A method for extracting the waviness of involute helical gears suitable for error tracing
By using an involute helical gear waviness extraction method, combined with least squares polynomial fitting and Fourier transform, the problem of mid-frequency waviness control in new energy vehicle gears was solved, and the reliability of error tracing and noise control was improved.
Patent Information
- Application Number
- CN202410111369.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-26
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2044-01-26
AI Technical Summary
Existing technologies make it difficult to reliably control the waviness of the gear surface in the mid-frequency range during gear processing for new energy vehicles, leading to difficulties in tracing the source of errors and affecting gear noise control.
An involute helical gear waviness extraction method is adopted. By measuring the waviness of the tooth profile and tooth direction, and combining least squares polynomial fitting and Fourier transform, the full tooth deviation curves are separated and connected. The curve coincidence quality and fitting quality are considered to reduce the error.
A method for characterizing the waviness of the entire tooth surface is provided, which reduces curve fitting errors, provides a data basis for error tracing during worm gear grinding, and improves the reliability of gear noise control.
Smart Images

Figure CN118149752B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of gear measurement and evaluation, and in particular to a method for extracting the waviness of involute helical gears suitable for error tracing. Technical Background
[0002] New energy vehicles, with their advantages of simple structure, high energy efficiency, and environmental friendliness, are currently experiencing explosive growth. Since the transmission systems of new energy vehicles lack the background noise of internal combustion engines, lower gear noise is required to improve passenger comfort. Under the high-speed operation conditions of new energy vehicles, gear morphology (tooth profile and direction error, roughness, and waviness) is the main factor affecting gear meshing noise. Currently, new energy vehicle gears are machined using continuous generating grinding, a method that ensures relatively stable and controllable tooth profile and direction error, roughness, etc., and the methods for detecting and evaluating these errors and roughness are relatively mature. However, mid-frequency tooth surface waviness, mainly generated by machine tool vibration, is currently difficult to reliably control. Therefore, reliable extraction of tooth surface waviness is crucial for new energy vehicle gear manufacturers.
[0003] Currently, tooth surface waviness separation is achieved by directly removing shape errors from the single tooth profile total deviation curve through parabolic regression to obtain the single tooth profile waviness deviation curve. Considering the continuity of tooth profile waviness, the value with the higher amplitude is selected as the deviation value of the overlapping part in the overlapping area of the curve, thus obtaining the complete tooth surface information of the gear with 360 degrees of continuity. The continuous tooth surface information is then further filtered to obtain the tooth profile waviness.
[0004] Regarding the above method, the inventors believe that the parabolic fitting quality was not considered when separating the single tooth profile waviness deviation curve. In addition, the subsequent direct selection of a higher deviation value in the overlapping area may lead to errors, which will hinder the traceability of errors in the gear processing. Summary of the Invention
[0005] In view of this, the present invention proposes a method for extracting the waviness of involute helical gears suitable for error tracing.
[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:
[0007] A method for extracting the waviness of involute helical gears suitable for error tracing, the method comprising the following steps:
[0008] (1) Obtaining the total deviation information of gear tooth profile / tooth waviness measuring points
[0009] The tooth profile curve and the tooth profile curve in the same normal plane are measured under the same diameter. The tooth profile / tooth direction measurement principle is based on the VDI VDE 2612Blatt 1 standard. The obtained one-dimensional coordinate data is the total normal deviation value of several tooth surface measuring points on the gear. The total deviation includes tooth profile / tooth direction bulge, shape error, waviness and roughness.
[0010] According to the tooth profile measurement principle described in the VDI VDE 2612Blatt 1 standard, the relative length of the involute normal family corresponding to each measuring point of a single tooth profile is taken as the abscissa of the single tooth profile measuring point, and the tooth width position of the tooth direction measuring point is taken as the abscissa of the single tooth direction measuring point.
[0011] According to the reference evaluation range described in VDI VDE 2612 Blatt 1 standard, the single-tooth waviness evaluation curve is extracted, and the tooth profile / tooth direction starting point is taken as the zero point, and the abscissa of the measurement point is translated.
[0012] (2) Removal of tooth profile / tooth bulge considering curve coincidence mass
[0013] By using the least squares polynomial fitting method, the tooth profile / tooth direction bulging modification is removed from the single tooth waviness evaluation curve, and the single tooth profile / tooth direction deviation curve is obtained.
