3.1 Cooling rate and hardness in single-sided die quenching
The heated boron steel sheet was quickly transferred to the single-sided quenching die to conduct the quenching process. According to the continuous cooling transformation (CCT) diagram of boron steel, the minimum cooling rate required to fully transform the austenite (A) into martensite (M) was in the range of 25–30°C/s [25, 26].
The cooling rate during the single-sided cooling process was calculated from the obtained cooling curve, as shown in Fig. 5. It was clear that the cooling rate of the non-contact zone was affected by the thickness of the sheet and the size of the non-contact zone. For a given sheet thickness, as the diameter of the non-contact circular zone decreased, the cooling rate of this zone increased. When the sheet thickness was 1.4 mm, the cooling rates of the central non-contact zone with diameters of 15 mm and 20 mm were 47°C/s and 30°C/s, respectively, which both satisfied the condition for complete martensite transformation. However, when the diameter of the non-contact zone increased to 25 mm, the cooling rate of the central zone was 23°C/s, which was already lower than the critical cooling rate. In addition, for sheets of wall thicknesses of 1.2 mm and 1.0 mm, the cooling rate in the central zone was greater than 25°C/s, which still satisfied the condition for complete martensite transformation. The results above demonstrated that for a boron steel sheet with a specific thickness, there existed a critical dimension of the non-contact zone that could achieve complete martensite transformation. Decreasing thickness of the sheets would enlarge the critical dimension of the non-contact zone.
To quantitatively characterize the mechanical properties of the sheet after quenching, the Vickers hardness of the sheet from the center of the non-contact zone to the boundary with the die after quenching was tested. The hardness distribution results are shown in Fig. 6.
In the 5 groups of experiments, the Vickers hardness measured in a non-contact zone diameter of 25 mm and a sheet thickness of 1.4 mm was significantly lower than the other experimental conditions. Under this condition, the hardness values in the radius range of 6 mm in the central zone were only between 435 ~ 440 HV0.2, while in the radius range of 6.0 ~ 12.5 mm, the hardness gradually increased to 470 HV0.2. In the other 4 sets of experiments, the hardness of the entire sheet remained between 473 ~ 485 HV0.2. The central non-contact zone and the die-contact zone had similar hardness values, indicating that the non-contact zone also achieved complete quenching, which was completely consistent with the judgments of complete quenching according to the cooling rate in the non-contact zone in Fig. 5 above.
3.2 Effects of process parameters on the corner filling
The cooling rates and hardness distribution of the central non-contact zone on the sheet were analyzed by single-sided die quenching experiments, and the cooling rate of the non-contact zone of the boron steel tube during the in-die quenching process was simulated. Based on the results of hot gas forming-quenching integrated process experiments for boron steel variable diameter tubular parts of a thickness of 1.4 mm, the effects of process parameters on corner filling, thickness distribution, microstructure, and mechanical properties would be explored in the following sections.
3.2.1 Effect of bulging temperature
Figure 7 shows the variable diameter tubular parts obtained under a pressurizing rate of 3 MPa/s and different bulging temperatures. It was clear that bulging had occurred in the middle zone of the three formed components. The greater the bulging temperature, the greater the bulging that occurred in the middle zone, and thus the greater the axial length of the effective bulging zone. The main difference within the three formed components was observed in the corner zone.
Contours of the formed tubular part measured by the 3D optical scanner were compared with the die cavity, and the differences were obtained. Figure 8a shows the dimensional deviation of the formed tubular parts under a pressurizing rate of 3 MPa/s and different bulging temperatures. In the figure, the threshold value for the gap is 0.2 mm. It was clear that as the bulging temperature increased, the zone with a gap value of more than 0.2 mm decreased rapidly. The entrance corner zone (AQ) of the component and most of the middle bulge zone had contacted the die although there were different degrees of dimensional deviation between the variable diameter corner zone (AP) and the adjacent parts of the bulging zone on the components. The proportion of the areas in the component corner zone (PAQ) with a dimensional deviation lower than 0.2 mm at bulging temperatures of 700 ℃, 800 ℃, and 900 ℃ was 73.33%, 72.20%, and 65.29%, respectively. Figure 8b shows the filling state of the corner zone for variable diameter tubular parts under different bulging temperatures. With the increase of bulging temperature, the corner zone gradually contacted the die, and the obtainable minimum corner radius became smaller. At the bulging temperatures of 700 ℃, 800 ℃, and 900 ℃, the obtainable minimum radius of the variable diameter corner (measuring at section AP in Fig. 8a) was 19 mm, 17 mm, and 16 mm, respectively.
