Abstract: In order to deeply analyze the load distribution and the variation law of plate crown between row of rolling bearings of four-high plate rolling mill, the radial load distribution, transverse thickness distribution and plate crown of each row of cylindrical roller bearings of four-high plate rolling mill were measured and analyzed by using the self-designed comprehensive testing device of rolling bearing force and temperature. The results show that the radial load of each bearing row increases with the increase of the reduction rate and the rolling width, and the load distribution between bearing rows is "M". After rolling, the crown of the plate first increases and then decreases with the increase of the reduction rate. When the reduction rate is the same, it increases with the increase of the original thickness of the rolled piece, decreases first and then increases and decreases with the increase of the width of the rolled piece. The research results can provide useful reference for the application of rolling bearings in strip rolling mill and the establishment of profile analysis model.

Key words: rolling bearing; Load test; Plate shape; Strip mill

With the development of plate and strip rolling technology, the rolling process is developing towards large-scale, continuous and high-speed, and the production efficiency and product quality of the rolling mill are constantly improved. As the main model of strip foil production, the work roll bearing is rolling bearing, and the support roll bearing is rolling bearing or oil film bearing. Due to the influence of technology, equipment and other factors, the stress situation of roll bearing in the rolling process is quite complicated. Literature studies have shown that uneven distribution of load between rows of multi-row rolling bearings and serious off-load are important factors affecting bearing life [1-3]. Most finite element simulation and simplified calculation methods treat bearing load as concentrated load or simple uniform load, which makes the prediction accuracy of rolled sheet shape different from the actual [4-5]. In this study, the load distribution of each row of cylindrical roller bearings in the rolling process was tested by using a self-designed four-high experimental rolling mill support roller bearing force and temperature comprehensive test device, and the influence of factors such as reduction rate, thickness and width on the load distribution between bearing rows and the crown of rolled plate was analyzed. The research results can provide useful reference for the application of rolling bearings and the establishment of the analysis model of plate shape.

1 Experimental Research

The theoretical and experimental studies on the radial load distribution of roll bearings show that the radial load of bearings in the rolling process is distributed in the arc range of the corresponding bearing center Angle of about 120° according to the cosine law [6-7]. In order to measure the actual distribution of bearing load between rows, a comprehensive test device for bearing force and temperature of support rolls under a four-high plate strip mill was designed. A groove with a width of 148mm was processed along the roll axis in the bearing area corresponding to the bearing seat, and four pressure blocks that could move up and down were installed in the groove corresponding to each row of cylindrical roller bearings. The upper arc part of each pressure block is in contact with the outer ring of the corresponding column bearing (as shown in Figure 1(a)), and together with the boring of the upper bearing box, it forms the inner hole of the bearing seat with a diameter of 170mm. The bottom of the pressure block is processed with two countersunk holes arranged symmetrically for installing two radial pressure sensors. In the rolling process, the bearing is stressed and the pressure is transferred to the bearing outer ring, and the bearing outer ring transfers the pressure to the pressure block. The pressure block deforms the radial pressure sensor, and the radial force borne by the two pressure sensors is the actual radial force received by the bearing.

In order to facilitate the recording and analysis of bearing load distribution, columns of cylindrical roller bearings or pressure blocks are numbered, with the bearing furthest from the drive side to the roller body as the first column, the bearing closest to the operating side to the roller body as the second, third and fourth columns, and the bearing closest to the operating side to the roller body as the fifth column, and the bearing successively sixth, seventh and eighth columns as the distance from the roller body increases. The specific numbers of each column bearing or pressure block are shown in Figure 1(b).

The rolling experiment was carried out on a four-roll reversible cold rolling mill in the laboratory (as shown in FIG. 2). The main equipment parameters were as follows: the work roll size was φ100 ×320mm, and the support roll size was φ220×320mm; Main motor power 90kW, speed 1500 ~ 2800r/min; Hydraulic cylinder pressure 5mm/s; Take-up/unwinding tension 0.5 ~ 20kN; Maximum rolling force 1200kN; The rolling speed is 0 ~ 60 m/min. Raw material thickness ≤6mm; Finished product thickness ≥0.2mm.

