undergoing peak amplitude detection. Once the peak amplitude measurement is performed with conversion to a digital value using an analog-to-digital converter (ADC), the tank capacitor fed by the peak detector is emptied of its charges by means of a MOS transistor. N-channel controlled by a microcontroller and bypassing the tank capacitor. This assembly can be used to measure the force or pressure of a fluid subject to closing the double wall, in other words to deploy a membrane connecting all the beams or by deploying a bulkhead connecting the sleeve and the outer wall. It should be noted that to increase the deformability of the partition it is advantageous for its diameter to be as large as possible. It is therefore not necessarily wise to close the end of the probe at the beams, but to do it further upstream, where the available section is larger. Then, to measure a pressure, it suffices to insert a probe in a pipe by rigidly fixing its external wall to the pipe whose internal pressure is to be measured and by measuring the relative displacement of the inner wall (sheath) relative to to the outer wall. FIG. 23 illustrates an alternative embodiment of a probe 98 comprising 5 tapered waveguides 10, and in particular a bayonet locking cap 55, elastically deformable, and serving to hold the guides in their housing. The holding structure 55 is made of porous or polymer material loaded for example with alumina and also dampens the echoes to avoid resonances in the waveguides. The hood is thus locked by a system of lugs 53 on the supporting structure. Elastic strips 54 are furthermore created in the deformable cover in order to impose a bearing force perpendicular to the base of the guides. Acoustic paths 56 between the support zone on the base of the guides and the periphery of the cowling are complex, for example S-shaped or zigzag-shaped and maximized in length in order to reduce acoustic cross-talk or short-circuits between the guides ( also called "cross-talk" in English).

F. WAVEGUIDE PROBE SOLID DISPERSIVE AND LIQUID JET

Fig. 24 is a schematic sectional view of an ultrasonic probe 99 for non-destructive testing in a situation where the sample and probe are in relative motion (Mov). The carrier structure comprises two guides 10 and 93 in series, the first guide 10 is solid and consists of a truncated conical dispersive tip guide 14, for lowering the radiation impedance and optimizing the transfer to the second guide 93, liquid, coupled to the first via a connection chamber 92 and opening onto an ejection nozzle of the liquid coupling in the form of a collimated jet 93 directional and high pressure. The liquid jet is protected from drafts by a skirt 95 in the form of a cylindrical multi-strand brush. Referring to FIG. 24, there is shown a dispersive waveguide probe 10, coupled to a liquid guide 93 constituted by a more or less viscous liquid coupling, providing a continuity of the liquid medium between the end 14 of the dispersive waveguide and the surface of the sample (60, 61, 62, 63, 64). It is recalled that a sample may have a hard 3D surface 60, constitute the surface of a viscoel medium as 61, be able to propagate shear waves (62, be very hot and constitute a molten magma 63, constitute the dermis 64 of the skin, the wall of a vein 65, or the surface of a liquid 66. There is therefore provided a conduit 91 for supplying the liquid coupler under constant hydrostatic pressure at the end 14 of the waveguide This duct is, for example, made in the outer wall 42 of the double-walled U-shaped support 40. The duct opens into a chamber 92 into which the end 14 of the conical tip also opens, which is preferably truncated. extends in the chamber over a length of the order of a millimeter.The ultrasonic wave then refracts from the tip to the liquid in two main directions not aligned similarly to the directions 6 and 7 of Figure 19, but closer , from the axis of the tip, with a high insertion rate round-trip given the lowering of the radiation impedance of the tip. The ultrasound-carrying liquid then exits via a nozzle whose inlet is oriented to receive the refracted waves in the liquid at one of the main angles and whose exit is of a straight or curved conical shape and is projected by a jet under pressure. 93 towards the surface of the sample. The liquid supply duct opens into the connection chamber 92 preferably at an angle to receive the other part of the refracted beam in the liquid and intended to be damped as it rises in the duct. brought to the inlet 90. The outer part of the ejection nozzle may have a thread 94 for screwing in an adapter piece of the diameter of the outlet jet. The variant of curved shape of the nozzle is not shown in FIG. 24, but its interest is obvious and this role can be provided by an adapter when it is desired to inspect the underside of pieces of complex shape, for example for crack detection under the mushroom of a railroad rail. The probe therefore comprises at least one broadband transverse wave transducer as previously described, coupled to the dispersive waveguide 10. The function of the dispersive waveguide is therefore to lower the peak radiative impedance and to optimize the ultrasonic transfer in the liquid. The coupling with the liquid is carried out in the connection chamber 92 of the conduit and the tip and occurs over an extent of about 1 mm and a characteristic volume of about 1 mm ³ .

