DE102017203765B4 - Method for determining mechanical stresses in substrates or circuit carriers that are formed with a ceramic material - Google Patents

Abstract

Method for determining mechanical stresses in substrates or circuit carriers (3), which are formed with a ceramic material, in which
a laser beam emitted by a laser source (2) is directed onto a surface area of a substrate to be examined and electromagnetic radiation reflected and scattered by the irradiated surface is detected in the form of a speckle pattern using a detector (1a) designed for spatially resolved detection; wherein the detection of speckle patterns is detected at a first and at least a second subsequent time, or
a static detection of an image of a detected speckle pattern of at least one area that is illuminated with electromagnetic radiation is checked for its properties and thereby characterized and parameterized if the evaluation is carried out with a correlation function with a gray value transfer matrix, and
the data recorded by the optical detector (1) are fed to an electronic evaluation unit and
in the electronic evaluation unit, a comparison is made with speckle patterns stored in an electronic memory and obtained through a calibration, and the respective mechanical stress of the ceramic material is determined at least in the area of the irradiated surface of the substrate with the comparison and
a substrate (3) which has metallic layers on its surface with which electrical conductor tracks or electrical connection contact points are inspected in an area of the surface of the substrate, the inspection being performed in an area where a metallic layer is formed and/or in an area located immediately next to a metallic layer.

Description

The invention relates to a method for determining mechanical stresses in substrates or circuit carriers that are formed with a ceramic material.

Power electronic assemblies are usually implemented on the basis of ceramic circuit carriers. The circuit carriers, which are generally flat, are provided with structured metallic conductor tracks and contact pads in order to connect the electronic components to one another electrically or thermally and mechanically. Due to the connections that are formed with different materials, mechanical stresses arise in the substrates or carriers formed from ceramic materials, which lead to cracks when overstressed. Overstressing occurs as a result of the manufacturing process in the ceramic-metal composites, during the manufacture of the power modules or as a result of operating loads during use. The cracks usually run in the ceramic and essentially separate the metallization from the ceramic. As a result, defects arise either immediately or at a later point in time, which can result in the functional failure of the assembly. When testing internal mechanical stresses, functional failures during operation of the assembly must be avoided on the one hand and costs saved on the other if the critical ceramic substrates or circuit carriers can be identified and sorted out early in the production value-added process.

The circuit carriers are usually composite materials made of at least one ceramic substrate or circuit carrier (e.g. Al2O3, AlN, Si3N4, etc.) and metallic layers (e.g. Cu, Al, Ni, etc.) formed on surfaces, which form electrical conductor tracks and electrical connection contacts. The layer structure is sintered under pressure and very high temperatures and then cooled. The different physical thermal expansion coefficients of the partner ceramics and the metal lead to mechanical layer stresses, which occur in particular in the joint or interface area. Other structures, e.g. electronic components, heat sink connections or housings, which are assembled, lead to changes in the residual stress states in the joining areas.

Typical circuit carrier types are: DCB - Direct Copper Bondings, AMB - Aluminum Metal Brazing, DAB - Direct Aluminum Bondings.

Very time-consuming and costly tests are currently being carried out in order to obtain information on the reliability of the layer bond between metal and ceramic. With passive cycles and ultrasonic measurements, it is possible to detect cracks that have already formed and to monitor further crack development. In this case, circuit carriers are exposed to cyclic temperature fluctuations in temperature change ovens for several days or weeks. After certain time intervals, the circuit carriers are removed and an ultrasonic test is carried out. The circuit carriers in the water bath are tested using a sequential scanning ultrasonic measuring head. Ultrasonic waves are coupled into the object and the ultrasonic waves reflected at the boundary surfaces of the layered structure are detected and evaluated. Additional echoes of the reflected ultrasonic waves appear when cracks or delaminations have occurred. It is therefore a destructive process, since the cracks have to be created in order to be able to demonstrate increased internal mechanical stresses.

