JPH06109444A - Noncontact measuring method for tooth-surface shape of gear - Google Patents

Abstract

PURPOSE:To measure a gear accurately and quickly without using a reference gear by a method wherein a gear under test is turned and set automatically in a proper measuring position and interference fringe data on the gear is taken directly into a computer. CONSTITUTION:When the central point of a tooth groove in a gear 6 under test is decided, a gear support device 7 selects a proper setting position on the basis of data which has been learnt, the gear is turned by a prescribed angle by a servo driving gear 9 and comes to a standstill, and taking the photograph of the gear 6 is started. That is to say, a laser beam which is radiated from a light source 1 is divided 2 into two beams, a beam of reference light R which has been fringe-scanned by a reflecting mirror 3 which is moved fine by a piezoelement 3a and a beam of object (measuring) light S with which a tooth surface 6a has been irradiated via a reflecting mirror 4 and which has been transmitted through a photographing lens 5 are picked up by a CCD camera 8 provided with a camera photodetection face 8a in an interference position. Interference-fringe data which has been photodetected as a light quantity is analyzed by a computer 12, the phase of each picture element is computed, a length (a height) is computed according to an effective wavelength by an incidence angle, and the shape of the tooth surface 6a is found.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a non-contact method for precisely measuring the tooth surface shape of a gear and, more particularly, to a free-form surface shape measuring apparatus for non-contact measurement of a free-form surface shape by an interferometry method. A non-contact free-form surface that does not require the reference surface of each gear that was required in the conventional hologram method in the method of measuring the tooth surface shape of various gears such as spur gears and helical gears. The present invention relates to a method for measuring the tooth flank shape of a gear using a measuring device, and an apparatus used for carrying out the method.

[0002]

2. Description of the Related Art Generally, the power output from a drive source such as a motor or an engine rarely uses the rotation as it is, and all the power is transmitted to a driven portion through a speed reducer or a speed increasing device incorporating gears. An output drive system is configured. However, in this drive system, there is a problem in that vibration and noise are generated due to the geometrical properties, mechanical properties of the gears, and whether or not they are processed / assembled. To solve these problems,
Various ingenuity was added to the gears to make complicated modifications to the geometrical shapes, and to produce a deceleration / acceleration device with less vibration and noise.
The current situation is that improvements are being made. Therefore, if we can find out what shape the machined gear has and what kind of vibration and noise it produces, and if we can obtain the resulting data, then we can obtain suitable machining data. Since such a measurement / analysis machine can contribute to the improvement of quality in gear manufacturing, there is a demand for such a measuring / analyzing machine.

Now, in the conventional measurement method using a contact-type stylus, that is, by tracing one line on the tooth surface, the angle transmission or torque transmission surface is essentially used. The information on the gears that function as is extremely small and takes a long time. The applicant's prior application (Japanese Patent Application No. 62-273492 and Japanese Patent Laid-Open No. 1-11) proposed to improve the deficiency of the contact type gear shape measuring method.
6403) and the like, which are non-contact measurement methods. That is, there is a method that allows non-contact measurement of the tooth surface by holographic interference and a free-form surface shape measuring device that can be used for implementing the method.

[0004]

However, in the measuring method using the holographic interference and using the free curved surface shape measuring apparatus, the tooth surface to be the reference surface is prepared, and the hologram image thereof is photographed. Since it is necessary to start measuring the tooth profile of the tooth surface to be measured, it takes a long time to manufacture the reference tooth surface, and it takes a considerable amount of time to prepare for imaging and the like. Further, since the measurement is performed by comparison with the reference surface, the processing accuracy of the reference surface also becomes a problem. Therefore, the main object of the present invention is to solve the above-mentioned problems of trouble and accuracy. Another object of the present invention is to modify the structure of the above-mentioned free-form surface shape measuring device so that the interference fringe data corresponding to the tooth surface shape of the gear to be measured mounted on the same machine is directly loaded into the computer and automatically. An object of the present invention is to provide a method for measuring the shape of a gear that obtains a measurement result and enables feedback to the gear machining process.

[0005]

SUMMARY OF THE INVENTION In view of the above-mentioned object, the present invention has a structure in which the free curved surface shape measuring apparatus according to the present application disclosed in Japanese Patent Application Laid-Open No. 1-116403 is modified to have the same structure. Make sure that the attached gear to be measured is automatically set to the proper measurement position, and the measurement mechanism is moved and set to the proper position, and the interference fringes on the tooth surface are directly measured using the high-density CCD camera. The measurement data of the tooth surface shape taken in by the computer is analyzed and obtained by calculation. Further, at the same time, the above-mentioned method is simultaneously used to automatically and successively measure a plurality of tooth flanks of the gear to be measured and continuously calculate the tooth flank shape error. Similarly, measure the tooth flank error of the gear to be meshed, and assume both gears composed of these tooth flanks, input them to the mesh state simulation program, and load the deceleration or acceleration gear system. In this way, it becomes possible to analyze the transmission error, the vibration force, and various stress states in advance to analyze the power transmission characteristics, and to appropriately analyze the problems of gear strength, vibration, and noise.

