DE10065277A1 - Internal stress analysis method for metallic material e.g. for turbine blade, involves limiting scattering volume of photon beam, such that beam is reflected from every crystal lattice of irradiated target portion - Google Patents

Abstract

A target (12) is rotated about the radiation axis of high energy photon beam to meet certain diffraction conditions for detecting Bragg reflection. Scattering volume of high energy beam from the target is limited, such that the beam is reflected from every crystallite of the irradiated target portion. Internal stress of target is detected based on width of the beam and measured angle.

Description

description

The invention relates to a method for analyzing Eigen tensions in metallic materials where high Get photons onto a target from the metallic plant and the crystal lattice of the factory scattered photons by a detector who the, being from the detected width and the Bragg reflections and / or the measured angle derives the residual stresses are cash.

Especially in the course of materials technology, these are the me inherent residual stresses, whether in the crystalline structure of the material itself or e.g.  for welds, the metallic workpieces or work combining fabrics of extraordinarily high scientific and practical interest for durability statements in with regard to heat influence and / or deformation of the work under load. Internal stresses are internal forces and Moments in the material, the property changes, in particular which can cause in the near-surface area. Own stresses are basically static multi-axis Tensions that often have the same directional character such as have main load voltages. Residual stresses are divided into 1st, 2nd and 3rd order internal stresses. Depending on the size of the material areas, this can result the maximum residual stress values as an overlay of egg gene voltages of the first and higher order can be drafted. It is known that in a heterogeneous metal structure Maximum values of the internal stresses are a multiple of the internal ones can achieve 1st order voltages. The origin matters of residual stresses can correspond to the three Main groups of materials, manufacturing and stress thing.

It is therefore for scientific, technological and practical questions of extraordinary importance accuracy, precise information about the internal stresses in metal materials.

Residual stress investigations of metallic materials have been in until recently and a. using neutro NEN carried out by means of so-called thermal neutro with wavelengths between approx. 0.1 nm and 0.3 nm the neutron scattering studies can be done for each  Measurement scattering angle of approx. 90 ° can be set, leading to in about the same width and the same depth leads. Neutron scattering studies are done on neutrons beam cross sections from typically 0.5 mm × 20 mm to approx. 2 mm × 50 mm. The systemic big ones Neutron beam cross sections have the disadvantage that changes changes in properties in workpieces, for example Weld away across, perpendicular to the beam direction, by means of Neutrons only allow a suitable local triggering to a limited extent sen. In addition, a suitable one must be used for such examinations Neutron source in the form of a reactor hen what besides the actual reactor also a high degree of security and safety measures.

For a few years now, high-energy synchrotron beams have been used lung, for example, photons with an energy of approx. 40 keV and some 100 keV, for residual stress investigations used (e.g. at HASYLAB (DESY / Hamburg) and at the ESRF (Grenoble)). The high-energy synchrotron radiation penetrates metallic work with only minor losses fabrics up to a few centimeters thick, for example aluminum um and titanium alloys. Stress analysis by metallic materials are distinguished from Untersu by means of neutrons in that the measurements perform with a much finer measuring beam let, typical values of a photon beam cross are for example 10 µm × 10 µm to 50 µm × 100 µm. There due to changes in the internal stresses in the material fen, perpendicular to the beam direction and, for example, over a Weld away with high-energy photons  much finer spatial resolution than determined with neutrons become. A disadvantage when determining residual stresses in metallic materials using high-energy pho However, toning is that only very small scattering angles occur that lead to a very stretched extension of the Scattering volumes lead to a much poorer lower resolution of the residual stress distribution in beam direction tion as perpendicular to it. The scatter volume (gauge-volume), this is referred to as part of the target or sample volume is the volume out of which is in front of each given arrangement of primary beam and aperture photons can get into the detector.

The scatter volume is used in investigations to determine the Residual stresses of metallic materials using high-energy table photons with apertures in front of the material target and confined behind the material target, specifically for loading limit the primary beam and limit the divergence of the reflected beam and to localize the litter volume. It has been shown that at 111- Lattice reflections from pure aluminum or an aluminum alloy alloy as a target for measurements with 100 keV photons average scattering angle of approximately 6 ° occurs and at egg ner typically chosen aperture width of 30 µm and at Expe Rims of this type are typical aperture distances of 260 and A typical scatter volume of 1,160 mm behind the target 0.68 mm is obtained. This for experiments with high typical photon scattering behavior shows that the Length of the scattering volume the beam width by more than one Order of magnitude. This resolution is for many scientific and technological questions not  sufficient.

