WO2000039574A1

WO2000039574A1 - Method and device for the quality control of a work piece

Info

Publication number: WO2000039574A1
Application number: PCT/DE1999/004098
Authority: WO
Inventors: Volker Sauer; Fawzi Attia; Andreas Frede; Ronny Jahnke
Original assignee: Robert Bosch Gmbh
Priority date: 1998-12-28
Filing date: 1999-12-27
Publication date: 2000-07-06
Also published as: JP2002533720A; DE19860471C2; EP1058842A1; DE19860471A1

Abstract

The invention relates to a method for the quality control of a work piece (1). According to said method, said work piece (1) is incited to vibrate (8), at least one vibration-related parameter is measured (18) and the work piece is approved of if said parameter is within a predetermined range.

Description

METHOD AND DEVICE FOR QUALITY CHECKING

WORKPIECE

Field of the Invention

The invention relates to the non-destructive quality inspection of workpieces.

State of the art

A large number of different processes have been developed for the non-destructive quality inspection of workpieces, more precisely the detection of externally invisible material defects and internal material defects in a workpiece.

For example, it is known to x-ray workpieces with x-rays or gamma rays in order to detect material defects. Errors that lead to inhomogeneities in the radiation absorption by a workpiece, such as bubbles, cavities or foreign bodies, can be clearly seen in the images obtained in this way. Detecting cracks in materials is more difficult because of the presence of a crack has no significant effects on the thickness of the absorbent material of the workpiece and can therefore only be recognized essentially by means of radiation scattered on surfaces of the crack. The evaluation of the X-ray images obtained requires a great deal of experience and is difficult to automate.

Ultrasound test methods are based on the reflection of ultrasound waves entering a workpiece from cracks or faults in the same. Although such methods are inherently highly sensitive to cracks hidden in the material, ultrasound reflections occur at any interface of a workpiece, which makes the evaluation of ultrasound reflection data that complicates and complexes on irregularly shaped or made of different materials.

Other methods, such as eddy current methods, in which eddy currents are generated in a material to be checked with the aid of a current-carrying coil and disturbances in the course of the current eddies induced by cracks in the material are detected by changing the impedance of the coil, or leakage flux measuring methods, in which from an agnetized workpiece externally displaced magnetic flux, the existence of a crack in the workpiece is only applicable to certain classes of materials.

From DE 4316473 AI a device is known which derives structure-borne noise from a workpiece which is generated when the workpiece is drilled. The derived structure-borne noise can be used in a manner not described further to monitor the drill for cutting edge wear, breakage, etc.

Advantages of the invention

With the present invention, as defined in the independent claims, a method and a device for quality inspection of a workpiece, in particular for the detection of cracks or defects in a workpiece, are created, the fast and uncomplicated quality inspection of workpieces made of practically any solid materials allow. The effort required to process the measurement signals obtained is low, in particular no complex image processing is required, and the processing effort is essentially independent of the complexity of the shape of the workpiece to be checked. The temporal development of the oscillation amplitude or the frequency of the oscillation can preferably be used as the parameter of the oscillation to be measured. The temporal development can in particular be characterized by the decay time of the oscillation.

In most workpiece designs, an excited vibration will always include a plurality of spectral components. In such a case, frequency and temporal development of individual spectral components can be recorded or monitored. The individual spectral components generally differ in their decay behavior and are influenced in different ways by possible errors in the workpiece. In general, higher-frequency components are more sensitive to errors of small extent than low-frequency ones. Furthermore, the sensitivity of a given spectral component to a disturbance differs depending on whether it is located in the region of an antinode or node of the vibration mode of the workpiece belonging to this component. By evaluating a plurality of spectral components with different distributions of antinodes and nodes, a more homogeneous one can be obtained achieve local distribution of detection sensitivity for defects in the workpiece.

The decay time of the oscillation and / or individual spectral components of the oscillation is preferably measured and the workpiece is found to be good if the decay time lies in a predetermined interval. This interval can be determined in advance on the basis of experimental parts known to be free of defects.