The tip is truncated at the point where its phase velocity is just greater than or equal to the velocity of the compressional waves in the coupling liquid, usually water with an antifreeze or an oil. The process thus makes it possible to transfer most of the ultrasonic energy from the dispersive guide 10 to the liquid guide and vice versa during the return path. The liquid guide is a collimated jet. The liquid is fed through a hose connected to the probe for example at the outer wall via a hose connection 90. For a duralumin tip, the diameter of the truncated section is close to 0.4 mm when the angle at the top Θ is 4.8 °. The apex angle may be even smaller or the tip shorter if it is desired to have a working frequency greater than 1 MHz. The tip is itself guided at its base and its end and inserted into force in the connection chamber 92 and ultrasonic transfer. The chamber opens into the open by an outlet nozzle. The outlet nozzle may be a mechanical part in itself fixed on the support or a cast piece or sintered by 3D prototyping and constituting with the double-walled support U 40 a monolithic part. The wall of the duct, when manufactured by sintering, can be previously treated with a solvent, for example acetone to smooth its surface state. The liquid injected under pressure thus serves as a vehicle for the ultrasonic waves transmitted by the conical tip. It conducts the waves to the surface of the sample, over a typical distance of a few millimeters to a few centimeters. The width of the liquid jet is typically 0.5 mm to 2 mm. In Figure 24, the outlet nozzle is directed downward and its main axis and is not quite aligned with that of the conical tip due to the refraction angle in the liquid. On the other hand, it may be useful to probe the sample from below and the outlet nozzle may have a hook shape to project the liquid from the side or from below. These nozzle shapes have applications in the non-destructive testing of railway tracks, while the train is moving, or the inspection of complex parts, for example in aeronautics or wind power. As can be imagined, inspecting a part is less expensive if it can be done at high speed. In the case of a railway line, it is sought that the inspection can be made while the train is traveling at its nominal speed of 300 km / h. But it would be interesting if the inspection could be carried out on regional express trains moving at maximum speeds close to 120 km / h. In this context, the rapid movement of the train and the associated air movement can lead to dislocation of the inspection jet. It would then be useful, on the one hand, to increase the surface energy of the liquid so as to make it more difficult to fragment or nebulize, but the surface energy of the water, about 72 mN / m is already typically three times that of most common liquids, so it is difficult to do better at low cost and without impacting the environment. Nevertheless, one can play on the viscosity of the liquid which can be more easily increased by a factor of 10 to 100 by using oils, so as to have a laminar flow at higher ejection speeds, typically higher at 10 m / s and up to several tens of meters per second. For example, a used frying oil with a typical dynamic viscosity of 35 mPa.s, 35 times more viscous than water. The viscous liquid can itself be heated and thus first fluidized by circulating in a heated conduit. It is then expelled with a smaller diameter and cools very quickly in contact with the air. This treatment has the effect of improving the cohesion of the jet during its journey through the air. To avoid excessive curvature of the jet, the increase in the ejection speed of the liquid must be all the greater as the relative speed of movement of the nozzle relative to the surface of the sample is greater and therefore that the braking forces responsible for the curvature and dislocation of the jet become more important. In this, it is advisable to put the jet away from braking forces over the greater part of its path, to the surface of the sample, by means of a protective skirt 95, made for example in a flexible material of rubber type or made of metal strands forming a screw-shaped annular and cylindrical brush 98 on the outer wall of the probe 99. The protective skirt comprises a large number of strands so that if a number of strands breaks under the effect of friction on the rail, the protection function is always provided by other strands along the entire train path. The damaged skirts can then be easily replaced during a maintenance operation of the probe.

