GB2121543A

GB2121543A - Method and apparatus for automatically sorting gearwheels with surface defects

Info

Publication number: GB2121543A
Application number: GB08310961A
Authority: GB
Inventors: Manfred Erhardt; Heinrich Fischer; Monchen-Wertherstr Piekenbrink
Original assignee: Hurth Verwaltungs GmbH
Current assignee: Hurth Verwaltungs GmbH
Priority date: 1982-05-18
Filing date: 1983-04-22
Publication date: 1983-12-21
Also published as: DE3218811A1; IT1197584B; FR2527104A1; GB8310961D0; IT8347778A0

Abstract

For sorting gearwheels with surface defects, the gearwheels 5 roll in two-flank abutment, under a slight radial pressure with a master gear 8. A measurement signal produced by changes in the inter-axial distance and in the form of voltage pulses which may be compared with respect to size, duration and frequency with variable threshold values is evaluated. Depending upon the test result the gearwheels 5 are automatically sorted by a switch 15 in accordance with predetermined criteria.

Description

SPECIFICATION Method and apparatus for automatically sorting gearwheels with surface defects The invention relates to a method according to the preamble of Claim 1 and to an apparatus for automatically performing the method.

Like otherworkpieces, gearwheels are always susceptible to the risk of damage during manufacture, in transit etc. Minor scratches, indentations, bulges and the like remain frequently concealed on the tooth flanks during the final inspection, but may subsequently give rise to annoying gear noises.

Visual inspection before fitting is very extravagant in terms of manpower and also gives no guarantee that all the defects will be discovered. Operating testing with a master gear or the counter gear does not always lead to the desired result, since the noise behaviour of a gearwheel or a pair of gearwheels in the testing machine is frequently different from the behaviour in the gear housing.

On the basis of these observations, the object of the invention is to develop a method and an apparatus suitable therefor, by means of which gearwheels may be subjected to a total inspection of their tooth flanks during series production and in the process unsatisfactory gearwheels may likewise be picked out.

This object is attained by a method which has the features of Claim 1. Any irregularity of or damage to one or more tooth flanks of the gearwheel to be tested results in a change in the inter-axial distance and thus a momentary change in position of the master gear. These changes are converted by a pickup into electrical measurement signals which give information on the size of the defect. The pickup may, according to preference, be a displacement, velocity or acceleration pickup. The nature of the defect may be deduced from the length of time that the measurement signal occurs and/or from the frequency of the same or similar signals. These measurement signals are compared with externally variable threshold values and the gearwheels are then sorted in accordance with the test result.For the purpose of a comparison of the measurement signals with the threshold values it is advantageous to ignore a possible concentricity deviation of the respective gearwheel. Since many defects become apparent in differing degrees depending upon the direction of rotation of the gearwheel, the inspection is preferably performed in both directions of rotation. For economic reasons the test result should differentiate at least between the three criteria set out in Claim 4. Furthermore, additional criteria are possible, e.g. the necessary reworking may be further specified. The use of a master gear with a number of teeth clearly different from that of the gearwheel, e.g. double the number, allows damage or a deposit of dirt on the master gear to be detected.

Although the tested workpieces are automatically allocated by way of the sorting diverter in accordance with the test result, an optical display of the test result is advantageous. In addition, it may be desirable to record the measurement result by way of a graph for calibration or statistical purposes.

The invention also provides apparatus for automatically testing gearwheels and sorting out gearwheels with surface defects, comprising a master gear and driving means for rotating it, holding means for holding rotatably a gearwheel to be tested in tight mesh with the master gear with their axes parallel whereby the gearwheel is driven by the master gear in operation, sensing means for providing signals responsive to variations in the distance between the axes of the master gear and gearwheel during such rotation, means for conveying the gearwheel to a selected one of a plurality of possible destinations, and electronic control means arranged to compare said signals with predetermined values and to control the conveying means for conveying the gearwheel to a destination selected in accordance with the result of the comparison.

The apparatus preferably has the features of Claim 12.