A well-known measuring method for the optical determination of distortions is gray value analysis using digital image correlation. The deformations are recorded using samples that the test object must provide. An evaluation of a distortion that has occurred in the respective pattern leads to a deformation field that can be used to identify mechanical strains. The main disadvantage of the method is that there must be a basic pattern and it can only change slightly within the distortion of the object. There is a risk that the pattern recognition will fail and no deformation field determination can be made. Representatives here are MicroDac or the ARAMIS system. The ARAMIS relies on an additional surface coloring of the test object.

Moire interferometry is known as another highly sensitive, optical method for determining voltage changes. The measuring method is based on a two-beam speckle interferometry. High-resolution interferometric grid patterns (fringes) are used to calculate the stresses. The method calculates the displacement/strain field factor, which is used to convert to stress values. To do this, the method compares the surface condition in an unloaded and loaded condition. Forces must therefore be exerted on an object to be examined which can lead to deformation and thus possible damage to the object examined in each case.

So are off U.S. 7,079,257 B1 a method and apparatus for evaluating mechanical and thermal stresses in electronic Materials, semiconductor materials and other structures are known.

WO 2017/030652 A1 relates to a device and a method in which photothermal speckle examinations are carried out.

From Bing Pan, et al. are from "Two-dimensional digital image correlation for in-plane displacement and strain measurement: a review"; 2009 Meas. science technol. 20 062001 known ways of determining stresses when bending samples.

It is therefore the object of the invention to specify possibilities for the non-destructive determination of mechanical stresses on circuit carriers, which consist of a substrate with layers formed on the surfaces, which form electrical conductor tracks and/or electrical connection contacts for electrical or electronic components, which can easily be carried out in a short time, can be carried out with sufficient measurement accuracy and little effort.

According to the invention, this object is achieved with a method having the features of claim 1. Advantageous refinements and developments of the invention can be implemented with the features specified in the subordinate claims.

In the method according to the invention, the procedure is such that a laser beam emitted by a laser source is directed onto a surface area of a substrate to be examined and, with a detector designed for spatially resolved detection, reflected and scattered electromagnetic radiation from the irradiated surface in the form of a speckle pattern to a first and at least a second subsequent time. The data recorded with the optical detector is supplied to an electronic evaluation unit and a comparison is carried out in the electronic evaluation unit with speckle patterns obtained by calibration and stored in an electronic memory. With the comparison, the respective mechanical stress of the ceramic material is then determined at least in the area of the irradiated surface of the substrate.

If the evaluation is carried out with a correlation function or with a gray value transition matrix (GLCM), a static detection of an image of a detected speckle pattern of at least one area that is illuminated with electromagnetic radiation is possible and its properties are checked and thereby characterized and parameterized. The parameter calculated using one of the equations (2) to (7) is compared using a previously determined calibration function and the value of the respective mechanical stress is determined from the comparison. Static acquisition does not require time-resolved detection and detection at a single point in time is sufficient.

The stored speckle patterns obtained through a calibration should be speckle patterns that have been determined under the same conditions, with the respective mechanical internal stresses having been determined in parallel with a different measurement method and the respective detected during the calibration have been assigned to speckle patterns. Thus, if there is sufficient agreement between the speckle patterns detected by the method according to the invention and the speckle patterns detected during a calibration on a sample of the same or a comparable material, the respective mechanical properties of the sample examined with the invention in the form of a substrate or Circuit carrier are assigned.

During the calibration, the same materials or composites, the same energy inputs and speckle patterns, which were detected and stored at the same times after the energy input, should be taken into account when determining the internal mechanical stresses of the respective sample examined (substrate, circuit carrier). The same lighting conditions and the same settings should also be maintained on the detector designed for spatially resolved detection.

Of course, more than two speckle patterns detected one after the other at a time interval can be taken into account in a time-resolved detection. This can be carried out up to a course, ie a detection of a large number of speckle patterns that are detected one after the other at small time intervals, so that a detection can be carried out in a manner similar to that of a film. When evaluating using GLCM, a single detection at a time is sufficient.

