Study and optimization of radial-axial rolling process for large 40Cr13 stainless steel rings
The rolling of large 40Cr13 stainless steel rings is complex. In the radial-axial rolling process, there are often problems such as instability, offset, and abnormality caused by unreasonable parameter settings in the rolling process. Aiming at this kind of problem, a large-size ring with a target outer diameter of Φ2952mm was taken as the research object, a 4-stage rolling curve was designed, and key parameters such as the initial temperature of the ring, the speed of the driving roll, and the speed of the ring’s outer diameter increase were selected during the rolling process. The rolling forming process was simulated by Deform-3D software to analyze the effects of different parameters on the radial rolling force, the equivalent effect strain, and the temperature distribution. The results show that in the 4-stage rolling process when the initial temperature of the ring is 1100℃, the speed of the driving roll is 20r.min-1, and the speed of the ring OD increase is 5.6mm.s-1, the finishing ring’s rolling force is suitable. The equivalent strain and temperature distribution are uniform.
What is Ring rolling?
Table of Contents
Ring rolling, also known as ring rolling or reaming, is a plastic process of rolling the blank ring through the ring rolling machine diameter-axial rolling so that its outer diameter increases, wall thickness decreases, and the cross-sectional profile of plastic processing technology. Compared with the traditional die forging and free forging process, the ring rolling process has the advantages of high precision, good mechanical properties, high production efficiency, good surface quality, etc. Therefore, the rolled ring parts are widely used in automotive, aviation, chemical industry, energy, and many other fields.
With the development of society, some areas of the ring for more and more high requirements; its size requirements are also growing, such: in nuclear power energy, high precision and high corrosion resistance ring demand are growing; in the field of aviation, the requirements of a larger size, higher performance of the ring to ensure that can be used at high temperatures, high pressures, high corrosion, and other environmental stability. Compared with other metal materials rings, the market for strong corrosion resistance, wear resistance, and high strength 40Cr13 martensitic stainless steel ring is in short supply.
In order to improve the precision and mechanical properties of the rolled ring, researchers and scholars have conducted in-depth studies on the ring rolling process and its theoretical aspects. To guarantee the stability and uniformity of batch forging of aviation ring parts, Liu Jun et al. planned a 5-segment core roll feed curve. They gave the engineering experience values of time and speed of each stage, finally completing the refinement design of the ring rolling process. Guo Lianggang et al. analyzed the effect of driving roll speed on recrystallization by establishing a finite element model. WangM et al. simulated the ring pieces and found that lowering the driving roll speed, increasing the core roll feed speed, and the initial temperature of the ring pieces can improve the β-phase distribution and the uniformity of the dimensions. Ning Xiangjin et al. proposed an adaptive control method for the stability and roundness of the rolling process. They experimentally showed that this method reduced the ring offset by 48.3% and improved the roundness by 61.8%. Li Guanguo, through the use of rolls with different feeding methods to simulate the ring rolling, using the outer diameter to increase the speed of the design of the rolling curve of the feeding method will make the ring rolled after the strain and temperature distribution are uniform.
The above research has very important guiding significance to the ring rolling forming; however, the combination of roll feeding mode and ring different parameters to study its impact on the forming of the research is less. The ring rolling process of various parameters on the final shape of the ring has a vital role, and the current means of numerical simulation technology based on computer-aided technology is gradually replacing the traditional experimental trial and error method. Therefore, this paper, through the design of a 4-stage rolling curve, and according to the size of the finished ring size design blank size, in the Deform software special module Rill Rolling Establishes the model and uses the control variable method to study the impact of the drive roller speed, ring initial temperature, ring O.D. increasing speed on the ring rolling, to obtain the optimized parameters, and through other sizes of the ring to verify the feasibility of the optimized parameters, to obtain excellent performance of the finished ring, providing technical support for the actual processing and production of the enterprise.
