A method for manufacturing a high-quality electric furnace hot work die steel

Abstract

The application discloses a manufacturing method of high-quality electric furnace hot-working die steel, which comprises the following steps: initially refining die steel raw materials by an EAF furnace; refining the obtained metal raw materials by an LF furnace; vacuum degassing by a VD furnace; improving carbide by adopting magnesium cored wire; pouring the obtained molten steel into 13t or 20t steel ingots by using a protective slag and hanging; using a heating agent after pouring; demoulding after mould cooling; hot sending the steel ingot to a forging area for heating or hot charging annealing after demoulding; forging by adopting a programmed heating mode; and finally heat treating. The die manufactured by processing and heat treating the forged rod is used for producing 6 series aluminum alloy profiles by a 5000t and above extruding machine, the die does not crack and fail early, the service life is equivalent to that of an electric slag 4Cr5MoSiV1 (H13), the manufacturing cost is greatly reduced, and the effect is good.

Description

Technical Field

[0001] This invention relates to the field of mold steel technology, specifically to a method for manufacturing high-quality electric furnace hot work mold steel. Background Technology

[0002] Molds are the main forming tools in the manufacturing industry, widely used in die casting, extrusion, stamping, hot forging and other production processes. With the rapid development of industrial technology and the transformation and upgrading of products, mold manufacturing and mold steel manufacturing are also constantly iterating and upgrading.

[0003] Hot work die steel is used to manufacture dies for forming metals or liquid metals heated to above their recrystallization temperature. It is commonly used to manufacture die-casting dies, hot extrusion dies, hammer forging dies, and press forging dies. For example, in the production of aluminum alloy profiles, the hot extrusion die is a complex thermo-mechanical coupling process involving multiple deformations such as elastic-plastic, rigid-plastic, and viscoplastic deformation. During operation, the die cavity surface is heated to 300-600°C, which can easily cause a decrease in die hardness and wear and cracking. At the same time, the cross-section of aluminum alloy profiles is complex, and there are many sharp corners in the die cavity. Stress concentration is easily formed during the die extrusion process, leading to die cracking. Therefore, the die steel for extrusion is required to have not only sufficient high-temperature strength but also high toughness to prevent die cracking.

[0004] Currently, the most widely used hot work die steels both domestically and internationally are the Cr5 series, represented by 4Cr5MoSiV1 (H13) and 4Cr5MoSiV (H11). By adding a certain amount of alloying elements such as Si, Mo, and V, Mo and V easily form M6C and MC type carbides, which can refine the austenite grains, thereby improving the steel's toughness and resistance to thermal fatigue. During tempering, a certain amount of M2C and MC type dispersed carbides can precipitate, increasing the secondary hardening effect and providing certain high-temperature strength and thermal stability.

[0005] To improve the performance of hot work die steels for extrusion, such as 4Cr5MoSiV1 (H13), and to manufacture them in a more economical way, metallurgists at home and abroad have successively developed different smelting, hot working methods, and heat treatment processes. For example, in smelting and casting, they can be divided into converter continuous casting, electric furnace die casting, and electric furnace die casting + electroslag remelting. Hot working methods can be divided into forging and rolling. Heat treatment processes can be divided into general annealing, spheroidizing annealing, solution treatment + spheroidizing, etc. The processes are continuously optimized at each manufacturing stage, thereby improving high-temperature strength and toughness, and reducing manufacturing costs. Among them, continuously cast billets with lower manufacturing costs can meet the requirements of small-sized flat dies and bushings with lower specifications. However, for high-quality 4Cr5MoSiV1 (H13) large forging bars with a specification of Φ350mm or more used to manufacture multi-hole die cores, there are currently two main manufacturing methods on the market. One is simply called electric furnace 4Cr5MoSiV1 (H13), which involves electric furnace smelting, ladle refining and vacuum degassing to cast steel ingots of more than 10 tons, and then hot-transferring the ingots to fast forging for forming finished products. However, due to the large size, segregation is more serious, the uniformity of the structure is poor, the forging bar has a loose center defect, and even large chain-like liquid carbides exist. Flaw detection often shows excessive phenomena (such as...). Figure 1 (As shown). To improve the segregation, uneven microstructure, and poor strength and toughness of large-size electric arc furnace (EAF) steel forgings, many companies often use small-size EAF die casting + forging forging, and then reforging to expand the size. This method can improve the problems to some extent, but the reforging process increases annealing and consumption, raises quality risks, increases costs, and extends the production cycle. Another method, abbreviated as electroslag 4Cr5MoSiV1 (H13), uses EAF die casting + electroslag remelting + forging to directly produce large-size forgings. Compared with EAF steel, the purity and microstructure uniformity of the steel can be improved, but the addition of electroslag remelting and annealing processes significantly increases the cost of the die steel, approximately 3,000 yuan per ton of steel. Furthermore, regardless of whether it's electric furnace steel 4Cr5MoSiV1(H13) or electroslag steel 4Cr5MoSiV1(H13) forging bars, without a high-temperature homogenization heating process and a microstructure refinement heat treatment process, due to their large size and slow cooling rate after forging, banded segregation will still occur. The microstructure after annealing will be very uneven, often exceeding the poor AS10 grade, and exhibiting unstable quality phenomena such as chain-like or large-block liquid carbides and non-compliant flaw detection. The resulting molds often suffer from early cracking and short service life (refer to...). Figure 2 (As shown). Clearly, neither of the two existing manufacturing methods can meet the increasingly demanding requirements of the aluminum extrusion industry's molds. Summary of the Invention