The energy input can take place before or during the detection of speckle patterns in the material of the respective substrate. An energy input can be achieved in particular by elastic deformation of the substrate and/or thermal excitation, for example by irradiation with electromagnetic radiation, which is preferably carried out in pulsed form.

The energy input can preferably by the action of mechanical stress or done by a focused laser beam, with the radiation reflected from the respective surface and impinging on the optical detector, the speckle patterns are also detected. During the time in which the energy input takes place, a power density in the focal spot should be maintained that is greater than during the times when speckle patterns are being detected with the optical detector. With a corresponding beam shaping, in which the size of the area of the focal spot is smaller during the time of energy input, than during the times in which speckle patterns are detected and used to determine the internal mechanical stresses. This can be achieved by simply changing the focal length by a translational movement of at least one optical lens that is arranged in the beam path of the laser beam. The laser power can be kept constant.

A CCD or CMOS camera can be used as an optical detector.

In the invention, substrates that have metallic layers on their surface, with which electrical conductor tracks or electrical connection contact points are preferably formed, are examined in a region of the surface of the substrate. In this case, the examination is carried out in an area in which a metallic layer is formed and/or in an area which is arranged directly next to a metallic layer. In this case, an area directly next to a metallic layer can have a distance of up to 2 mm.

The determination of internal mechanical stresses in the metal-ceramic circuit carriers or purely ceramic substrates is carried out using laser speckle photometry. This is a laser-optical, non-contact and fast measuring method with which geometric distortions of surfaces can be recorded non-contact and non-destructively. A geometric distortion on the surface that can be used here can be initiated by an energy input, which can be caused in particular by mechanical deformation or also by heating. Locally limited heating can be achieved with a laser beam which is emitted by a radiation source and is directed preferably in a focused manner onto a surface of the circuit carrier. A changing speckle pattern forms as a result of the reflection of the laser radiation and the specific surface roughness. It is then a single-beam speckle method without interferometric generation of patterns (fringes). The scattered laser radiation reflected from the irradiated surface with the formed speckle pattern is spatially resolved using a CCD camera as a suitable optical detector and recorded and digitized at specific times. In contrast to digital image correlation, a specific speckle pattern on the surface is not tracked or its deformation observed, but a comparison of at least two completely different stochastic speckle patterns caused by external state changes is carried out based on the spatial relationship of a point to neighboring points. Or the quality of the speckle pattern is taken from a static image of the detected surface area using the image processing logarithms.

1. a) Histogramm -Anisotropie, Entropie
2. b) Grauwertübergangsmatrix - Energie zweiter Ordnung, Kontrast, Homogenität, Korrelation
3. c) Bildintensität - Mittelwert, Standardabweichung
4. d) Differenz-Korrelationsfunktion (Aussagen über den zeitlichen Zusammenhang der stochastischen Intensitätssignale)

With the detected speckle patterns and their change due to external changes in state, systematic conclusions can be drawn about the internal mechanical stresses, at least in the material of the circuit carrier. This requires calibrations that take into account a correlation between the respective internal mechanical stress and the spatially resolved detected laser speckle signals. The spatially resolved and digitized speckle pattern signals can be evaluated using various image processing algorithms. Suitable image processing logarithms are:

1. a) Histogram -anisotropy, entropy
2. b) Gray level transition matrix - second order energy, contrast, homogeneity, correlation
3. c) Image intensity - mean, standard deviation
4. d) Difference correlation function (statements about the temporal relationship of the stochastic intensity signals)

The four image processing algorithms mentioned that are suitable for this purpose are explained in more detail below.