This configuration of a liquid coupling probe has the advantage of consuming only a small amount of liquid coupling, of the order of 0.1 to 10 milliliters per second, ie 0.36 to 36 liters per hour and per conical tip, and to have a good sensitivity considering the effective transfer of ultrasound to the solid-liquid interface 92. The couplant is projected at a distance of a few millimeters to a few tens of centimeters from the surface of the sample. By way of example, a pressure of the order of 2 to 3 bar makes it possible to create a water jet with a diameter of 0.8 mm at a speed of approximately 10 m / s with a physical continuity of the jet reaching a few tens centimeters in a quiet room. This speed of expulsion is already in a turbulent regime and it would be better to use oils to stay in lamellar regime. In fact, the turbulence is characterized by the fact that concentric layers of liquid change their position inside the jet, which generates distortions in the acoustic signal. A speed of 10 m / s is sufficient to bend the skin safely to a depth of about one millimeter. Fig. 24 illustrates a probe having only one solid-liquid guide system. Nevertheless, it is obvious that this configuration can be extended to a bipod, tripod or pentapode. In these latter cases, the protective skirt 95 naturally surrounds all the points. It will be noted that the coupling under pressure is expelled by the nozzle in direction of the sample, but it will be understood that the fact that it is under pressure also allows it to completely fill the connection chamber 92.

Alternatively, the liquid jet probe may consist of several parts possibly made of different materials. The ejection nozzle with its nozzle is a critical part that can be damaged and must be replaced without changing the entire probe. It is therefore possible to provide the possibility of screwing it into the coupling chamber.

In addition, the insertion of the metal tip into the coupling chamber requires force insertion and perfect centering of the tip. This being fragile, an opening in the sheath is provided to see the insertion operation. Furthermore, the fact of screwing the tip into the coupling chamber does not facilitate its fine positioning relative to the end of the tip. The positioning of the exit cannula is correct only for a whole number of turns of the screw.

Alternatively, the thread of the tip may be removed to allow the tip to slide along the walls of the coupling chamber. Once the optimal position found corresponding to the maximum transmission of the ultrasonic beam in the outlet cannula, it is frozen by gluing or side locking screw (not shown). The tip must satisfy several constraints: The first relates to the alignment of the main axis of the tip with the wall of the outlet cannula, the second the positioning of a reflective surface covering an angular sector of 180 ° and of which the generator is at an angle to the main axis between 17 ° and 25 °. This angular variation continuously and gradually corrects the angle of refraction that increases towards the end of the tip. This places one of the two ultrasonic beams refracted by the tip in the axis of the cannula. The other beam is reflected on another surface occupying the remaining angular sector and intended either to return the beam to the cannula or to deflect the beam in the direction of the liquid supply duct so that it is gradually damped in the leads. The last constraint is that the tip does not touch the tip. It must get closer to the beak of the first reflective surface without touching the cone.

