RU2371700C1 - Method for detection of steel product hardness - Google Patents

Abstract

FIELD: machine building. ^ SUBSTANCE: steel product is exposed to local action of powerful laser radiation, as a result of which non-stationary thermal field is optically excited. Space-time modulation of thermal waves from metal surface is transferred to optical area with application of a reading laser beam with low power in visible spectrum. Using CCD camera, dynamics of sample surface speckle images is registered. Breaking speckle images into several fragmentary sections, function values F(T, r)=1-cor(T, r) are calculated, where cori(T1, r) is a correlation function. For a separated fragment, using produced values, they build a curve of function F(T, r), and calibration curves are used to detect mechanical hardness of this portion in steel product. ^ EFFECT: lower error of measurements due to exclusion of negative effect of parasite infrared sources of radiation. ^ 13 dwg

Description

The invention relates to the field of mechanical engineering and can find application for non-destructive optical control for remote determination of the mechanical hardness of steel products.

Widespread methods for measuring the hardness of metals and alloys, based on the resistance of the sample to mechanical indentation, do not satisfy the increasing requirements of modern industry. The main complaints are that they are destructive contact methods. The problem of controlling curved surfaces and the difficulty of measuring the profile of hardness in depth are also noted. The method of non-destructive testing of the mechanical properties of steels, based on the application of the magnetic method (coercive force) [eg 1], is applicable only to ferromagnetic materials. The development of non-destructive non-contact methods for determining the hardness of the hardening surfaces of parts made of steel alloys in industry is a very urgent task.

The physical basis of the known thermo- and elasto-optical methods of non-destructive testing and diagnostics is the use of weakly attenuated acoustic waves and unsteady thermal fields for sensing the surface and internal areas of metals. In these methods, spatio-temporal modulation, localized in the amplitude-phase profiles of acoustic or thermal waves, is transferred to the metal surface, and information is collected using a read beam. For example, in [2], laser infrared photothermal radiometry was used to measure the thermophysical properties (thermal diffusion and conductivity) of coatings deposited on a metal. Known "flash" method [3] for measuring the coefficient of thermal diffusivity, heat capacity and thermal conductivity of metals (copper, silver, iron, nickel, aluminum, tin, zinc). Here, a powerful light pulse was absorbed by the front surface of a thermally insulated sample, and the dynamics of the temperature of the rear surface were measured with a thermocouple.

It is known that there is a correlation between mechanical hardness and thermophysical characteristics of the alloy [4]. Using infrared photothermal radiometry, the authors of [4] conclude that there is a good correlation between thermophysical parameters, their dynamics, and mechanical hardness profiles, which is suitable for assessing the transverse inhomogeneity of hardness of the hardened layer.

In the photothermal method for determining the hardness of steel [5] (prototype) it is shown that one of the thermophysical characteristics suitable in practice for determining hardness can be the coefficient of thermal diffusion k. This coefficient determines, in particular, the propagation velocity of the heat flux, which is formed under the action of an external heat source Q and depends on the density ρ and specific heat capacity from the metal. Assuming that the density of steel is constant, a conclusion is made in [5] on the correlation of hardness and thermal conductivity. Thermal conductivity, in turn, is proportional to the coefficient of thermal diffusion. Calibration curves between the hardness of the steel sample and the coefficient of thermal diffusion provide the basis for determining hardness. The physical essence of the photothermal method for determining the hardness of steel products consists in laser excitation of the heat flux in the product and registration of temperature dynamics using an infrared infrared sensor. Photothermal measurement of the amplitude and phase difference using the multiple smoothing procedure allowed the authors [5] to restore the profile of thermal conductivity of steel. According to the calibration curves, the thermal conductivity profile was “translated” into the hardness profile and compared with the traditional Vickers profile. One of the disadvantages of the photothermal method for measuring hardness is the negative effect of “spurious" infrared radiation sources, which are very difficult to get rid of under production conditions.

The objective of the invention is the ability to use in an optical non-destructive method for determining the hardness of steel products, which eliminates the negative influence of "spurious" infrared radiation sources, thereby reducing the measurement error.