[0006] In order to achieve the purpose of the background art description, the technical solution of the present invention is as follows: The present invention provides a method for manufacturing high-quality electric furnace hot work die steel, including the following steps: (1) using an EAF furnace to perform primary refining of die steel raw materials;

[0007] (2) The metal raw materials obtained from the primary refining are then refined in an LF furnace. Lime is added during the refining process to form slag, and then deoxidation is carried out.

[0008] (3) After refining in step (2), the steel material is then degassed in a VD furnace under vacuum, and magnesium cored wire is used to improve the carbide content.

[0009] (4) The molten steel obtained in step (3) is poured into 13t or 20t steel ingots. If nodules are found at the sprue, they are removed immediately. Protective slag is used and the ingots are hung up. After pouring, a heating agent is used. After the mold cools, the ingots are demolded. After demolding, the steel ingots are hot-sent to the forging area for heating or hot-charging annealing.

[0010] (5) Forging: Forging is carried out by programmed heating. The temperature of the first stage is 650-700℃, the temperature of the second stage is maintained at 650-700℃, the temperature of the third stage is raised to 840-860℃, the temperature of the fourth stage is maintained at 840-860℃, and the temperature of the final stage is 1235-1265℃.

[0011] (6) Heat treatment: After forging, air cool to 300-350℃ on the surface, and then solution treatment and spheroidizing annealing.

[0012] Furthermore, in step (1), the raw materials for the mold are selected from scrap steel with a phosphorus mass percentage of less than 0.025% and a carbon mass percentage of less than 0.37%, H13 or CrMoV mold steel cut-offs and high-quality pig iron, and the tapping temperature is ≥1610℃.

[0013] Furthermore, in step (2), the LF heating temperature is 1670-1700℃, lime is added 2-3 times, and the amount of lime added each time is 100-300kg. Deoxidation is carried out using SiFe powder and C powder, with an addition amount of 20-40kg of SiFe powder and C powder. The Ar gas pressure is controlled to ensure that the molten steel does not overflow the slag surface. The mass percentage of aluminum in the mold raw materials is controlled at 0.02-0.05%. After refining, the slag with a mass percentage of 30-50% of the steel ladle is discarded. The temperature of the VD ladle is 1610-1640℃.

[0014] Furthermore, in step (3), the ladle pumping time is ≤8min, the vacuum degree reaches 66.7Pa for ≥20min, the Ar pressure is maintained at 0.20-1.00Mpa under vacuum conditions, and the bottom blowing argon weak stirring time must be ≥10min. The liquidus temperature of the steel is approximately 1483℃, and the ladle temperature is 1520~1540℃. The magnesium-clad wire used has a magnesium mass percentage of 9%, an aluminum mass percentage of 29%, and the balance is iron. The magnesium-clad core wire is fed in two batches, 60 meters each time, with a 30-second pause in between, and the feeding speed is 150m / min.

[0015] Furthermore, in step (4), argon gas is used for protection during the casting of the steel ingot, and the argon gas flow rate is controlled at 2 Nm. 3 / h-5Nm 3 / h, the exothermic agent used is (SiO2-CaO-Al2O3-Na2O-K2O), the demolding time for 13t is ≥9 hours, and the demolding time for 20t steel ingot is ≥12 hours.

