Keywords

1 Introduction

As industrial production is increasingly changing from manual labor to highly automated and demanding work, the application of modern measurement technology is a main part of this production. One of the most important tasks in the technical production industry is the inspection of products. Devices with contact measurement methods of the obtained parameters are still used for quality control of machined surfaces. In these contact measuring methods, the measuring contact is in direct contact with the machined surface. This direct contact of the measuring contact with the machined surface is one of the negative effects of the measurement. On the one hand, the contact surface of the measuring device is damaged, and on the other hand, the surface to be measured is damaged. One of the many possible uses of contactless sensors is to measure deviation of surface from workpiece diameter after process of turning. This method allows rapid detection of defects on the workpiece surface without sending the product to quality control after production [1,2,3,4,5,6,7,8].

The chapter main thesis was the application of optical contactless sensors in the measurement of the deviation parameter from the workpiece diameter of C45 steel after the turning process. As a comparison to optical sensors, a mechanical micrometer was used in the experiment as a standard for the measured values. In this experiment, we demonstrated the feasibility of using these sensors to directly detect surface damage and other undesirable features of processed materials in processing centers, eliminating the need for post-processing quality control increase.

2 Optical Sensors Measuring Literature Review

In contrast to contactless testing, many traditional distance measuring devices such as micrometers and calipers make contact with the workpiece surface. This method involves contact that damages not only the instrument but also the object being measured [9].

If the object to be measured is not fixed or intended to move, the measurement results will be inherently unstable and inconsistent on subsequent measurements. The current market for sensors is growing dynamically. With over 1000 different sensor types in different countries, choosing the right sensor type for a particular application can be difficult. Machining technology is realized in the system machine, tool, and workpiece. In relation to the machine, Kishore et al. applied optical sensors vibration detection sensors to monitor the life of a diesel engine, based on the amount of measured vibrations (distance deviations) it produced [10, 11, 12] (Fig.1).

Fig. 1
A diagram of how the laser light is passed through the device and the emitted lights are detected at different angles by the lens and detector.

Laser displacement triangulation sensor scheme [12]

Full size image

Sensors are defined as units capable of detecting errors to defined principles. The head of sensor as the key part of the device is the interface between the external stimulus and the internal components for processing and further analysis of the information measured during the application [13, 14]. The main principle of this measuring method is application of polychromatic white light. This light is projected from the sensor head through a multi-lens onto the workpiece surface. The so-called confocal lens arrangement is used, in which the light is divided into colored components with different wavelengths by the natural diffraction of light. The reflected radiation to the confocal aperture of the sensor, through which only focused radiation of a specific wavelength reaches the sensor. The light quantity returned to the optical receiver changes depending on the position and distance of the workpiece. In a study, Zou et al. describe the application of non-contact sensors using a neural network in the detection of points on the surface of the workpiece, on the basis of which they have developed an active control of machining during the process. Based on the elimination of spindle deflection and accurate measurement of the center axis of the workpiece, they were able to create a 3D model of the measured workpiece and compare it with the resulting workpiece after the process of turning. In this way, they were able to detect the error in serial processing. Yu et al. used a chromatic confocal sensor in their study to measure the unevenness (thickness) of transparent materials. This study showed that with proper illumination, it is possible to use confocal sensors to measure not only metallic materials, but also materials such as glass and liquids with a measurement error rate of only 2% [15,16,17,18,19].

This type of sensor can be used to optimize, control and diagnose production reliability based on workpiece features on the surface. Laser scanners are ideal for these applications due to their high speed, repeatability, and accuracy of up to 1 micron. Fu et al. designed a measurement scheme using two confocal sensors facing each other, with the help of which he measured the topographic changes of the surface of the machined material directly after the machining process. The measurement system was implemented in a robotic arm, and the distance and angle measurements were adjusted during the measurement based on the computer-aided design model of the respective component. Experimental data confirm the ability of this system to measure surface roughness in the range Ra 0.002–0.007 mm (λc/λs 300, accuracy was 5%). Cheng et al. used a monochromatic confocal sensor in the measuring process of the workpiece surface with the method of a straight line along the surface, creating a topographic map of the machined product. Measured data proved that the proposed system can achieve accuracy comparable to mechanical laboratory conditions. The measurement error is 0.001 mm. Since the current measuring mechanisms are mainly used in development processes, the present chapter can be used for new applications [20,21,22,23] (Fig. 2).

Fig. 2
An illustration represents how the light source is passed through the beam splitter, the scattered lights are passed through the pinhole and lens to obtain the signal and monochromatic images.