1. Kinetic finite element model
1.1 Finite element model
Radial-axial rolling is different from the previous radial rolling, the process through the upper cone roll and the core roll at the same time to do the feeding movement so that the ring can be fully plastic deformation and volume transfer to obtain high-quality ring generally used in the manufacture of large-sized ring parts. The principle is for the drive roller to do a constant speed of rotation, driven by the ring rotation, the role of friction so that the core roll and follow the passive rotation of the guide roll to maintain the surface linear velocity of the cone roller and the ring surface linear velocity is equal to the rolling process to follow the ring back, followed by the core roller and the upper cone roller were done radially and axially feeding movement, which produces continuous plastic strain on the ring, so that the blank ring outer diameter increased This results in an increase in the outer diameter of the blank ring, a decrease in wall thickness, and shaping of the cross-section profile. Figure 1 for the diameter-axial rolling schematic diagram.
Figure.1 Diagram of radial-axial rolling
In Deform finite element software, the rill rolling algorithm has a unique stability control function that can ensure that the calculation process of the ring is in the horizontal and vertical direction of the movement and rotational stability, so this paper does not study the effect of the angle of the following guide roller on the ring, simplify the model, in the special module to enter the two-dimensional parameters, through the rotary axis to generate a three-dimensional model, the generated model is shown in Figure 2.
Fig.2 Finite element model of ring piece rolling
In the simulation, each roll is set as a rigid body, the ring blank material adopts the X40Cr13 parameters in Deform database, and the ring mesh division adopts Deform’s unique ALE adaptive mesh re-division technology, which can automatically re-divide the mesh during the ring rolling process, which improves the computational accuracy and reduces the computation time at the same time. Due to the large size of the ring, a description file is added before starting the simulation to make the cone roll adaptively follow the center of the ring backward. In the rolling process, the friction factor and thermal convection coefficient are two important parameters; Wu Andong studied the rolling process of a 40Cr13 stainless steel ring and obtained a reasonable value of the friction factor, combined with the actual processing of similar rings in the enterprise’s basic parameters of the actual rolling process of the ring in this paper, the specific parameters of the simulation of the ring rolling process is shown in Table 1.
1.2 Design of ring blank size
According to the size of the finished ring, this paper adopts the rolling process of wall thickness thinning, height reduction, and diameter expansion, as shown in Figure 3. In the design of blank size, the rolling ratio and radial-axial deformation distribution ratio are based on the rectangular cross-section of the ring parts of the radial-axial rolling blank size design method. Based on this method, combined with the finished product size to design the blank size, the specific parameters are shown in Table 2.
Table.1 40Cr13 stainless steel ring rolling parameters
| Parameter | Numerical value |
| Drive roller radius R1/mm | 425 |
| Core roller radius R2/mm | 145 |
| Cone angle of cone roller/(°) | 35 |
| Roll temperature/℃ | 150 |
| Environmental temperature/℃ | 20 |
| Friction coefficient between the driving roller and the ring | 0.88 |
| Friction coefficient between core roller and ring component | 0.5 |
| Friction coefficient between cone roller and ring | 0.6 |
| Thermal convection coefficient between the roller and the ring/(N · (s·mm·℃)-1) | 0.02 |
| Heat transfer coefficient between roller and ring/(N·(s·mm·℃)-1) | 5 |
Figure.3 Before and after rolling ring cross-section schematic diagram
Table.2 Specific size parameters of the ring piece (mm)
| Ring components | Parameter | Numerical value |
| Finish product | Outer diameter Df | Φ2952 |
| Wall thickness bf | 163 | |
| Height hf | 305 | |
| Blank | Outer diameter D0 | Φ1640 |
| Wall thickness b0 | 245.5 | |
| Height h0 | 405 |
1.3 Design and calculation of roll movement law
1.3.1 Speed of driving roll
In the rolling process, the rotational speed of the driving roll determines the rotational speed of the ring, and the quality of the finished product will be affected if the rotational speed of the ring is too large or too small. Too small drive roll speed will make the radial feed per revolution increase, resulting in the ring cannot bite into the hole shape; a drive roll speed is too large will make the ring rotational speed too large, which will make the rolling unstable, affecting the quality of the ring. Therefore, according to the actual production experience, the ring in the rolling process to maintain the linear speed in the range of 0.4-1.6m.s-1 is more appropriate. Assuming that no relative sliding occurs between the ring piece and the roll, the range of the driving roll speed nr is obtained according to the geometric relationship:
0.2/πR1≤nr≤0.8/πR1 (1)
1.3.2 Roll feeding speed
In this paper, the roll feeding method adopts the way that the speed of increasing the outer diameter of the ring piece is constant; for this reason, under the premise of ensuring that the core roll and the upper cone roll feed are satisfied, a concave rolling curve is used to set up the core roll and the upper cone roll feeding motion curve. The instantaneous height h calculation formula is shown in equation (2):
h = (h0-hf)/(b0-bf)2⋅(b-bf)2+hf (2)
In the formula: b is the instantaneous wall thickness.