[0016] Furthermore, in step (5), the first stage temperature is 650-700℃ and maintained for 1.5h. After 2-3h, the temperature is raised to 840-860℃ and maintained for 1-2h. After 5-6h, the temperature is raised to 1235-1265℃ and maintained for 15-28h. The starting temperature is ≥1100℃, the stopping temperature is ≥850℃, the forging ratio is ≥5, and the electric furnace ingot adopts a 2-upsetting and 2-drawing process. The first upsetting should be upsetting to less than 40% of the original ingot height to ensure internal forging. After the first forging is completed, the heating temperature of the remaining forgings is reduced to 1200±10℃ and held for ≥2h. After forging and upsetting, the ingot is widened, and the deformation in the last forging is ≥40%. The defects at the head and tail are cleaned by hot cutting after forging.

[0017] Furthermore, in step (6), the solution treatment process involves maintaining a temperature of 350-450℃ for 4-5 hours, raising the temperature to 830-870℃ at a rate of ≤80℃ / h, maintaining it for 3 hours, raising the temperature to 1040-1060℃ at a rate of ≤80℃ / h, maintaining it for 10 hours, and finally cooling the surface with water mist to 300-400℃, followed by air cooling to 200-300℃.

[0018] Furthermore, in step (6), the spheroidizing operation is to maintain the temperature at 600-640℃ for 2 hours, and then raise the temperature to 870-900℃ at a rate of ≤30℃ / h. At this time, the amount of spheroidizing treatment M is M=3+0.2Q+2D, where D is the diameter of the round steel or the thickness of the flat steel in decimeters, and Q is the actual amount of steel loaded into the furnace in tons. Then, the temperature is lowered to 720-740℃ at a rate of ≤30℃ / h. The spheroidizing is stopped when the amount reaches 10D. Finally, the temperature is lowered to 350℃ at a rate of ≤40℃ / h before the steel is removed from the furnace.

[0019] Preferably, the process also includes, after the hot work die steel is manufactured, performing defect detection on the hot work die steel, and obtaining qualified hot work die steel when the detection results meet the requirements.

[0020] The defect detection of hot work die steel includes:

[0021] Select any hot work die steel as the hot work die steel to be tested; acquire images of the hot work die steel under different perspectives, and take the image under each perspective as the target image. The target image is marked with the target position and target flow direction corresponding to each target gate in the target gate set of the hot work die steel.

[0022] Based on Hough line detection, straight line edges in the target image are obtained, and straight line edges with a length greater than a preset length threshold are determined as the first connected edges.

[0023] The edges in the target image other than the first connecting edge are regarded as non-linear edges. Any non-linear edge is encoded based on the chain code method to obtain a chain code encoding sequence. When there is a local chain code group composed of at least two continuously distributed chain codes in the chain code encoding sequence and the number of consecutive occurrences of the local chain code group is greater than a preset regularity threshold, the non-linear edge is determined as the second connecting edge.

[0024] The target location and flow direction of the target gate are analyzed to identify molten metal confluence points, thus determining a set of target confluence points. Abnormal connection points are extracted from the first and second connecting edges to identify these points. The target image is then identified as an initial defect region. Based on the target confluence point set, abnormal connection points, and the initial defect region, initial seed points are determined. The target image is divided into several image blocks, and the integrity coefficient of each block is calculated. Based on the abnormal connection points and the initial defect region, the defect confidence score of each pixel within the initial defect region is calculated. Based on the initial seed points, the grayscale values, gradient direction angles, and defect confidence scores of the pixels within the initial defect region, the initial defect region is segmented using region growth to determine the target defect region. An image of standard hot work die steel from the same perspective is acquired, and the local features of the target defect region are used to verify whether it is a true defect region. If the verification result is negative, qualified hot work die steel is obtained.

[0025] Preferably, based on the target set of convergence points, anomalous connection points, and initial defect regions, the initial seed point is determined, including:

[0026] If there is a target convergence point in the target convergence point set that is located within the initial defect region, then the target convergence point located within the initial defect region is determined as the initial seed point;

[0027] If there are no target convergence points located within the initial defect region in the target convergence point set, then for each target convergence point in the target convergence point set, the angle formed by the two target lines intersecting at that target convergence point is determined as the target angle, and the angle bisector of the target angle is determined as the target feature line. All target feature lines form the target feature line set. The pixels where the target feature line set intersects with the initial defect region, as well as the abnormal connection points, are jointly determined as candidate seed points, forming the candidate seed point set.