wobei nmax die maximale Anzahl von mit dem zur ortsaufgelösten Erfassung ausgebildeten Detektor, τ die Zeitverschiebung oder die Anzahl der Rahmenintervalle, S (n, x, y) die Intensität eines jeweiligen Bildpunktes (Pixels), dessen Position durch die Koordinaten x und y im n-ten Rahmen ist. Diese Korrelationsfunktion ergibt sich aus dem sogenannten Semivariogramm, bei dem es sich um ein geostatistisches Werkzeug handelt, mit dem die Beziehung zwischen ortsaufgelöst detektierten Intensitäten in Abhängigkeit von Abstand/Verschiebung zu einem Zeit- und/oder Raumpunkt untersucht werden kann. Das Ergebnis dieser Korrelationsfunktion bei den Zeitpunkten zeigt die Akkumulation der Intensitätsdifferenz von Intensitätspaaren einer zeitlichen Folge. Bei der Erfassung mechanischer Spannungen konzentriert sich die Korrelationsfunktion jedoch auf den Unterschied zwischen der ersten und anderen detektierten Abbildungen eines beleuchteten Oberflächenbereichs in Abhängigkeit von der Zeitverschiebung. Daher wurde eine geringfügige Änderung der Gl. (8) vorgenommen, wobei n zu einem festen Wert n = 1 festgelegt wird. Der Exponent wurde als 1 gesetzt, um die Richtungsänderungen von mechanischer Spannung zu verfolgen. Die neue Korrelationsfunktion ist gegeben durch: $$C\left(\mathrm{\tau }\right)=\frac{1}{N\langle I\rangle }\langle I\left(1+\tau ,x,y\right)-I\left(\mathrm{1,}x,y\right)\rangle$$

wobei N der Anzahl der detektierten einzelnen Abbildungen und I (n, x, y) die Intensität eines jeweiligen Bildpunktes (Pixels) entspricht.The dynamic evaluation can be carried out on the basis of the evaluation of surface area images of an illuminated surface area detected successively at a number of points in time. The intensity detected in a time-resolved manner can be examined for its statistical properties using the correlation function (1). $$C\left(\tau \right)={\displaystyle \sum _{n=1}^{n_{max}}{\left(S\left(n+\tau ,x,y\right)-S\left(n,x,y\right)\right)}^2}$$

where nmax is the maximum number of detectors designed for spatially resolved detection, τ is the time shift or the number of frame intervals, S (n, x, y) is the intensity of a respective Pixels whose position is given by the x and y coordinates in the nth frame. This correlation function results from the so-called semivariogram, which is a geostatistical tool that can be used to examine the relationship between spatially resolved detected intensities as a function of distance/displacement from a point in time and/or space. The result of this correlation function at the points in time shows the accumulation of the intensity difference of intensity pairs in a time sequence. However, when detecting mechanical stresses, the correlation function focuses on the difference between the first and other detected images of an illuminated surface area as a function of the time shift. Therefore, a slight change in Eq. (8) where n is set to a fixed value n=1. The exponent was set as 1 to track changes in direction of stress. The new correlation function is given by: $$C\left(\mathrm{\tau }\right)=\frac{1}{N〈I〉}〈I\left(1+\tau ,x,y\right)-I\left(\mathrm{1,}x,y\right)〉$$

where N corresponds to the number of individual images detected and I (n, x, y) corresponds to the intensity of a respective picture element (pixel).

• Energie:
  
• Homogenität:
  
• Kontrast:
  
• Korrelation: $$Correlation=\frac{{\displaystyle \sum {}_{i,j=0}^{width}\left(i-u_x\right)\left(j-u_y\right)c_{ij}}}{s_x{s}_y}$$
  
  mit $$-1\le \mathrm{<figure-callout\; id="1"\; label="R\; \le "\; filenames="DE102017203765B4\_0016.png,DE102017203765B4\_0017.png"\; state="\{\{state\}\}">R</figure-callout>}\le 1$$
  

In addition to evaluating the correlation function of the time signal of the stochastic signal, the detection of speckle patterns of the respective sample can be characterized using image processing methods for texture analysis (cooccurrence matrix or gray value transition matrix) and used to evaluate the respective mechanical stress state of the respective substrate. This includes the following parameters:

• Energy:
  
• Homogeneity:
  
• Contrast:
  
• Correlation: $$CO\mathrm{right}\mathrm{right}elatiOn=\frac{{\displaystyle \sum {}_{i,j=0}^{wi\mathrm{i.e}tH}\left(i-{\mathrm{and}}_x\right)\left(j-{\mathrm{and}}_y\right)c_{ij}}}{s_x{s}_y}$$
  
  With $$-1\le R\le 1$$
  

wobei h(i) die Häufigkeit und g der Grauwert sind.In addition, the entropy of the histogram of a static image can be evaluated according to equation (10):

where h(i) is the frequency and g is the gray value.