That is, according to the present invention, a gear to be measured is attached to a rotation spindle of a non-contact free curved surface measuring apparatus, coherent light emitted from a light source is separated into reference light and object light, and the object light is measured. By irradiating the target tooth surface of the gear, an interference fringe image of the target tooth surface is obtained from the reference light and the reflected light of the object light, and the shape error of the target tooth surface is measured. In the measurement method, a pinhole optical path of a part of an interferometer is used to automatically set a measurement origin with respect to a rotation center of the gear to be measured, and then based on a tooth surface shape specification of the gear to be measured in advance. According to the input measurement parameters,
The measurement positioning is performed from the measurement origin, and the object light is irradiated to one tooth surface, and the reflected light is CC having a large number of pixels
An image is formed on the light receiving surface of the D camera, and at the same time, the reference light separated previously and the reflected light of the object light are directly interfered with each other on the light receiving surface of the CCD camera to obtain interference fringes on the tooth surface. ,
A mirror inserted in the middle of the reference light is finely moved by a driving means such as a piezo element in several divisions of 1/2 wavelength of the coherent light, and an interference fringe is moved by one pitch by a so-called fringe scan method. By this, the movement of stripes for each division is taken out for each pixel of the CCD camera and the phase is analyzed. In addition, the camera is determined in advance from the specifications of the gear to be measured, the gear mounting position relationship on the measuring machine, and the irradiation position of the object light. The image that is expected to be captured is simulated by a computer to create a theoretical tooth surface shape, and the difference in each pixel is calculated by comparing and operating with the created theoretical tooth surface shape, and the incident angle at that point is calculated. By converting the effective wavelength by the height of the surface shape to obtain the tooth surface error for the entire tooth surface shape, the obtained tooth surface error is placed on the basic shape of the gear, and the contour height by map coloring is calculated. In was to perform error display,
It is intended to provide a non-contact measuring method of a tooth flank shape of a gear, which includes a measuring step.

More specifically, the coherent light emitted from the light source is separated into reference light and object light, and the object light is applied to a preselected reference tooth surface of the gear to be measured. From the holographic interference obtained by replacing the reference curved surface with the tooth surface to be measured, and then using the non-contact free curved surface measuring machine to obtain the surface accuracy of the measured tooth surface. In the surface shape measurement method,

First, 2 corresponding to the left and right tooth flanks of the optical probe.
Make a fine hole at the middle point between the polarizing prisms so that the light goes straight to the CCD camera without passing through the polarizing prism.
When the gear to be measured is set on the measuring machine, the light is transmitted through the tooth space. The measuring instrument is previously verified by a jig so that this hole is at the point O of the coordinates. By doing so, when the gear to be measured is rotated, the light beam is blocked by the left and right tooth surfaces. Then, slowly rotate the gear to be measured and measure this hole on the CCD camera.
The rotation angle of the gear from the time the tooth groove appears to the time it disappears is measured, and the center position of this angle is the origin position when measuring the tooth surface shape of the gear.

The analysis of the ideal tooth surface is performed by using the rotation angle τ from the tooth groove center to the position of the optimum state where the tooth surface is irradiated as a parameter. Therefore, in the present invention, the measurement of τ is important for both the setting on the measuring machine and the simulation, and the appropriate value of τ is determined by the empirical value stored in the computer from the gear specification. Next, in the present invention, a high-density CCD light-receiving surface is placed as a sensor for measuring the interference fringe brightness between the reference light and the object light coming from the tooth surface on the hologram surface in the non-contact free-form surface shape measuring device, and the object is The interference image of the light reflected from the tooth surface and the reference light is directly captured by the computer.

That is, in the conventional tooth surface shape measuring method using the non-contact free curved surface shape measuring apparatus, the method of forming the hologram of the reference surface once and making it interfere with the reflected light of the object light from the measuring surface is used. In other words, it is a comparison measurement with a reference surface, but in the present invention, the interference image of the measurement tooth surface is directly taken into a computer, the shape is analyzed, and the comparison is made with a theoretical tooth surface obtained by a simulation method by a built-in function of the computer. Is used to obtain the tooth surface error.