[0028] For each candidate seed point in the candidate seed point set, select two adjacent candidate seed points that are on the same target feature line as the candidate seed point, and use them as the first reference point and the second reference point, respectively. Calculate the distribution feature value of the candidate seed point based on the gray values ​​of the candidate seed point, the first reference point, and the second reference point. Count the number of pixels in the initial defect area whose distribution feature value is the same as that of the candidate seed point, and use this as the feature quantity. Calculate the Euclidean distance between the candidate seed point and each target merging point in the target merging point set, select the smallest Euclidean distance as the first distance, and select the target merging point closest to the candidate seed point as the reference merging point. Calculate the Euclidean distance between each target merging point in the target merging point set and each pixel in the initial defect area, and select the largest Euclidean distance as the reference distance. Calculate the target optimization value of the candidate seed point based on the reference distance, the gray value of the reference merging point, the feature quantity, the first distance, and the gray value of the candidate seed point. Select the candidate seed point with the largest target optimization value from the candidate seed point set, and use it as the initial seed point.

[0029] Compared with existing technologies, this invention has the following advantages: According to this manufacturing method, 4Cr5MoSiV1(H13) is smelted and its composition is precisely controlled, with the Mo content controlled above 1.35%. After vacuum degassing, magnesium-core wire is used for treatment, thereby improving the high-temperature strength and carbide characteristics of the mold steel. 13.5-ton and 20-ton large steel ingots are cast and hot-sent for high-temperature homogenization heating, optimizing the forging process, increasing the upsetting and refining operation requirements, reducing the central porosity during solidification of the large steel ingots, and further improving the segregation of liquid carbides and microstructure. After forging into Ø360-700mm forging bars, post-forging air cooling is performed. When cooled to 300-350℃, a solution + spheroidizing annealing heat treatment process for ultra-refining microstructure is used, thereby improving the uniformity of the microstructure. High-magnification and mechanical analysis revealed that the microstructure uniformity rating was no greater than AS9, with no large or chain-like liquid carbides observed. The average unnotched impact value was approximately 260 J, indicating good toughness, and the flaw detection was satisfactory. Dies manufactured from this forged bar after machining and heat treatment were used in 5000-ton extrusion presses and above to produce 6-series aluminum alloy profiles. No die cracking or premature failure was observed, and the service life was comparable to that of electroslag 4Cr5MoSiV1(H13), significantly reducing manufacturing costs and demonstrating good performance. Attached Figure Description

[0030] Figure 1 This study describes the cracking phenomenon observed during the modification of forged bars using the existing electric furnace 4Cr5MoSiV1(H13) method.

[0031] Figure 2 This is a schematic diagram showing the failure of flaw detection in 4Cr5MoSiV1 (H13) forged bars produced using existing electric furnace casting and electroslag remelting techniques.

[0032] Figure 3 The forging process for 20-ton H13 steel ingots.

[0033] Figure 4 Water mist cooling process for refining the microstructure of H13 forgings.

[0034] Figure 5 It is a large-diameter H13 forged bar with a diameter of Ø500mm. Detailed Implementation

[0035] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] Example 1: One batch of electric furnace steel 4Cr5MoSiV1 (H13) forging bars with a diameter of Ø400-500mm was produced according to this manufacturing method. The batch number was 465B0586. The chemical composition is shown in Table 1. The details of the qualified forging bars and their warehousing are shown in Table 2. Some production process photos are as follows. Figure 3-5 :

[0037] Table 1. Composition of H13 electric arc furnace steel (furnace number 465B0586)

[0038]

[0039] Table 2 Production details of H13 forged bars for large-scale electric furnaces (furnace number 465B0586)

[0040]

[0041] The specific method is as follows: (1) Use an EAF furnace to perform primary refining on the mold steel raw material;

[0042] (2) The metal raw materials obtained from the primary refining are then refined in an LF furnace. Lime is added during the refining process to form slag, and then deoxidation is carried out.

[0043] (3) After refining in step (2), the steel material is then degassed in a VD furnace under vacuum, and magnesium cored wire is used to improve the carbide content.

[0044] (4) The molten steel obtained in step (3) is poured into 13t steel ingots. If nodules are found at the sprue, they are removed immediately. Protective slag is used and the ingots are hung up. A heating agent is used during pouring. After heating, the ingots are demolded. After demolding, the ingots are hot-sent to the forging area for heating or hot-loading annealing.

