JPS6338657B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a microhardness meter that automatically measures the hardness of metal materials and the like.

A hardness tester that measures the hardness of metal materials, etc. in a microscopic area presses a square pyramid-shaped diamond indenter to form a recess (hereinafter referred to as an indentation) 1 as shown in Figure 1, and then magnifies this indentation 1 with a reading microscope. The apex angles A, B and C, D are sandwiched between two hairlines 2 and 3 placed in a reading microscope, and the lengths of the diagonals are read, and the average length of these diagonals is d.
The length d is then substituted into the formula (1) below to calculate the Vickers hardness Hv.

Hv=1.8544×F/d ² Here, F is the pressing load applied to the diamond indenter. For reading such diagonals, 2
In order to measure the indentation 1 by holding the indentation 1 between the hairlines 2 and 3 of the book, it is necessary to read at the point where the apex angles A, B or C, D of the indentation 1 touch the hairlines 2 and 3, and therefore, skill is required. At the same time, the influence of differences among people is also very large. For example, there are cases where the apex angle of the indentation and the hairline are in contact with each other but it is not obvious, as shown in Figure 2, and there are cases where the apex angle and the hairline are far apart but do not appear to be in contact as shown in Figure 3. This can cause measurement errors. These errors are unavoidable when measurements are performed by humans. Furthermore, in long-term measurements,
There were drawbacks such as very large variations due to eye fatigue and correspondingly large measurement errors.

The present invention was made for the purpose of eliminating the above-mentioned conventional drawbacks, and provides a microhardness meter that electrically measures the length of the diagonal line of an indentation, and also calculates and prints out the Vickers hardness. It is something.

The present invention will be described in detail below based on an embodiment of the accompanying drawings.

In FIG. 4, the television camera 11 is connected to the hardness meter 1.
0, and images an indentation 1 (FIG. 5) formed by a diamond indenter (not shown) of the hardness meter 10 while scanning, for example, in the direction of arrow F, and outputs a video signal e _v , It is added to the shedding correction circuit 13 of the image processing section 12. The shedding correction circuit 13 corrects the unevenness of the video signal e _v caused by uneven sensitivity of the image pickup tube, etc., and outputs it as a signal e _v ', which is applied to the signal processing circuit 14. Signal processing circuit 1
4 differentiates the input signal e _v ', detects the boundary between the indentation 1 and the other part 1' (Fig. 5), cuts it at a predetermined threshold level, and converts it into a binary signal.
e _b is output and added to the storage device 15. Storage device 1
For example, 5 is a (32×512) bit memory 15a,
15b are arranged to cross each other at right angles in the vertical and horizontal directions, and each edge portion of the impression 1 is stored in these cross memories 15a and 15b.
In addition, the above center alignment and each apex angle A,
To center B, C, and D in the width direction, the operator displays an image displayed on the cathode ray tube (a composite image of the cross memory and edge images as shown in Fig. 6), and through operator operations, the cross memory and edge images are ) will be moved on the screen.) Do this as appropriate. In this case, it is not necessary that the center O of the indentation 1 coincides with the center P of the memory, and that each apex angle A, B, C, D coincides with the center of the memory in the width direction, and θ = 45 degrees. It is sufficient that the vertices A, B, C, and D are located in four memory areas. Also, the boundary of the indentation is set by a predetermined number of bits, for example, 3.
It is stored in the width of the number of bits (Figures 6 and 8).
This Fig. 8 is an enlarged explanatory diagram of the edge part including the apex angle A in Fig. 6. Each square represents a memory element, the blacked out part stores the signal e _b , and the white part stores the signal. It shows that it has not been done. In this way, the cross memories 15a, 15
The edge portion including the apex angles A to D of the indentation 1 is stored in the memory 15b so as to form an angle of 45° with these memories 15a and 15b. In addition to the impression 1, shading may occur in the image due to polishing scratches, inclusions, dirt, etc.
These are also stored in memory and are represented by 25 in FIG. 8, for example.