3.2.2 Effect of pressurizing rate
Figure 9 shows the formed variable diameter tubular parts under a bulging temperature of 900°C and different pressurizing rates. Under this bulging temperature and with the increase in pressurizing rate, the bulging deformation of tubes was more sufficient. This is because the use of a high pressurizing rate could build up the internal pressure to the required value more quickly after closing. Thus, the component could be formed under a relatively higher temperature and lower deformation resistance.
Figure 10a shows the dimensional deviation distribution of the formed components under a bulging temperature of 900°C and different pressurizing rates. At the pressurizing rates of 1 MPa/s, 3 MPa/s, and 5 MPa/s, the entrance corner zone (AQ) had fully contacted the die, while the variable diameter corner zone (AP) and part of the bulging zone did not fully contact the die. The width of the region that was not in contact with the die between the bulge zone and point A decreased with the increase of pressurizing rate: from 13.75mm at the pressurizing rate of 1 MPa/s to 8.31mm at the pressurizing rate of 5 MPa/s. Figure 10b shows the filling state of the corner zone under different pressurizing rates. Corresponding to the pressurizing rates of 1 MPa/s, 3 MPa/s, and 5 MPa/s, the radius of the variable diameter corner was 24 mm, 16 mm, and 12 mm, respectively. Therefore, increasing the pressurizing rate would obtain a greater degree of filling at the corner zone and a smaller variable diameter corner radius.
It should be noted that the maximum gas pressure provided by the pressurizing unit was 35 MPa. Due to the limitation of pipeline diameter and safety factors, 5 MPa/s was already the highest pressurizing rate that could be achieved in the experiments. In the process of pressurizing, the hot tube would cool down rapidly under the action of the outer cold die and the inner room-temperature gas. When the material strengthening caused by “rapid cooling” was greater than the load increase caused by “pressurizing”, the tube would no longer deform, and therefore it would be meaningless to continue with pressurizing.
3.3 Thickness distribution
The wall thickness uniformity is another major factor in evaluating the performance of tubes after forming. The formed variable diameter tubular part was cut along the longitudinal symmetrical section, and thickness distribution at each position in the axial direction is shown in Fig. 11. It can be seen that along the axis direction, the wall thickness of the clamping zone was unchanged. The wall thickness decreased rapidly from the entrance corner zone to the variable diameter corner zone. The minimum wall thickness of 1.12 mm appeared at the junction of the variable diameter corner zone and the bulging zone, and the maximum thinning ratio was 20%. Compared with the junction, the bulging zone was thicker with an average wall thickness of approximately 1.14 mm.
Studying comprehensively the different bulging results in Fig. 8 and Fig. 10, changes in the tubular part section's shape and size, as well as the contact gap to the die of the component can be used to understand the bulging and die-contacting process of the tubes. For the variable diameter tubular parts, the deformation was concentrated in the bulging zone and the corner zone. Under internal gas pressure, bulging first occurred in the middle of the bulging zone. As the diameter in this region increased and the wall thickness decreased, the circumferential stress increased gradually, and the deformation would continue until the tube contacted the die. Due to the temperature of the tubes reducing sharply and the deformation resistance increasing when the bulging zone contacted the die, the deformation zone would extend to the corner zone on both sides. As mentioned previously, when the pressurizing rate was slow, the deformation at a low speed would no longer continue and thus the corner filling process would stop. This process has essential differences from thermostatic forming processes such as superplastic forming and hot gas forming with hot dies.
3.4 Mechanical properties and microstructural features
During the forming process of the variable diameter tubular parts, large plastic deformation occurred in the corner zone and bulging zone, and the microstructure and properties would change significantly. The Vickers hardness of the two regions (see Fig. 4) was measured and the tensile properties of the bulging zone were tested. The results are shown in Fig. 12.
The Vickers hardness and tensile strength of the corner zone (Position A) and bulge zone (Position B) were affected by the bulging temperature and the pressurizing rate. Under the pressurizing rate of 3 MPa/s and bulging temperature of 700 ℃ and 800 ℃, the Vickers hardness of the corner zone was 262.4 HV0.2 and 372.9 HV0.2, respectively, which were both lower than the hardness of the material in a fully quenched state of 550 HV0.2. When the bulging temperature increased to 900 ℃, the Vickers hardness of the corner zone was 576 HV0.2, indicating that the material had completed martensite transformation. When the bulging temperature was 900 ℃ and the pressurizing rate was 1 MPa/s, the Vickers hardness of the corner zone was only 432 HV0.2, with incomplete martensite transformation. This fully illustrates the comprehensive effect of bulging temperature and pressurizing rate on deformation and quenching effect.