The problem is solved as follows. The investigated steel product is subjected to local exposure to laser radiation with the power necessary for the optical excitation of a non-stationary thermal field. Then the object is additionally irradiated with laser radiation and the dynamics of speckle images of the surface of the product is recorded. Dividing speckle images into a series of fragmented sections, the values of the function are calculated

Where

cor _i (T ₁ , r) - correlation function (CF),

I is the intensity of the reflected radiation,

r = (x, y) - coordinates of the point of the speckle image fragment,

T ₁ , T _m - the temperature of obtaining speckle images,

i = 1 ... M is the number of the speckle image fragment,

m = 1 ... N is the number of speckle images received from the object at a temperature T _m ,

ΔS is the area of the ith fragment of the speckle image.

For the selected fragment, the function F (T, r) is plotted using the obtained values and the mechanical hardness of this section of the product is determined from the calibration curves.

The essence of the invention is illustrated by drawings, where:

figure 1 shows a diagram of a device for non-stationary speckle photometry;

figure 2-5 - dependence of the correlation function C (T, r) for samples No. 1 and No. 2 from time t at fixed values of z;

figure 6 - dependence of the correlation function C (T, r) for sample No. 3 at a fixed point in time t depending on the coordinate z;

in Fig.7 - the dependence of the thermodynamic temperature T on time t, obtained from a numerical solution of the spatially one-dimensional heat equation for steel at three fixed z;

on Fig - dependence of the function F (T, r) on time t for sample No. 1 at a fixed value of z;

in Fig.9 - the dependence of the derivative of thermodynamic temperature T on time t for cases corresponding to Fig.7;

figure 10 - dependence of the function F (T, r) on time t for sample No. 3 with fine shot peening at three fixed values of z;

figure 11 - dependence of the function F (T, r) on time t for sample No. 3 with rough rolling at three fixed values of z;

on Fig - dependence of the derivative of the function F (T, r) on time t for sample No. 3 with fine bead-blasting and rough rolling at a fixed value of z;

on Fig - dependence of the derivative of the thermodynamic temperature T on time t, obtained from a numerical solution of the spatially one-dimensional heat equation with values of the coefficient of thermal diffusion k for steel, found on the basis of experimental data from Fig. 12.

The performance of the proposed method for determining the hardness of steel products was confirmed by experimental studies. The experimental setup for unsteady speckle photometry, using spatially inhomogeneous heating of the object of study, is presented in figure 1. On the surface of the product 1 was directed the radiation of a laser diode 2 with the power necessary for optical excitation of a non-stationary thermal field. In our experiment, the power of the laser diode was 15 watts. The radiation was focused by a cylindrical lens 3. Due to inhomogeneous heating along the surface and in the volume of the sample, an unsteady heat flux 4 was excited. When probing the region of thermal disturbance 5 by 10 mW He-Ne laser radiation, a dynamic speckle image of the sample surface was formed using an optical system 7 recorded by a high-speed CCD-camera 8. In the experiment, we used the following CCD-camera modes: frequency of speckle image registration 100 Hz, number of resolution elements 1392 × 1040pi pix with size 6.4 × 6.4 mm. The studies were carried out on three samples: 1) two of Steel 45 Steel in the form of cylinders 18 × 8 mm in size with various degrees of hardening - HRC = 21.5 and HRC = 28.7; 2) a plate made of steel 18MnСr05 with a size of 55 × 40 × 10 mm with four sections differing in the types of mechanical hardening of the surface: a) thin bead-blasting HV = 376, b) thin rolling HV = 427, c) rough bead-blasting HV = 447, g ) rough rolling HV = 481.

We experimentally realized the simplest case from the point of view of the accuracy of speckle-photometric measurements when the thermal field is quasi-flat. Such a field was formed when the laser beam was focused by a cylindrical lens 3. To process the recorded speckle image of the sample surface, an analysis of the temporal and spatial dynamics of speckle fields was used. Note that, when measuring the speckle field at various distances from the source, their dynamics in accordance with the heat equation is determined by the thermal diffusion coefficient k. Thus, the proposed method is focused on extracting the coefficient k of steel from the correlation function KF (2) of speckle images obtained at different points in time and at different distances.