Now, from the cross memories 15a and 15b, the impression 1
When detecting the apex angles A to D, the signal processing circuit 17
reads the vertical and horizontal addresses of the cross memories 15a and 15b downward 45 degrees to the right (direction of arrow R in Figure 6) and downward 45 degrees to the left (direction of arrow L in Figure 6), and if there is data at this address. is counted as 1. In this way, as shown in FIG.
Obtain DA ₁ , DB ₁ , DA ₂ , DB ₂ . And data
Address of DA ₁ , DB ₁ Midpoint address of Aa ₁ , Ab ₁
Aa (Figure 7c), data DA ₂ , DB ₂ addresses
From the midpoint address Ab of Aa ₂ , Ab ₂ to vertices A and B respectively
Ask for the address. Similarly, addresses Ac and Ad of vertices C and D are obtained from the vertical cross memory 15b. The signal processing circuit 17 also performs dust removal processing. As mentioned above, as a result of addition in the downward right and downward 45° directions, the data DA ₁ to Dd ₂ for each edge is 10.
There is an amount of ~16 bits. That is, when the apex angle of the indentation 1 is completely right angle, there are 16 bits, but as the indenter wears and the apex angle becomes rounded, the number decreases to about 10 bits (the pixel portion marked with * in FIG. 8). For example, when examining data with an amount of 10 bits or more, the part 25 due to dust shown in Figure 8 is
The data n _b and n _a (Fig. 9 a, b) are output. Therefore,
Among the data DA ₁ , n _a , DB ₁ , n _b , memory 15a
The address Aa ₁ that is closer to the center P (Fig. 6) of
Ab ₁ is determined to be the tip of the indentation and the data is based on dust.
Distinguish between n _a and n _b . This means that even if there is dust in the indentation, the pressing pressure is made uniform by the indenter, so that it cannot be detected as a dust signal, and therefore it can be assumed that there is no dust in the indentation. In other words, it may be assumed that the signal due to dust is always outside the indentation. Therefore, edge data is always located closer to the center of memory than garbage data. In this way, the signal processing circuit 17 outputs address data Aa to Ad of each edge A, B, C, and D. The adder 18 (FIG. 4) displays the original image of the indentation 1 on the cathode ray tube using the signal e _v ' from the shedding correction circuit 13, and masks 30 (FIG. 10) using the address data Aa to Ad from the signal processing circuit 17.
Take a picture. Addresses of vertices A and B of original image 1
The difference between Aa and Ab and the address AC and Ad of vertices C and D are the lengths of the diagonals on the imaging plane. Therefore, by multiplying these values by the actual size coefficient of one pixel in the memory, the indentation 1 can be calculated. diagonal,
You can find the actual size of a CD. The length of the diagonal line obtained in this way can be used as is when the apex angles A to D of the indentation are sharp, that is, when they are completely right angles, but if the apex angles are rounded due to wear of the indenter. Fig. 8) There is a difference from what you see with the naked eye. Therefore, in such a case, the operator, while looking at the cathode ray tube, determines the apex angles A to D of the indentation 1 on each side of the mask 30 '', '', '', ''.
The size of the backbone of the mask 30 is adjusted so that it matches. This adjustment changes the sharpness coefficient δ by the apex angle A,
In addition to the addresses Aa and Ad of D, it is subtracted from the addresses Ab and Ac of the apex angles B and C (Fig. 7 c, d, Fig. 9 c, d). This sharpness coefficient δ can be set to an appropriate value depending on the degree of wear of the indenter. The signal processing circuit 17 performs the arithmetic processing according to the operator's operation, outputs corrected address data Aa' to Ad' for each apex angle A to D, and adds the corrected address data Aa' to Ad' to the adder 18 to correct defects in the mask 30. do. Further, each of these data Aa' to Ad' is applied to the arithmetic circuit 21 of the data processing section 20. Arithmetic circuit 21
calculates the diagonal lengths d ₁ and d ₂ of the indentation 1 as described above using the input data Aa' to Ad', and calculates the average value d of these lengths d ₁ and d ₂ . Next, the average length d of the diagonal line is used to calculate the Vickers hardness Hv according to the above equation (1), and the corresponding signal _VH is output and applied to the print section 22.
The printing section 22 prints out the Vickers hardness Hv in response to the input signal _VH .

As explained above, according to the present invention, measurement errors can be extremely reduced, eye fatigue during measurement can be reduced, and Vickers hardness can be measured accurately and quickly. .
In addition, by using a cross memory to memorize only the edge part of the indentation, and reading out and adding the memorized data diagonally at 45 degrees to the edge, measurement accuracy is improved without being affected by dust, etc. be able to. Furthermore, since it is an interactive method with the operator, there are excellent effects such as consistency with the naked eye.

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram of the measurement of Vickers hardness, Figs. 2 and 3 are explanatory diagrams of the causes of measurement errors in Vickers hardness, and Fig. 4 is a block diagram showing an embodiment of the microhardness meter according to the present invention. , FIG. 5 is a diagram showing an example of the scanning direction of the indentation by the television camera in FIG. 4, FIG. 6 is a diagram showing an example of the storage device in FIG. 4, and FIGS. FIG. 8 is an enlarged explanatory diagram of the memory in FIG. 6, FIGS. 9 a to d are explanatory diagrams when determining the address of the edge of an indentation, and FIG. It is a figure which shows the relationship between the original image of an indentation, and a mask in the addition part of a figure. 1... Indentation, 10... Hardness meter, 11... Television camera, 13... Shedding regeneration circuit, 14,
17...Signal processing circuit, 15...Storage device, 16
... Memory control device, 18 ... Addition section, 21 ...
Arithmetic circuit, 22...Printing section.

Claims (1)

[Scope of Claims] 1. Imaging means for capturing an image of a square indentation made by an indenter of a hardness tester and outputting a corresponding image signal; and contour extraction means for extracting an axial portion of the imaged square indentation; It has a binarization means for binarizing the image from which the outline has been extracted, and a storage area corresponding to a cross-shaped image area formed by intersecting two rectangular image areas at right angles. a cross memory that stores a part of the binarized image so that each vertex of the square outline is located in each of four image areas other than the intersection area; and a binarized image stored in the cross memory. The image is scanned along the pixel rows in the positive and negative 45-degree diagonal directions, and the coordinate positions corresponding to the pair of contour parts are respectively obtained from the four image areas, and the midpoint position of each of these coordinate positions is determined. a signal processing means that calculates the coordinate positions of the four vertices of the square contour by calculating for each pair; and a signal processing means that calculates the length of the diagonal line of the square indentation based on the coordinate positions of the four calculated vertices; A microhardness meter comprising a data processing means for calculating hardness based on the calculated value.