Assuming that the volume before and after rolling is constant, it is obtained:
π(D20-d20)h0 = π(D2-d2)h (3)
In the formula: D is the instantaneous outer diameter; d is the instantaneous inner diameter; d0 is the inner diameter of the ring blank.
Therefore, the instantaneous outer diameter D is:
D = (D0-b0) b0h0/bh + b (4)
According to the constant volume before and after rolling can be obtained, the core roll feed speed Vf and the upper cone roll feed speed Va are calculated as.
Vf = Vd/[(D0-b0)b0h0(1/b2h+(h0-hf)/(b0-bf).1/bh2)-1]
Va = 2Vf⋅(h0-hf)/(b0-bf)2⋅(b-bf) (5)
In the formula: Vd is the speed of increasing the outer diameter of the ring member.
Under the conditions of radial hole type biting in and radial hole type forging through, it is concluded that the range of core roller feeding speed is.
6.55×10-3nrR1(D0-d0)(1/2R1+1/2R2+1/D0-1/d0)/πD0 ≤ Vf ≤ 2nrβ2R21/(1+R1R2)2⋅(1+R1/R2+2R1/D0– 2R1/d0)/πDf (6)
The formula: β is the friction angle between the ring member and the core roller. Combining the above equations (5) and (6) to obtain the range of values of the outer diameter increasing speed Vd is:
Vf min[(D0-b0)b0h0(1/b2h+nr⋅(h0-hf)/(b0-bf)⋅1/bh2)-1]≤ Vd ≤ Vf max[(D0-b0)b0h0(1/b2h+nr⋅(h0-hf)/(b0-bf)⋅1/bh2)-1] (7)
In the formula: Vf min is the minimum value of the core roller feeding speed; Vf max is the maximum value of the core roller feeding speed.
According to the above calculations can be obtained on the cone roll and core roll feed speed. According to the volume before and after rolling is unchanged, so the total radial feed Δs and total axial feed Δh of the ring are:
Δs = bf-b0 = ∫T0Vd(t)dt
Δh = hf-h0 = ∫T0Va(t)dt (8)
In the formula: t is the time; T is the total rolling time.
In order to ensure the feed quantity and avoid the precision of the ring parts being affected by too large or too small feed quantity, the whole ring rolling process adopts a 4-stage formula: the 1st stage is the accelerating stage; the 2nd stage is the stage in which the speed of increasing the outer diameter of the ring parts is constant due to the movement of the rollers, i.e., the main rolling stage; the 3rd stage is the decelerating stage; and the 4th stage is the idling shaping stage of the ring parts, i.e., the idle stage, which guarantees the roundness and precision after the ring parts are rolled. The schematic diagram of the rolling process is shown in Fig.4.
Figure.4 Schematic diagram of the rolling process
To guarantee that the outer diameter of the rings increases at a uniform rate during the main rolling stage within the rolling curve, this paper sets the accelerating stage within the radial feed Δs/10 (at this time, the wall thickness is b1), and the time of the accelerating stage, T1, is:
T1 = 2b1/Vf1 (9)
In the formula: Vf1 is the instantaneous feed speed of the core roll when the instantaneous wall thickness of the ring is b1.
In this paper, we set the feed amount of the main rolling stage as 8/10.Δs; at this time, the wall thickness of the ring piece is b2, then the time of the main rolling stage T2 is:
T2 = [(D0-b0) b0h0(1/b2h2-1/b1h1) + b2 – b1]/Vd (10)
In the formula: h1 and h2 are the instantaneous height of the ring member when the instantaneous wall thickness of the ring member is b1 and b2, respectively.
