DE102009004981B4 - A method of determining erosion on a surface of a workpiece as a result of applying groove structures by an ECM method - Google Patents

Abstract

The inventive method for determining the erosion on a surface of a workpiece as a result of applying groove structures by means of an ECM method comprises the steps of: scanning the surface along a measurement path, wherein at least two adjacent groove structures are detected by the scan, detecting a depth profile of the surface from a multiplicity of measured values recorded along the measuring path, alignment of the measured values with respect to a reference line, identification and selection of regions of the depth profile between the groove structures, determination of a theoretical depth profile to be expected along the measuring path by mathematical approximation, determination of the expected theoretical erosion function by superimposition theoretical depth profiles, fitting the expected theoretical erosion function to the selected measurement data between the groove structures through a mathematical algorithm, and Er averaging the maximum erosion from the result of the adaptation.

Description

Field of the invention

The The invention relates to a method for determining erosion a surface a workpiece in Result of applying preferably microfine groove structures by means of an ECM procedure. In particular, a method for Determination of erosion in bearing surfaces of a fluid dynamic Camp described.

State of the art

fluid Dynamic Bearings typically include at least two rotatable relative to each other Bearing components, the bearing surfaces associated with one another with a bearing fluid, for. B. air or bearing oil, filled bearing gap form. In a known manner the storage areas are assigned and on the Bearing fluid acting bearing grooves provided. In fluid dynamic Store the bearing grooves usually in the form of depressions usually on single or both storage areas applied. These are arranged on corresponding bearing surfaces of the bearing partners Bearing grooves serve as pumping structures, with relative rotation the bearing components within the bearing gap a hydrodynamic Create pressure. For radial bearings, for example, parabolic, sinusoidal or herringbone Used bearing grooves perpendicular to the axis of rotation of the bearing components on the Scope of at least one bearing component are distributed. Axial bearings become spiral-shaped or herringbone species arranged along a circular line (herringbone) bearing grooves used.

The Bearing grooves are usually made by electrochemical machining (ECM: Electro Chemical Machining) applied to the bearing surfaces. ECM is an abrasive manufacturing process for electrically conductive workpieces. at ECM can depend on it from the number and the distance of the bearing grooves to the so-called Erosion on areas not covered by bearing grooves (so-called "land" area) come; this means two adjacent bearing grooves merge into each other in the "land" area, where between the edges of the bearing grooves an undesirable Material removal stops. This is the actually unstructured "land" surface through The ECM method also partially removed. This effect of Erosion can worsen the storage properties.

Further is an inadmissible size or uneven erosion also a sign for a wear on ECM tool or for erroneous process parameters such. B. electrolyte concentration, Duration and / or magnitude of voltage pulses or orientation of the ECM electrode.

Therefore The aim is to randomly determine erosion in ECM production to detect faulty ECM electrodes or wear early to quantify. Furthermore It has to be determined if the given specifications regarding the maximum erosion are maintained.

The quality specifications for modern Fluid dynamic bearings require maximum erosion that is so small is (<0.45 μm) that They are very difficult from the roughness of the bearing surfaces and is to be distinguished from the noise of the measurement. conventional Evaluation methods, such as In patent application US 2004/0025577 A1 support partly based on subjective user input and therefore do not deliver the needed Accuracy and in particular no reproducible results. There is the bearing surface mechanically scanned along a measuring section and the depth profile the bearing surface detected. By aligning the measurements with a reference line and Identification and selection of the areas of the groove structures can draw conclusions the erosion will be pulled.

US Pat. No. 6,782,338 B2 discloses a method for determining the groove structure on a surface of a workpiece, wherein the surface is scanned along a measurement path and at least two adjacent groove structures are detected by the scan. From a plurality of measured values taken along the measuring path, a depth profile of the surface can be detected, wherein the measured values of the depth profile are aligned with respect to a reference line. Identification and selection of regions of the depth profile between the groove structures and characterization of the groove structure are performed.