[0045] (5) Forging: Forging is carried out by programmed heating. The temperature in the first stage is 650℃, the temperature in the second stage is maintained at 650℃, the temperature in the third stage is raised to 840℃, the temperature in the fourth stage is maintained at 840℃, and the temperature in the final stage is 1235℃.

[0046] (6) Heat treatment: After forging, air cool to 200°C on the surface, and then solution treatment and spheroidizing annealing.

[0047] Furthermore, in step (1), the raw materials for the mold are selected from scrap steel with a phosphorus mass percentage of less than 0.025% and a carbon mass percentage of less than 0.37%, H13 or CrMoV mold steel cut-offs and high-quality pig iron, and the tapping temperature is ≥1610℃.

[0048] Furthermore, in step (2), the LF heating temperature is 1670℃, lime is added twice, each time 100kg of lime is added, deoxidation is carried out using SiFe powder and C powder, the amount of SiFe powder and C powder added is 20kg, the Ar gas pressure is controlled to ensure that the molten steel does not overflow the slag surface, the mass percentage of aluminum in the mold raw materials is controlled at 0.02%, after refining, the slag with a mass percentage of 30% of the steel ladle is poured off, and the temperature of the VD ladle is 1610℃.

[0049] Furthermore, in step (3), the ladle pumping time is ≤8min, the vacuum degree reaches 66.7Pa, and the time is ≥20min. Under vacuum conditions, the Ar pressure is maintained at 0.20Mpa. The magnesium-coated wire has a magnesium mass percentage of 9%, an aluminum mass percentage of 29%, and the remainder is iron. The magnesium-coated wire is fed in two batches, 60 meters each time, with a 30-second pause in between. The feeding speed is 150m / min, and the bottom blowing argon weak stirring time must be ≥10 minutes. The liquidus temperature of the steel is approximately 1483℃, and the ladle temperature is 1520℃.

[0050] Furthermore, in step (4), argon gas is used for protection during the casting of the steel ingot, and the argon gas flow rate is controlled at 2 Nm. 3 / h, the exothermic agent used is (SiO2-CaO-Al2O3-Na2O-K2O), the demolding time for 13t is ≥9 hours, and the demolding time for 20t steel ingot is ≥12 hours.

[0051] Furthermore, in step (5), the first stage temperature is 650℃ and maintained for 1.5h, then the temperature is raised to 840℃ after 2h and maintained for 1h, then the temperature is raised to 1245℃ after 5h and maintained for 15h. The starting temperature is ≥1100℃, the stopping temperature is ≥850℃, the forging ratio is ≥5, and the electric furnace ingot adopts a 2-upsetting and 2-drawing process. The first upsetting should be upsetting to less than 40% of the original ingot height to ensure internal forging. After the first forging is completed, the heating temperature of the remaining forgings is reduced to 1200±10℃ and held for 2h. After forging and upsetting, the ingot is widened, and the deformation of the last forging is ≥40%. The defects at the head and tail are cleaned by hot cutting after forging.

[0052] Furthermore, in step (6), the solution treatment process involves maintaining a temperature of 350°C for 4 hours, raising the temperature to 830°C at a rate of ≤80°C / h, maintaining the temperature for 3 hours, raising the temperature to 1030°C at a rate of ≤80°C / h, maintaining the temperature for 10 hours, finally cooling the surface to 300°C with water mist, and then continuing to air cool the surface to 200°C.

[0053] Furthermore, in step (6), the spheroidizing operation is carried out at 600℃ for 2 hours, then the temperature is increased to 870℃ at a rate of ≤30℃ / h. At this time, the amount of spheroidization treatment M is M=3+0.2Q+2D, where D is the diameter of the round steel or the thickness of the flat steel in decimeters, and Q is the actual amount of steel charged into the furnace in tons. Then the temperature is decreased to 720℃ at a rate of ≤30℃ / h. The process is stopped when the amount of spheroidization reaches 10D. Finally, the temperature is decreased to 350℃ at a rate of ≤40℃ / h before the steel is removed from the furnace. The manufacturing process is as follows: Figure 3-5 As shown.

[0054] After final annealing and finishing, flaw detection, sampling, sawing, peeling, grinding, etc. are carried out according to contract requirements. All products pass inspection and are then put into storage.