In this paper, the last radial feed Δs/10 of the core roller is set as the deceleration stage, which makes the core roller gradually decelerate to 0, then the time T3 of the deceleration stage is:
T3 = 2(b2-bf)/Vf 2 (11)
In the formula: Vf 2 is the instantaneous feed rate of the core roller when the instantaneous wall thickness of the ring is b2.
The 4th stage is the idling stage, set the idling λ turns after the end of the feeding motion, according to the speed of the driving rollers can be obtained as the idling time T4 of the ring member:
T4 = λR/nrR1 (12)
In the formula: R is the instantaneous outer radius of the ring member.
The design process of the upper cone roller feed speed curve is the same as the above calculation while ensuring the feed quantity. According to formula (5) to calculate the ring outer diameter increase speed of 5.6mm.s-1 when the core roll and the upper cone roll feed speed, and combined with formula (9), formula (10), formula (11) and formula (12) shown in the four stages of the rolling time and the corresponding amount of feed to get the feed speed curve, shown in Figure 5.
Figure.5 Roll feed speed curve
According to the above roll feed speed design method, measuring the simulation of the ring diameter increases speed because the cone roll in the rolling process follows the ring center backward, so the collection of cone roll in the rolling process of the backward path, Figure 6 for the cone roll in the rolling of the backward distance with the time of the curve, can be seen in Figure 6, the main rolling stage of the simulation of the cone roll backward distance, and the theoretical cone roll backward distance coincide.
Figure.6 Variation curve of the backward distance of the cone roll with time
2. Simulation analysis and parameter optimization
According to the method given in Chapter 1 to calculate the rolling parameters, this paper is based on Deform-3D software for finite element simulation of 40Cr13 stainless steel ring, using the control variable method to select the initial temperature of the ring, the drive roll speed and the ring outer diameter of the speed of the ring to simulate the simulation of the three parameters, to analyze the equivalent strain, temperature, and radial rolling force change rule, the standard deviation of the equivalent strain of the ring and the standard deviation of the temperature after rolling. The standard deviation of equivalent strain and temperature of the ring after rolling is expressed by SDP and SDT, respectively, and the optimal parameters are selected by analyzing these rules of change.
Select the above design of the ring blank for simulation of rolling, Figure 7 for the rolling process under different moments of the ring structure changes, and the actual finished ring map. As seen from Fig. 7, the outer diameter of the ring becomes larger, the wall thickness decreases, the height decreases during the simulated rolling process, and the ring stays round during the whole rolling process. In terms of accuracy, the simulated finished ring size and the theoretical ring size, although there is a certain gap within the allowable range, indicate that the design of the ring blank size and simulation of the rolled ring can meet the technical requirements of the product. Figure 7e shows the actual finished ring of the same size.
Figure.7 Structural changes in the ring at different moments in the forming process and the finished rings
(a) 45s; (b) 205s; (c) 350s; (d) simulated finished ring piece; (e) actual finished ring piece
2.1 Influence of the initial temperature of the ring on the rolling results
Combined with the nature of 40Cr13 stainless steel and the actual production of forging temperature, selected four groups of ring initial temperature: 1000, 1100, 1200, and 1300 ℃, this paper to maintain the other parameters remain unchanged under the conditions of the simulation analysis of 40Cr13 stainless steel ring.
The SDP and SDT of different ring pieces at the initial temperature after rolling are shown in Fig. 8, from which it can be seen that, with the increase of the initial temperature of the ring pieces, the standard deviation of the equivalent strain of the ring pieces after rolling gradually decreases. In contrast, the standard deviation of the temperature gradually increases because, with the increase of the rolling temperature, the mobility of the metal becomes better. The forging permeability further improves so that the equivalent strain at the ring pieces’ surface decreases while the equivalent strain at the center increases. The equivalent strain at the surface of the ring is reduced, while the equivalent strain at the center is increased; therefore, with the increase in the initial temperature of the ring, the uniformity of the distribution of the equivalent strain of the ring after rolling is also improved.
Fig.8 Variation of SDP and SDT at different initial temperatures of ring pieces
Fig.9 Variation of radial rolling force during rolling at different initial temperatures of the rings
However, the standard deviation of the temperature of the ring member increases after rolling, which is due to the following reason: with the increase of the initial temperature of the ring member, the temperature at the surface of the ring member decreases in the rolling process, while the temperature at the center of the ring member decreases less, which leads to a larger temperature difference between the surface and the interior of the ring member.