The experimental material is: H24 pure aluminum plate with thickness of 2mm and uniform thickness distribution. The experimental scheme is as follows: the length of 250mm and the width of 80mm, 120mm, 160mm, 200mm, 240mm and 300mm were rolled, and the reduction rate was set at 40%. The sample with length of 250mm and width of 240mm was tested by rolling, and the reduction rates were set at 10%, 20%, 30%, 40% and 50%, respectively. The rolling speed is 0.2m/s. The radial load distribution of each bearing is obtained by processing the output data of radial pressure sensor collected by data acquisition instrument in rolling process, and the thickness distribution of aluminum plate along the width direction after rolling is measured, and then the crown of aluminum plate is obtained.

2 Analysis of radial load distribution between bearing rows

In order to analyze the uneven distribution of radial load among all columns of bearings, bearing load percentage α(i) is defined as the proportion of radial load of column i bearing to the total radial load, that is, α(i)=F(i)/F×100%.

Where, α(i) is the bearing percentage of column i bearing, F(i) is the radial load of column i bearing, and F is the total radial load.

2.1 Influence of reduction rate on radial load distribution

When the width is the same, the change and distribution of radial load of each column bearing with the reduction rate are shown in Figure 3. It can be seen from FIG. 3(a) that with the increase of the reduction rate, the radial load of each bearing column basically increases linearly, which is the result of the increase of the total rolling force caused by the increase of the reduction rate. As can be seen from FIG. 3(b), the bearing percentage of radial load between bearing rows is generally "M" shaped distribution, the bearing percentage of the second and seventh columns where the bearing seat fulcrum is located is the largest, exceeding 25%, the bearing percentage of the first, third, sixth and eighth columns near the fulcrum is larger, and the bearing percentage of the fourth and fifth columns far from the fulcrum is the smallest. Less than 5%; When the reduction rate is less than 43%, the bearing bearing percentage of the second and seventh columns gradually decreases with the increase of the reduction rate. When the reduction rate is greater than 43%, the bearing bearing percentage of the second and seventh columns increases with the increase of the reduction rate. With the increase of the reduction rate, the bearing bearing percentage of the first, second, seventh and eighth columns changes greatly, that is, the bearing radial load distribution mainly occurs between the fulcrum position and the two columns away from the rolling center line, and the bearing bearing percentage of the other columns changes little. This is because the deflection and flattening deformation of the support roll are the main causes of uneven load distribution of each column of bearings. When the reduction rate is less than 43%, the deflection of the support roll becomes larger with the increase of the reduction rate. Due to the self-aligning adjustment of the column pad at the second and seventh columns, the bearing bearing percentage of the second and seventh columns gradually decreases. When the reduction rate is greater than 43%, the total rolling force is basically close to the maximum rolling force of the mill, and the self-aligning effect of the cylinder pad is reduced, making the bearing percentage of the second and seventh columns increase.

2.2 Influence of plate width on radial load distribution

When the incoming material thickness and the reduction rate are the same, the change and distribution of radial load of each bearing row with the plate width are shown in Figure 4. It can be seen from FIG. 4(a) that the radial load of all columns of bearings increases with the increase of the rolling piece width, and the radial load of the second and seventh columns of bearings slows down with the increase of the plate width, the radial load of the first and eighth columns of bearings basically increases linearly, and the radial load of the third, fourth, fifth and sixth columns of bearings does not change much. It can be seen from FIG. 4(b) that the bearing percentage of radial load between bearing rows is generally distributed in an "M" shape. With the increase of rolling piece width, the bearing percentage of the first and eighth columns gradually increases, while the bearing percentage of the other columns gradually decreases. This is because the degree of deflection and flattening deformation of the support roll is the main reason that affects the uneven distribution of bearing load. When the thickness of the input material and the reduction rate are the same, the rolling force increases in direct proportion to the width of the rolled piece, making the deflection of the support roll larger, but the rolling force also widens the range of action, and the deflection of the support roll becomes smaller. The above phenomenon is the result of the joint action of the two.