The invention is described below with reference to one exemplary embodiment and illustrated in Figures 1 to 11, in which Figure lisa simplified general view of the most important elements of the testing device; Figure 2 is a block circuit diagram of the measurement electronics; Figure 3 is a flow chart to explain the mode of operation of the measurement electronics; Figure 4 is the displacement signal of a gearwheel, as recorded in the case of two-flank contact-rolling testing, and Figures 5to 11 are various forms of the measurement signal.

A stationary holding device 2,3 is disposed on a stand 1, which has only been suggested in Figure 1, for the purpose of holding in a rotatable manner the gearwheel 5 which is to be tested and which is brought forward by way of a channel 4 or other conveying device. The channel 4 is provided with a separating device (not shown) in order to bring only one of the workpieces 5' up to the holding device 2, 3. A shaft 6 for a pivot arm 7 is mounted in the stand 1 and its free end holds a master gear 8 in a rotatably driven manner. For the purpose of rotatably driving the latter a motor 9, which is connected to the master gear 8 by way of gearwheels, a chain or a belt drive (not shown), is provided coaxially with the shaft 6.

The master gear 8 is in two- flank tooth engagement - under slight radial pressure produced by a spring 11 - with the gearwheel 5 to be tested. In order to change the gearwheel the pivot arm 7, is raised against the force of the spring 11, for which purpose a cylinder-piston unit 10 is provided, the piston rod of which is articulated on the pivot arm 7 and the housing of which is articulated on the stand 1. When it has been released by the holding device 2, 3, the tested gearwheel 5 reaches, by way of a delivery channel 14, a sorting diverter 15, by way of which it reaches one of three magazines 17, 18, 19, depending upon the test result which will be described in detail below. A cylinder-piston unit 16, the piston rod of which is articulated on the sorting diverter 15 and the housing of which is articulated on the stand 1, is provided for driving the sorting diverter 15.

A bracket arm 7', which cooperates with a pickup 20, is mounted on the pivot arm 7. In this connexion the pickup 20 may be a displacement, velocity or acceleration pickup, which picks up the radial deviations of the master gear 8 and converts them into voltage signals. These voltage signals for example pulses are fed to an electronic measuring device 21 where they are processed accordingly and compared with externally supplied threshold values. The sorting diverter is set in accordance with the test result derived therefrom. At the same time the test result is indicated visually. In the example illustrated three criteria are distinguished: "satisfactory", "rework" and "reject", which are symbolized by corresponding lights 40,41,42 on a display 28.A recording mechanism 29 may be additionally connected to the electronic measuring device in order to record the test result, which is useful particularly for calibrating the apparatus or for statistical purposes.

The essential parts of the electronic measuring device 21 are illustrated in a block circuit diagram in Figure 2 and their mode of operation is shown in a flow chart in Figure 3. The voltage pulses generated by the pickup 20 are fed in succession to an amplifier 24, a rectifier 25, a level meter 26 and two comparators 31,32. The pass thresholds (threshold values) of the comparators may be adjusted externally on potentiometers 34, 35 on an operating panel 22. The rectifier 25 is necessary since defects on the tooth flanks of the gearwheel 5 in the positive and negative direction may produce voltage pulses of different magnitude. Since the measurement voltage should not be affected by a possible concentricity error of the gearwheel 5, concentricity errors are suppressed by means of a filter 23.The pulses are registered, e.g. counted, integrated or the like (counting device 27), over a set predetermined time, e.g. two revolutions of the gearwheel 5 in each rotational direction and this amplitude is fed to a fur- ther comparator 33 which likewise comprises a pass threshold which may be set on a potentiometer 36.

If one or more comparators 31,32,33 respond, trigger stages are set which are associated with the comparators and which control relays 37, 38, 39 or suitable switching transistors for positioning the sorting diverter 15.

Reference should also be made to the possibility of providing a further comparator, the pass threshold of which may likewise be set on a potentiometer on the panel, for the purpose of detecting defects in the tooth-to-tooth rolling error.