DE 10 2007 007 459 A1 discloses a method of measuring a bore in a bushing and associated fluid dynamic bearing shaft. The bearing bush and the shaft are provided with corresponding bearing groove structures for generating a fluid dynamic pressure within the bearing gap, wherein the shaft defined at a defined speed relative to the bearing bush in a fluid Visko sity is rotated so that sets a fluid dynamic effect in the camp. A torque acting on the bearing bush and also a pressure generated in the bearing gap by the fluid-dynamic effect are measured, from which the width of the bearing gap and errors in the bearing groove structures can be determined.

Disclosure of the invention

It The object of the invention is a method for determining the Specify erosion that is automated and without subjective user input feasible is and delivers reproducible results with the required accuracy.

These The object is achieved by a method having the features of the claim 1 solved.

preferred Embodiments of the invention and further advantageous features are indicated in the dependent claims.

The method according to the invention for determining the erosion on a surface of a workpiece as a result of the application of groove structures by means of an ECM method comprises the steps:
Scanning the surface along a measuring path, wherein at least two adjacent groove structures are detected by the scanning,
Detecting a depth profile of the surface consisting of a plurality of measured values taken along the measuring path,
Aligning the measurements with respect to a reference line,
Identifying and selecting areas of the depth profile between the groove structures,
Determining a theoretical depth profile to be expected along the measurement path by mathematical approximation,
Determining the expected theoretical erosion function by overlaying theoretical depth profiles,
Fitting the expected theoretical erosion function to the selected measurement data between the groove structures by a mathematical algorithm,
Determine the maximum erosion from the result of the adaptation.

The Procedure allows a completely automated determination of erosion and supplies therefore reliable and comparable results.

In A preferred embodiment of the invention is for scanning the surface a mechanical scanning method used. It can also be one optical or another scanning method can be used the needed Accuracy in the sampling provides.

The measured depth profile consists of a large number of measured values, wherein a feed step along the measurement path in each case an x-coordinate and an associated depth measurement, a y-coordinate of a measurement specify.

According to the invention Aligning readings to a reference line using the method of smallest error squares performed.

The Identify and select areas of the depth profile between the groove structures (in the "land" area) is preferably carried out by means of averaging.

The theoretical depth profile is achieved by approximating the groove structures predetermined by one or more mathematical equations.

By the mathematical overlay dependent on two depth profiles the distance between adjacent groove structures becomes the expected erosion function determined in the region between the adjacent groove structures.

By Adaptation of the erosion function to the previously identified measurement data between the groove structures (in the "land area") a mathematical algorithm determines the maximum erosion.

Further Features and advantages of the invention will become apparent from the drawings and the following description of a preferred embodiment.

Brief description of the drawings

1 shows a section through a fluid dynamic bearing, in which the radial bearing areas are measured,

2 shows a diagram of a measured depth profile,

3 schematically shows the alignment of the measurement data of the depth profile on a reference line (eg x-axis)

4 schematically shows a first step in the identification of measurement data in the land areas of the depth profile,

5 schematically shows a further step in the identification of measurement data in the land areas of the depth profile,

6 schematically shows a further step in the identification of measurement data in a selected land area of the depth profile,

7 schematically shows a diagram of the identified measurement data in the land areas of the depth profile,

8th shows schematically a section through the bearing bush and the bearings of the bearing groove structures,

9 shows schematically the determination of the apex of the groove structures,

10 shows a diagram of the cross section of a measured bearing groove compared to a cross section of an idealized bearing groove;

11 schematically shows a diagram of the height contour of two adjacent bearing grooves with strong overlap;

12 schematically shows a diagram of the identified land areas of the depth profile, the calculated erosion curve and the position (slope straight line) of the erosion curve with respect to the X-axis.