To further illustrate the microstructure evolution of each region after deformation and its relationship with mechanical properties, the metallographic structures of the corner zone and bulge zone under different deformation conditions are shown in Fig. 13.
At a pressurizing rate of 3 MPa/s and the bulging temperature of 700 ℃, no clear martensite (M) transformation had occurred in the corner zone, and only a small amount of martensite appeared in the bulging zone. When the bulging temperature increased to 800 ℃, a full martensite microstructure appeared in the bulging zone, although the corner zone was still dominated by ferrite (F) and pearlite (P). When the bulging temperature was increased to 900 ℃, the complete martensite microstructure formed in the corner zone. Under a bulging temperature of 900 ℃, if the pressurizing rate reduced to 1 MPa/s, no full martensite microstructure existed in the corner zone. These results were completely consistent with the above-mentioned hardness and tensile properties.
3.5 Process window for B1800HS tube
As mentioned above, it is critical to complete the hot deformation and quenching processes in a coordinated manner by reasonably matching the bulging temperature and pressurizing rate in the hot gas forming-quenching integrated process for boron steel tubes. Otherwise, incompatible deformation, such as local bulging and even rupture may occur due to excessively high temperatures and unreasonable temperature distribution on the tube. On the other hand, defects such as insufficient filling at the corners will appear because of the rapid cooling and high deformation resistance of the tube.
Figure 14a shows the relationship of tube temperature and gas pressure with time in the hot gas forming-quenching integrated process for boron steel tubes. It is assumed that the cooling of the hot tube is not affected by pressurizing and deformation. After the tube is sealed at both ends, rapid gas filling and pressurizing commence. When the gas pressure reaches the minimum internal pressure Pmin required for material deformation, tube deformation occurs. Meanwhile, the phase transformation takes place with the decrease of tube temperature. Importantly, the deformation of the tube is advised to occur earlier than the phase transformation and be completed before the phase transformation is complete. The reason for this requirement is that when the phase transformation is completed, the tube temperature is relatively low (approximately 280°C) and a large amount of austenite has been transformed into martensite microstructure, leading to a sharp increase in deformation resistance. Although the gas pressure is further increased, it is increasingly difficult to fill small features and the maximum gas pressure is Pmax when the tube deformation is finished. The rapid cooling rates limit the range of process parameters by which the tube can be deformed.
To improve the filling capacity, especially to accomplish forming at small corners, it is necessary to extend the deformation time of the tube at elevated temperature or increase the maximum bulging gas pressure within the deformation time. This can be achieved by decreasing the cooling rates of the tube in the early stage of deformation and increasing pressurizing rates, respectively. For the two initial parameter curves (cooling and pressurizing curves) in Fig. 14a, with the adjustment of the process parameters, the curve trajectory will be changed. By decreasing the cooling rates before deformation or at the initial stage of deformation, the cooling curve “O” will move left to curve “A” and tDS will advance to t*DS, which means the tube has a greater initial deformation temperature. By increasing the pressurizing rates, the curve “O” will move left to curve “B”, thus the tube will have a higher maximum bulging gas pressure of P*max.
By combining the microstructural results in the corner zone, appropriate processing parameters can be determined to obtain specific phase compositions. Such a processing window for boron steel tubes with a triangular shape is shown in Fig. 14b, which combined with the literature [27], is based on whether the axial corner zone has achieved martensite transformation as the established standard. The above adjustments of the process parameters in Fig. 14a provide a possibility to extend the reasonable range of process parameters for the hot gas forming-quenching integrated process. Therefore, the process window in Fig. 14b of boron steel tubes can be further enlarged.
It should be noted that the hot gas forming-quenching integrated process involves the complex combination of heat transfer, deformation, and microstructure evolution. Therefore, this process is highly coupled with hot deformation and quenching, which will inevitably shrink the range of feasible process parameters and narrow the processing window. In this way, it is not appropriate to evaluate the quality of the components and determine the process parameters from a single process. Instead, the reasonable range of process parameter can be optimized or extended according to the relationship from multi-dimensional mapping between the process parameters and the critical evaluation methods. For instance, based on the minimum diameter of the non-contact zone that can realize complete quenching measured in the experiments of this paper, the local die-contacting requirements can be decreased within the allowable range of part accuracy. For another, due to the existing extreme value of the corner radius that can be obtained in the hot gas forming, the gas pressure fluctuation has little influence on the results in the later stage of forming. Hence, the seriously control of the pressurizing rate in the later stage of forming can be reduced. The research results on this aspect are subject to further study.