In the method of the present invention, each tooth of a pair of driving and driven gears is automatically measured one after another, and one tooth surface having a typical error is selected from the results, and this data is vibrated. Put it into an analysis software program and perform meshing vibration analysis by simulation. In this way, it is possible to obtain in advance vibration and noise data for the tooth surface that actually has an error, so that the gear manufacturing process should be performed so that such vibration and noise data fall within the specified range. By feeding back the measurement data of the tooth surface shape and correcting the gear manufacturing data, it is possible to manufacture a gear in which vibration and noise levels are sufficiently reduced during the meshing motion under load conditions. The present invention provides a method for measuring and analyzing the tooth flank shape of a gear by the above-mentioned non-contact free curved surface measuring machine. Hereinafter, the present invention will be described in more detail based on the embodiments shown in the accompanying drawings.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a mechanism diagram showing an optical configuration of a non-contact free curved surface measuring apparatus used for carrying out a tooth surface shape measuring method according to the present invention, and FIG. 2 is a block diagram of a curved surface shape measuring machine according to a prior application. 3 is a perspective view showing an overall configuration of a measuring apparatus used for carrying out the measuring method according to the present invention, and FIG. 4 is an interferometer surface plate which is a part of the measuring apparatus. FIG. 5 is a mechanical three-dimensional view that schematically shows the rotation mechanism that irradiates the measured tooth surface with object light. FIGS. 6A and 6B are prisms for guiding the optical path of the object light to the camera. And a schematic mechanism diagram showing a servo mechanism for driving the mirror, FIG.
(B) is an explanatory diagram showing parameters necessary for associating a tooth surface analysis diagram by simulation with an actual captured image,
FIG. 8 is a flow chart of measurement / analysis, FIG. 9 and FIG.
0 is a measurement flow chart showing two different methods for comparing a simulation image and a captured image in the measurement flow, and FIGS. 11A and 11B show left and right of the measurement gear error for easy understanding. Explanatory drawing showing an example of displaying on the reference tooth surface, FIG. 12 is a perspective view showing the structure of an optical probe for measuring the center of gear setting, and FIGS. 13 (a) and 13 (b) show the center of gear setting. A plan view and a front view of the optical probe to be measured, FIG.
4 (a) to (c) are explanatory views showing a relationship for obtaining a center by using the probe, FIG. 15 is a diagram showing a relationship between a computer and its related interface and peripheral devices, and FIG. (A) and (b) are explanatory views showing the development view of the tooth surface and the positional relationship of meshing, FIG. 17 is a graph showing the load state of the tooth surface on the operating surface, and FIG. 18 is FIG.
Explanatory drawing which shows the connection relation of 9-FIG. 21, FIG. 19-FIG.
3 is a flowchart of an analysis process of a power transmission state of a gear. First, in explaining the principle and operation of the non-contact type measurement method of tooth surface shape according to the present invention, regarding the configuration of the non-contact type free-form surface shape measuring apparatus, an example of an apparatus directly used for carrying out the method according to the present invention will be described. Figure 1 shows an example of an already proposed device.
From now on, description will be made based on the illustration of FIG.

The optical configuration of the non-contact type free-form curved surface shape measuring apparatus used in the non-contact type measuring method according to the present invention is, as shown in detail in FIG. A beam splitter 2 that splits a laser beam that is interference light into a reference laser beam R (hereinafter, simply referred to as reference beam R) and a measurement laser beam S (hereinafter, referred to as object beam S), and the reference beam R
And the reference light R that has reached the object light S through the reflection mirrors 3, 4 and the reflection mirror 3, and the tooth surface 6a of the gear 6 to be measured is irradiated through the reflection mirror 4 to reflect the reflected light of the object light S. A CCD camera 8 having a camera light-receiving surface 8a placed at an interference position of both the object light S transmitted through the photographing lens 5 for photographing, and a supporting device 7 having a spindle for rotatably supporting the gear 6 to be measured at both ends. Is coupled to the reflection mirror 3 for the reference light R described above,
The reflection mirror 3 is provided with a piezo element 3a or the like that gives a predetermined fine movement.

In the present invention, the above-mentioned apparatus of FIG. 1 is a further improvement of the measuring apparatus shown in FIG. 1 of Japanese Patent Application No. 3-20084 by the present applicant. The conventional non-contact type free-form surface shape measuring device shown in FIG. 1 has a basic configuration shown in FIG. For ease of understanding, in FIGS. 1 and 2, the same structural elements used in both devices are designated with the same reference numerals. That is, FIG.
The apparatus used for the measurement of the present invention shown in FIG. 2 is directly connected to the CCD camera 8 at the position of the light receiving surface 5 in the conventional measuring apparatus of FIG.
The light receiving surface 8a is placed, and the image of the specularly reflected light from the tooth surface is formed by the taking lens 5 on the light receiving surface 8a.