It is therefore an object of the present invention to provide a ver to train driving of the type mentioned at the beginning, that achieved a much better resolution than before becomes and changes of internal stresses substantially free ner than previously possible can be determined and a we significant improvement in the distinction between Eigen tensions of the first, second and third kind is possible also a variation of the texture in small target or sample areas can be detected and the method essentially based on previous measurements ser type carried out by means of high-energy photons can be.

The object is achieved according to the invention in that the Scattered volume of the high-energy photons inserted in this way is limited that a separation of single crystallites of the metallic material is made possible.

The advantage of the solution according to the invention is essentially Lichen in that by the solution, d. H. through the inventions Restriction of the scattering volume of the high-energy table photons compared to previous ones, in the state of Technology known scattering geometries a much better localization of the spreading volume itself is achieved, d. H. the single crystallite can in the beam direction with egg with an accuracy of <0.1 mm. Also the The width of the Bragg reflexes can thus be much more accurate than have so far been determined so that taking advantage of the inventions completely new perspectives for Eigen  stress analysis of metallic materials are possible. The Analyzes are more precise and of higher quality because different internal stresses of different degrees can be.

Analy carried out by means of the generic method are subject to the restrictions listed above due to the scattering geometry used up to that point the resulting limitations and disadvantages another limitation, namely a sufficient number Crystallites in the target in a reasonable measuring time with suitable to find orientation. This limitation results from the fact that the measurements so far only with one location-insensitive detector could be performed. To remedy this disadvantage, it is advantageous to the inventive method in the beam path of the Target reflected photons and the detector one as Insert ring-shaped aperture.

It is also advantageous that the ringför moderate aperture passed through with a local temp detect sensitive area detector, so that by both Measures, namely the detector as a location-sensitive area train detector and the aperture as a ring aperture form, so that a raster measurement (scan) then by a Grid measurement (scan) of the aperture system in the beam direction (X direction) can be carried out, with this Wei almost all reflections in an elongated scattering volume, whose cross-section corresponds to that of the "gauge volume" can be detected by the detector.  

Basically, you can finally choose two different ones Assign all crystallites in the scattering volume in reflex posi tion and thus determined. It can be done before be particulate, the target relative to the beam axis of the high-energy photons in at least one freedom degree to fulfill the diffraction conditions for detection the Bragg reflexes turn, then from all crystal some reflexes are briefly oriented so that they meet the diffraction conditions and become a reflex on the respective Debye-Scherrer ring. In the Rotation of the target becomes the scattering volume within the Sample rotated. All crystallites in a previously fixed Partial volumes placed in the sample are recorded by many Measurements are carried out one after the other in a grid pattern and from grid measurement to grid measurement the target in the Rich perpendicular to the primary beam and the axis of rotation tion is shifted by a distance that is approximately the pulp te and corresponds to the diameter of the primary beam.

The other type, the crystallites in the scattering volume in reflex to position it is that advantageously the Analysis of residual stresses at different energies of the high-energy photon beam is carried out. This is equivalent to the analysis of the Eigen voltages at different wavelengths of the photons of the high-energy photon beam is carried out. These two variants are preferably at high energy table photons used, for example, by synchrotron radiation in elementary particle accelerators (DESY, CERN) be generated. A continuous change in energy gie or  the wavelength λ of the primary beam does not require a rotati on the target. At such a facility for generation high-energy photons can be measured by a λ grid be performed. Here λ is continuously changed and In this way, many crystallites in the scattering volume are briefly stimulated to one or more reflexes, if the λ grid measurement has a sufficiently large range covers.

The invention will now be described with reference to the accompanying schematic drawings based on an exemplary embodiment les described. In it show:

Fig. 1 a diffusing geometry as haemostasis previously for self-clamping was used by means of high-energy photons (prior art),

Fig. 2 is a schematic representation of the inventive SEN scattering geometry tone by means of high-energy Pho,

Fig. 3 is a schematic representation of residual stresses 1st, 2nd and 3rd type (according Macherauch and Kloos)

Fig. 4 shows the analysis of the 111 reflex of a crystallite in pure aluminum at different temperatures and

Fig. 5 is a schematic representation of a beam path of high-energy synchrotron radiation, which is set up for residual stress measurements in the scattering geometry according to the invention with ring diaphragms and surface detectors.

