Method for determining expected noise emissions of gear pairs.

Abstract

The invention is a method for determining the expected noise emission of a gear pair, in which the tooth flank of one gear of the gear pair is measured, from the measurement data obtained thereby the deviation of the measured tooth flank surface from a predefined target profile of this surface is determined, and based on this deviation a simulation in the form of a contact analysis under load is performed for the gear pair, the contact analysis being not based on the entire deviation but only on a specifically selected part of the deviation in the complementarity space.

Description

[Technical field]

[0001] The present invention relates to the field of transmission engineering and gearing technology. [Background technology]

[0002] It is well known that the meshing of the two torque transmitting gears, i.e. gear pairs, in a transmission creates a risk of unwanted noise emissions, especially in the gear drives of electric vehicles. It is well known that if the gears used have a high level of precision, it is possible not only to extend their service life but also to improve noise through reducing the excited vibrations (so-called noise vibration-harshness, NVH performance). Therefore, vehicle manufacturers are now in their strategies basically demanding high to very high quality manufacturing according to ISO 1328 for the manufacture of gears.

[0003] There are many factors that influence noise generation, but when designing gear pairs for a transmission, undesirable noise generation can still occur, despite the required high quality. It is therefore very common for undesirable noise generation to be detected after designing and manufacturing gears that move within the desired tolerances, simulating the transmission in use under load by contact analysis or after actual testing. Summary of the Invention

[0004] In this respect, the invention relates in particular to a method for determining the expected noise emission of a gear pair, in which the tooth flank of one gear of the gear pair is measured, and from the measurement data obtained thereby, a deviation of the measured tooth flank surface from a predetermined target profile of this surface is determined, and on the basis of this deviation a simulation in the form of a contact analysis under load is carried out for the gear pair.

[0005] Due to the problems mentioned above, the present invention is based on the object of contributing to provide a low-noise gear pair for a transmission.

[0006] From a process engineering point of view, the invention is solved by the fact that the contact analysis is not based on the deviation as a whole, but only on a specific selected part of the deviation in the complementary space.

[0007] Within the scope of the present invention, it is recognized that the data provided by the gear testing machine (which data represent the surface deviations from the ideal target tooth flank) in this form is not very suitable to provide conclusions about the expected noise emission generation. Therefore, according to the present invention, it is provided to work using the complementary space. One possible design option is for example the frequency domain by Fourier transformation, taking into account the resulting spectral weights for the individual spatial frequencies, but also other transformations for decomposing the measurement data into spectral components. Furthermore, the present invention is based on the recognition that waviness with a wavelength on the order of magnitude of the tooth shape, such as the height or the face width in the profile direction, plays an important role in noise generation. This has been discussed in modern research for some time and is still controversial. The influence of waviness does not seem to be clear. It certainly seems logical that waviness of the tooth flank can excite undesirable vibrations and thus can or should be the cause of noise generation. However, it is not clear how accurately and quantitatively such excitation is related to the amplitude and length of the waviness, and there are seemingly experiments in which specifically tailored waviness can also have a positive effect on noise generation.

[0008] Finally, the invention is based on the recognition that a pure simulation only at the level of calculations and simulations of targeted tooth surface modifications, including waviness, is of limited informational value since it has no specific reference to the actual manufactured teeth, while, even with specific reference to the manufactured teeth, the holistic data does not provide a good starting point for targeted analysis due to the simultaneous and coordinated influence of a large number of influencing factors in the form of tooth defects, which are also included in the manufacture for many reasons and whose influence overlaps in the test data of tooth measurements.

[0009] However, with the approach according to the invention, the part specifically used for further analysis is derived from test data for this toothing and is therefore specifically linked to the toothing manufacturing process, and furthermore, the influence of the specifically selected part in the complementary space can be more reliably determined since the influence of other parts is at least partially suppressed.

[0010] Further advantageous configurations of the method are set forth in the dependent claims.

[0011] For example, a two-dimensional complementary space can be considered for surface-measured tooth measurement data of the surface (3D topography data). However, the measurement can also directly provide measurement data along a (one-dimensional) path (2D test data). In a preferred design, a one-dimensional complementary space is used. In the spatial domain, a path is provided over the tooth surface for this purpose.

[0012] In one variant of the method, the direction of the path can be determined with reference to the geometric dimensions of the tooth flank, for example via the angle relative to the tooth flank line direction or profile direction (see also below). In another preferred variant, the path direction is set in the direction in which the waviness is expected. In this context, the direction is preferably selected depending on the specifications resulting from the micromachining, in particular the physical micromachining of the teeth, such as feed marks and galling, i.e. depending on the meshing conditions during precision machining.