The spatially resolved detected signals, which represent the respective detected speckle pattern, can be linked to mechanical variables that occur on an examined substrate or circuit carrier in the irradiated and detected surface area. Mechanical variables can be forces, moments and deformations. Mechanical secondary variables, in particular mechanical stresses and strains, can be determined by means of numerical calculations, into which knowledge of the mechanical material properties of the ceramic to be tested is incorporated. In a final calibration function, the respective mechanical stresses and one of the above signal variables can be brought into a mathematical relationship. This makes it possible to determine the internal mechanical stresses on any circuit carrier that consists of a metal-ceramic compound. The respective ceramic material of the respective substrate or circuit carrier examined should be the same or a material of the same type of ceramic as was taken into account in the calibration function.

The invention represents a non-destructive, non-contact optical method and an application and evaluation methodology with which mechanical internal stresses that have occurred in a substrate or circuit carrier formed with a ceramic material can be determined. As a result, tests can be carried out in every process step and at any time during the processing of the ceramic circuit carriers with regard to residual stresses that have arisen. Errors and defects that have occurred can therefore be recognized immediately during the manufacturing process in a short time. In particular, circuit carriers that have not been processed or only slightly processed, in which none or only some of the components have already been mounted, can be sorted out, which significantly reduces the costs can. It is even possible to detect errors in the manufacturing process and counteract them by adjusting process parameters, for example changing the temperature profile. Metal-ceramic circuit carriers or ceramic substrates can be determined after their production and after each subsequent process step with their respective mechanical residual stress state.

Since metal-ceramic substrates or other ceramic substrates can be tested non-destructively with regard to their internal mechanical stress, a tested circuit carrier can be returned to the manufacturing process. So it can be avoided as a result of a test.

The optical measurement does not require any tools, such as a water bath for ultrasonic microscopy or surface staining for ARAMIS. This means that the respective tested circuit carrier remains free of contamination. In addition, mechanical stresses and internal stresses in the material can be determined without damage occurring, for example in the form of a crack. It is therefore possible to evaluate the state of a substrate or a circuit carrier, with which preventive detection of possible failures or defects that may develop later can be achieved with sufficient certainty.

The method can be carried out with sufficient accuracy in a measurement in a very short time. In this way, a circuit carrier or substrate can be measured and checked in a period of < 15 s, which is conducive to use directly during the production process.

Using a similar setup, other speckle pattern parameters can be calculated to construct the calibration function.

The measurement setup for determining the mechanical stress can differ in the type of activation/excitation. In this case, forces can act at a distance from the irradiated surface area on the substrate or the circuit carrier. The surface changes as a result of a mechanical force (load) or a change in temperature. The speckle parameters, calculated according to equations (2-7), capture these changes over time.

The position and brightness (intensity) of the speckles change over time, due to the changes in the expansion fields of the surface during the propagation of the locally introduced energy (deformation energy, heat). The surface condition and thus the specifically detected speckle images are functions of the strain fields εi,j(t,R) (t - the time, R the location of the measurement and i,j - the coordinate axes x,y in the detected and influenced by energy input surface area). These in turn are linked to the temperature T(t,R) via equation (8) of thermal expansion and to the mechanical stresses σi,j(t,R) via the elastic laws: $$e_{x,y}\left(t,R\right)=\frac{{\sigma }_{k,l}\left(t,R\right)}{E_{x,y,k,l}}+a_{x,y}\left(T\left(t,R\right)-T_0\right)$$

The mechanical strain is determined by Hooke's law with mechanical stresses σi,j(t,R) and the modulus of elasticity E and the relative change in length of the respective sample (substrate, circuit carrier) due to thermal expansion over the temperature range (T(t,R)-T0 ) with a known coefficient of thermal expansion α.