The dependences of the correlation functions of KF cor _i (T ₁ , r) on time t obtained experimentally for hardened steel (sample No. 1 — curve 1, sample No. 2 — curve 2) are shown in FIGS. 2-5. The CF data was obtained at different distances (the z coordinate is given in CCD pixels in FIGS. 2-5) from the heat source. Figure 2-5 shows that the correlation function in the region close to the source for sample No. 2 is lower (the hardness of sample No. 2 on the HRC scale is higher) than for No. 1. In this case, the difference in CF for these samples with increasing distance from the source first increases, and then at a sufficiently remote distance z≈500 decreases again. The behavior of the correlation functions at a fixed time t depending on the distance z for sample No. 3 of steel with various types of processing is shown in Fig.6. It can be seen that the behavior of the correlation functions “harmonizes” with the change in hardness of the selected sections of steel, measured by the classical method of indentation on the HV scale. Thus, as follows from figure 2-6, there is an unambiguous relationship between the hardness of steel and the correlation function obtained from the dynamics of the speckle image of the surface of the sample. Moreover, the unambiguity of the relationship between these parameters increases with an appropriate choice of the temporal and spatial areas of the CF construction, i.e. while providing in experiments the necessary CCD speed and the distance from the heat source to the measurement point.

It is known and tested by us on the studied samples of steel products that at low heating temperatures ΔT (in the experiment ΔT was about 25 ° C), the correlation function is linearly related to the magnitude of the movement of speckles. The thermal expansion of steel is proportional to the first degree of temperature at low heating temperatures. And it is precisely because of thermal expansion that the speckles shift. As can be seen, due to the saturation effect of CF (falling to zero), the heating ΔT is associated with the average speckle size and is significantly lower than the temperature range of linear thermal expansion of steel. Taking into account the above remarks, as a characteristic of the thermal field, we introduced the function F (T, r) = 1-cor (T, r), which can be used as an optical analog of the thermodynamic temperature for a steel sample (conventional name - “optical temperature”) . The similarity of the "optical temperature" and thermodynamic is illustrated in Fig.7-13. Figure 7 shows the time dependence of the thermodynamic temperature obtained from the numerical solution of the spatially one-dimensional heat equation for steel at three fixed values of the z coordinate, and Fig. 8 shows a dependence of the form (1) F (T, r) = 1-cor ( T, r) obtained on the basis of experimental data. The first significant similarity of the thermodynamic temperature function T and the function F (T, r) versus time t is that they have an inflection point, i.e. the maximum point of the first time derivative. It follows from the spatially one-dimensional heat equation that the coefficient of thermal diffusion k is equal to z ² / 2t (provided that the beam width is neglected), where (z, t) are the coordinates of the maximum of the function dT / dt. The position of this maximum shifts to the right in time (lag) when the point of temperature measurement is removed (by increasing the z coordinate) from the heat source. Figure 9 shows the dependence of the derivative of thermodynamic temperature T on time t for the cases corresponding to Fig. 7 (the graph number grows with increasing z).

Figure 10 and figure 11 shows the experimentally obtained temporal dependences of the "optical temperature" at three different values of the z image of steel No. 3 for fine bead-blasting and rough rolling, respectively. Taking the derivative of the "optical temperature" at a fixed z for the indicated types of machining (for fine bead-blasting and rough rolling), we obtain dome-shaped curves with a pronounced maximum of Fig. 13. As can be seen, the previously indicated effect of the displacement of the maximum of the derivative of the “optical temperature” is observed in case of a change in the hardness of the selected fragment of the steel product. If we assume that the similarity coefficients K _z and K _t (these are peculiar characteristics of the experimental design of the device) are known, then, by determining, from Fig. 13, the coordinates (z, t) of the maximum of the function dF (T, r) / dt, we can calculate the value of the coefficient k. The thermal diffusion coefficient k obtained in this way from Fig. 13 for fine shot blasting is k ₁ = 0.112 cm ² / s, for coarse rolling k ₄ = 0.105 cm ² / s. You can use the relative determination of the coefficients of thermal diffusion. In this case, knowledge of the true values of the coefficients K _z and K _{t is} not required. On Fig presents the results of a theoretical calculation of the derivative of the thermodynamic temperature of steel products of famous brand from the one-dimensional heat equation for values of k obtained on the basis of experimental data from Fig.13. There is a good agreement between theoretical and experimental results. Using, for example, the calibration curve from [5], for the obtained values of the coefficients k ₁ and k ₄ we find the hardness of the corresponding sections of the product No. 3 of steel. The obtained hardness values for thin bead-blasting and rough rolling are equal to HV ₁ = 380 and HV ₄ = 480, respectively. The found hardness values correspond to the parameters of the samples used.

