Optimization of electroslag remelting technology through the use of a digital model of the process

INTRODUCTION

Currently, the metallurgical products market is experiencing a growing demand for metal obtained through electroslag remelting (ESR). This process is used to manufacture ingots, particularly from 45KhN2MFA-Sh alloy steel, which is employed in the production of critical components subjected to torsional stresses under dynamic and cyclic loads.

The tasks of improving product quality and increasing yield are paramount for industrial enterprises. One of the measures to help address these issues is the search for optimal technological process parameters and operating regimes for production equipment. The practical study of thermal regimes in ingot formation through ESR in an operating metallurgical plant presents significant limitations, including the difficulty of measuring the temperatures of the ingot, slag, and consumable electrode, as well as the need to prevent the production of defective parts and the risk of equipment failures during any modifications. In this context, the use of digital models for analyzing solidification processes serves as an efficient and highly informative diagnostic tool, enabling the study of ingot manufacturing with different dimensions in a cost-effective manner without disrupting production processes.

The development and validation of an appropriate mathematical model for numerical calculations open the possibility of investigating the individual characteristics of the solidification process, establishing the dynamics of the distribution of the liquid phase of the slag and metal, and determining the remelting conditions that ensure an optimal combination of productivity and product quality.

Problem Description and Methodology

The metallurgical complex of JSC Metallurgical Plant "Petrostal" has an ESR installation equipped with a bell-type system that allows for the process to be carried out in an inert gas (argon) environment, as well as automatic process control. The basic production technology for ESR ingots with a square cross-section of 500 mm per side and a nominal mass of 4 t of 45KhN2MFA-Sh steel, using the company’s available installation, did not guarantee that the ingots would be free of slag inclusions in their heads. Regardless of the specific steel grade, the heads of the finished ingots, beneath the open shrinkage cavity, contained porosities and closed cavities filled with slag inclusions (Fig. 1).

Fig. 1. Schematic cutting layout of the ingot for separating the scrap portion, including metal evaluation along its longitudinal section (a), and a typical result from inspection for slag inclusions and shrinkage depth in the technological trimming section (b)

To ensure that the ingots enter the subsequent deformation processing without slag inclusions, a standard was established requiring the removal of 4.65% of the metal from the ingot head before its transfer to the rolling workshop. When producing rolled products from ESR ingots, there is a technological need to remove the ends of the rolled product, which ultimately leads to an increase in the overall consumption factor. Under conditions where the removal of technological waste is possible during the rolling process, the prior cutting of the ingot head may be a redundant operation.

The objective of the present work was to develop a set of technological solutions aimed at preventing the formation of slag-filled cavities and porous zones in ingots produced by ESR. The main cause of this type of defect is the excessive depth of the liquid metal bath[1–4].

Experiments conducted under production conditions did not provide a sufficiently clear picture of the impact of varying melting parameters. For this reason, it was decided to combine real production experiments with mathematical modeling of ingot formation processes. This approach significantly reduces the number of physical experiments, substantially increases the amount of data extracted, and enables evaluating the progression of solidification processes not only indirectly, through analysis of the metal in finished ingots, but also by directly observing the dynamics of changes in the thermo-phase field during remelting in a digital ingot model.

The ESR process comprises three main periods:

• Initial phase (according to the solid-start technology) for the formation of the slag bath, the initiation of electrode melting, and the establishment of the required process speed.

• Main electrical remelting regime, during which most of the ingot’s body is formed.

• Shrinkage cavity removal phase, in which shrinkage compensation conditions must be created to ensure a minimal height of the zone occupied by porosities and cavities in the upper part of the finished ingot.

The modeling of ingot growth was investigated using a digital model implemented in the PoligonSoft simulation software environment. When implementing the process model, the following main assumptions were adopted:

• The ingot’s taper was neglected, as defined by the geometry of the copper mold.

• Electrode melting was not included in the model structure; therefore, disturbances caused by variations in its size and cross-sectional shape were not considered.

• The initial heating phase was not considered, as its influence is insignificant compared to the total duration of the remelting cycle and the formation of most of the ingot during the main electrical regime and the shrinkage cavity removal phase.

• A constant temperature of the incoming metal from the top was assumed.