Reference is first made to the representation of FIG. 1, in which an examination / analysis arrangement 10 is shown schematically, as was used in previous self-stress analyzes using high-energy photons 13 . By means of an aperture 15 (A ₂ ), a cross section of the beam of high-energy photons 13 is set to, for example, 10 μm × 10 μm to approximately 50 μm × 100 μm. In the beam path of the high-energy photons 13 , a target 12 made of a metallic material to be examined is arranged. Behind the target 12 is to limit the divergence of the reflected photon beam 19 and to localize the scattering volume an aperture 17 (A ₃ ) and at the greatest possible distance X ₃ from the target 12 and with the width a ₃ and an aperture 16 (A ₄ ) in the smallest possible From X _{4 was provided} by the target 12 with a width a ₄ . With the simplified assumption of diaphragms of the same width (a = a ₂ = a ₃ = a ₄ ), the following designation for the length of the scattering volume L _gv and the divergence δ _{2 θ} , _sg of the photon beam passing through the detector diaphragm 17 is obtained in a good approximation 19th The high-energy photon beam 13 can be both monochromatic and white. The scattering angle between the primary high-energy photons 13 and the photons 19 reflected on the target 12 is denoted by 2θ _sg . This angle is formed by the primary beam of high-energy photons 13 with the photon beam 19 reflecting on a grating plane, for example the 111 plane of a metallic target. This angle is generally very small and is, for example, only ~ 6 ° for 100 kev photons when these are reflected on a 111 plane of an Al alloy.

The numerical example

δ _{2 θ} , _sg = 2a / (X ₃ -X ₄ ) and
l _gv ~ 2a / (2θ) + x ₄ δ _{2 θ , sg} / (2θ)

shows the dimensions that occur in experiments with high-energy photons in the examination / analysis arrangement 10 shown in FIG. 1. With reflections in the 111 plane of pure aluminum and an Al alloy, an average scattering angle of 2θ ~ 6 ° occurs when measuring 100 kev photons. If an aperture width of a = 30 microns is assumed and it is still assumed that the apertures 17 (A ₃ ) and 16 (A ₄ ) are mounted at distances of X ₃ = 260 mm and X ₄ = 1660 mm from the target 12 , you get a length of the scattering volume of l _gv ~ 0.68 mm.

This scattering behavior, which is typical for investigations with high-energy photons 13, shows that the length of the scattering volume l _gv exceeds the beam width by more than an order of magnitude, ie; that in general even with very small lens values of 1 _gv <0.5 mm can be achieved. The arrangement with the known in the prior art examination / analysis. FIG. 1 achievable resolution is not sufficient therefore for many scientific and technological issues.

In the method presented here for analyzing residual stresses in metallic materials, the distinction between residual stresses of the 1st, 2nd and 3rd types is the goal, cf. also the representation acc. Fig. 3. In simple terms, the internal stresses of the

• 1. Kind of over many crystals in general lite averaged residual stresses that only over relatively large distances, for example over a distance Weld away, change the
• 2. Kind averaged over about a grain Tensions, because from grain to grain can be strong Changes in the voltage values that occur the edges of crystallites in most plant strong gradients of mechanical properties let appear and the
• 3. Kind that only over parts of Kristal liten averaged residual stresses caused by star ke stress on a material in its manufacture position or during its use can or as a result of creep processes that Relocation networks have been set up ha ben. As a result, the angular width in particular the reflexes impaired.

The distinction between residual stresses of the 1st, 2nd and 3rd types is difficult if the examinations or measurements are carried out with an examination / analysis arrangement 10 , as described above in connection with FIG. 1. Reflections of different crystallites, each with slightly different angles 2θ and angle widths δ _{2 θ of} the reflections, will then overlap and the measured intensity in the angular range (2θ _sg -δ _{2 θ sg} / 2, 2θ _sg + δ _{2 θ sg} / 2) contribute.

An examination / analysis arrangement 11 according to the invention is shown in Fig. 2, to which reference is now made. The same reference numerals and other identical information in FIG. 2 correspond to those which have been set forth in connection with the description of FIG. 1. Since the separation of Bragg reflections of different crystallites is the actual goal for determining the residual stresses, it is assumed that the position of the crystallites is the inside of the target 12 . The inventive method was carried out with a very well monochromatized and highly colli mated beam of high-energy photons with a divergence of the primary beam of ~ 10 ^-5 rad. The distances between the aperture 16 (A ₄ ) and the aperture 17 (A ₃ ) from the target 12 were X ₃ = 260 mm and x ₄ = 1.66 m, where the beam height through the apertures 15 (A ₂ ) and 17 (A ₃ ) in front of the target to ~ 0.03 mm and in front of the detector 18 (not shown in FIG. 2), to 1 mm. On an approx. 3 mm thick Al sample as target 12 , Bragg reflections were determined and measured in four grid measurements (grid measurement = measurement series = scan). It applies that each crystallite has a certain characteristic Reflexionsver hold.