[0013] In principle, measurement data taken on one flank, for example the left or right flank, is sufficient. However, with respect to the spatial region, the tooth measurement data is preferably used correspondingly for tooth flanks with different names, not only for one tooth flank, but for several tooth flanks, in particular for all tooth flanks (each with the same name). The path over the tooth flank should preferably have a directional component in the profile direction, in particular a path that extends with a dominant directional component, in particular in the profile direction. That is to say, profile form deviations are of particular interest for the present invention, since they have been found to have a greater influence on noise emission than flank form deviations. On the other hand, the influence of the tooth flank form deviations must be taken into account, and in this respect the present invention also offers the possibility of an essential directional component of the path in the tooth flank line direction, which directional component may also be the dominant directional component. It is also envisaged to use the contact path on the gears of the meshing of the gear pair with a path. As a method for displaying the contact path, instead of displaying the contact path in the direction of the profile by following the profile directly over the course, it is also possible to select the method of displaying the rotational length (length on the meshing line) known from gear engineering.

[0014] The frequency domain is preferably used as the complementary space, preferably via a Fourier transformation, e.g., via a Fourier series expansion, which results in corresponding spectral weights of the measured deviation data for specific frequencies in the frequency domain.

[0015] As a result, the specifically selected portion may include a single frequency or a group of selected frequencies, such as harmonics, but also ghost frequencies, which will be described later. Since the selection is based on real data actually measured, a very sharp centering of the individual frequency peaks in the frequency domain, as is usually the case in simulations, does not usually occur. Nevertheless, directly adjacent frequencies can preferably be included, provided that their spectral weight is above a certain limit. A person skilled in the art can easily select this threshold appropriately from a diagram in the frequency domain. For example, for this purpose, adjacent frequencies, possibly also the next adjacent frequencies, even up to the third or fourth next adjacent frequencies, are taken into account.

[0016] In one variant, so-called meshing frequencies with some desired harmonics, i.e. frequencies that can be derived from the motion frequencies of the meshing gear pairs and are expected, can be included, since some of them will inevitably be present anyway. These frequencies are included with their actual amplitudes, i.e. their actual spectral weights determined from the measured deviations. However, the meshing frequencies can also be partially or completely excluded and other frequencies specifically selected to test the effect of the meshing frequencies.

[0017] Additionally, so-called ghost frequencies may also be included, i.e. frequencies unrelated to the frequency / wavelength that can be derived from the motion frequency of the meshing gear pair and from the geometry of the teeth of the gear pair.

[0018] In another preferred variant, it is envisaged to specifically select one or more frequencies depending on feedback from noise measurements of real transmissions, for example from a line termination test station of the transmission manufacturer. It may happen that the manufacturer wants to experimentally find a particular interference frequency in a real transmission and establish its origin. In the simplest case, where this frequency can already be clearly identified from the measurement data, the origin is relatively clear. However, the origin may also be in the form of another frequency or in the form of multiple frequencies (beats, convolutions). The method may therefore include a targeted search for frequencies and frequency combinations that result in the feedback frequency obtained (by superposition / convolution), and a check of the measured spectrum for the presence of these individual frequencies in one or more combinations. These can then be selected to test their effect, independently of the frequencies that were not taken into account because they were not selected. This may lead to frequencies also being selected whose amplitudes appear less pronounced in the measured spectrum and therefore are not easily recognized as harmful in the spectrum.

[0019] With regard to the operation, it is expressly intended that the operator personally determines, for example by corresponding input, all or part of a particular selected portion, which may be done graphically via a touch screen or a multi-touch screen, for example by setting intervals or windows.

[0020] However, it is also envisaged that a selection program is provided that determines part or all of the specifically selected portion. Such a selection program can scan the Fourier spectrum for frequencies that lie above a threshold in terms of their spectral weight, determine these frequencies and select them individually one by one or in pairs or in multiple combinations for further analysis. This means that a corresponding test grid can be specified that is then automatically checked. Identification of frequency portions that exhibit a measurable and potentially repeatable influence on noise emissions can be made at the end of such an automatic or semi-automatic contact analysis.

[0021] In a preferred embodiment, the transmission error is determined as part of a load-based contact analysis, however, other characteristics may also be determined, such as force excitation, Hertzian pressure, losses, etc. The mating gear for simulating the contact analysis may be an ideal "master gear" or a measured mating gear.

[0022] In a further variant, a simulated portion, for example another frequency with amplitude, wavelength and phase, can be superimposed on the portion selected according to the invention.

[0023] Furthermore, the vibration excitation is preferably determined by the so-called NVH performance and a concrete representation of the expected noise emissions is also created, preferably by means of a visual representation, such as for example a Campbell diagram.

[0024] A conventional gear testing machine having a non-contact or tactile sensor can be used to generate the measurement data.