3. Analysis of influencing factors of plate crown

3.1 Influence of reduction rate on plate crown

When the width is the same, the transverse thickness distribution of rolled pieces and the change of plate crown with the reduction rate after rolling are shown in Figure 5. It can be seen from FIG. 5(a) that the rolled piece has a relatively obvious non-uniform deformation along the width direction after rolling, and the rolled piece changes from the rectangular section before rolling to a convex section. As can be seen from FIG. 5(b), with the increase of the depressurization rate, the plate crown first increases and then decreases. When the depressurization rate is 30%, the plate crown reaches the maximum value. The flexural deformation of work rolls is the main factor affecting the shape of roll gap and the crown of rolled parts, and the factors affecting the flexural deformation of work rolls are mainly rolling force and inter-roll pressure. For aluminum plates with the same original thickness and the same width of rolled parts, the rolling force is proportional to the reduction rate, and the inter-roll pressure mainly depends on the coordinated deformation between work rolls and support rolls [8]. In the experimental rolling mill studied in this paper, the roll diameter ratio of the support roll to the work roll is 2.2, which reaches the upper limit of the roll diameter ratio of the strip mill [9], that is, the stiffness of the support roll is large, when the reduction rate is less than 30%, the rolling force is small, the flexure deformation of the support roll is small, and the pressure between rolls is mainly concentrated in the middle of the roll. At this time, with the increase of the reduction rate, the increase of work roll deflection caused by the increase of rolling force plays a major role, so the crown of the plate increases. It can be seen from FIG. 3(a) that the difference of bearing loads in the first and third columns near the fulcrum in the bearing seat has little change, which also indicates that the support roll deflection is small. When the reduction rate is greater than 30%, the rolling force is large and the support roll has obvious flexure deformation. With the continuous increase of the reduction rate, the flexure of the support roll becomes larger, while the rolling force increases, the action range of the pressure between rollers becomes wider. Under the combined action of the two, the flexure deformation of the work roll becomes smaller, resulting in a decrease in the crown of the plate [10], as can be seen from Figure 3(a). The difference of bearing load between the first column and the third column near the fulcrum in the bearing seat becomes larger, indicating that the support roll has a large deflection deformation.

3.2 Influence of plate width on plate crown

When the reduction rate is the same, the transverse thickness distribution of rolled pieces and the change of plate crown with plate width after rolling are shown in Figure 6. It can be seen from FIG. 6(a) that the rolled piece has a relatively obvious non-uniform deformation along the width direction after rolling, and the rolled piece changes from the rectangular section before rolling to a convex section. It can be seen from FIG. 6(b) that with the increase of the plate width, the crown of the plate first decreases and then increases and then decreases, in which the minimum value appears at the plate width of 120mm and the maximum value appears at the plate width of 240mm. The crown of the rolled piece is mainly determined by the bending deformation of the work roll caused by rolling force and inter-roll pressure. For the same original thickness of aluminum plate, the rolling force is proportional to the width of the rolled piece when the reduction rate is the same. When the width of the rolling piece is small and the reduction rate is large, the rolling force is small. With the increase of the width of the rolling piece, the rolling force increases, and the flexural deformation of the work roll increases. At the same time, the working range of the rolling force also widens, and the elastic deformation of the work roll decreases. When the width of the rolled piece is 120mm, the crown of the plate reaches the minimum value. As the width of the rolled piece continues to increase, the rolling force continues to increase, resulting in a certain degree of flexural deformation of the support roll, but at this time, the rolling force still plays a major role in increasing the influence of the work roll flexural deformation, so the crown of the plate increases again. When the width of the rolled piece is 240 mm, the crown of the plate reaches the maximum value. When the width of the rolled piece further increases, the rolling force continues to increase, but at this time, the rolling force's range of action becomes wider, and the support roll also has obvious deflection deformation, and the crown of the plate begins to decrease again [11-12]. As can be seen from FIG. 4(a), the difference of bearing load between the first and third row bearings near the fulcrum in the bearing seat becomes larger. It shows that the support roll has a large deflection deformation.

3.3 Influence of plate thickness on plate crown

In order to analyze the influence of plate thickness on plate convexity, the H24 pure aluminum plates with a width of 240mm, thickness of 2mm, 3mm and 5mm, and uniform thickness distribution, were rolled under the setting of 30% reduction rate (the actual reduction rate was 30%, 32% and 30% respectively). The transverse thickness distribution of rolled pieces and the change of plate crown with plate thickness after rolling are shown in Figure 7. It can be seen from FIG. 7(a) that the rolled piece has a relatively obvious non-uniform deformation along the width direction after rolling, and the rolled piece changes from the rectangular section before rolling to a convex section. It can be seen from FIG. 7(b) that the crown of the plate increases with the increase of the original thickness of the rolled piece.