This comparator is not shown in Figures 2 and 3.

The displacement signal of a gearwheel without damage is shown in Figure 4. It has the typical characteristic of a two-flank rolling diagram with the sinusoidal concentricity deviation and the superimposed tooth-to-tooth rolling error. The displacement signal is the electrical representation of the change in distance in terms of displacement, as it is generally measured and indicated in two-flank rolling testing. In comparison, Figure 5 shows the same measurement signal of the same gearwheel, as processed in the case of the method according to the invention.This measurement signal is the electrical signal which is generated by the pickup 20, with the concentricity and the absolute value of the inter-axial distance between the gearwheel 5 and the master gear 8 suppressed and thus shows the changes emphasized, such as the tooth-to-tooth rolling error from the tooth engagement and - for example in Figure 6 - particularly abrupt changes in the interaxial distance, which are produced by defects.

In Figure 5 the lower threshold value 48 and the upper threshold value 49 are marked in addition to the signal shape. Starting from a base line 47 they may be set for example at 10 Fm and 15011m. Since the lower threshold value is not reached, there is no damage or no damage worth mentioning. The gearwheel is thus classified as "satisfactory".

In the case of the signal shape in Figure 6, on the other hand, there is damage, as indicated by peaks designated 50. Although they exceed the threshold of 10 Fm, they do not reach the upper threshold which has been set. The gearwheel is therefore classified as "to be reworked".

A very similar signal shape is shown in Figure 7.

The peak 50 is again visible twice corresponding to two revolutions of the gearwheel 5. In addition, a greater peak 51 is evident, which only occurs once.

This is a sure indication of the fact that there is damage on the master gear 8 or that dirt or the like has been deposited on a tooth of the master gear 8.

The gearwheel is only classified as "to be reworked" on the basis of the peaks 50 resulting from gearwheel defects.

In contrast to Figure 6, several peaks, which project above the specified upper threshold of 150 Fm, are visible in Figure 8. Such a gearwheel cannot as a rule be readily reworked and is therefore classified as "reject". In many cases, however, the parts designated as "rejects" may be subjected to a thorough visual inspection and a decision may be made in the case of each part as to whether reworking is possible, whether by hand or with more machine working.

Defects in manufacture may also be detected with the method according to the invention. The possibility of detecting defects in the tooth-to-tooth rolling error has already been mentioned. Another case is illustrated in Figure 9. The numerous peaks 53 occurring in direct succession and which exceed the lower threshold may suggest a defect in the finemachining of the tooth flanks. The pulses are registered, i.e. integrated, and if the integration amplitude exceeds a specified threshold value, the gearwheel is likewise classified as "reject".

Defects in the toothing size may also be recognized with the measurement signal. Where a gearwheel has been made too large the mean signal level 54 is above the zero line 47 (Figure 10) and where a gearwheel has been made too small the means signal level 55 is below the zero line 47 (Figure 11). In both cases the gearwheel is classified as "reject".

The lower threshold value 48 and the upper threshold value 49 may also be set invariably in the sorting apparatus. In this way these limits would be uniformly set for all workpieces, irrespective of their dimension (e.g. tooth size or modulus) and their purpose. By virtue of the setting facility, however, the sorting apparatus may be simply changed over and so may be used universally, i.e. adapted to the conditions in question.

The gearwheels classified as "to be reworked" may be reworked by grinding or honing or other fine-machining method suitable for hardened gearwheels and may be put into a proper condition. As a rule the upper threshold value 49 is adjusted in such .

a way that gearwheels with defects within this limit can still be reworked economically with the abovementioned method. A careful inspection of the gearwheels designated "reject" may provide information as to basic defects in manufacture (e.g.

insufficient overmeasure for fine-machining, cf. Figure 9) or when conveying the workpieces (e.g.

damage owing to insufficient care taken to prevent workpieces from striking one another), which may be eliminated.