Equation (8) can be summarized as follows to set up the calibration function: $$Pa\mathrm{right}amete{\mathrm{right}}_{LSP}\left(t\right)=a\cdot \frac{\sigma \left(t\right)}{E}+b\cdot a\cdot \Delta T+c$$

und durch Definition der Parameter P1 - P3: $$P_1=\frac{E}{a}$$

$$P_2=-b$$

$$P_3=-c$$

zu Gleichung (11) vereinfacht. $$\sigma \left(t\right)=P_1\left(Paramete{r}_{LSP}\left(t\right)+P_2\alpha \Delta T+P_3\right)$$

For the stress change calculation, equation (9) was converted into a suitable form: $$\sigma \left(t\right)=\left(Pa\mathrm{right}amete{\mathrm{right}}_{LSP}\left(t\right)-b\ast a\ast \Delta T-c\right)\ast \frac{E}{a}$$

and by defining the parameters P1 - P3: $${\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="DE102017203765B4\_0016.png,DE102017203765B4\_0017.png"\; state="\{\{state\}\}">P</figure-callout>}}_1=\frac{E}{a}$$

$${\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="DE102017203765B4\_0016.png,DE102017203765B4\_0017.png"\; state="\{\{state\}\}">P</figure-callout>}}_2=-b$$

$${\mathrm{<figure-callout\; id="3"\; label="P"\; filenames="DE102017203765B4\_0016.png,DE102017203765B4\_0017.png"\; state="\{\{state\}\}">P</figure-callout>}}_3=-c$$

simplified to equation (11). $$\sigma \left(t\right)=P_1\left(Pa\mathrm{right}amete{\mathrm{right}}_{LSP}\left(t\right)+P_{2}a\Delta T+P_3\right)$$

The variables P1, P2 and P3 are initially unknown and must be determined using a suitable calibration method.

The invention can therefore be used to determine the internal stress in metal-ceramic substrates in electronics (hybrid electronics, power electronics) but also to determine the internal stress in monoceramic layers (e.g. membranes, covering shells).

The invention will be explained in more detail below by way of example.

• 1 in schematischer Form beispielhaft einen Aufbau mit dem eine Laser-Speckle-Photometrie (LSP) während einer Messung an einem Schaltungsträger DCB durchführbar ist;
• 2a eine Darstellung zur Realisierung einer Kalibrierung unter Anwendung der Laser-Speckle-Photometrie-Technik und einer Biegevorrichtung zur elastischen Verformung;
• 2b eine Datenverarbeitung, in dem die detektierten LSP-Signale parametrisiert und die mechanische Belastungsangaben für die Durchführung von FEM-Berechnungen zur Eigenspannungsbestimmung an der Messposition verwendet werden.
• 2c eine Aufstellung von Kalibrierungsfunktionen anhand der ermittelten Parameter bzw. Eigenspannungswerte aus einer FEM-Berechnung.
• 3a ein Beispiel der umgerechneten Spannung aus der LSP-Messtechnik, entnommen aus dem Biegeversuch an Aluminiumoxid, im Zusammenhang mit der ermittelten Eigenspannung aus FEM-Berechnung und
• 3b eine Darstellung der ermittelten relativen Fehler in % zwischen der mechanischen Eigenspannung aus FEM-Berechnung und der ermittelten Eigenspannung aus LSP.

show:

• 1 in schematic form, by way of example, a structure with which laser speckle photometry (LSP) can be carried out during a measurement on a circuit carrier DCB;
• 2a a representation of the realization of a calibration using the laser speckle photometry technique and a bending device for elastic deformation;
• 2 B data processing in which the detected LSP signals are parameterized and the mechanical load information is used to carry out FEM calculations to determine the residual stress at the measuring position.
• 2c a list of calibration functions based on the determined parameters or residual stress values from an FEM calculation.
• 3a an example of the converted stress from the LSP measurement technology, taken from the bending test on aluminum oxide, in connection with the residual stress determined from the FEM calculation and
• 3b a representation of the determined relative error in % between the mechanical residual stress from FEM calculation and the determined residual stress from LSP.