Figure 2 shows a 3D finite element model used for calculating ingot formation during the ESR process. To simulate the progressive filling of the mold, a displacement definition tool available in PoligonSoft was used. This tool is typically employed for calculating heat transfer under sliding contact conditions between bodies, in the analysis of continuous casting processes, among others. In the case of modeling the ESR process, a method was implemented that gradually incorporates the predefined geometry of a full-size ingot into the calculation. The finite elements located above the height reached by the ingot at each instant remain inactive.

Fig. 2. 3D model for analyzing the remelting process: (a) finite element model; (b) portions of the ingot actually involved in the calculation at different process times (shown: one of the initial steps, halfway through the process, and upon completion).

Within the numerical analysis, when defining the remelting regime to be studied, the ingot solidification calculation was carried out taking into account heat exchange with the water-cooled mold, the slag bath, the base plate, and the surrounding environment.

Each technology variant was studied in two stages:

• Main remelting period and shrinkage cavity removal phase, using a specific remelting speed and a corresponding regime for reducing the supplied power.

• Cooling of the ingot in the mold until the moment of its extraction.

During a series of calculations, the influence of the process’s technological parameters on the depth and shape of the metal bath was investigated. These parameters were varied within the limits set by the production equipment’s capabilities. The calculation results provided a visualization of the thermo-phase field distribution in the ingot body during and after remelting, as well as a prediction of shrinkage,including macro and microporosity. According to a known recommendation [2, 4], the optimal depth of the metal pool h_mb should be half the width of the mold a_cr, This value was taken as the target when analyzing and comparing the results of the calculations for different ingot formation regimes.

Results and Discussion

Model verification was performed by comparing temperatures measured with a pyrometer on the surface of a real ingot immediately after it was extracted from the mold with those observed at the same time in the computational model. The calculations showed good agreement with the measurement data (Fig. 3).

Fig. 3. Temperature distribution along the height of the finished ingot according to pyrometry results (dots) and calculation (solid line).

Modeling of ingot formation during the ESR process and in the final cooling stage in the mold was carried out for different mass remelting speeds:

• The speed during the main electrical regime G, corresponding to the base technology previously approved by the company.

• Reduced speeds: 0.84G and 0.94G.

• Increased speeds: 1,1G and 1,2G.

For each remelting speed, appropriate regimes for the shrinkage cavity removal period were selected according to the need for a gradual reduction in applied power (P ) and, consequently, in the remelting speed and the depth of the liquid metal bath.

According to [5], the relative reduction in power during the shrinkage cavity removal phase is defined by the following expression:

Where P₀ and P_f  represent the power at the initial and final moments of the shrinkage cavity removal period, respectively. According to the established recommendations, this reduction should be within the range of 30% to 50%.

Figure 4 presents the power reduction curves during the shrinkage cavity removal period as a function of the remelting speed prior to this phase. These curves were developed based on the aforementioned recommendations.To facilitate comparison, all curves represent the ratio of the current power during the shrinkage cavity removal period (i.e., after remelting at 0.84G, 0.94G, G, 1.1G, and 1.2G) to the initial power of the baseline remelting technology at speed G: P/P₀^G

Fig. 4. Power reduction trajectories during the shrinkage cavity removal phase for different remelting speeds.

The graphical representation of the bath depth (compared to the recommended level h_mb ^opt = a_cr/2 ) and the configuration of the liquid metal bath at different remelting speeds is shown in Fig. 5. The profile and depth of the liquid metal bath, established during the main electrical remelting regime, constitute the initial dataset for the process carried out during the shrinkage cavity removal period. This process involves ensuring solidification conditions (both during this phase and during subsequent cooling) that promote the most complete possible extraction of shrinkage from the upper part of the ingot to the open cavity on its surface.

Fig. 5. Distribution of the liquid phase in the solidification zone upon stabilization of the liquid metal bath during the main electrical remelting regime and at the end of the shrinkage cavity removal phase.

Table 1 presents the main quantitative characteristics of the electroslag remelting (ESR) process and ingot quality, obtained from the remelting modeling results at different speeds. An increase in the remelting speed leads to a logical increase in the depth of the liquid metal bath. This increase is primarily due to a larger volume of the liquid phase, while the width of the two-phase transition zone remains practically constant during the shrinkage cavity removal period and only experiences a slight increase with higher remelting speeds (from 0,10a_cr to 0,13a_cr). The solidification time during the ingot’s final cooling stage shows an insignificant increase as remelting speed increases.