FIG. 3 is a perspective view showing the overall structure of the non-contact type free-form curved surface measuring apparatus used in the present invention shown in FIG. 1 from the rear side of the measuring apparatus. Drives the gear to be measured 6 supported by the support device 7. Further, on the side opposite to the gear supporting device 7 that supports the gear 6 to be measured, a radial direction setting table 15 that can adjust the measurement portion by a servo mechanism is provided corresponding to the size of the gear, and various types of gears having different dimensions can be provided. Corresponding to the measurement gear, left and right,
The interferometer surface plate 16 can be moved up and down.

Here, as shown in FIG. 4, the interferometer base 16
Includes a laser light source 1, a half-wave plate 19, a beam splitter 2, and a mirror 4 (in this example, a pair of mirrors 4a and 4a).
b), piezo element 3a for fringe scanning, mirror 3, quarter wave plate 21, beam splitter 20, optical path length adjusting mirror group 22, mirror 35 for changing beam traveling direction, CCD camera 8 and rotating mechanism 17 Etc. are attached.

The rotating mechanism 17 is arranged between the mirror 4 and the taking lens 5 in FIG.
It is provided for the purpose of irradiating the object light S on the. The gear 6 to be measured is generally designed to have various right twists and left twists, but as shown in FIG. 5, the rotating mechanism 17 has a linear mechanism so that the tooth flank 6a to be measured can be irradiated in accordance with the twist angle. −
Although it can be rotated by a servo mechanism 26 (FIG. 4) equipped with a servo motor, the object light S enters the fixed mirror 27a at the center of the mechanism from the mirror 4, and the spatial filter 24, a pair of expander lenses from the rotating mirror 28a. After passing through 29, the optical probe 30 is entered. That is, FIG.
With the deflection angle prism 31 shown in FIG. 13, the tooth surface 6a to be measured is measured.
After passing through the deflection angle prism 32, the light is transmitted from the rotating mirror 28b and the fixed mirror 27b, which are coaxial rotation centers, to the taking lens 5.

As described above, since the fixed mirrors 27a and 27b are placed at the center of rotation, even if the mechanism 17 rotates by an angle corresponding to the twist angle of the gear 6 to be measured, the object light S can be reliably photographed. It can pass through the lens 5. An optical tracing stylus 30 is provided between them, which can correspond to the right tooth surface and the left tooth surface, as will be described later, and the measurement is performed by switching between the motor 33 and the gear mechanism 33a. The reflected light from the tooth surface 6a forms an image on the light receiving surface 8a of the CCD camera 8 by the taking lens 5 through this optical path system and mechanism. Further, the piezo element 3a on the interferometer base 16 finely moves the mirror 3 in response to a command from the computer 12 to perform fringer scanning.

By superimposing the reference light R on this image,
The tooth surface image data of the generated interference fringes is stored in the frame memory 10 as shown in FIG.
Alternatively, a computer 12 including a microcomputer
Is input, stored, and calculated. The image captured by the CCD camera 8 is displayed on the screen of the CRT 13 as needed.

The computer 12 has a storage device having various capacities necessary for image processing and various output devices. On the other hand, the computer 12 uses the servo motor 9a as a drive source via the interface 11 and precisely controls the rotation angle amount of the worm gear mechanism. The servo drive device 9 having the encoder device 9b for measuring and outputting can be controlled, and further, as shown in FIG.
The measuring device can be set to an appropriate measuring position on the x, y, z coordinate system of the gear by the servomotors of the driving devices 9c, 9d, 9e, respectively.

The twist angle of the rotating mechanism 17 can be set within ± 45 ° by a servo drive mechanism having a linear servo motor 26, as shown in FIG. Further, the prisms for the right and left tooth surfaces of the optical probe 30 are switched by the motor 33 and the gear mechanism 33a. In order to guide the object light S that slightly changes the reflection direction to the lens 5 and the CCD camera 8 by the twisted surface of the helical gear, as shown in FIG. 6A, the small prism 34 is rotated around the X and Z axes. There are mechanisms 34a and 34b for controlling the fine movement rotation, and servo mechanisms 25a and 25b for controlling the fine movement rotation of the half mirror 25 around the y axis and the xz axis as shown in FIG. 6B, both of which are provided with a gimbal mechanism. The structure is such that it can be controlled from the computer 12. The specifications of these gears are determined, and when the measurement program of the computer is started, the specifications are determined based on the empirical values in the computer 12, and each is moved to an appropriate position to set the above-mentioned prism, mirror, etc., and measured. Prepare for

Since the CCD camera 8 outputs a signal output according to the brightness for each pixel of the light receiving surface 8a, the interference fringes received as the amount of light are analyzed by the computer 12 by, for example, the well-known fringe scan method, The phase of each pixel is calculated, and the length (height) is calculated according to the effective wavelength depending on the incident angle to obtain the shape of the tooth surface 6a. This calculation process can be performed in the same manner as that disclosed in Japanese Patent Application No. 3-20084 of the prior application, and since it is not directly related to the present invention, its description is omitted.