Moreover, in another field of applications concerning the medical diagnosis by pulse elastography and therefore corresponding to a case where the surface is deformable, as is the case of organic tissues, the hydrostatic ejection pressure can be variable and modulated in time in amplitude and / or in phase. The modulation amplitude can reach an intermittent ejection regime with physical discontinuity of the jet and a rate ejection can vary between 1 Hz and 10 kHz. Knowing that the round-trip transit time of the ultrasonic wave propagating in the guide to the sample is of the order of 100 for a throw of a few centimeters, it suffices that the spatial continuity of the jet between the end of the solid guide and the sample lasts for a period of at least 100 for ultrasound measurement to be possible. Nevertheless, the duration of a jet may more advantageously be between 5 ms and 50 ms to allow the ultrasonic waves to penetrate deeply into the sample. This mode of operation generates a rhythmic percussion of the surface of the sample and approaches, in principle, the low frequency pulse elastography with here a major difference is that the low frequency shear wave is not generated by a vibratory pot or a bulky electromagnetic percussion device juxtaposed with the ultrasonic source, but results directly from the modulation of the ejection velocity or the intermittent impact of the liquid jet conveying the ultrasonic wave. And, given the speed of propagation of shear waves in human tissues, of the order of a few meters per second, it is therefore necessary to visualize the displacement of the shear wavefront generated by the impact that the Spatial continuity of the jet is ensured over a period of at least 10 to 100 ms. Moreover, while the jet or jets regularly strike the surface of the sample at a rate of eg 50 Hz (with a duty cycle of 50%), the ultrasonic pulses are generated at a much higher rate of the order from 0.5 to 10 kHz and the central ultrasound frequency is preferably between 300 kHz and 3 MHz. The diameter of the collimated jet 93 is typically 0.5 mm to 2 mm. The jet thus constitutes a quantity of movement displacing the surface of the skin to a depth that depends directly on its ejection speed and its duration of impact. The amplitude of the deformation is typically 1 mm. If water is used as coupling, its viscosity being of the order of 1 mPa.s, a Reynolds number close to 500 is obtained for a nozzle outlet speed of 1 m / s and a diameter of 0.5 mm nozzle. The flow is then laminar. It remains so up to a Reynolds number of 2300, which corresponds to approximately a nozzle exit velocity of 4.6 m / s. In order for the flow to remain laminar at a higher ejection speed, it is sufficient to increase the viscosity of the liquid coupling and the ejection pressure applied to the liquid. This type of probe and this mode of operation thus has the advantage of being able to be easily moved during a medical examination by pulse elastography. In addition, and it is the interest of the present invention, the support structure can be designed so that the jets are collinear or fan in order to be able to play on the lateral bulk of the ultrasonic beam by playing on the distance to the sample. The jet associated with each point can also be associated with a law of modulation of amplitude and of different phase of the neighboring jet. It is then possible, for example, in the case of a five-jet pentapode, or more generally of a multi-jet system, to control the radiation pattern of the shear wave generated in the biological tissue, according to the amplitude and the phase of the pressure applied to each jet. Finally, in the case of a medical check, it is desirable to recover the liquid coupling sprayed on the skin of the patient, even if it is in small quantities. The protective skirt here will not function to avoid a dislocation of the jet, which is not likely to occur here, but to confine the splash caused by the impact of the jet. In addition, since the movement of the probe during an examination is rather slow, of the order of a few millimeters per second, the skirt may be made of a less aggressive material than a wire brush, for example in the same plastic material sintered or injected as that of the guide support or in another smooth material having non-allergenic properties. An additional suction function of the liquid spray can be integrated on the periphery of the probe for example for the sake of cleanliness and comfort of the medical examination. For this, the skirt may consist of a double wall, the suction flow being carried out in the annular space between the two walls of the skirt. Since the suction flow itself is turbulent and generates low-frequency waves that can disturb the low-frequency shear waves generated by the jets, the suction is not carried out continuously, but only when the amount of couplant liquid is too important and requires vacuum cleaning. In this application, the viscosity of the liquid coupling may advantageously be greater than that of water to approach that of an oil to prevent the couplant from running too easily on the skin of the patient.

Moreover, as indicated previously, in order to maintain the collimated jet, and therefore in laminar flow, it is essential to control the value of the Reynolds number, defined according to the formula: μ in which p denotes the density of the liquid, _v la flow velocity of the liquid at the exit of the ejection nozzle, d the exit diameter of the ejection nozzle and μ the dynamic viscosity of the liquid. Since the Reynolds number must not exceed a defined value according to abacuses adapted to the shape of the preferably tapered flow conduit in order to preserve the collimated appearance of the jet, it appears that increasing the viscosity constitutes an effective solution for speeds ejection and to reach more distant inspection areas, or to avoid a dislocation of the jet for example by a strong side wind, or to facilitate the use of a liquid having a density greater than that water and thus be able to generate a more intense shock wave intermittent while retaining the collimated aspect of the jet.

Figure 25 illustrates a dispersive tip 10 held in a bearing structure 40 whose end is bent hook-shaped according to one embodiment. Preferably, the curvature compensates for the increase in the refraction angle as the wave progresses toward the end of the tip so that the ultrasound beam diffracts into the medium at a constant angle. Preferably, to refract at least one half of the beam at 90 ° from the main axis as shown by an arrow in this figure, it is expected that the curvature of the tip is greater than 60 ° at the tangent to the tip of the tip.