Since the initial depth of the liquid metal bath varies in each numerical experiment during the shrinkage cavity removal phase, along with the specific characteristics of the developed power reduction regimes, the development of shrinkage in the ingot head exhibits unique features in each case. However, the trend of increasing defect depth with higher remelting speed is quite pronounced. At the same time, the ingots produced under the analyzed regimes formally comply with current regulations regarding the required technological trimming level.

At remelting speeds of G to 1.2G, the liquid metal bath acquires a moderate or clearly excessive depth, favoring radially oriented crystallization, hindering the refining of metal from non-metallic inclusions, and leading to an increase in the zone affected by shrinkage defects. An excessively low remelting speed of 0.84G results in an insufficient bath depth, creating risks such as deteriorating separation conditions at the metal-slag interface, reducing refining efficiency, and promoting the formation of surface defects such as constrictions, among other issues.

As the solidification process reaches completion during the ingot cooling stage in the mold, following the shrinkage cavity removal phase (Fig. 6), a small region of liquid metal is observed in all cases beneath the solid-phase bridge, which forms below the open shrinkage cavity.

Fig. 6. Change in the liquid phase content during the ingot cooling stage in the mold after the shrinkage cavity removal phase (example for remelting speed G).

A remelting speed of 0.94G could be advisable for achieving the optimal liquid metal bath depth; however, it does not meet the requirement of at least maintaining process productivity.As suitable options for further optimization, remelting regimes at speeds of G and 1.1G were selected, accompanied by power reduction strategies during the shrinkage cavity removal phase, validated through modeling, within a range of 44% to 50%, to achieve an efficient reduction of the liquid metal bath depth  (h_mb). In general, additional constraints on ΔP_rel within the interval recommended in the literature must be defined on a case-by-case basis, considering factors such as ingot geometry, steel composition, and remelting speed, among others.

The subsequent optimization and adjustment of the selected ESR technological parameters were carried out in practice through the use of additional methods and techniques aimed at further reducing the liquid metal bath depth. As part of these activities, a series of experimental remeltings were conducted, involving modifications to the flux’s chemical composition, a reduction in the immersion depth of the electrode tip in the slag bath, and an increase in the working flux mass. Among the three measures mentioned, the latter two proved to be the most effective. When combined, upon completion of this computational and practical work, a closed shrinkage cavity, free of slag, was obtained, and the depth of its location was reduced to 3.7% (Fig. 7).

Fig. 7. Closed, slag-free shrinkage cavities in the technological trimming section of the experimental ingot (both parts shown after cutting).

Conclusion

Using the digital model of electroslag remelting (ESR) developed in the PoligonSoft simulation system, various technological regimes for ingot formation were analyzed, and several patterns were identified for controlling liquid metal bath parameters, with the aim of reducing the tendency for slag inclusion in the upper part of the ingot and minimizing the area affected by shrinkage defects. Based on the results of computational simulations, along with subsequent adjustments and special modifications at the plant, remelting regimes were approved that improved product quality and optimized economic efficiency and productivity indicators.

• The tendency for slag penetration in the upper part of the ingot was eliminated, removing the need to cut off the upper portion before plastic deformation and generally eliminating the associated risks during rolling.

• An increase of approximately 10% in the productivity of the electroslag remelting (ESR) process was achieved.

• The reduction in shrinkage cavity depth in the upper part created conditions for lowering the technological trimming requirement for commercial ESR ingots by about 20% from the current level.

The approach applied to solve the production problem is characterized by a rational balance between the complexity of developing and using the digital model and the number of experimental remeltings required to refine the identified promising technological solutions.

REFERENCES

[1] Glebov, A.G. Electroslag Remelting. Moscow: Metallurgiya, 1978. — 333 p.
[2] Vorobyev, A.A., Pozhidaev, Yu.V. Electroslag Remelting. Novokuznetsk: SibGIU Publishing, 2002. — 116 p.
[3] Protokovilov, I.V., Porokhonko, V.B. Methods of controlling metal crystallization in ESR ingots. // Modern Electrometallurgy, 2014, No. 3, pp. 7–15.
[4] Medovar, B.I., Shevtsov, V.L., Marinsky, G.S., Demchenko, V.F., Makhnnenko, V.I. Thermal Processes in Electroslag Remelting. Kiev: Naukova Dumka, 1978. — 304 p.
[5] Pavlov, V.A., Lozovaya, E.Yu., Babenko, A.A. Special Electrometallurgy of Steel and Alloys. Yekaterinburg: Ural University Publishing, 2018. — 168 p.

Translated by A.J. Camejo Severinov
Original text in Russian
‍