Figure 9 shows the change of radial rolling force at different ring initial temperatures. It can be seen from Figure 9 that the maximum radial rolling force required in the rolling process decreases with the increase of initial temperature of the ring, which is due to the following reasons: with the increase of temperature, the metal mobility becomes better, the deformation resistance of the ring reduces, and the plasticity is further improved, so that the rolling force required in the rolling process is reduced with the increase of initial temperature of the ring. And decreases.
In summary, with the increase in the initial temperature of the ring parts, the rolling of the ring parts of the equivalent effect of the standard deviation of deformation and radial rolling force will be improved. Still, the temperature standard deviation will be reduced. At the same time, too high a temperature will cause a waste of resources, so the comprehensive consideration of selecting the ring parts at the initial temperature of 1100 ℃.
2.2 Impact of driving roll speed on rolling results
Under the condition of keeping other parameters unchanged, combined with the actual production experience, select the drive roller speed were 16, 20, 24, and 30r.min-1 to analyze its impact on the ring rolling.
After rolling, the SDP and SDT under different driving roll speeds are shown in Fig. 10. It can be seen from Fig. 10 that the standard deviation of the equivalent strain of the rolled rings increases gradually with the increase of driving roll speed, which is because with the increase of driving roll speed, the rotational speed of the rings increases, the number of nibbles under the same core roll feeding speed increases, the equivalent strain of the ring surfaces increases. The deformation of the inner rings is relatively small. Thus, the inner and outer rings have the same core rolls. The equivalent strain on the surface of the ring increases, while the internal deformation is relatively small, so the difference between the equivalent strain on the inner and outer surfaces of the ring member is larger, which makes the standard deviation of the equivalent strain of the ring member increase.
Fig.10 Variation of SDP and SDT of ring members at different speeds of driving rollers
As the speed of the drive roll increases, the standard deviation of temperature after rolling gradually decreases. Still, when the speed of the drive roll is 30r.min-1, the standard deviation of the temperature of the ring piece will increase, this is because as the speed of the drive roll increases, the heat of plastic deformation of the ring piece increases, and due to the friction will produce a certain amount of heat to offset with the heat lost, so the temperature of the ring piece is distributed more uniformly. However, at 30r.min-1, due to the large speed of the drive roll, the contact time between the ring and the roll is increased, and the heat dissipation at the edge of the ring is faster, which makes the surface temperature of the ring drop faster, resulting in uneven temperature distribution.
Figure 11 shows the radial rolling force change of the ring piece under different drive roll speeds; from Figure 11, with the increase of the drive roll speed, the maximum rolling force of the ring piece in the rolling process gradually decreases, the reason for this is that: with the increase of the drive roll speed, the ring piece rotational speed is also accelerated, and under the same core roll feed speed, the rolling force required for the ring piece is reduced.
Figure.11 Radial rolling force change curve under different drive roll speeds
In the analysis of different drive roll speed ring parts of the equivalent strain standard deviation, temperature standard deviation, and radial rolling force changes, the comprehensive consideration of the drive roll speed of 20r.min-1 under the performance of all aspects of the relatively more appropriate so drive roll speed 20r.min-1.
2.3 The effect of the speed of increasing the outer diameter on the rolling results
Under the condition that other parameters remain unchanged, 4 groups of OD increasing speeds are selected: 3.6, 4.6, 5.6, and 6.6mm.s-1, and their effects on ring rolling are analyzed.
The SDP and SDT at different OD increasing speeds are shown in Fig. 12. It can be seen from Fig. 12 that the standard deviation of the equivalent strain of the rolled rings decreases with the increase of OD increasing speed, which is because with the increase of OD increasing speed of the rings, the forging permeability of the rings gets better, which makes the equivalent strain of the ring surface decrease and the equivalent strain of the ring increase so that the standard deviation of the equivalent strain of the rings decreases with the increase of OD increasing speed and the equivalent strain of the rings increases. Therefore, the standard deviation of the equivalent strain of the rolled ring will be reduced with the increase of the ring diameter increasing speed.