[0014] The order of the highest term of the polynomial affects the profile / direction deviation curve of a single tooth. Therefore, the order of the highest term is determined iteratively by the difference in the overlapping region of adjacent tooth curves. The basic form of the polynomial is as follows.
[0015]
[0016] Where m is the order of the highest term and p is the coefficient of the p-th term.
[0017] Since the tooth surface bulge shaping and other features are curves, the initial value of the highest term order m is 2.
[0018] (3) Multi-tooth / full-tooth deviation curves coincide
[0019] The overlapping area of adjacent tooth curves is obtained by calculating the tooth profile / tooth direction overlap, and the zero point of the rear tooth deviation curve is shifted to the starting position of the front tooth curve overlap.
[0020] Through the above steps, two results emerge regarding the overlapping region of the final full-tooth profile / tooth direction deviation curves. Result 1: The deviation values and trends corresponding to the overlapping portions of the abscissas of adjacent teeth differ significantly. Result 2: The deviation values and trends corresponding to the overlapping portions of the abscissas of adjacent teeth are roughly the same.
[0021] In the continuous generating grinding process of gears, there are situations where two or more positions are processed simultaneously. Under the same rotation angle, there are similar ripples formed by factors such as processing vibration. Therefore, the fitting quality is judged by the difference in the overlapping area of the curves, and the final form of the fitting formula is determined.
[0022] If result one occurs in the overlapping region of tooth profile / tooth direction curves, then return to least squares polynomial fitting, increase the order of the highest term of the drum-shaped fitting polynomial, and continue to complete the full tooth curve overlap until result two occurs.
[0023] The average deviation is taken in the overlapping part of the tooth profile / tooth direction horizontal coordinate to obtain the final full tooth deviation connection curve for one revolution of the gear.
[0024] (4) Obtaining the order and amplitude of tooth profile / tooth direction waviness
[0025] Based on the tooth profile and tooth direction curve formed by one revolution of the gear, the tooth profile / tooth direction waviness is extracted by Fourier transform.
[0026] The Fourier transform frequency is converted to the gear order using the following formula.
[0027] O = L × f
[0028] Where O is the order of the tooth profile / tooth direction, L is the length of the full tooth connection curve, and f is the geometric frequency after Fourier transform.
[0029] The beneficial effects of this invention are as follows:
[0030] 1. This invention combines the measurement principle of involute cylindrical gears and considers the overlap of adjacent tooth curves to connect the full tooth deviation curve, thus providing a method for characterizing the waviness of the full tooth surface of involute helical gears.
[0031] 2. This invention considers the gear waviness formation mechanism in worm wheel generating grinding. When separating the waviness deviation curve of a single tooth profile, it takes into account the curve fitting quality, reducing the error caused by inaccurate curve fitting, and providing a data basis for tracing the source of errors in the worm wheel gear grinding process. Attached Figure Description
[0032] Figure 1 This is a flowchart illustrating a method for extracting the waviness of involute helical gears suitable for error tracing in this invention.
[0033] Figure 2 Layout diagram of tooth profile / tooth waviness measurement location;
[0034] Figure 3 Schematic diagram of tooth profile waviness measurement principle;
[0035] Figure 4 This is a curve fitting diagram of the total deviation of a single tooth profile.
[0036] Figure 5 This is a curve fitting diagram of the total deviation of a single tooth along the tooth direction.
[0037] Figure 6 Schematic diagram for calculating tooth profile overlap ratio;
[0038] Figure 7 This is a schematic diagram for calculating the overlap ratio of tooth profile curves.
[0039] Figure 8 This is the result of the tooth profile / tooth direction deviation coincidence curve appearing after step G of the present invention;
[0040] Figure 9 This is the second result of the tooth profile / tooth direction deviation coincidence curve appearing after step G of the present invention;
[0041] Figure 10 This is the final tooth profile deviation connection curve obtained in this invention;
[0042] Figure 11 This is the final tooth deviation connection curve obtained in this invention;
[0043] Figure 12 The results of extracting the tooth profile waviness of this invention;
[0044] Figure 13 This is the extraction result of the tooth waviness in this invention. Detailed Implementation
[0045] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.
[0046] A method for extracting the waviness of involute helical gears suitable for error tracing, the method comprising the following steps:
[0047] A: Use a gear measuring machine (such as a Klingberg P40) to measure the tooth profile curves under the same diameter and the tooth profile curves under the same normal plane (such as...). Figure 1 The measurement principle of tooth profile / tooth direction is based on VDI VDE 2612 Blatt 1 standard, such as... Figure 2 As shown.