Thus, by optically exciting a non-stationary thermal field and additionally irradiating the object with a low-power laser, the values of the function F (T, r) = 1-cor (T, r), a kind of optical analog of thermodynamic temperature, are calculated from the dynamics of speckle images of the sample surface. By measuring the position of the maximum dF (T, r) / dt and its displacement and using the calibration procedure, we can extract the value of the coefficient of thermal diffusion k and determine the mechanical hardness of steel products from it. In our experiments, the maximum displacement dF (T, r) / dt was small and, according to Fig. 12, amounted to two time intervals, which also corresponded to theoretical calculations. Therefore, to increase the reliability of the specified algorithm for measuring the coefficient k, it is necessary to increase the speed of the speckle image registration system. It is also clear that this approach will significantly reduce the error in the results of determining hardness for steel products with a lower value of the coefficient of thermal diffusion k.

Note that in the proposed method, the measured coefficient of thermal diffusion is sufficient to determine the hardness of steel products. For this, additional recalculation of these parameters is necessary, using the empirical relationship between them. It is noted in the literature that these correlation dependencies must be established separately for each steel grade. At present, a general theory has not been developed on the basis of which one can analytically calculate the relationship between the coefficient of thermal diffusion and hardness. Therefore, the approach consists in establishing the empirical dependencies of these parameters and their subsequent application to calibrate the optical measuring device.

The proposed method for determining the hardness of steel has certain advantages in comparison with the radiometric method (prototype) based on the measurement of thermal radiation. In particular, the optical method does not require external thermal insulation of the measuring device, and is also based on the use of cheaper and easier to use optics in the visible range.

Information sources

1. G.V. Bida, A.P. Nichipuruk. Coercimetry in non-destructive testing. / Defectoscopy. 2000. No. 11. S.-1-29.

2. J.A. Garcia, A. Mandelis, B. Farahbakhsh, C. Lebowitz and I. Harris. Thermophysical Properties of Thermal Sprayed Coatings on Carbon Steel Substrates by Photothermal Radiometry./International Journal of Thermophysics, 1999, v.20, N5. p. 1587-1602.

3. W.J. Parker, R.J. Jenkins, C.P. Butler and G. L. Abbott. Flash Method of Determining Thermal Diffusivity, Heat Capacity, and Thermal Conductivity./ Appl. Phys., 1961, v. 32, N.9. p.1689-1684.

4. Y. Liu, N. Baddour, A. Mandelis. Transverse depth-profilometric hardness photothermal phase imaging of heat treated steels. / Appl. Phys., 2003, v. 94, N.9. p. 5543-5548.

5. H. G. Walther, D. Fournier, J. C. Krapez, M. Luukkala, B. Schmitz, C. Sibilia, H. Stamm, J. Thoen. Photothermal Steel Hardness Measerements- Results and Perspectives // Analytical Sciences, 2001, v. 17 Special Issue, p. 165-168.

Claims (1)

A method for determining the hardness of steel products, including local exposure to the product of laser radiation with the power necessary for optical excitation of a non-stationary thermal field, and finding a parameter correlating with hardness of this field parameter, characterized in that the product is additionally irradiated with laser radiation, the dynamics of speckle images are recorded and , dividing them into a series of fragmented sections, calculate the function F (T, r) = 1-cor (T, r),

- correlation function,
I is the intensity of the reflected radiation,
r = (x, y) - coordinates of the point of the speckle image fragment,
T ₁ , T _m - the temperature of obtaining speckle images,
i = 1 ... M is the number of the speckle image fragment,
m = 1 ... N is the number of speckle images received from the object at a temperature T _m ,
ΔS is the area of the i-th fragment of the speckle image,
and from the obtained values from the calibration curves determine the hardness of the selected fragment of the steel product.

Images