[0055] Example 2: One batch of electric furnace steel 4Cr5MoSiV1 (H13) forging bars with a diameter of Ø450-700mm was produced according to this manufacturing method. The batch number was 365B0197. The chemical composition is shown in Table 3. The details of the qualified forging bars and their warehousing are shown in Table 4.

[0056] Table 3 Composition of H13 electric arc furnace steel (furnace number 061C0241)

[0057]

[0058] Table 4 Production details of H13 forged bars for large-scale electric furnaces (furnace number 061C0241)

[0059]

[0060] The specific preparation method is as follows: (1) The mold steel raw material is initially smelted using an EAF furnace;

[0061] (2) The metal raw materials obtained from the primary refining are then refined in an LF furnace. Lime is added during the refining process to form slag, and then deoxidation is carried out.

[0062] (3) After refining in step (2), the steel material is then degassed in a VD furnace under vacuum, and magnesium cored wire is used to improve the carbide content.

[0063] (4) The molten steel obtained in step (3) is poured into 20t steel ingots. If nodules are found at the sprue, they are removed immediately. Protective slag is used and the ingots are hung up. After pouring, a heating agent is used. After the mold cools, the ingots are demolded. After demolding, the steel ingots are hot-sent to the forging area for heating or hot-loading annealing.

[0064] (5) Forging: Forging is carried out by programmed heating. The temperature in the first stage is 700℃, the temperature in the second stage is maintained at 700℃, the temperature in the third stage is raised to 860℃, the temperature in the fourth stage is maintained at 860℃, and the temperature in the final stage is 1265℃.

[0065] (6) Heat treatment: After forging, air cool to 350°C on the surface, and then solution treatment and spheroidizing annealing.

[0066] Furthermore, in step (1), the raw materials for the mold are selected from scrap steel with a phosphorus mass percentage of less than 0.025% and a carbon mass percentage of less than 0.37%, H13 or CrMoV mold steel cut-offs and high-quality pig iron, and the tapping temperature is ≥1610℃.

[0067] Furthermore, in step (2), the LF heating temperature is 1700℃, lime is added 3 times, and the amount of lime added each time is 300kg. Deoxidation is carried out using SiFe powder and C powder, with an addition amount of 40kg of SiFe powder and C powder. The Ar gas pressure is controlled to ensure that the molten steel does not overflow the slag surface. The mass percentage of aluminum in the mold raw materials is controlled at 0.05%. After refining, the slag with a mass percentage of 50% of the steel ladle is poured off, and the temperature of the VD ladle is 1640℃.

[0068] Furthermore, in step (3), the ladle pumping time is ≤8min, the vacuum degree reaches 66.7Pa for ≥20min, the Ar pressure is maintained at 1.00MPa under vacuum conditions, the magnesium-coated wire has a magnesium mass percentage of 9%, an aluminum mass percentage of 29%, and the remainder is iron. The magnesium-coated wire is fed in two batches, 60 meters each time, with a 30-second pause in between, at a feeding speed of 150m / min, and the bottom-blowing argon weak stirring time must be ≥10 minutes. The liquidus temperature of the steel is approximately 1483℃, and the ladle temperature is 1540℃.

[0069] Furthermore, in step (4), argon gas is used for protection during the casting of the steel ingot, and the flow rate of argon gas is controlled at 5 Nm. 3 / h, the exothermic agent used is (SiO2-CaO-Al2O3-Na2O-K2O), and the demolding time of 20t steel ingot is ≥12 hours.

[0070] Furthermore, in step (5), the first stage temperature is 700℃ and maintained for 1.5h, then the temperature is raised to 860℃ after 3h and maintained for 2h, then the temperature is raised to 1265℃ after 6h and maintained for 28h. The starting temperature is ≥1100℃, the stopping temperature is ≥850℃, the forging ratio is ≥5, and the electric furnace ingot adopts a 2-upsetting and 2-drawing process. The first upsetting should be upsetting to less than 40% of the original ingot height to ensure internal forging. After the first forging is completed, the heating temperature of the remaining forgings is reduced to 1200±10℃ and held for 2.5h. After forging and upsetting, the ingot is widened, and the deformation of the last forging is ≥40%. The defects at the head and tail are cleaned by hot cutting after forging.

[0071] Furthermore, in step (6), the solution treatment process involves maintaining a temperature of 450°C for 5 hours, raising the temperature to 870°C at a rate of ≤80°C / h, maintaining the temperature for 3 hours, raising the temperature to 1060°C at a rate of ≤80°C / h, maintaining the temperature for 10 hours, and finally cooling the surface to 300°C with water mist, followed by air cooling to 300°C.