Description of a preferred embodiment the invention

1 schematically shows the surface of a bore of a bearing bush of a fluid dynamic bearing with one in a bearing bush 12 freely rotatable shaft 10 , In each case one of the mutually facing surfaces of shaft 10 or bearing bush 12 in the case shown, it is the surface of the bearing bush 12 , has cylindrical zones with incorporated bearing grooves 14a and 14b on which part of each are a radial bearing. The radial bearings are through a separator area 15 separated from each other. There are several bearing grooves per radial bearing 14a respectively. 14b over the circumference of the bearing bush 12 arranged distributed. The bearing grooves 14a and 14b are formed, for example, sinusoidal or parabolic. The bearing bush 12 is at its lower end face with an annular recess for receiving a pressure plate 16 Mistake. As well as the wave 10 in the bearing bush 12 , which rotates with the shaft 10 firmly connected pressure plate 16 in the recess. The lower opening of the bearing bush 12 is through a cover plate 18 hermetically sealed. In the storage gap 20 , which is the opposite surfaces of the shaft 10 , the printing plate 16 and the bearing bush 12 separates from each other, is a lubricant such. As oil introduced. Through the described bearing grooves 14a and 14b it comes with rotation of the shaft 10 to a pumping action on the lubricant, resulting in both pressure buildup and distribution of the lubricant in the bearing gap 20 leads. The bearing grooves 14a and 14b can be symmetrical or asymmetrical in itself, so that, for example, a pumping action can be generated in a defined direction. In the example shown, the upper radial bearing 14a formed asymmetrically by the upper, the bearing opening adjacent bearing grooves are formed longer than the lower bearing grooves of the upper radial bearing, resulting in a directed into the bearing interior pumping action on the bearing fluid and thus an increase in pressure within the bearing. Furthermore, there is a thrust bearing 22 between the pressure plate 16 and the bearing bush 12 , Also between the pressure plate 16 and the cover plate 18 there may be another thrust bearing.

The method according to the invention can be advantageous for measuring the erosion on the bearing surfaces the fluid dynamic bearing can be used. In this case, preferably, the cylindrical bearing surfaces of the shaft 10 and / or the bearing bush 12 measured.

The measurement of the depth profile of the bearing surface is preferably carried out by means of mechanical surface scanning with a button. Here, for example, commercially available systems can be used. There are preferably three measurement series each along a straight measurement path 24 in the direction of the bearing bore or the axis of rotation of the shaft 12 carried out. The three measuring sections 24 For example, each offset 120 ° to each other.

The Difficulty is the erosion of other surface defects, such as roughness and machining defects not caused by ECM to be distinguished. The inventive method allows a automated evaluation of measurement data, regardless of the arbitrary Choice of vertices by a user.

When Default will be a certain size of this surface defects assumed the bearing surface to be measured. For example, surface defects become from a maximum of about 0.8 to 1.5 microns (Peak-to-peak) assumed. The assumed amount of surface defects will be used as the offset value below.

How out 2 can be seen, you get a depth profile 26 in the form of a measurement data series that can be displayed in an xy diagram. The feed along the measuring section defines the x-direction and the depth measurement defines the y-direction. The greater the depth, the greater the y-value. You can see left and right the storage areas with peak values, the individual bearing grooves 14a . 14b correspond, and intermediate flat areas, which are the "land" areas 14c correspond. Between the storage areas 14a . 14b there is a deeper separator area 15 that separates the storage areas from each other.

step 1: Alignment of the measured data

As a rule, the x-axis of the measured data does not agree with an assumed axis-parallel baseline of the bearing bush 10 match. To facilitate further evaluation, the baseline is determined and the coordinate system adjusted accordingly. For this purpose, the following steps are performed.

As it is in 3 is shown, the measurement data of each depth profile 26 through a straight line 28a approximated. For this example, the method of least squares (least-square-fit) is used.

In a next step, all measurement points that are above this line 28a + Offset value (eg ~ 1.5 μm) are excluded from further consideration.

The remaining measurement data are now again by a straight line 28b approximated and all measuring points excluded from further consideration, those above this line 28b + Offset value lie.