Figure.12 Variation of SDP and SDT of rings with different ring O.D. increasing speeds
The standard deviation of the temperature of the rolled ring gradually decreases with the increase of ring O.D. increasing speed, but it increases again after 5.6mm.s-1; the reason is: when the ring O.D. increasing speed increases, the feed per revolution of the ring increases, which makes the heat of the outer surface of the ring to generate strain increase, while the internal heat dissipation is less, so the standard deviation of the temperature of the ring gradually decreases; however, when the ring O.D. increasing speed exceeds 5.6 mm.s-1, the feed per revolution of the ring is too large, resulting in faster heat generation, which is difficult to diffuse to the outside, thus making the temperature distribution of the ring uneven after rolling.
Figure 13 for different ring outer diameter increase speed radial rolling force changes, as seen in Figure 13, with the increase in ring outer diameter increases speed, the ring required radial rolling force increases, the reason is: ring outer diameter increases speed, making the ring per turn feed amount increases, the whole rolling process time is shortened, therefore, need to complete the rolling required radial rolling force increases.
Figure.13 Different ring outer diameters increase the radial rolling force change speed
Considering the equivalent strain standard deviation, temperature standard deviation, and radial rolling force change of the ring parts after rolling, the outer diameter increase speed of 5.6mm.s-1 is adopted.
3. Parameter analysis and verification
Through the above parameter optimization, the optimization results are as follows: the initial temperature of the ring piece is 1100°C, the speed of the driving roll is 20r.min-1, and the speed of increasing the outer diameter of the ring piece is 5.6mm.s-1. Fig. 14 compares the finished ring pieces before and after optimization. It can be seen from Fig. 14 that the temperature distribution and the equivalent strain of the optimized finished ring piece are more even than that before optimization. The maximal rolling force decreases from 4250kN before optimization to 5.6mm.s-1, and the maximum rolling force decreases from 4,250kN before optimization to 4,650kN. The maximum rolling force is reduced from 4250kN before optimization to 3580kN, which is a big improvement.
Figure.14 Comparison of finished rings before and after optimization
(a) Before optimization, equivalent strain; (b) After optimization, equivalent strain; (c) Before optimization, temperature; (d) After optimization, temperature
According to the optimization results of the above parameters, two kinds of finished ring sizes produced by Yaang Pipe Industry Co., Limited. are selected, the ring blank size is designed, and the above optimization process is selected under the 4-stage rolling curve. The specific parameters are shown in Table 3.
Rolling results show that the equivalent strain distribution and temperature distribution of the two ring parts are more uniform, the shape of the regular and less ellipticity, measurement of the size of the rolled ring parts, ring 1 and ring 2 of the dimensional accuracy of 0.37% and 0.24%, respectively, the diameter error is less than 15mm.
4. Conclusion
(1) The design of the 4-stage rolling curve, based on this rolling curve, in the main rolling stage of the ring growing speed is unchanged, and with the theoretical ring increasing speed is the same after rolling the ring shape regularity, ellipticity is small.
(2) Optimize the initial temperature of the ring part, the speed of the driving roll, and the speed of increasing the outer diameter of the ring part using the control variable method, and comprehensively analyze the standard deviation of the equivalent variation of the ring part, the standard deviation of the temperature and the change of the radial rolling force, and then arrive at the optimized parameters: the initial temperature of the ring part is 1100℃, the speed of the driving roll is 20r.min-1, and the speed of increasing the outer diameter of the ring part is 5.6mm.s-1.
Table.3 Ring rolling parameters
| Parameter | Ring 1 | Ring 2 |
| Finished ring Df × Bf × Hf/(mm × Mm × Mm) | 3255×257.5×215 | 2404×98.0×194 |
| Ring blank D0 × B0 × H0/(mm × mm × mm) | 1801×367.5×315 | 1077×182.5×244 |
| Initial temperature of the ring/℃ | 1100 | |
| Drive roller speed/(r·min-1) | 20 | |
| Increase speed of outer diameter of ring/(mm·s-1) | 5.6 |
(3) The optimized rings’ overall equivalent strain and temperature distribution are better than before optimization. Based on the existing finished ring size, the simulated rolling verification was carried out, and the results showed that the overall performance and accuracy of the optimized rolled rings were also excellent.
Author: Shi Zhiqi