Principle of confocal measuring sensor [23]

Full size image

3 Research Methodology

As mentioned in the previous chapter, two types of optical deviation sensors were used in the experiment, both of which operate on the same principle, namely the insensitivity of the light reflected from the workpiece back to the sensor. Although the principle of these technologies is almost the same, the circuit diagram for these sensors is not.

3.1 Confocal Sensor

In the case of confocal (CS) sensor, there is no need to application of additional hardware components. The data are written directly into the software database, which is the biggest advantage in comparison to other measuring devices that require the manual input of measured data. With the increasing distance from the sensor to the measuring workpiece, the inaccuracy of the measurement also increases, what presents the biggest issue of these types of sensors. Besides the limitation of distance in the measuring process, other disadvantages are see-through materials or glossy types of surfaces, which affect the measurement accuracy.

From Fig. 3 the wiring scheme of the measurement system can be briefly seen. Applied measuring system consists of the measurement head of the Keyence CL – P070 (A) with sensor installed on the rotary holder of the (Leadwell-T5) cutting tool, to achieve a dynamic type of measurement for the machined workpiece. All digitized values of the sensor head were transmitted by means of an amplifier Keyence CL – 3000 (C) to the communication module Keyence CL – P070N (B), where microscopic irregularities of the workpiece diameter could be observed via Ethernet to the software (Navigator).

Fig. 3
Three photos of the rotary holder on the left, the measuring system Keyence C L P 070 in the middle, and Keyence C L 3000 on the right with several connecting probes.

Measuring scheme of confocal sensor (CS)

Full size image

3.2 Laser Triangulation Sensor

In the experiment part, we prepared a measurement system, which scheme is shown in Fig. 4.

Fig. 4
Four photos of the laser sensor Keyence I L 100 on the left, followed by the measurement devices with connecting probes and L E D lights.

Measuring scheme of triangulation sensor (LTS)

Full size image

The measurement system consists of measuring head (A) Keyence IL100. Its measuring range is 70–130 mm, measuring spot diameter (400 × 1350) μm which depend on the workpiece distance, repeatability 4 μm, linearity ±20 mm, and weight 75 g.

The communication unit (B), DL – PN 1. These elements are connected to the sensor head via an extension cable and are connected to the PLC (C), Siemens 1511C via the profinet module. The switch unit (D), Siemens XB005 is used as a boundary (connection) via ethernet between the PLC and the evaluated unit (PC–TIA Portal software).

4 Experimental Measuring

The first objective of this chapter was to test the possibility of using optical displacement sensors to measure the deviation of the diameter of round C45 steel machined under the same turning conditions but with another measuring devices (CS, LTS) (Fig. 5).

Fig. 5
Two photos of the screw gauge like device with measuring scale, holder, and L C D monitor.

Diameter measuring of the C45 sample

Full size image

The basic parameters of the experimental workpiece before machining were diameter d = 14.99 mm and length l = 88.8 mm.

The machining conditions in the turning process were feed (0,28 mm), cutting speed (60 m/min), and cutting depth (0,1 mm) (Fig. 6).

Fig. 6
Two photos illustrate how the machine with the conical shaped rotating blade drills the cylindrical shaped sample piece of material.

Figure of the workpiece working zone during the turning process

Full size image

After the machining, the first TLS was inserted into the tool holder, which we aligned exactly with the center of the workpiece and then defined the zero point at the base of its measuring range (100 mm). In the next step, we divided the sample into 6 sections of 10 mm in length and a circumference of 72 degrees and measured the deviation from the diameter at each section (Fig. 7).

Fig. 7
Two photos illustrate how the T L S and C S are inserted into the tool holder that is placed on the center of the workpiece for measuring the diameter.

Measuring of the diameter deviation by ((a) LTS, (b) CS)

Full size image

The same measurement principle was used at CS when we reinserted the sensor into the tool holder, set the zero point for CS to (70 mm), and then measured the deviation values from the workpiece diameter. For the purposes of this experiment measuring holders were constructed for both TLS and CS sensors using a 3D printing with Filalab PLA filament.

A LIEVE micrometer with 0–25 mm measuring range and an accuracy of 1 μm were used in this experiment as a standard for the measured values (Fig. 8).

Fig. 8
A photo of the round shaped machine with a cylindrical shaped workpiece in the center that is measured by using the L I E V E micrometer with an L C D monitor.