Since the phase of the tooth flank of this measurement result shows the actual shape of the tooth flank 6a to be measured, an error is obtained as compared with the simulation of the ideal tooth flank using the above parameters. As mentioned above, the phase difference of each coordinate point is
It is obtained by replacing the length with the effective length of the point, that is, the shape error amount. FIG. 7 shows the relationship between the coordinate positions and parameters of the photographed tooth surface image 6a in the gear coordinate system and the simulated theoretical tooth surface image 18. The superposition at this time is based on the tooth tip of the gear tooth surface and the upper and lower limits of the tooth width, but the angle of incidence on the tooth surface of a helical gear that has an involute helicoid surface is on the tooth surface. Since there is a considerable change at each point, the following methods A and B can be used in order to improve the measurement accuracy.

Method A : When a measurement error is obtained by superimposing the image taken by the CCD camera 8 and the simulation image, if the overlay accuracy does not fall within the required value,
Reverse ray tracing is performed from the obtained surface shape, the parameters of the first ideal value of the tooth surface simulation, such as the sensitivity of the fringes as the first approximation, are corrected, and the corrected simulation image is calculated again to obtain the captured image. Overlap. At this time, if the error is converged from the previous error, the same work is repeated to improve the accuracy. This process is shown as the flowchart of FIG. Method B The image captured by the CCD camera 8 and the simulation image are superposed, the phase difference in each pixel is sampled from the obtained surface shape, and the first-order approximation value of the error is obtained based on the principle of parallel film interference. A simulation image is created for a new tooth profile curved surface that is slightly shifted in the normal direction from each pixel of the simulation image. If the error when this is superimposed on the captured image is more convergent than the previous error, the process is repeated until the required error accuracy is reached. In this way, the tooth surface error is obtained, and the process thereof is described in the flowchart of FIG.

Even if the measured gear 6 is not limited to a spur gear or a helical gear and has a twisted surface, the measurement result is shown in FIG. 11 (a) so that the operator can easily understand the error. ),
As shown in (b), contour lines and colors like a map are displayed on the shape of the spur gear surrounded by p, q, s, and r. p
The error is easy to grasp because the tooth width is between -q and the width between pr is proportional to the rolling length or rolling angle on the basic cylinder, as in the conventional measuring device. The right flank and the left flank are shown for easy viewing in their respective directions. Since the action line passing through the pitch circle is drawn according to the twist angle, the error distribution around this line can be grasped at a glance.

FIG. 12 and FIG. 13 show the shape and configuration of the optical probe 30. Servo mechanism 9
9a, 9b, 9c, 9d, 9e and 26 of FIG.
Is operated to move so as to sandwich the gear 6 to be measured between the measuring element bodies 30 having a U-shape, and irradiate the tooth surface with the object light. Reference numerals 31L, 32L and 31R, 32R are deflection prisms for the left tooth surface and the right tooth surface, respectively. Figure 1
The incident angle θ of the object light on the tooth surface at is determined by the angle α of this prism. The selection of the left and right tooth flanks, that is, whether the measured tooth flank of the measured gear 6 is the right tooth flank or the left tooth flank, is determined by the measurement program of the computer, and is switched by the motor 33. A pinhole 30 is provided in the central portion of the probe main body 30.
The object light S is straightly transmitted through the central pinhole 30a. When the tracing stylus body 30 is in the center, it is adjusted so as to be at the origin position on the y-axis, so that a pinhole is projected on the CRT of the computer.

When the tracing stylus 30 is set in a state where the pinhole is at the center as shown in FIG. 14A and the gear 6 is rotated, the object light S passing through the pinhole is blocked by the tooth surface. Is captured by the CCD camera 8 and observed on the CRT. Looking at the output when passing through the CCD camera 8, since the pinhole 30a has a certain size, the output changes with rotation as shown in FIG. 14B. This data is binarized as shown in FIG. 14 (c) to determine the u and v points of the rotation angle, and the center point thereof is the center point on the coordinate of the gear,
That is, it is defined as the center point of the tooth space. The rotation angle from the center point to the measurement position is the parameter τ on the simulation.