The main interest of this geometry resides in the possibility that it gives to couple acoustically two points with very close bases and very distant ends. The interest is obvious in the case of a differential flow time flow probe with two points inserted into the pipe by a single small hole (for example less than 20 mm). Note that to refract a half of the beam along the main axis of the tip, it is sufficient that the curvature of the tip is close to 30 ° at the tangent to the tip of the tip.

Fig. 26 shows an application of pressure measurement of a device according to one embodiment. In this figure, the pressure sensor comprises a sleeve 454 and a metal tip coupled to the sheath by crimping and brazing. The sensor is mounted on the wall of a pressure vessel 450. The tip held in the upper part by a guide 455 opens into a cylindrical cavity 451 closed by the reflector. The latter has at least one hole 453 or an opening for pressurizing with the medium. The sleeve 454 ends with a thread allowing it to be mounted on the wall of the pressure vessel. The end of the tip radiates radially by reflecting one or more times on the side walls before reflecting on the plane reflector and then back to the end. The planar reflector is housed and crimped into a shoulder at the end of the sheath.

In the following claims, the terms used are not to be construed as limiting the claims to the embodiments set forth in this specification, but should be interpreted to include all the equivalents that the claims are intended to cover because of their formulation and whose forecast is within the reach of those skilled in the art in applying his general knowledge to the implementation of the teaching just disclosed to him.

In addition, it has been proposed here various embodiments of the invention, applied to different fields of application. These different embodiments can of course be combined with one another to obtain new advantages in terms of accuracy and measurement performance. In particular, provision may be made to use tips having bent hook-shaped ends in all the embodiments proposed above, as well as in all of the applications also mentioned above, to compensate for the increasing the angle of refraction as the wave progresses towards the end of the tip so that the ultrasonic beam diffracts in the medium at a constant angle.

Claims

1. System (8, 9, 87, 88, 89, 97, 98, 99) for characterizing the mechanical and thermal properties of a solid or fluid medium (60, 61, 62, 63, 64, 65, 67), comprising at least one broadband ultrasonic transducer (20, 21, 22) coupled to a first end of at least one tapered solid waveguide (10, 17, 18) so as to generate in said tapered solid waveguide at least one dispersive bending wave (A) in continuous or pulsed mode which propagates, the tapered solid waveguide being in contact, on the side of a second end, of lower section than that of the first end, with the solid or fluid medium, in which:

said at least one dispersive bending wave transmitted in the waveguide is refracted into an ultrasonic elastic wave in said medium,

the maximum lateral extent of the surface (11, 14, 15, 16) of contact between the tapered solid waveguide and said medium is less than the wavelength of said ultrasonic elastic waves generated in the medium,

the at least one tapered solid waveguide is housed in a carrier structure (40), at least one wall (41) of which allows acoustic insulation and the holding (43) of said tapered guide as well as the damping of said ultrasonic waves; said guide being held by the carrier structure on its periphery at its base on the broadband ultrasonic transducer side and, on the other hand, upstream (44) of its tapered end,

the device further comprises a device for measuring the amplitude or phase variations of the reflected (Ar (x, y) and / or transmitted (At (x, y)) waves in the medium and a device for calculating mechanical and / or thermal properties of said medium from the measurements provided by the measuring device.

The system (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1, wherein the tapered solid waveguide (10, 17, 18) has a cylindrical axis of symmetry, has a solid conical profile (10) or hollow (17) having a section which is refined until the phase velocity in the refraction zone is less than or equal to the speed of the longitudinal or transverse waves or guided waves generated in the medium or that its radiation impedance at the point of contact is between 1% and 100% of the transverse impedance of the material constituting the guide.

3. System (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1 or 2, wherein the computing device determines the mechanical impedance Z ₂ of the medium determined by the formula:

- Zair is the mechanical impedance of the air,

Ar2 is the reflection coefficient at the tapered waveguide interface - medium to be characterized,

Arl is the reflection coefficient at the tapered waveguide-air interface,

Zi is the contact impedance of the tapered waveguide.