Measuring of the diameter with LIEVE micrometer

Full size image

4.1 Experimental Results

Since it was necessary to create custom measurement software for the LTS, the data measured by the LTS was written by the PLC to a text block in the form of measurement time, measurement value, and write index. During the measurement sections, 200 values were measured, from which an arithmetic mean was formed. All data were subsequently processed from text block to Excel program for further analysis (Fig. 9).

Fig. 9
An illustration depicts the measured values of L T S. Some of them are 00:00.031; 55, 185; 12, 336; 12336; 00:00.046; 55, 181; 12, 336; 12336, and more.

Example of measured values from LTS

Full size image

In the case of CS, data processing was somewhat easier because Keyence also provides the sensor with software to analyze and process the measurement data, so it was possible to measure more values with one confocal sensor, exactly 1000 measurements per second with an accuracy of 1 μm. The measurement data can be evaluated directly in the CLNavigatorN-N program or exported to Excel (Fig. 10).

Fig. 10
A screenshot of the C L Navigator software window, which has a fluctuating continuous curve with zero, training, reset, and save buttons.

Example of the measured data from CS in CLNavigatorN-N software

Full size image

As mentioned in the previous part, measured values from the sensor head were saved from the software or PLC to a table in Excel. In Table 1, data of LTS in the form of arithmetic mean values are processed, which were determined from all values for a certain section. In Table 2, data from the same workpiece are processed, but using CS. The last data in Table 3 were processed data measured from micrometer measurements.

Table 1 Measured data for LTS
Full size table
Table 2 Measured data for CS
Full size table
Table 3 Measured data for micrometer
Full size table

The measuring system determines only the zero position deviation, for the real values the following equation must be applied:

$$ {\Delta}_d=l+x\pm y $$
(1)

where

  • Δd – real deviation of diameter,

  • l – measured deviation,

  • x – the required value,

  • y – sensor error.

Calculation example of the real measured dimension:

  • x: 10 mm,

  • l: 0.005 mm,

  • y: 0.003 mm,

Δd = 10,005 ± 0,003 mm.

When measuring the workpiece with both sensors the lowest deviation was about - 0.18 mm and the highest deviation was 0.01 mm, due to the machined surface quality after turning.

From Fig. 11, we can see that the CS is a very accurate representation of the readings from the mechanical micrometer. The largest difference occurred in the 30 mm section, when the difference in the average of the measured values was 0.019 mm. An even greater difference between the LTS readings occurred when the difference between the mechanical gauge and the LTS exceeded the difference by 0.039 mm, again in the 30 mm section. Regarding the measured values of the circumference of the workpiece from Fig. 12, we can again confirm that CS measured almost identical values as a mechanical gauge, the difference between a mechanical gauge and LTS increased on the contrary. At an angular rotation of 144°, the difference of the measured values reached up to 0.095 mm.

Fig. 11
A line graph plots the deviation versus the measured length. It plots 3 fluctuating curves for the L T S, micrometer, and C S, while the micrometer has a high deviation of 0.058 millimeters.

Graphical comparison of individual measuring systems by deviation of diameter on length

Full size image
Fig. 12
A spider chart compares the diameter and angle of the workpiece. It plots the data in the rotating pattern that provides data for the L T S, micrometer, and C S.

Graphical comparison of individual measuring systems by deviation of diameter on workpiece angle

Full size image

5 Discussion

In this chapter, a new experimental measurement system focused on contactless optical sensors was presented. Analyzing the diagrams, we can see that the average total deviation of the workpiece diameter is between - 0.18 mm and 0.01 mm. Based on these deviations, we can determine the surface errors of the machined material by identifying defectively machined workpieces in turning process. However, these measuring systems also have their limits, namely the measuring distance.

6 Conclusion

From the results of the experiment, it can be concluded that the confocal sensor copied the micrometer measurement data very accurately, with the highest deviation from the measured value being 0.019 mm. This is in contrast to the triangulation sensor, whose deviation from the diameter of the workpiece in comparison with the mechanical micrometer was at one point up to 0.095 mm. With this experiment we have clearly proved that the use of optical sensors has its application in the machine industry, especially in the measurement of surface properties of workpieces after turning.

The second part of research at the Faculty of Manufacturing Technologies in Prešov is the practical implementation of optical sensors in manufacturing operations as a substitute for post-machining quality control. Contactless sensors are one of the best technologies in the field of optical distance and position measurement. A disadvantage of small and medium-sized automated plants is the high purchase price of the device compared to the more common diffuse sensors or single-channel optical sensors, but in large-scale production and advanced production with automation processes, where they belong, where very high measurement accuracy must be achieved, these sensors are essential.