[0048] B: For tooth profile measuring points, the relative lengths of the involute normal family corresponding to each measuring point of a single tooth profile are obtained according to the following formula and used as the abscissa of the single tooth profile measuring point.
[0049]
[0050] Where, x pi d represents the relative length of the involute normal family (the x-coordinate of the measurement point on the single tooth profile). s The diameter of the measuring circle at the starting point is represented by d.e N indicates the final measurement point for the diameter of the circle. p n represents the number of measurement points on a single tooth profile. pi Indicates the sequence number of the tooth profile measuring point.
[0051] D: According to the reference evaluation range described in VDIVDE 2612Blatt 1 standard, the single-tooth waviness evaluation curve is extracted, and the tooth profile / tooth direction starting point is taken as the zero point, and the horizontal coordinate of the measuring point is translated.
[0052] F: Using the least squares polynomial fitting method, the single-tooth waviness evaluation curve is modified to remove tooth profile / tooth direction bulging, etc., to obtain the single-tooth tooth profile / tooth direction deviation curve, such as... Figure 3-4 As shown. The polynomial must satisfy the following formula, the form of which is determined iteratively by the meshing quality (curve overlap) described in step H below.
[0053]
[0054] Among them, y i The total deviation of the measurement point (the one-dimensional coordinate data described in step A), x i a is the abscissa of the tooth profile / tooth direction measuring point mentioned in step D. p Let be the coefficient of the p-th term of the polynomial, and m be the order of the highest term of the polynomial.
[0055] The initial highest term order m of the fitted polynomial is 2.
[0056] G: The full tooth profile / tooth direction deviation curves coincide. (This is achieved through...) Figure 5-6 The principle shown is used to calculate the tooth profile / tooth direction overlap to obtain the overlap area of adjacent tooth curves, and then shift the starting point of the rear tooth deviation curve to the starting position of the front tooth curve overlap.
[0057] In step G, the formula for calculating the tooth profile / tooth direction overlap ratio is:
[0058]
[0059] Among them, e p Indicates the degree of tooth profile overlap, e l The gear indicates the tooth contact ratio, z represents the number of teeth on the measured gear, and α... Fe The final evaluation of the tooth profile pressure angle, α e β is the tooth profile pressure angle at the starting point of evaluation, b is the evaluation tooth width of the measured gear, and β b To measure the fundamental helix angle of a gear, d b To measure the base circle diameter of the gear.
[0060] In step G, after all the rear teeth are translated to the starting position where they coincide with the corresponding front teeth, the tooth profile / tooth direction abscissa of each tooth is obtained by the following formula.
[0061]
[0062] Where, x i k The z-coordinate represents the x-coordinate of the i-th measuring point in the tooth profile / tooth direction of the k-th tooth. k Let x represent the k-th tooth. Fa The x-coordinate value represents the final evaluation point. E The x-axis value represents the starting point, and e represents the tooth profile / tooth direction overlap.
[0063] In step F, the fitting polynomial form (highest term order) for removing the drum shape differs, resulting in two different outcomes for the final tooth profile / tooth direction deviation coincidence curve obtained after step G. Result 1: The curves corresponding to the overlapping portions of the abscissas of adjacent teeth have poor overlap (e.g., ...). Figure 7 ); Result 2: The curves corresponding to the overlapping portions of the abscissas of two adjacent teeth have a good degree of overlap (e.g., Figure 8 ).
[0064] The degree of overlap of the above curves is measured by the following formula:
[0065]
[0066] Where Adv represents the degree of curve overlap, and y i k y represents the deviation value of each point on the front teeth in the overlapping region of the curve. j k N represents the deviation value of each point corresponding to the tooth in the overlapping area of the curve. o This represents the number of points in the overlapping region of the curves.
[0067] If Adv is less than or equal to 60%, result 1 is determined; if Adv is greater than 60%, result 2 is determined.
[0068] If result one occurs after step G, return to step F to increase the number of terms in the drum-shaped fitting polynomial, and continue executing step G until result two occurs.
[0069] H: If result two occurs (e.g.) Figure 9 As shown), the average deviation of the overlapping portion of the tooth profile / tooth direction abscissa is taken to obtain the final full tooth deviation connection curve for one revolution of the gear, as shown. Figure 10 .
[0070] I: Based on the tooth profile and tooth direction curve formed by the gear in one revolution, the tooth profile / tooth direction waviness is extracted by Fourier transform.