In return, the workpiece is found to be bad if the decay time is within a second predetermined interval.

This interval need not be complementary to the first; if inadequately long or short decay times are measured, this is more an indication of a measurement error than of an actual defect in the workpiece being examined. In such a case, it is advisable to check the workpiece using another method. This can also be expedient if the measured value lies between the two predetermined intervals.

Measured decay times of the oscillation or of individual spectral components can be different for one and the same workpiece if the Vibration excitation occurs with different strengths. In order to achieve reproducible measurements, it is therefore important that the vibration is reproducibly excited, in particular by striking the workpiece with an object at a fixed speed.

A to some extent unavoidable spread of the excitation energy or - which is equivalent - the velocity can be compensated for by the normalization of measurement data, in particular vibration amplitudes, to the excitation energy or velocity.

This object is preferably a pendulum deflected with a fixed amplitude.

After the pendulum bumps into and bounces off the workpiece, it is convenient to stop to avoid bumping it a second time with reduced energy while the workpiece is still vibrating, thereby interfering with the measurement.

The excited vibration of the workpiece is preferably measured interferometrically. For this purpose, for example, a laser beam is divided into two partial beams, one of them on one Surface area of the workpiece is directed. Laser light reflected from the workpiece is brought into interference with the other partial beam and the interference pattern is evaluated. In this way, the movement of the irradiated surface area of the workpiece can be measured with high accuracy, without the vibration behavior of the workpiece being influenced in any way.

As an alternative or as a supplement, it can also be provided that the vibration is measured using sound waves emitted by the workpiece through air, for example with the aid of a microphone.

An arrangement for carrying out the method according to the invention comprises, in addition to devices for excitation and for measuring the vibration, an oscillatable bearing for the workpiece. Because this mounting is capable of oscillation itself, it is ensured that the vibration of the workpiece is not suppressed at any point, and that consequently errors can be detected in the same way at any point on the workpiece.

This storage expediently supports the workpiece at three points. This makes it stable Fixation of the workpiece with minimal restriction of its oscillation ability.

The bearing preferably comprises projections, for example in the manner of arms, columns or knobs, made of a rubber-elastic material which supports the workpiece.

A simple means of exciting the vibration of the workpiece is a pendulum. In order to achieve an exactly reproducible excitation with the help of the pendulum, a stop element should be provided which defines a maximum deflection of the pendulum. In a simple configuration of the arrangement, the pendulum has an actuating lever which extends upward beyond the pendulum axis and which an operator can tilt to move the pendulum against the stop element and then release it so that the pendulum can hit the workpiece. After the pendulum has struck the workpiece, the operator can also hold it on the operating lever to prevent a second impact on the workpiece.

Further features and advantages of the present invention result from the following description of an exemplary embodiment with reference to the figures. characters

Show it:

FIG. 1 shows a schematic representation of a first part of an arrangement for carrying out the method according to the invention, which comprises a device for exciting the vibration;

Figure 2 is a schematic representation of the arrangement for performing the method according to the invention including a device for measuring the vibration;

Figure 3 is a schematic representation of a variant of the arrangement of Figure 2;

FIG. 4 shows a graphical representation of results of a method according to a first embodiment of the invention; and

Figure 5 is a graphical representation of the results of a method according to a second embodiment.

Figure 1 shows a side view of a device for exciting a vibration in a workpiece, which a first part of the arrangement for Implementation of the method according to the invention forms. On a base plate 3, three columns 2 made of, for example, silicone are mounted, which support a workpiece 1 to be tested, here a cam ring of a radial piston injection pump, at three points near its circumference. The base plate 3 also carries an arm 4 which holds a pendulum 6 rotatable about an axis 5 perpendicular to the plane of the figure. The pendulum comprises an impact body 8 which, in the position shown in solid lines in FIG. 1, just abuts the workpiece 1. The impact body 8 is connected to the axis 5 by a flexible sheet 9, for example made of spring steel. An operating lever 7 above the axis 5 is firmly connected to the sheet 9.