The measurement procedure is summarized as follows, and its flow is shown in FIG. Also, computer 1
The detailed configuration of the device 2 and peripheral devices is clearly shown in FIG. 15, and the relationship between the imaging of the tooth surface image of the CCD camera 8 and the servo drive system are controlled by the computer 12 via the interface 11, respectively. First, the gear 6 to be measured is attached to the gear supporting device 7 on the machine base 14 of the measuring device. Next, the gear specifications are input to the computer 12. If it is a new gear, the necessary items are sequentially input according to the instruction of the computer 12. The specifications registered in advance are called by the tooth number. In order to irradiate the tooth surface with the object light S, there is an appropriate position on the coordinates of the non-contact free curved surface measuring device, and the computer 12 has learned the optimum position in advance by the ray tracing method.
By the gear diameter direction y axis, that is, the gear pitch circle diameter Do, y
The gear 6 and the optical probe 30 of the measuring device move using the basic cylindrical twist angle β _g , which is a rotation about the axis, and the irradiation position x offset in the x-axis direction as parameters, and the irradiation is performed at the calculated position. The measuring device is set. At the same time, a simulation calculation of a three-dimensional stereoscopic image of the tooth surface to be captured by the camera 8, which is a measurement reference, is performed.

Then, using the pinhole 30a at the center of the optical probe 30, the center of the tooth groove of the gear is set as described above, and the origin of the rotation angle τ is determined. When the center point of the tooth groove is determined, the supporting device 7 of the measuring device selects an appropriate measurement position based on the data learned in advance, stops at the point rotated by the angle τ, and starts imaging the tooth surface. To be done. By the fringe scan method, the piezo element 3 is driven n times by λ / 2n, and a total of n 1-pitch stripe-shifted images are sequentially captured by the computer 12. next,
The phase of the tooth surface at each coordinate position is calculated in the phase calculation process, but the fringe scan method can also perform finer interpolation between the fringes if necessary. As described above, the calculation result is compared and calculated with the result of the phase simulation to obtain the difference. For example, if it is necessary to measure a certain portion with higher accuracy, the error can be converged by the method of FIGS. 9 and 10 to obtain higher accuracy. Next, since the effective wavelength at that position can be known according to the angle of incidence of the object light S on the tooth surface, it is converted to the height, the error at each point on the entire tooth surface is calculated, and these points are connected. Then, the three-dimensional image of the measurement tooth surface is completed. When the colors are assigned according to the measurement error amount step, these data are shown in FIG.
On the basic tooth surface such as 1, the highland and the lowland are color-coded like a linear contour line or a map, and can be displayed on the display as visually obvious data or can be color printed. .

Next, the procedure for executing the simulation analysis of the vibration and noise important for the gear from the measured tooth surface shape of the gear by using the present measuring device will be described.
FIG. 16 shows the measured contour data arranged according to the meshing state of the gears. According to the principle of meshing of gears, when meshing, each tooth appears every 2π / z and participates in meshing. Therefore, where two pieces of data overlap, there is meshing of two teeth, and one area is meshing of one tooth. For the two-tooth meshing part, the same part of both data is compared, and the tooth with a positive error participates in the meshing.

However, when transmitting power, the teeth bend under load like a cantilever. Even in the case of a geometrically ideal gear that has no static error, when a load is applied during power transmission, it flexes, and when meshing ends, it is released from the load and repeatedly returned to its original state, causing vibration. become. Therefore, the transmission accuracy of the gear cannot be evaluated only by the static error. That is, if the situation can be predicted in advance from the measurement result by simulation, it will greatly contribute to the design and manufacture of a gear with less noise. Therefore,
After performing the above-mentioned measurement for both the driven and driven gears, select the tooth data with typical error content from them and simulate it with a computer,
Analysis is performed in consideration of the bending of the tooth due to the load, and under a certain load, detailed vibration starting force / transmission error and various stress states are analyzed.

In this analysis method, the load distribution on the contact line is calculated from the number of simultaneous contact lines and their positions from the contour lines of the error of the tooth surface shape obtained by the above measurement, and the contact deformation and the degree of convergence of the load distribution are calculated. Then, a load correction loop is repeatedly used to calculate the total load by stacking the calculations, and finally the excitation force index value and various generated stresses are output. In FIG. 17, the action plane of one tooth is shown by a rectangle in the figure surrounded by the root and both side ends of the tooth, and if the contact line at a certain rotation angle position is L _j , the meshing is from point S to point E. Proceed to. More specifically, when the distributed load at the coordinate points in a loaded state with a tooth surface of a helical gear having an error is Pj (ξ), the vibration excitation force index value is given by the following equation (1). It is obtained by solving under the condition of (2).