In this case, a circuit carrier 3, here a conventional DCB ceramic, is illuminated by means of the laser diode as the radiation source 2. A CCD camera as a detector 1 designed for spatially resolved detection with detector 1a records the speckle pattern at the measurement position in an irradiated surface area and is connected to an electronic evaluation unit, in this case a personal computer.

A region of the surface of the circuit carrier 3 is illuminated with the radiation source 2 and electromagnetic radiation reflected and/or scattered from there is detected with the detector 1a in a spatially resolved manner. As 2 B it can be seen that individual detected speckles change in terms of their position and intensity of the electromagnetic radiation reflected and/or scattered by them. From these changes and also from the change in direction of the individual positions of speckles after energy has been introduced into the material of the respective circuit carrier 3, statements about the internal mechanical stresses on a circuit carrier 3, as a particularly interesting example of a ceramic substrate, can be obtained.

For the calibration test, the circuit carrier 3 is placed on a support 5 and loaded with a movable fin 4 and thereby elastically deformed. The change in the speckle pattern is detected using the camera as detector 1a. The spatially resolved intensities from the force device and the LSP technique are then used to calculate the parameters P1, P2 and P3 according to equation (11). The determined parameters P1 to P3 are used in the back calculation on the same materials to determine internal mechanical stresses at the examined measurement position within the illuminated surface area of the circuit carrier 3.

Claims (5)

Method for determining mechanical stresses in substrates or circuit carriers (3), which are formed with a ceramic material, in which a laser beam emitted by a laser source (2) is directed onto a surface area of a substrate to be examined and is used with a detector ( 1a) electromagnetic radiation reflected and scattered by the irradiated surface is detected in the form of a speckle pattern; wherein the detection of speckle patterns is detected at a first and at least a second subsequent time, or a static detection of an image of a detected speckle pattern of at least one area that is illuminated with electromagnetic radiation is checked for its properties and thereby characterized and parameterized , if the evaluation is carried out with a correlation function with a gray value transfer matrix, and the data recorded with the optical detector (1) is supplied to an electronic evaluation unit and a comparison is carried out in the electronic evaluation unit with speckle patterns stored in an electronic memory and obtained through calibration and with the comparison, the respective mechanical stress of the ceramic material is determined at least in the area of the irradiated surface of the substrate and a substrate (3), which has metallic layers on its surface, with which n electrical conductor tracks or electrical connection contact points are formed, are examined in an area of the surface of the substrate, the examination being carried out in an area in which a metallic layer is formed and/or in an area which is arranged directly next to a metallic layer, is carried out. procedure after claim 1 , characterized in that before or during the detection of speckle patterns, energy is introduced into the material of the respective substrate, in particular by elastic deformation of the substrate or on the substrate (4, 5) or by irradiation with electromagnetic radiation. Method according to one of the preceding claims, characterized in that a CCD or CMOS camera (1) is used as the optical detector. Method according to one of the preceding claims, characterized in that an image processing algorithm selected from an image intensity, histogram, gray value transition matrix evaluation, image intensity analysis and/or a difference correlation function is carried out in the electronic evaluation unit. Method according to one of the preceding claims, characterized in that the same materials or material composites, the same energy inputs and speckle patterns that have been detected and stored at the same times after the energy input, as well as the same lighting conditions and the same settings on the detector designed for spatially resolved detection (1a) are taken into account and adhered to when determining the mechanical residual stresses of the respective substrate or circuit carrier (3) examined and the calibration.

Images