By tilting the operating lever counterclockwise in FIG. 1, the pendulum reaches the position shown in broken lines in FIG. 1, in which its lower end abuts a stop element 10. This stop element 10 defines the maximum possible deflection of the pendulum. The pendulum can be released from this position and then strikes the workpiece 1 at a precisely defined, reproducible speed. This speed is selected so that it is not sufficient to overcome the static friction of the workpiece 1 on the columns 2 and this to be placed on the pillars slide. Thus, the workpiece 1 and the columns 2 are excited to oscillate with a precisely predetermined energy by the impact with the pendulum.

After hitting the workpiece 1, the pendulum rebounds and is stopped by the operator before it can hit the workpiece 1 a second time. In this way, the vibration excited by the first impact with the pendulum can end undisturbed.

The position of the stop element 10 can be displaceable in order to be able to reproducibly set different pendulum deflections for different types of workpieces which differ, for example in terms of their weight.

The support of the workpiece can also be displaceable in order to avoid a second stop directly.

According to a variant, an electromagnetic coil can be mounted on the stop element 10, which can have different functions. On the one hand, it can be energized before the start of a measurement in order to hold the impact body 8 on the stop element 10 until the current is interrupted is and thus the pendulum is released to hit the workpiece 1. A short time later, the coil is expediently energized again so that it exerts a magnetic attraction on the pendulum 6 rebounding from the workpiece 1 and pulls it back to the stop element 10.

FIG. 2 shows the entire arrangement for carrying out the method, the device for exciting the vibration described with reference to FIG. 1 being shown schematically in a top view.

A continuous wave laser 11, for example a HeNe laser, emits a beam which is split into two partial beams at a partially reflecting mirror 12, one of which is reflected on a reflector 13, here a 90 ° prism, and finally reaches a sensor 18, and the second hits a lateral surface of the workpiece 1, which is arranged on the columns 2 (not visible in this figure) on the base plate 3. The reflected beam is fanned out on the convex outer surface of the ring-shaped workpiece 1, part of the reflected light passes through an aperture 14 and returns to the partially reflecting mirror 12 and is reflected in the direction of the sensor 18. A sensor 18, the light reflected by the workpiece 1 is superimposed with the beam reflected by the reflector 13. The reflector 13 is positioned such that the optical path lengths of the two partial beams do not differ by more than the coherence length of the laser light, so that an interference pattern arises at the sensor 18.

Before the actual measurement is carried out, the amount of light reflected back from the workpiece 1 to the sensor 18 can be maximized. For this purpose, the interferometer arm, which contains the reflector 13, is temporarily blocked, which prevents the formation of an interference pattern on the sensor 18. The measurement signal from the sensor 18 is then a direct measure of the amount of light reflected by the workpiece. This can be optimized, for example, by moving the base plate 3 perpendicular to the direction of the beam coming from the laser 11 or by tilting the base plate.

When the workpiece 1 is excited to vibrate by the impact body 8, the outer surface of the workpiece 1 moves back and forth in the direction of the laser beam, which leads to constant changes in the optical path length in an arm of the interferometer structure, which is caused by the mirror 12, 15 and the reflector 13 is formed. These path length changes conditions lead to shifts in the interference pattern formed on the sensor 18, which are detected by the sensor and processed by evaluation electronics connected to it.

The evaluation electronics comprise, on the one hand, an oscilloscope 19, which directly displays brightness differences occurring at the sensor 18 and thus gives an operator an immediate impression of the course of the vibration excited in the workpiece 1.