[0071] J: Convert the geometric frequency of the Fourier transform to the gear order, such as Figure 11-12 The conversion formula is as follows:
[0072] O = L × f
[0073] Where O is the order of the tooth profile / tooth direction, L is the length of the full tooth connection curve, and f is the geometric frequency after Fourier transform.
[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A method for extracting the waviness of involute helical gears suitable for error tracing, characterized in that, The method includes the following steps: 1) Obtaining the total deviation information of gear tooth profile / tooth waviness measurement points: Based on the measurement principle and measurement range, the tooth surface position correspondence of the one-dimensional coordinate data results is realized, and the single tooth waviness evaluation range curve is obtained according to the evaluation range. The acquisition of the total deviation information of gear tooth profile / tooth waviness measurement points involves measuring the tooth profile curve of the same diameter or the tooth profile curve of the same normal plane using a gear measuring machine. The obtained one-dimensional coordinate data includes the total normal deviation value of several tooth surface measurement points on the gear. This total deviation includes tooth profile / tooth direction bulging, shape error, waviness, and roughness. Based on the normal measurement principle and measurement range, the length of the involute normal family corresponding to each measurement point of the single tooth profile is supplemented, and the tooth width position of the tooth direction measurement point is used as the abscissa of the single tooth tooth direction measurement point. The single tooth waviness evaluation range curve is extracted according to the evaluation range, and the tooth profile / tooth direction starting point is used as the zero point. The abscissa of the measurement point is then translated. 2) Tooth profile / tooth direction bulge removal considering curve overlap quality: Using the least squares polynomial fitting method, tooth profile / tooth direction bulge is removed from the single tooth waviness evaluation curve to obtain the tooth profile / tooth direction deviation curve for each tooth. The polynomial form is determined based on the subsequent curve overlap quality. 3) Full tooth deviation curve coincidence: The coincidence area of adjacent tooth curves is obtained by calculating the tooth profile / tooth direction coincidence. The starting point of the rear tooth deviation curve is shifted to the starting position of the front tooth curve coincidence. The form of the fitting polynomial is determined iteratively based on the degree of curve overlap. Two results were found in the overlapping region of the full tooth profile / tooth direction deviation curves: Result 1: The overlap of the curves in the overlapping part of the horizontal coordinates of two adjacent teeth was poor; Result 2: The overlap of the curves in the overlapping part of the horizontal coordinates of two adjacent teeth was good. If Result 1 was found in the overlapping part of the curves, the highest degree of the fitting polynomial was increased until Result 2 was found. Then, the mean deviation of the overlapping part of the horizontal coordinates was taken to obtain the full tooth deviation connection curve. The degree of overlap of the above curves is measured by the following formula: Where Adv represents the degree of curve overlap. For the overlapping region of the curve, the first tooth of the front tooth Deviation value of measuring point For the overlapping region of the curve, the first tooth of the rear tooth Deviation value of measuring point This represents the number of points in the overlapping region of the curves. If Adv is less than or equal to 60%, result 1 is determined; if Adv is greater than 60%, result 2 is determined. 4) Obtaining the order and amplitude of tooth profile / tooth direction waviness: Based on the full tooth profile / tooth direction deviation connection curve, the tooth profile / tooth direction waviness is extracted by Fourier transform, and the spatial frequency is converted into the gear order.
2. The method for extracting the waviness of involute helical gears suitable for error tracing according to claim 1, characterized in that, The tooth profile / tooth direction bulge removal considering curve overlap quality is achieved by using the least squares polynomial fitting method to remove the tooth profile / tooth direction bulge from the single tooth waviness evaluation curve, resulting in the tooth profile / tooth direction deviation curve for each tooth. The degree of the highest term in the fitting polynomial is determined iteratively by the degree of curve overlap, with the initial highest term degree being 2.
3. The method for extracting the waviness of involute helical gears suitable for error tracing according to claim 1, characterized in that, The tooth profile / tooth direction waviness order and amplitude are obtained by extracting the tooth profile / tooth direction waviness through Fourier transform based on the full tooth profile and tooth direction deviation connection curve formed by the gear rotating one revolution, and converting the frequency after geometric Fourier transform into the gear order.
Citation Information
Patent Citations
Modeling method for principle error of tooth surface in forming grinding of longitudinal modified helical gear
CN107908857A
Tooth surface waviness evaluation method
JP2018185166A