[0025] Of course, the specified target profile of the tooth flank may already include flank modifications, such as crowning, setback, etc.

[0026] The invention is also protected not only in terms of process engineering, but also in the form of a software program which, when executed, performs the contact analysis according to the invention, as well as by a measurement and analysis apparatus comprising a corresponding analysis device and a tooth testing device for determining measurement data, the tooth testing device being able to be located in the same place as the analysis device, but not necessarily being located there. [Brief description of the drawings]

[0027] Further features, details and advantages of the invention can be found in the following description, with reference to the accompanying drawings. [Figure 1] FIG. 4 is an explanatory diagram of a profile deviation on a tooth surface. [Diagram 2] FIG. 13 illustrates a fictitious waveform deviation. [Diagram 3] FIG. 1 shows the measured shape deviation and the superposition of three frequency parts. [Figure 4] A representation of the measurement signal taking into account all tooth surfaces and in the complementary space for the left and right tooth surfaces is shown. [Diagram 5] We show a purely schematic representation of the expected noise, which corresponds to the complementary space representation in the lower right corner of FIG. [Figure 6] FIG. 1 shows a schematic diagram of a test apparatus for measuring gear tooth flanks by laser scanning. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0028] To briefly explain the so-called profile deviation, this is clearly shown in Figure 1. Thus, the tooth flank 4 of a tooth 2 of a toothing section extends in the tooth width direction and in the profile direction (height direction) (Figure 1, left), the deviation from the profile can be determined at the head K and at the root F (Figure 1, center), and for a theoretically ideal tooth flank shape TH, the profile measurement value M shown as an example should be within a specified tolerance window T (Figure 1, right). The difference between the curves M and TH therefore represents the deviation between the measured tooth flank and the theoretically specified target tooth flank.

[0029] If a fictitious hypothetical waveform deviation is entered into a measurement diagram as shown in Fig. 2, it can be seen that the waveform deviation can certainly be within the tolerance range (outer limit line in Fig. 2). Furthermore, in Fig. 2, the three indicated dimension arrows represent parameters that can define the waveform and its position, such as the wavelength (vertical distance arrow from about 18 to 14 mm in Fig. 2), the position of the high point relative to the head (upper distance arrow from about 18 to 20 mm in Fig. 2), and the (double) amplitude of the wave (horizontal distance arrow in Fig. 2). In the diagram of Fig. 2, the vertical axis corresponds here to the rolling length of the left tooth flank (length on the approach line).

[0030] On the left side of Fig. 3, the shape deviations measured in the profile direction and compared to the target profile are shown. After Fourier transformation, coefficients with corresponding amplitudes are obtained for the frequencies f1, f2 and f3 given in the center of Fig. 3, and their inverse transformation into the spatial domain results in a sine wave with constant amplitude and only one frequency, shown in the center of Fig. 3. For example, a contact analysis can be based only on the part consisting of these three frequencies. The representation of the corresponding superposition of these three frequencies, inverse transformed and displayed in the spatial domain, results in the curve shown on the right side of Fig. 3. Further irregularities and parts that are still recognizable in the measurement signal on the left side of Fig. 3 are no longer included therein.

[0031] For example, if all tooth profile deviations are measured in one section and combined according to the rotation path, the deviations of the bold lines in FIG. 4 for the upper left left tooth surface and the lower left right tooth surface are obtained after removing the overlapping areas, and after Fourier analysis, the related expressions in the complementary space (frequency domain) are obtained. Prominent first to fourth harmonics are found on both tooth surfaces, although with different amplitudes. Furthermore, it can be seen that for a particular selection, directly adjacent coefficients for mapping the harmonics can be included. In this selected exemplary embodiment, they are two directly adjacent frequency coefficients in the frequency domain.

[0032] Furthermore, it can be seen from the frequency domain representation on the right hand side of Figure 4 that there are coefficients with a spectral weight above a certain threshold that lie outside the harmonics and are circled in Figure 4. These frequencies, in contrast to the so-called "meshing harmonics", are called ghost frequencies in the context of this application since they are not related to the motion frequencies of the meshing gear pairs and the frequencies / wavelengths that can be derived from the geometry of the teeth of the gear pairs.

[0033] For example, if four prominent harmonics are selected as the specifically selected portions, and optionally one or more or none of the ghost frequencies are selected, then the impact of each individually recorded ghost frequency on noise emissions can be obtained individually by basing the simulated contact analysis only on the corresponding specifically selected portions, and using the amplitudes as they exist in the actual gear and are known after measurement.

[0034] In the diagram of Fig. 5 a comparison is shown between the spectral weights as in the bottom right diagram of Fig. 4 and the results of the determination of the noise emission. The bottom diagram of Fig. 5 as a Campbell diagram is shown only very diagrammatically for the purposes of illustration, but with oval outlined areas shows the connections to the three considered ghost frequencies that are not shown in the Campbell diagram of the left tooth surface (top diagram of Fig. 4).