Scan-1 was carried out to turn a crystallite into the reflecting position. In this scan, the target aperture was open to approximately 1 mm and the detector aperture to approximately 5 mm in width. In Scan-1, the target 12 was rotated about the vertical ω axis by approximately 1 ° in steps of approximately 0.02 °. A pronounced reflex _{maximum was generally} found at an angular position ω = ω _max . The reflex is caused by only one (or a few) crystallites that have been brought into reflex position in Scan-1.

Following Scan-1, ω = ω _{max was} set and the target aperture A ₂ was narrowed to a width of a ₂ ~ 30 µm. This was followed by a second scan of the target 12 in the horizontal direction perpendicular to the primary beam (y direction), and the target region 12 irradiated in scan-1 was scanned. Maximum intensity was found when the crystallite standing at ω = ω _max in reflection position was irradiated, its width and position y _s = y _{3, max} was determined (usually only a pronounced maximum occurred).

Then the target 12 was moved to the position y _s = y _s.max and the detector aperture A _{3 was} scanned in the horizontal (y) direction in front of the detector. In this way, the position y ₃ = y _{3, max} and the width of the reflection in the detector plane could be determined very precisely.

For the last scan, the detector aperture was opened again and an edge of aperture A ₄ was scanned by the reflex. This was weakened by 50% at y ₄ = y _4.50% .

In this way, for the angle of the Bragg reflex 2 θ _rb ~ tdn (2θ _rb ) = (y _{3, max} - y _4.50% ) / (X ₃ - X ₄ )

X _cryst = X ₃ - y _4.50% / tan (2θ).

For the errors in the determination of the angle Szerb and the localization of the crystallite in the beam direction SXcrys one obtains

δ _{2 θ rb} ~ ((δy _{3, max} ) ² + (δy _4.50% ) ^{2) 1/2} / (X ₃ - X ₄ )
δx _crys ~ ((x ₄ δ _{2 θ rb} ) ² + (δy4.50%) ^{2) 1/2} / tan (2θ _rb ).

In Fig. 4a and 4b examples of scans 3 and 4 are performed, which were measured at the ( 111 ) reflex of an oriented crystal in a pure Al target 12 at 19 ° C and 49 ° C. This temperature increase leads to an increase in the lattice spacing d by only δd / d ~ 6.9 × 10 ^-4 . Very pronounced shifts of y _{3, max} and y _{4.50% were} measured. The errors of y _{3, max} and y _4.50% were estimated in the present case at δy _{3, max} <0.02 mm and δy _4.50% <0.01 mm ( Fig. 4a and 4b). With these values one obtains δ _{2 θ rb} / (2θ) <1.5 10 ^-4 and δx _crys <0.1 mm (with x ₃ = 260 mm, x ₃ - x ₄ = 1.4 m).

Thus, with a very high measuring accuracy of δ _{2 θ rb,} the position of the crystallite parallel to the beam direction could be determined much more precisely (δx _crys <0.1 mm) than is the case in conventional scattering geometry, where the accuracy of the localization in the beam direction coincides that corresponds to the length of the spreading volume (l _gv ~ 1 mm).

A restriction of analyzes that are carried out with the examination / analysis arrangement 10 according to Fig. 1 be carried out, it still gives itself from it to be able to find a sufficient number of crystallites in a reasonable measuring time with a suitable orientation. The limitation results from the fact that the measurements could previously only be carried out with a location-insensitive detector. According to the invention, the diaphragm 16 (A ₄ ) is formed by an annular diaphragm and the photons which have passed through it, ie that of the reflected photon beam 19 , are placed on a position-sensitive area detector. The above grid measurement (scan 4), cf. Fig. 2, is then replaced by a raster measurement (scan) of the diaphragm system in the beam direction (x direction). In this way, almost all reflections from an elongated scattering volume, the cross section of which corresponds to that of the scattering volume 14 of FIG. 1, are detected on the detector 18 . Finally, all crystallites in the scattering volume 14 can be brought into the reflex position in two different ways.

The sample is rotated around one or more axes (e.g. ω, Fig. 2). Then crystal planes in all or almost all crystallites are briefly oriented so that they meet the diffraction conditions and contribute to a reflection on the respective Debye-Scherrer ring. With the rotation of the target 12, the scattering volume 14 is also rotated within the Tar gets 12th All crystallites in a previously defined partial volume of the target 12 are finally captured by performing many grid measurements one after the other and from grid measurement to grid measurement the target 12 is shifted in the direction perpendicular to the primary beam 13 and to the axis of rotation by a distance which is approximately corresponds to the width or the diameter of the primary beam 13 .