[Equation 1]

This is shown in a flow chart in FIG.
The loop as shown in the flow chart of the gear power transmission state analysis program of FIG. 21 will be solved. The result of the simulation can be output as an analysis value of a tooth contact state diagram, a root stress, a load distribution chart, a tooth surface contact stress, an instantaneous tooth surface temperature rise, and the like, and a vibration excitation force transmission error diagram. Further, these data can be Fourier analyzed and displayed. It is already known that the transmission error and the vibration excitation force have a correlation with the meshing noise of the gears, so it is possible to know the vibration situation under a certain torque by the simulation of this method in advance. Appropriate processing information including cutting tools and grinding tools can be promptly fed back to the processing department, so that large time and economic benefits can be obtained.

The contents of the loop of the analysis flow shown in FIG. 18 are as follows. Loop A: A repetitive routine in which one meshing pitch is divided into N parts. Loop B: A routine for matching the transmission load calculated based on the rotation delay angle Δ and the input load. C loop: A routine for obtaining the load distribution on the contact line of each tooth surface that is meshing at the same time. Loop of D: A routine for determining the true contact area on one contact line. Loop E: Convergence of non-linear compliance to tooth deformation and calculation routine for load distribution. Here, n is the number of teeth simultaneously meshed. x is the coordinate of the point where the deflection of the pair of teeth is considered (tangential line L _j ). ξ is a point considering the load distribution on the contact line L _j . C _bj (x, ξ) is the bending and shear deformation of the x point caused by the unit load acting on the ξ point, and the deflection due to the inclination of the tooth base. C _cj (x) is the amount of contact approach to the unit load at point x. Δ is a value obtained by converting the rotation delay angle of the driven gear with respect to the driving gear into the tooth surface movement distance on the line of action. e
_j (x) is the sum of deviations (combined error) of the driven gear from the ideal involute helicoid surface relative to the driving gear in the direction of the normal to the front cross section. β _g is the twist angle on the base cylinder. W is the transmission normal force. T _Q1 and T _Q2 are transmission torques of driving and driven gears. r
_g1 and r _g2 are the basic circle radii of the driving and driven gears.

[0035]

As is apparent from the above description of the embodiments, according to the conventional method, the reference tooth surface has to be manufactured. However, according to the present invention, the interference image of the tooth surface is obtained. By comparing with theoretical values simulated by a computer, precise and extremely rapid measurement is possible, and especially with the remarkable development of computers and CCD cameras, the measurement time and data processing method are speeded up. Can expect more. In addition, by inputting the measurement results to the vibration simulation, it will be possible to perform unprecedented epoch-making analysis in a short time, provide valuable data for gear design and manufacturing, and greatly contribute to quality improvement. It will be.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a basic mechanism of a non-contact free-form curved surface measuring apparatus used for carrying out a non-contact tooth surface shape measuring method according to the present invention.

FIG. 2 is a block diagram showing a configuration of a conventional non-contact free-form curved surface measuring device, and a diagram for clearly showing a difference from the device used in the present invention.

FIG. 3 is a three-dimensional view showing the overall configuration of the non-contact free-form surface measuring device used in the present invention.

FIG. 4 is a block diagram showing a configuration of an interferometer surface plate which is a part of the measuring apparatus.

FIG. 5 is a perspective view showing an internal configuration of a rotation mechanism of the device.

6A and 6B are perspective views showing a small prism gimbal servo device and a final wedge-shaped half mirror gimbal servo device of the same measuring apparatus.

FIG. 7 is an explanatory perspective view showing parameters necessary for associating the coordinate of the gear by simulation and the coordinate of the tooth surface actually photographed.

FIG. 8 is a flowchart showing a measurement flow.

FIG. 9 is a partial flowchart showing a part of the measurement flowchart and showing another kind of method for calculating an error by comparing a simulation image with an image captured by a camera.

FIG. 10 is a partial flowchart showing a part of the measurement flowchart and showing another kind of method different from the flow of FIG. 9 for comparing the simulation image with the image captured by the camera and calculating the error.

11A and 11B are views showing a reference tooth surface which allows a gear engineer to understand the condition of the tooth surface at a glance, and a state in which an error is displayed on the reference tooth surface.

FIG. 12 is a mechanism perspective view showing a structure of an optical probe and a pinhole for measuring a reference point of a deviation angle prism and a gear for measuring left and right tooth surfaces to be measured.

13A and 13B are a top view and a front view of the optical probe shown in FIG.

FIG. 14 is an explanatory diagram showing a procedure for setting the origin of the gear,
FIG. 6A is a relational diagram in which an optical tracing stylus is set along a twist angle of a gear and a reference point of the gear is measured. (B) is a diagram showing the relationship between the gear rotation angle and the amount of received light, and shows that the light that has passed through the pinhole is captured by the light receiving surface of the camera and cut by the tooth surface as it rotates, resulting in a change in the amount of light. ing.
FIG. 6C is a diagram showing that the center of the tooth groove is set by binarizing the light amount.