This Procedure is repeated until the results are nothing more changes or a predetermined maximum number of iterations is reached. The offset value is necessary because otherwise the surface defects unconsidered and lose important information in the land area.

yi:

radiale Messwerte (Tiefenwerte)

yi':

radiale Messwerte (Tiefenwerte) nach Angleichung

xi:

axiale Messwerte (Vorschub)

a1, a2:

Koeffizienten der Ausgleichsgerade

The end result is a balance straight 28c determine which is a more accurate reference line. When aligning the measurement data is the best-fit line 28c shifted to the X-axis of the coordinate system and adjusted the surrounding data points accordingly. The following formula describes this mathematically: yi '= yi - (xi · a1 + a2)

yi:

Radial measurements (depth values)

yi ':

radial measured values (depth values) after approximation

xi:

axial measured values (feed)

a1, a2:

Coefficients of the balancing line

step 2: Identification of the land areas between the grooves. Only there Erosion can occur.

As in 4 is shown, first a straight line 30 with a coarse, the surface defects reflecting offset (eg, ~ 1.5 μm) defined parallel to the x-axis. All measured data above the line 30 are cut off.

In 5 the remaining measured values are shown. The measured values are now divided into separate sections that cover both the sought-after land areas 14c as well as remnants ("necks") of the individual bearing grooves 14a . 14b contain. Each section, for example the section according to 6 (corresponds to the outlined area in 5 ), is now considered separately.

It now become the "necks" of the bearing grooves away.

aller Messwerte eines Abschnitts gebildet, daraus wird eine Gerade y = p (32) parallel zur x-Achse definiert und mit einem weiteren, feineren, vordefinierten Offset beaufschlagt, dessen Betrag im Bereich der Oberflächenrauhigkeit liegt und beispielsweise 0.3 μm betragen kann. Somit wird sichergestellt, dass die Oberflächenrauhigkeit weiterhin in den Daten enthalten ist und keine Information verloren geht.This is the mean

of all measured values of a section, it becomes a straight line y = p ( 32 ) defined parallel to the x-axis and applied to a further, finer, predefined offset whose amount is in the range of surface roughness and may be, for example, 0.3 microns. This ensures that the surface roughness is still contained in the data and no information is lost.

All measuring points above this line + offset (y> p + Offset) are no longer considered for the following iterations. Analogous to the method step: Alignment of the measurement data, several iterations are performed in which the average of the other measurement data is formed and then all measurement data above the mean + offset are cut off until the result of averaging nothing changes or a predetermined maximum number of Iterations is reached.

Is the straight line 32 fixed, all measured data, which lie above these straight lines, are eliminated on both edges of the originally considered section from the outside, and that until such time as the straight line is undershot for the first time. This effectively eliminates the "necks" of the bearing grooves without losing information from the land areas:
What is left of valid measurement data is considered the actual land area 14c interpreted. These remaining land areas 14c are in 7 shown. An essential aspect and advantage of the invention is that the result of this algorithm depends primarily on the previously selected offset values and not on an arbitrary choice of start and end points by a user, such as e.g. B. at US 2004/0025577 A1 ,

The Values for the different offset values are calculated for each storage type by Experiences determined.

step 3: Adapt the data of the depth profiles to the theoretical positions the bearing grooves

Around To correctly calculate the erosion, the measured data must now be scaled and with the theoretical positions of the bearing grooves to cover to be brought. For this it is necessary to locate the edge areas of the storage areas, and the respective Apex to identify the storage areas.

In 8th the edge regions of the bearings are designated by the positions Z1, Z2 and Z3, Z4. The distances between these positions are determined by the respective design of the bearing bush and can be projected onto the measured data. First, the positions Z1 and Z4 are identified with the start and end of the measurement data series. Then the measured data can be scaled accordingly and the positions Z2 and Z3 in the measured data can be approximately defined from the known distances and the position of the separator which can be seen from the measured data.