The system (8, 9, 87, 88, 89, 97, 98, 99) according to any one of claims 1 to 3, wherein the waveguide is of a material having a characteristic impedance greater than or equal to that of the medium, for example a low density metal such as aluminum or titanium, or a hard plastic such as plexiglass or polycarbonate.

The system (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1, wherein the tapered solid waveguide is a solid cone whose end is truncated (14) to the where the phase velocity is equal to the velocity of the longitudinal acoustic waves generated in the medium.

The system (8, 9, 87, 88, 89, 97, 98, 99) according to any one of claims 1 to 5, wherein:

the first end of the tapered waveguide (10, 17, 18) is exposed at a first temperature below the Curie temperature Te of the ultrasonic transducer,

the tapered waveguide (10, 17, 18) is in a superalloy, for example in inconel,

the second end of the tapered waveguide is in contact with a medium of a second temperature greater than the first and has an area at least 100 times smaller than that of the first end of the tapered waveguide,

- The tapered waveguide (10, 17, 18) has a length which maintains a temperature difference between the first and second ends of said guide, the temperature difference resulting from a phenomenon of convection or conduction free or forced by a heat transfer fluid cooling system (70), and regulates the temperature of said transducer at the first temperature.

The system (8, 9, 87, 88, 89, 97, 98, 99) according to any one of claims 1 to 5, further comprising:

a second and a third conical tapered tapered waveguide (10) aligned to form a differential tripod, further comprising two receiving points arranged at equal distances from a bending wave emitter tip, and aligned with that ci, the receiving tips being arranged angularly with respect to their central axis so that the received signals are in phase opposition.

8. System (8, 9, 87, 88, 89, 97, 98, 99) according to any one of claims 1 to 5, characterized in that it further comprises four tapered waveguides (10) in shaped conical points to form a pentapode (98) or two tripods oriented at 90 ° relative to each other and sharing the same central point, the central tip of the system operating as a receiver and peripheral points transmitter (Ex -, Ex +, Ey-, Ey +), further comprising a single electronic amplifier connected to the single central receiver (R).

The system (8, 9, 87, 88, 89, 97, 98, 99) according to claim 8, wherein the acoustic transducers associated with the tapered waveguides are four-quadrant PZT ceramic disks (22) operating in linear combination of two dipoles oriented at 90 ° to each other to be combined in amplitude and / or in phase so as to impose an acoustic vibration direction in the contact plane.

The system (8, 9, 87, 88, 89, 97, 98, 99) according to claims 1 and 7, further comprising fourth and fifth tapered waveguides (10) for forming a pentapode (98). arranged in two tripods oriented at 90 ° to one another, for depth measurement of longitudinal (1, 3) or transverse (2, 4) velocity variations.

The system (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1 or 2, further comprising four additional tapered tips (10) for forming a pentapode (98), a central point operating in transmission mode while the other four peripheral peaks operate in reception mode, the central tip (10, 16) operating in precession mode, while the four peripheral peaks (10, 11) operate in pairs of receivers diametrically opposed with respect to the transmitter, the measuring device detecting the differential transit time per pair of receivers.

The system (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1 or 2, further comprising four tapered-tip guides (10) forming a pentapode (98) in time measurement mode. transit between a central emitter tip and the other four peripheral tips, the emitter tip producing a horizontal transverse wave (S) in pulse precession vibration, the two pairs of peripheral tips being oriented to detect said horizontal transverse wave (S), the measuring device giving a propagation speed of this wave in two perpendicular directions.

The system (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1 or 2, adapted for contact with the epidermis (64), wherein the conical tip transducers are truncated (14). ) and lightly pressed against the epidermis (64) so that the entire section of the guide is contact with the epidemic and that, in the contact zone, the section of the guide (14) is such that the phase velocity (Vd) in the tapered solid waveguide is equal to that of the longitudinal waves of the medium, or about 1500 m / s.