FIG. 15 is a diagram showing the configuration of a computer, an interface, and peripheral devices.

FIG. 16 is an explanatory diagram showing a situation in which meshing conditions of both driving and driven gears are included and combined continuous transmission action error surfaces of tooth surfaces are used as data for vibration analysis of gears.

FIG. 17 is an explanatory diagram showing a load distribution on a tooth surface when a load is applied on one tooth surface.

FIG. 18 is a view for explaining the connection relationship of FIGS. 19 to 21.

FIG. 19 is a flowchart of a vibration analysis by a computer, which is a former part of an explanatory flowchart showing a situation of an analysis loop.

FIG. 20 is the middle part of the flowchart.

FIG. 21 is a latter part of the flowchart.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Light source 2 ... Beam splitter 3 ... Mirror 3a ... Piezo element 4 ... Mirror 6 ... Gear to be measured 7 ... Gear support 8 ... CCD camera 8a ... Light receiving surface 9 ... Servo drive 10 ... Frame memory 11 ... Interface 12 ... Computer 13 ... CRT 14 ... Measuring device base 15 ... Radial direction setting base 16 ... Interferometer surface plate 17 ... Rotation mechanism 19 ... 1/2 wavelength plate 20 ... Beam splitter 21 ... Quarter wavelength plate 22 ... Optical path length adjustment Mirror 23 ... Expander 24 ... Spatial filter 25 ... Wedge-shaped half mirror 26 ... Linear servo motor 27 ... Fixed mirror 28 ... Rotating mirror 29 ... Expander lens 30 ... Optical probe 31 ... Deviation prism 32 ... Deviation prism 33 ... Left / right tooth surface switching motor 34 ... Optical path conversion prism 35 ... Optical path conversion mirror

Claims (3)

[Claims] 1. A gear to be measured is attached to a rotation spindle of a non-contact free-form surface measuring device, coherent light emitted from a light source is separated into reference light and object light, and the object light is a target tooth of the gear to be measured. By irradiating the surface, to obtain the interference fringe image of the target tooth surface from the reference light and the reflected light of the object light, in the non-contact measurement method of the tooth surface shape to measure the shape error of the target tooth surface, Using a part of the pinhole optical path of the interferometer, automatically set the measurement origin for the center of rotation of the gear to be measured, then, the parameters of the measurement entered in advance based on the gear specifications of the gear to be measured According to the above, the measurement positioning is performed from the measurement origin, and the object light is irradiated to one tooth surface, and the reflected light is guided to the CCD camera light-receiving surface having an extremely large number of pixels. At the same time, on the CCD camera light-receiving surface, Reference light and the object light To directly interfere with the reflected light of the reference light so as to obtain interference fringes on the tooth surface, and a mirror inserted in the middle of the reference light is divided into several wavelengths of the coherent light by a driving means such as a piezo element. By finely moving and moving the interference fringes by one pitch,
The movement of the stripes for each division is taken out for each pixel of the CCD camera and the phase of the stripes is analyzed. Also, from the specifications of the gear to be measured, the gear mounting position relationship on the measuring machine, and the irradiation position of the object light in advance. A theoretical tooth surface shape is created by simulating an image expected to be captured by a camera by a computer, and the phase difference of the stripes in each pixel is calculated by comparing and calculating with the created theoretical tooth surface shape, Converting the effective wavelength due to the incident angle at that point into the height of the surface shape, the tooth surface error for the entire shape of the tooth surface is obtained, and the obtained tooth surface error is applied to the basic shape of the gear and colored in a map shape. A non-contact measuring method of a tooth flank shape of a gear, comprising a measuring step for displaying an error by contour lines according to. 2. A comparison between a photographed image of the tooth surface of the gear to be measured and a theoretical tooth surface shape obtained by the simulation.
2. The high precision tooth surface error information is obtained by applying the ray tracing method and the inverse ray tracing method of the photographed image to the error of the comparison positions of the two occurring at the time of calculation. Method for measuring the non-contact shape of the tooth flanks of gears. 3. The tooth surface shape of all the teeth of the gear to be measured by measuring the tooth surface error of the one tooth surface for each tooth surface of a pair of driven and driven gears to be measured. By measuring the error, selecting the combination of the typical error shapes of the driving and driven tooth flanks, and inputting it to the computer simulation analysis program for the gear power transmission state, Transmission error during power transmission,
The non-contact measuring method of the tooth flank shape of the gear according to claim 1, wherein characteristics such as vibration excitation force and various stress states are analyzed.