This is because the width of the rolled piece and the reduction rate are basically the same. With the increase of the original thickness of the rolled piece, the reduction amount increases, although the unit width rolling force increases, which leads to the decrease of its influence on the crown of the plate, but the increase of the rolling force makes the deflection of the work roll larger and plays a leading role in the influence on the crown of the plate, so the crown of the plate continues to increase, but the increasing trend gradually slows down.

4 Conclusion

(1) The original thickness of rolled parts is the same, when the reduction rate or the width of rolled parts increases, the load between bearing rows generally increases, and the bearing percentage between bearing rows presents an "M" type distribution, the bearing percentage of the second and seventh columns where the bearing seat fulcrum is located is the largest, and the bearing percentage of the first, third, sixth and eighth columns near the fulcrum is larger. The bearing percentage of the fourth and fifth columns far away from the fulcrum is the smallest.

(2) With the increase of the reduction rate, the bearing load distribution mainly occurs between the second and seventh column bearings at the fulcrum position and the first and eighth column bearings away from the rolling center line, and the bearing load percentage of the other columns changes little; With the increase of the width of the rolling piece, the bearing percentage of the first and eighth columns gradually increased, and the bearing percentage of the other columns gradually decreased, but the bearing percentage of the fourth and fifth columns changed little.

(3) As the stiffness of the support roll is large and the reduction rate is small, the flexural deformation of the support roll is small. With the increase of the reduction rate, the flexural deformation of the support roll gradually increases, and the pressure distribution between rolls changes significantly. Under the comprehensive action of rolling force and inter-roll pressure, the crown of the rolled sheet increases first and then decreases.

(4) When the rolling width is small, the rolling force distribution has a greater influence on the work roll deflection. When the rolling width is large, the rolling force has a greater influence on the work roll deflection. With the further increase of the rolling width, the pressure distribution between the rolls also changes significantly after the support roll has obvious deflection. When the reduction rate and the rolling width are basically the same, the crown of the plate increases with the increase of the rolling thickness.

【 Reference 】

[1] LIU DE-YI, SHEN GUANG-XIAN, YU CHUN-XIAO, et al. Study on Multiple Boundary Element for Three-dimensional Elastic Contact Problem with Friction [J]. Advance in Boundary Element Techniques IV, 2003,34 (9):155-161.

[2] Shi Rong, Shen Guangxian, Chen Zhanfu, et al. On-line Detection of Load Condition of roll bearing in CVC Mill [J]. Journal of Mechanical Engineering, 2003,39 (1):107-110.

[3] EBAUGH, ROGERL, NOWAK, EDWARD. Chock and roll maintenance and its effect on work roll bearings. Iron and Steel

Engineer [J]. Iron and Steel Engineer, 1978,55 (11):64-68.

[4] Li Ming, Lv Zhaoyu. Load Characteristics Optimization and Boundary Element Analysis of Rolling Mill Bearing [J]. Heavy machinery, 201

[5] Jin Xiaohong, Tong Yunchun, Zhao Xianqiong. Study on the bending Stiffness of work roll diameter in Hydraulic Bending Control [J]. Journal of Metallurgy, 2009,13 (1):1-3.

[6] Ma Yuping, Chen Peng. Measurement and Analysis of Bearing Radial Contact Force Photoelasticity [J]. Journal of Bearing, 1981(1):10-12.

 Lateral Stiffness Analysis of Full width roll Gap in Four-high strip Mill [J]. Forging Technology, 2014,40 (9):111-116.

[8] Liu Aimin, Liu Guangming. Analysis of Roll Elastic Deformation in Four-high CVC Mill [J]. Steel Rolling, 2011,28 (3):21-23.

[9] HUANG Qingxue, Xiao Hong, Sun Binyu. Steel rolling machinery [M]. Beijing: Metallurgical Industry Press, 2007.

[10] Xu Jianzhong, Gong Dianyao, Liu Xianghua, et al. Effect of unit width rolling force on Crown of hot rolled strip [J]. Journal of Northeastern University, 2003,24 (3):280-283.

[11] Li Yue, Zhang Xiaoping, Liu Guangming, et al. Analysis of Load Distribution of roll diameter and Influence Factors on Plate Shape of Four-high strip Mill [J]. Steel Rolling, 2014,31 (6):1-4.

[12] Zhang Xiaoping, Guo Huiguang, Zhang Xuena, et al. Experimental study on the influence of plate width on plate shape [J]. Journal of Taiyuan University of Science and Technology, 2009,30 (3):225-227.