Claims

1. A method of sorting gearwheels to detect gearwheels with surface defects, in which the gearwheels are conveyed in succession to a holding device in which each successive gearwheel is held individually so as to be stationary but freely rotatable, the held gearwheel is brought into engagement with parallel axes with a rotationally drivable master gear mounted radially resiliently and is driven by the latter while being acted upon radially with two-flank abutment, characterized in that radial deviations in position of the master gear occurring during the rotation of the master gear and the held gearwheel are detected, converted into electrical signals and compared with threshold values in an electronic measuring device, and the gearwheel is then conveyed selectively to a destination selected from a plurality of destinations in accordance with the result of the said comparison.

2. A method according to Claim 1, characterized in that deviations produced by possible concentricity errors are filtered out.

3. A method according to Claim 1 or 2, characterized in that the master gear and the gearwheel rotate first in one direction of rotation and then in the opposite one.

4. A method according to any one of Claims 1 to 3, characterized in that the comparison differentiates at least between the criteria "satisfactory", "rework" and "reject", and a corresponding destination is provided for each criterion.

5. A method according to any one of Claims 1 to 4, characterized by the use of a master gear with a number of teeth clearly differing from that of the gearwheel.

6. A method according to any one of Claims 1 to 5, characterized in that the comparison result is indicated visually.

7. A method according to any one of Claims 1 to 6, characterized in that the comparison result is indicated by way of a graph.

8. A method as claimed in any of the preceding claims in which the threshold values are adjustable.

9. A method as claimed in any of the preceding claims in which the comparison is effected with respect to signal size, duration and frequency.

10. A method of sorting gearwheel substantially as herein described with reference to the accompanying drawings.

11. Apparatus for automatically testing gearwheels and sorting out gearwheels with surface defects, comprising a master gear and driving means for rotating it, holding means for holding rotatably a gearwheel to be tested in tight mesh with the master gear with their axes parallel whereby the gearwheel is driven by the master gear in operation, sensing means for providing signals responsive to variations in the distance between the axes of the master gear and gearwheel during such rotation, means for conveying the gearwheel to a selected one of a plurality of possible destinations, and electronic control means arranged to compare said signals with predetermined values and to control the conveying means for conveying the gearwheel to a destination selected in accordance with the result of the comparison.

12. Apparatus for performing the method according to Claim 1, characterized by a) a stationary holding device for the gearwheels to be tested; b) a conveying device, which is provided where appropriate with a separating device; c) a pivot arm which is articulated at one end and the free end of which receives the master gear which can be driven rotatably; d) a drive device which effects a pivoting movement of the pivot arm; e) a biasing deviceforthe pivot arm; f) a removing device with a displaceable sorting diverter; g) a pickup which converts deviations of the master gear into electrical signals; h) a device for optionally varying the permissible threshold values; and i) an electronic measuring device with devices for evaluating the electrical signals and which is connected to the threshold-varying device, the pickup and the sorting diverter.

13. Apparatus according to Claim 12, characterized in that the electronic measuring device is provided with a filter for filtering out the electrical signals produced by concentricity deviations.

14. Apparatus according to Claim 12 or 13, characterized in that the master gear can be driven in both rotational directions.

15. Apparatus according to any one of Claims 12 to 14, characterized in that the devices for evaluating the electrical signals comprise two comparators for comparing the band width of the electrical signals with an externally variable lower and upper threshold value.

16. Apparatus according to any one of Claims 12 to 15, characterized in that the devices for evaluating the electrical signals comprise a counting or adding device for detecting the frequency of the electrical signals which occur, and a comparatorforcompar- ing this frequency with an externally variable threshold value.

17. Apparatus according to Claim 11 or 12, characterized in that the electronic measuring device or the comparators respectively are connected in such a way that they differentiate the tested gearwheels at least according to the criteria "satisfactory", "rework" and "reject" and move the sorting diverter into corresponding positions.

18. Apparatus according to any one of Claims 12 to 17, characterized in that an indicating device is connected to the electronic measuring device for visual display of the test result.

19. Apparatus according to any one of Claims 12 to 18, characterized in that a recording mechanism for recording the test result as a graph is connected to the electronic measuring device.