In a device which generates high-energy synchrotron radiation and which enables a continuous change in the energy or the wavelength λ of the primary beam 13 , it is also possible to carry out a λ grid measurement. Here, λ is continuously changed and in this way many crystallites in the scattering volume 14 are briefly stimulated to one or more reflections if the λ grid measurement covers a sufficiently large area.

In the two measurements shown above, scattering intensities have to be measured with a large number of target positions. This can also be carried out, since only short measuring times of 1 s each are required for each individual measurement. However, this requires the use of location-sensitive surface detectors for optimal process management, which can be read out in less than 1 s. So-called gas detectors (multi-wire chambers or micro-strip proportional counters) with a downstream data acquisition system are proposed. Such Meßanord voltage enables the implementation of continuous grid measurements, in which, for. B. the target 12 is rotated sufficiently slowly, but with constant Ge, about the already mentioned ω axis. During the rotation, the adjustment unit supplies one or more signals to the data acquisition system mentioned, so that it is able to position the photons detected at locations (y _i , z _i ) at an interval δω _i = (ω _i - δω, ω _i + δω). The intervals δω _i and the rotational speed are determined before the measurement so that the change in the scattering intensity is detected with sufficient resolution and intensity. Raster measurements can also be carried out for aperture parameters in an analogous manner. By means of the detectors designed as gas detectors, the possibility of energy discrimination has also been created, which was not possible as detectors 18 in the previous so-called image plate and CCD systems. With the use of gas detectors, higher-order photons and those that have lost energy through Compton scattering can be discriminated against those that have been scattered purely elastically.

In the form of gas detectors in the form of gas detectors 18 in the examination-analysis arrangement according to the invention, another possibility of using the method according to the invention for analyzing residual stresses opens up. There are thus z. B. stroboscopic measurements possible, which can be carried out for in-situ analyzes on moving components. So it is conceivable to analyze residual stresses occurring under load in running machines (e.g. motors, turbine blades). In such measurements, the rotating system sends a signal to a data acquisition system when a certain position has been passed. The data acquisition system can record the smallest time intervals of ~ 1 µs, so that the position of the components to be examined is known very precisely even with fast-running systems. From the scattering intensity integrated in many runs, the residual stresses are then determined depending on the location of the moving component. For example, a turbine blade could be shone through at the level of its foot. The foot moves through the beam at high speed, whereby fast data acquisition enables a quick grid measurement. In one cycle, the measured intensity will not be sufficient for an interpretation, since in this time there is generally only one measuring time of δτ «1 ms per volume element in the foot. The integration of many grid measurements over a longer period of time ultimately enables reliable residual stress analyzes.

Reference list

10th

Examination / analysis arrangement (state of the art)

11

Examination / analysis arrangement (invention)

12th

Target

13

High energy photons / photon beam / primary beam

14

Scatter volume (gauge-volume)

15

cover

16

Aperture / ring aperture

17th

cover

18th

detector

19th

Reflected photon beam

Claims (8)

1. A method for analyzing residual stresses in metallic materials, in which high-energy photons are directed onto a target made of a metallic material and the scattered photos on the crystal lattice of the material are detected by a detector; whereby the residual stresses can be derived from the particular detected width of the Bragg reflections and / or the measured angle ge, characterized in that the scattering volume of the high-energy photons is restricted in such a way that a separation of individual crystallites of the metallic material is made possible. 2. The method according to claim 1, characterized in that into the beam path of the photons reflected on the target and the detector has an aperture designed as a ring aperture is inserted. 3. The method according to one or both of claims 1 or 2, characterized in that the detector as a location sensitive cher area detector is formed. 4. The method according to claim 3, characterized in that the area detector is designed in the form of a gas detector is. 5. The method according to one or more of claims 1 to 4, characterized in that the target relative to Beam axis of the high-energy photons in at least a degree of freedom to meet the diffraction conditions is rotated to record the Bragg reflexes. 6. The method according to one or more of claims 1 to 5, characterized in that the analysis of the chip removal at different wavelengths of the photons high-energy photon beam is performed.   7. The method according to claim 6, characterized in that the wavelength change to analyze the residual stresses done continuously. 8. The method according to one or more of claims 1 to 7, characterized in that the analysis of the Eigen voltages carried out in the manner of a grid measurement solution (scan or series of measurements), especially when rotating quickly the components, especially turbine blades, engines or engine parts.