The system (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1 or 2, adapted for contact with an air medium (67), further comprising four additional tips for forming a pentapode ( 98) with a central emitting tip (10) vibrating in precession mode and four receiving points (Rx-, Rx +, Ry-, Ry +) sensitive to a bending mode, the central tip (10) having an end (16). ) round, preferably of section diameter at its end less than 0.1 mm, and vibrating in a combination of two modes of bending oriented at 90 ° to each other.

15. System (9) according to claim 1, wherein the carrier structure is sealed and the tapered solid waveguide is pointed and partially immersed in a fluid medium to where its phase velocity is equal to the rate of propagation of the pressure waves in the medium.

16. System (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1 or 2, wherein the carrier structure comprises a U-shaped double wall (40) providing:

- Controlling a set temperature of an inner wall of the carrier structure (41), for example by means of a resistive coil (72) for heating the conical tip (10) by Joule effect;

- The damping of successive echoes passing in the zone (43, 44) of mechanical coupling of the waveguide to the inner wall of the carrier structure;

- The thermal insulation of the inner wall by the outer wall of the supporting structure, separated from the inner wall by a thermal insulator (73);

the relative displacement of the internal wall of the bearing structure relative to the external wall of the bearing structure under the effect of a force (F) of contact of the tip on the sample or of a deforming pressure exerted by the middle on a part of the supporting structure or the holding sleeve of the tip;

- intermittently measuring the value of a capacitance (81) of at least one capacitor accounting for the relative position of the inner wall relative to the outer wall of the supporting structure or the inner wall of the supporting structure relative to the tapered solid waveguide, and one of the electrodes is integral with the waveguide and the other electrode is integral with the inner wall or the outer wall, the value of the capacity being deduced from a resonance frequency or oscillations of a relaxation oscillator, or a bridge measurement divider of a reference capacity and the law of variation of the capacitance being connected by a correspondence table, or a mathematical formula such as a polynomial interpolation law, to the force of support of the tip on the sample, or at the middle pressure; or

a safety function for the tip and the sample, imposing a limit on the maximum penetration at the zero bearing force of the tip in the sample by a maximum value (PA) of between 0 and 10 mm, and preferentially less than 0.5 mm.

17. System (8, 9, 87, 88, 89, 97, 98, 99) according to one of claims 7, 8, 15 or 16, wherein the carrier structure (40) is composed of a part ensuring the positioning, orientation (41, 44), isolation and damping (42, 43, 73, 74) of the waveguides and a bayonet cowling (55) consisting of elastically deformable portions (54, 58) providing a holding force of the probes in the inner wall (41) and the locking (53, 57) of the flexible elements (54, 58) on the guide element (52).

The system (98) according to one of claims 1 or 2, further comprising:

a conduit (91) for feeding a pressurized liquid waveguide at the end (14) of the tapered solid waveguide (10),

one end (14) of the solid waveguide immersed partially in the liquid waveguide, in a connection chamber (92), on a limited extent, of the order of 1 mm and a volume of the order 1 mm ³ ,

a nozzle of ejection of the liquid waveguide, of straight conical or curved shape in the form of a hook allowing the inspection of external, lateral or internal surfaces of a sample, such as a railroad rail,

a protective skirt (95) of the liquid waveguide (93) against dislocations or fragmentation by a stream of air over at least a part of its length to the surface of the sample or confinement and suction of splashing of the liquid jet projected on the surface of the sample,

in which :

the hydrostatic ejection pressure of the liquid varies over time to ensure the generation of acoustic waves of low frequency shear,

the acoustic waves are controlled in amplitude and / or in phase, in particular between the jets of a configuration with several liquid waveguides,

the hydrostatic pressure is modulated to reach an intermittent ejection regime with an ejection rate of between 1 Hz and 10 kHz.

19. System (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1, characterized in that the end of the tapered solid waveguide is curved.

20. System (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1, wherein the second end of the tapered solid waveguide is inserted into force in a sheath formed in the supporting structure, the assembly rigidly coupled mechanically to be sealed by crimping the sleeve or metal solder between the base of the tip and the sleeve if it is metallic.