According to 8th At first, the apex is determined approximately from the now known positions Z3 and Z4 (approximation A1 *). For symmetrical bearing grooves, for example, the center between the positions Z1 and Z2 or Z3 and Z4 is simply assumed. The exact position of the bearing grooves on the right and left of the approximate value is then determined from the measured data (for example, by searching for local maxima or fitting the measured data to a theoretical groove profile and calculating the exact apex A1 or A2 therefrom:

step 4: Determination of the theoretical erosion function

vorgegeben, wobei r den Radius der Lagerbuchse bezeichnet. Y₀ bezeichnet die Oberflächenkoordinate in Umfangsrichtung und Z₀ die Oberflächenkoordinate in Längsrichtung der Lagerbuchse. Die Parameter ϕ und a legen den genauen Verlauf der Lagerrillen fest und können beispielsweise die Werte

The mathematical function describing the course of a sinusoidal bearing groove on the surface of the bearing bush is known and, for example, by a function

given, where r denotes the radius of the bearing bush. Y ₀ denotes the surface coordinate in the circumferential direction and Z ₀ the surface coordinate in the longitudinal direction of the bearing bush. The parameters φ and a determine the exact course of the bearing grooves and can, for example, the values

By Calculation of the normal at each point of this curve, the absolute Distance S between two bearing grooves at each point of the curve numerically be calculated. Naturally can the course of the bearing grooves of the illustrated shape differ. The course of the bearing grooves then follows a different function. The procedure remains the same.

10 schematically shows a diagram of the cross section of a bearing groove produced by the ECM method and the depth contour of a bearing groove. If you miss the depth contour of a bearing groove, corresponding measured values result 36 in the diagram of 10 are shown as measuring points. These measurement points can be described, for example, by the standardized function h (x) given below, so that the cross-section of the bearing groove as a function T · h (x) ( 38 ), where T denotes the maximum groove depth. If the center of the bearing groove lies at x ₀ = 0, then the normalized function h (x) results from the assumed current density distribution in the ECM method:

The diagram shows the depth contour of the bearing groove. In the example shown, the maximum depth T of the bearing groove is approximately T = 6 μm. The width w corresponds to the width of the conductive structure regions on the ECM electrode, through which the bearing groove structures are incorporated into the bearing surface. The cross section 36 The bearing groove produced by the ECM manufacturing process has rounded edges typical of the ECM process.

In the 1 shown bearing grooves, such as the bearing grooves 14a , are at a distance S from each other across the circumference of the shaft 12 arranged. This distance S is in the apex, that is at the top of the bearing grooves maximum. It can be seen that the distance between the bearing grooves 14a reduced to the edge. If the distance S is too small, then the bearing grooves can touch or overlap in the edge region, which leads to a partial erosion of the land area (erosion) and results in a deterioration of the bearing properties.

11 shows a schematic of the depth contours of two adjacent bearing grooves 38a and 38b , which have a relatively small mutual distance S to each other, whereby the depth contours overlap. In the overlap area, the depth of the bearing grooves is no longer reduced to the level of the bearing surface, but there remains a certain false depth T _F , which corresponds to the erosion to be determined. The erosion is greater, the smaller the distance S of the bearing grooves 38a and 38b from each other. To ensure optimum function of the bearing grooves, the erosion T _F must not exceed a specified value.

abhängig vom Abstand der Rillen S, die ihr Minimum im Apex des Lagers besitzt. Diese Erosionsfunktion kann nun an die zuvor identifizierten und selektierten Messdaten im Land-Bereich zwischen den Lagerrillen angepasst werden. Für die Anpassung der Erosionsfunktion an die Messdaten wird ein Simplex-Algorithmus, zum Beispiel die bekannte Nelder-Mead Methode, verwendet. Da der Abstand S der Lagerrillen und die Breite der Strukturen auf der ECM-Elektrode w bekannt sind, ergeben sich als Parameter aus der Überlagerung:

• – Der Abstand d zwischen ECM-Elektrode und Werkstück,
• – Die Lage der Erosionskurve

To calculate the expected erosion between two adjacent bearing grooves, the theoretical depth contours, which are described by the normalized function h (x), are superimposed. This results in a theoretical erosion function

depending on the distance of the grooves S having their minimum in the apex of the bearing. This erosion function can now be adapted to the previously identified and selected measurement data in the land area between the bearing grooves. For adapting the erosion function to the measured data, a simplex algorithm, for Example the well-known Nelder-Mead method used. Since the distance S of the bearing grooves and the width of the structures on the ECM electrode w are known, the parameters resulting from the superposition are:

• The distance d between the ECM electrode and the workpiece,
• - The location of the erosion curve

12 shows the processed measurement data of the land areas 14c , and the calculated curve 40 describes the erosion (= unwanted removal). The possibly oblique position of this erosion curve compared to the x-axis, its position by a straight line 42 results from production-related deviations of the bore of the bearing bush of the ideal cylindrical shape, measurement inaccuracies and inaccurate alignment of the bearing bush relative to the meter. The largest distance between the line 42 and the erosion curve 40 is defined as maximum erosion.

By Comparison of the calculated maximum erosion with given Specifications, can now be detected automatically which bearings meet the specification and which not, as well can, through comparing different bearings using the same ECM electrode were processed, quality problems how wear or misalignments are detected.

By temporally resolved Measurements, can Electrodes are sorted out before they produce faulty bearings.

There the measurement is not based on judgment or others, similar subjective inputs of the user is required, it is ensured that the measurements are reliable and, above all, are comparable to each other.

10

wave

12

bearing bush

14a

raceways

14b

raceways

14c

"Country" area

15

separator area

16

printing plate

18

cover

20

bearing gap

22

thrust

24

measuring distance

26

depth profile

28a-c

Just

30

Just

32

Just

34

cross-section Bearing groove (idealized)

36

cross-section Bearing groove (measured)

38a, b

raceways

40

erosion curve

42

Just (Erosion curve)

D

diameter the storage area

S

distance the bearing grooves

T _F

erosion (Incorrect depth)

T

maximum Depth of the bearing grooves

α

angle the longest Branches of the bearing groove with respect to the circumferential direction of the bearing surface (for radial bearings) respectively.

w

minimum Width of the conductive structures on the ECM electrode

d

distance between ECM electrode of bearing surface

h (x)

normalized Function for the depth profile of the individual bearing grooves

x, x _i

position along the circumference of the warehouse

y _i , y _i '

readings (Depth values)

a1, a2

coefficients the equalizer mean

p

Average

Y ₀ , Z ₀

surface coordinates

r

radius the bearing bush

φ, a

parameter

A1, A1 *

apex position of the first camp

A2

apex position of the second camp

Z1-Z4

edge positions

Claims (8)

Method of determining erosion on a surface a workpiece due to the application of groove structures by means of an ECM Procedure, with the steps: Scanning the surface along a measuring section, wherein at least two adjacent groove structures through the sampling will be detected Capturing a depth profile the surface consisting of a variety of taken along the measuring section Readings Aligning the measurements with respect to a reference line, Identify and selecting regions of the depth profile between the groove structures, Determine a theoretical depth profile to be expected along the measuring section through mathematical approximation, Determine the expected theoretical erosion function by overlaying theoretical depth profiles, To adjust the expected theoretical erosion function to the selected Measurement data between the groove structures through a mathematical Algorithm, and Determine the maximum erosion from the result the adaptation. Method according to claim 1, characterized in that that for scanning the surface a mechanical or optical scanning method is used. Method according to one of claims 1 or 2, characterized that the depth profile consists of a large number of measured values, wherein a feed step along the measurement path in each case an x-coordinate and an associated depth measurement, a y-coordinate of a measurement specify. Method according to one of claims 1 to 3, characterized in that Aligning readings to a reference line using the Method of least squares is performed. Method according to one of claims 1 to 4, characterized that identifying and selecting areas of the depth profile between the groove structures by means of averaging he follows. Method according to one of claims 1 to 5, characterized that the theoretical depth profile by approximation of the groove structures is given by a mathematical equation. Method according to one of claims 1 to 6, characterized that adjusting the expected theoretical erosion function to the selected measurement data between the groove structures a simplex algorithm. Method according to claim 7, characterized in that that adjusting the expected theoretical erosion function to the selected measurement data between the groove structures The Nelder-Mead method is done.