Another essential component of the evaluation electronics is a computer 20, which receives digitized brightness values from the sensor 18 via an AD converter 21 and is programmed to convert them into instantaneous deflections of the workpiece and to calculate the decay behavior of the entire oscillation therefrom, or possibly also the oscillation break down into their individual spectral components and determine their frequencies and their temporal behavior. Since the vibration modes of the workpiece 1, which correspond to the individual spectral components, are coupled to one another, the temporal behavior of these individual components can be quite complex, for example, depending on the workpiece, beats between individual components can occur, the intensity of individual components can gradually increase after the workpiece 1 has been hit, since vibration energy is only gradually coupled into the corresponding vibration mode, etc. Amplitudes and frequencies of the individual spectral components can be obtained in a simple manner using known processing techniques, such as the fast Fourier transformation, and from one Any deviations of the measured values from standard values previously determined on workpieces known to be error-free can easily be concluded that there is an error in workpiece 1. A screen 22 connected to the computer 20 displays the results of the evaluation, for example as numerical values of the measured frequencies, decay times, etc. Of course, the screen 22 can also display a decision made by the computer 20 about the defectiveness or freedom from defects of the workpiece.

FIG. 3 shows a variant of the device in which the vibration of the workpiece 1 is not detected optically-interferometrically, but acoustically with the aid of a microphone 23. The processing of the signal supplied by the microphone 23 is essentially the same as that of the signal supplied by the sensor 18 and is carried out using the same devices 19 to 22. The acoustic examination of the workpiece can also accompany tending to be interferometric. The two variants differ essentially in that the optical signal evaluated in the structure shown in FIG. 2 is obtained only from a locally limited area of the workpiece, while the acoustic signal is an “averaged” signal to which the entire workpiece contributes.

The implementation of the method according to the invention will now be described. An operator places a workpiece 1 to be checked, here a cam ring, on the silicone columns 2. These support the ring at three points, so that its spatial position is fixed. If necessary, the operator adjusts the position of the base plate 3 in order to maximize the intensity of the light reflected from the workpiece 1 to the sensor 18. Then she uses the lever 7 to deflect the pendulum towards the stop element 10. Simultaneously with the release of the lever, it triggers, for example via a switch, the device for measuring the vibration, which includes the oscilloscope 19, the AD converter 21 and the computer 20. Thus, the computer 20 is able to measure the time between releasing the pendulum and the vibration of the workpiece 1 and, if this time deviates too far from an expected value, a warning to output that the measurement is probably unsuitable.

If, as described above as a modification, the stop 10 contains a magnetic coil, the triggering of the measuring device and the interruption of the current of this coil can take place with a common switch.

The computer 20 then calculates the instantaneous deflection of the workpiece 1 from its rest position and the amplitude of the oscillation from the brightness values provided by the sensor 18. The decay time is determined by continuously monitoring the amplitude. Previous measurements on the cam rings used here as a workpiece have shown that they normally have a decay time in a range around 800 ms. The computer is therefore instructed to assess a cam ring as correct when the measured decay time is between 600 and 1500 ms. Significantly larger deviations from the normal value are permitted upwards than downwards because a long decay time is generally regarded as an indication of a good, crack-free and trouble-free material structure. If a decay time of more than 1500 ms is measured, there is probably a measurement error. In such a case the measurement can be repeated from the beginning, or ^' the cam ring in question is checked by another method.

If the decay time is between 100 and 400 msec, the computer 20 decides that the cam ring is bad. A decay time shorter than 100 ms is in turn an indication of a lack of measurement.

If the decay time is between 400 and 600 ms, the assessment is not completely certain. In such a case, a low-value workpiece could simply be discarded; for higher-quality workpieces, it can be economical to carry out a check beforehand using another method.

FIG. 4 shows a typical result of a check of a batch of 300 cam rings in the form of a diagram, on the axes of which the ordinal number of the respective cam ring or the measured decay time is plotted in ms. The overwhelming majority of the cam rings have decay times that fall within a band between 600 and 1000 ms, which is shown hatched in the figure, and are therefore within the dashed interval of 600 to 1500 ms within which the cam rings are found to be in order. Individual rings, represented by cross 30 in the diagram, have decay times between 100 and 400 ms and are therefore sorted out as faulty.