21. System (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1, wherein the tapered solid waveguide is inserted into a sheath formed in the support structure provided at its end with a reflective collar for orienting the emitted ultrasound beam or for shielding the received ultrasonic beam from a predefined direction.

22. System (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1, further comprising a second solid waveguide tapered to form a bipod and a reflector integral with the carrier structure disposed under the points, perpendicular to their main axis, at a distance between 75 and 125% of the inter-tip distance.

The system (8, 9, 87, 88, 89, 97, 98, 99) of claim 1, further comprising three additional tapered solid waveguides to form two bipodes inserted perpendicular to a pipe at two different depths, each bipode being aligned with the direction of the flow to be characterized, one of the bipods having its ends arriving at a first depth in the pipe so as to access the flow velocity at the center of the pipe, while the another bipod at its ends reaching a second depth of the pipe, distinct from the first depth, in the vicinity of the inner wall of the pipe so as to have access to the flow velocity close to the wall.

The system (8, 9, 87, 88, 89, 97, 98, 99) according to claim 18, wherein the liquid ejection nozzle is separate from the carrier structure and is bonded to the carrier structure in a position where at least one of the ultrasonic beams refracted in the liquid from the end of the tip is reflected on at least one of the faces of the tip and aligns after reflection with the main axis of the cannula d ejection of the tip.

25. System (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1, wherein the tapered waveguide is inserted in a sheath formed in the supporting structure, the end of the sheath with its open tapered solid waveguide being covered or filled with a anti-adherent and / or fungicidal protective coating, for example based on Teflon (PTFE or FEP or PFA) or silicone.

26. System (8, 9, 87, 88, 89, 97, 98, 99) according to claim 1, wherein the computing device deduces the pressure of a medium directly from the amplitude of an echo or a signal transmitted in the medium in combination with a capacitive measurement of the deformation of the carrier structure or a sheath formed in the carrier structure.

27. A method for characterizing the mechanical and thermal properties of a solid or fluid medium (60, 61, 62, 63, 64, 65, 67) using a device according to any one of claims 1 to 26 , comprising the following steps:

generating in said solid tapered waveguide at least one dispersive bending wave

(A) in continuous or pulsed propagation mode, said at least one dispersive bending wave transmitted in the waveguide being refracted into an ultrasonic elastic wave in said medium,

measuring amplitude or phase variations of the reflected (Ar (x, y) and / or transmitted (At (x, y)) waves in the medium, and

calculate the mechanical and / or thermal properties of said medium from the measurements provided by the measuring device.

The method of claim 27, comprising the steps of:

generating and detecting a reference echo in the medium by bipode (89), tripod or pentapode (98) probe successively in contact with the medium on a set of reference positions along a reference curvilinear path,

storing said reference echoes associated with each position of the probe on the medium along the curvilinear reference path

recognizing contact positions of the probe on the medium by calculating the minimum distance or by calculating the maximum of an inter-correlation function between a measurement echo obtained during a second passage on the curvilinear path and the one of the echoes corresponding to the learned positions of reference, and

provide amplitude and phase information on the spatial variations of impedance along the curvilinear reference path.

The method of claim 27, further comprising a step of driving the waveguide (10) into the medium (63) of a characteristic depth for which the amplitude of the internal echo (Ar) in the waveguide is decreased from 50% to 75% of its value with respect to conditions where the end (16) of the waveguide is free.

The method of claim 27, further comprising the steps of:

moving a differential tip tripod to the surface of the medium (61) in the direction perpendicular to the segment connecting the three points

measure the amplitude and the phase on the spatial variations of impedance.

The method of claim 27, further comprising:

a step of constituting a reference database characterizing different media or the different states of a medium and associating with each medium or at each reference state the measurement of at least one of its effusivity values, longitudinal and transverse impedances as well as angular anisotropy of said impedances in at least two different directions, for example orthogonal directions, from bipod, tripod or pentapode type probes,

a step of recognizing a medium or the state of a medium by measuring its effusivity values, longitudinal ZL and transverse Ζτ impedances as well as angular anisotropy of said impedances in at least two different directions, using a bipod, tripod or pentapode type probe by searching for a corresponding element closest to a reference database on this medium or on the states of this medium.