[0035] Thus, an analysis of the noise emissions is possible based on actual measurement data with specific reference to a toothing that is not affected by other types of disturbances, since not the deviation as a whole but only a specifically selected part of the above-mentioned deviation is taken as the criterion.

[0036] 6 shows the basic principle of a gear testing machine, in which the tooth flank 12 of a toothing 10 is clamped on a workpiece spindle 20 so as to be rotatable about the toothing axis W and is measured by a laser sensor 6 arranged with its field of view 8 in the area of ​​the toothing while being rotated in a direction R. It is understood that other testing machines can also be used, for example machines in which the laser sensor 6 can be arranged so as to be adjustably positionable in up to three spatial directions, as described in WO 2019 / 083932 A1, which is incorporated herein by reference in this respect.

[0037] The load-based contact analysis itself can be carried out using evaluation software known to the person skilled in the art, for example the KISSsoft software by the applicant. The software implementation can for example be realised via a software interface, in which the reduction of the above-mentioned two-dimensional measurement data to one dimension, the subsequent transformation into Fourier space and the specific selection of only a part of the transformed deviations in Fourier space are carried out. Of course, instead of such a software interface, existing known evaluation software can also be extended for the contact analysis, but still requiring the measurement data of the gear testing machine as input data. However, variants are also envisaged in which some of the above-mentioned steps can still be carried out, for example in the testing machine itself, which already creates and transmits the resulting frequency spectrum.

[0038] The present invention is not limited to the exemplary embodiments; instead, the features of the following claims and the above description may be essential, individually or in combination, to realize the invention in its various embodiments.

Claims

1. 1. A method for determining the expected noise emission of a gear pair, comprising measuring a tooth flank of one gear of said gear pair, determining from the measurement data obtained thereby a deviation of said measured tooth flank surface from a predetermined target profile of said surface, and performing a simulation in the form of a contact analysis under load for said gear pair on the basis of said deviations, The method of claim 1, wherein said contact analysis is not based on the entire deviation, but on only a specifically selected portion of said deviation in complementary space.

2. 2. The method according to claim 1, wherein the deviation data along a path on the tooth flanks is determined from the surface deviations taking into account several, in particular all, tooth flanks of the gears having the same name.

3. 3. The method according to claim 2, wherein the path has a directional component, in particular a predominant directional component, in a profile direction, in particular extending in the profile direction or is a contact path of the gear pair.

4. The method of claim 2 or 3, wherein the complementary space is one-dimensional and the associated spatial region corresponds to the path.

5. 5. A method according to any one of claims 1 to 4, wherein the surface deviations are subjected to a Fourier analysis, in particular one-dimensionally with respect to the path according to any one of claims 2 to 4.

6. 6. A method according to any one of claims 1 to 5, wherein the specifically selected portion comprises a spectral portion of a selected frequency or a selected group of frequencies.

7. The method of claim 6, wherein for a selected frequency, the spectral weights of its immediate surroundings are also selected.

8. 8. The method according to claim 1, wherein at least one, in particular several nearest harmonics are selected for a fundamental frequency / wavelength.

9. 9. The method according to claim 1, wherein the selected one or more frequencies are input together with their amplitudes obtained from the converted measured deviation.

10. 10. The method of claim 1, wherein a frequency is selected or is also selected that is independent of the frequency / wavelength derivable from the motion frequency of the meshing gear pair and from the geometry of the teeth of the gear pair.

11. 11. A method according to any one of claims 1 to 10, wherein the deviation is displayed to an operator, who determines, by an input, the specifically selected portion, partly or in whole.

12. 12. The method according to claim 11, wherein the display is graphical, in particular via a touch screen, in particular a multi-touch screen, and the input is made in particular by setting intervals or windows.

13. 13. The method of claim 1, wherein the specifically selected portion is determined in part or in whole by a selection program.

14. The method of claim 1 , wherein a rotational path error is determined during the contact analysis.

15. 15. A method according to any preceding claim, further comprising using the contact analysis to create a visual representation of the expected noise emission, such as by means of a Campbell diagram.

16. 16. The method according to any one of the preceding claims, wherein the measuring is performed by a touch sensor and / or contactlessly, e.g. optically, e.g. by laser scanning.

17. A software program for carrying out a catalytic analysis according to any one of claims 1 to 16 when the analysis device is executed on a computing unit.

18. 17. An analysis device configured to perform a contact analysis of a gear pair according to any one of the preceding claims.

19. 20. A measuring and analysis apparatus comprising a gear testing device for measuring tooth flank surfaces and an analysis device according to claim 18.

Images