FIG. 5 shows the results of a model examination on two different batches of cam rings, which differ in their dimensions. The ordinal numbers of the cam rings are plotted on the axes of the diagram and their resonance frequency in Hz is plotted vertically. The first batch corresponds to the area of the diagram designated I, the second to that designated II. A group of defective cam rings is inserted in the section designated 31 in the first batch. The measured value of each cam ring in this group is indicated by a circle in the diagram. It can be seen that the measured resonance frequencies of the defective cam rings vary widely and are usually below the typical resonance frequency of about 3750 Hz for the batch. Outliers 32 in area I are due to individual defective cam rings or measurement errors. In addition, a small group 33 of cam rings of the second batch is inserted into the batch, the resonance frequencies of which are significantly higher at approximately 4050 Hz. The second batch also contains a group 34 of defective cam rings, the measured values of which, denoted by small circles, in comparison to the actual one Scatter the batch heavily and lie lower on average.

With the present invention, a less complex method for quality inspection of workpieces made of practically any solid materials is created, in which the evaluation of the measurement results is simple and reliable and the entire method can be automated easily and thus without difficulty in an automatic production of the workpieces can be included.

Claims

claims

1. Method for quality inspection of a workpiece, in which the workpiece is excited to oscillate, at least one parameter of the oscillation is measured and the workpiece is found to be good if the parameter is within a predetermined range.

2. The method of claim 1, wherein the temporal development of the vibration amplitude and / or

Frequency of the vibration serves as a parameter.

3. The method according to claim 1, in which the oscillation has a plurality of spectral components and in which the intensity distribution over different spectral components of the oscillation, the temporal development of spectral components and / or their frequency serves as parameters.

4. The method of claim 1, 2 or 3, in which the decay time of the vibration and / or spectral components of the vibration is measured and the 99

Workpiece is found to be good if the cooldown is within a predetermined interval.

5. The method of claim 4, wherein the workpiece is found to be bad if the decay time is in a second predetermined interval.

6. The method according to claim 5, in which the workpiece is subjected to a test according to another method if the decay time lies outside the two predetermined intervals.

7. The method according to any one of the preceding claims, wherein the vibration is excited by striking the workpiece with an object at a predetermined speed.

8. The method of claim 7, wherein the object is a pendulum (6) which is deflected with a fixed amplitude and released.

9. The method according to any one of the preceding claims, characterized in that measured data are normalized using the excitation energy of the vibration.

10. The method according to any one of the preceding claims, 'in which the vibration is measured interferometrically.

11. The method according to any one of claims 1 to 9, wherein the vibration is measured on the basis of sound waves emitted by the workpiece through air.

12. Arrangement for carrying out a method according to one of the preceding claims, each with a device for exciting (4,5,6) or for measuring (11-22; 23,19-22) of the vibration and with an oscillatable bearing for the workpiece .

13. The arrangement according to claim 12, wherein the storage supports the workpiece at three points.

14. Arrangement according to claim 12 or 13, wherein the bearing has projections (2) for supporting the workpiece made of a rubber-elastic material.

15. Arrangement according to one of claims 12 to 14, wherein a device for excitation is a pendulum (6) for abutting the workpiece (1) and a stop element (10) defines a maximum deflection of the pendulum (6).

16. The arrangement as claimed in claim 15, in which the pendulum (6) has an actuating lever (7) which extends upward beyond the pendulum axis (5).

17. Arrangement according to one of claims 12 to 16, which has a laser (11) for illuminating the workpiece

(1) and an interferometer assembly (12-15), the interferometer assembly superimposing light received directly from the laser with light from the laser (11) reflected from a surface of the workpiece (1).

18. Arrangement according to one of claims 12 to 17, with a microphone (23) for recording from

Workpiece (1) sound emitted by air.