[0072] Furthermore, in step (6), the spheroidizing operation is to maintain the temperature at 640℃ for 2 hours, and then raise the temperature to 890℃ at a rate of ≤30℃ / h. At this time, the amount of spheroidizing treatment M is M=3+0.2Q+2D, where D is the diameter of the round steel or the thickness of the flat steel in decimeters, and Q is the actual amount of steel loaded into the furnace in tons. Then, the temperature is lowered to 740℃ at a rate of ≤30℃ / h. The spheroidizing is stopped when the amount reaches 10D. Finally, the temperature is lowered to 350℃ at a rate of ≤40℃ / h before the steel is removed from the furnace.

[0073] After final annealing and finishing, flaw detection, sampling, sawing, peeling, grinding, etc. are carried out according to contract requirements. All products pass inspection and are then put into storage.

[0074] Example 3: One batch of electric furnace steel 4Cr5MoSiV1 (H13) forging bars with a diameter of Ø360-400mm was produced according to this manufacturing method. The batch number was 365B0197. The chemical composition is shown in Table 5. The details of the qualified forging bars and their warehousing are shown in Table 6.

[0075] Table 5 Composition of H13 electric arc furnace steel (furnace number 365B0197)

[0076]

[0077] Table 6 Production details of H13 forged bars for large-scale electric furnaces (furnace number 365B0197)

[0078]

[0079] The specific preparation method is as follows: (1) The mold steel raw material is initially smelted using an EAF furnace;

[0080] (2) The metal raw materials obtained from the primary refining are then refined in an LF furnace. Lime is added during the refining process to form slag, and then deoxidation is carried out.

[0081] (3) After refining in step (2), the steel material is then degassed in a VD furnace under vacuum, and magnesium cored wire is used to improve the carbide content.

[0082] (4) The molten steel obtained in step (3) is poured into 13t steel ingots. If nodules are found at the sprue, they are removed immediately. Protective slag is used and the ingots are hung up. After pouring, a heating agent is used. After the mold cools, the ingots are demolded. After demolding, the steel ingots are hot-sent to the forging area for heating or hot-loading annealing.

[0083] (5) Forging: Forging is carried out by programmed heating. The temperature in the first stage is 670℃, the temperature in the second stage is maintained at 670℃, the temperature in the third stage is raised to 850℃, the temperature in the fourth stage is maintained at 850℃, and the temperature in the final stage is 1250℃.

[0084] (6) Heat treatment: After forging, air cool to 325°C on the surface, and then solution treatment and spheroidizing annealing.

[0085] Furthermore, in step (1), the raw materials for the mold are selected from scrap steel with a phosphorus mass percentage of less than 0.025% and a carbon mass percentage of less than 0.37%, H13 or CrMoV mold steel cut-offs and high-quality pig iron, and the tapping temperature is ≥1610℃.

[0086] Furthermore, in step (2), the LF heating temperature is 1680℃, lime is added twice, and the amount of lime added each time is 200kg. Deoxidation is carried out using SiFe powder and C powder, with an addition amount of 35kg of SiFe powder and C powder. The Ar gas pressure is controlled to ensure that the molten steel does not overflow the slag surface. The mass percentage of aluminum in the mold raw materials is controlled at 0.035%. After refining, the slag with a mass percentage of 40% of the steel ladle is poured off, and the temperature of the VD ladle is 1625℃.

[0087] Furthermore, in step (3), the ladle pumping time is ≤8min, the vacuum degree reaches 66.7Pa, the time is ≥20min, the Ar pressure is maintained at 0.6Mpa under vacuum conditions, the magnesium core wire has a magnesium mass percentage of 9%, an aluminum mass percentage of 29%, and the remainder is iron. The magnesium core wire is fed in two batches, 60 meters each time, with a 30-second pause in between. The feeding speed is 150m / min, and the bottom blowing argon weak stirring time must be ≥10 minutes. The liquidus temperature of the steel is about 1483℃, and the ladle temperature is 1540℃.

[0088] Furthermore, in step (4), argon gas is used for protection during the casting of the steel ingot, and the argon gas flow rate is controlled at 3 Nm. 3 / h, the exothermic agent used is (SiO2-CaO-Al2O3-Na2O-K2O), the demolding time for 13t is ≥9 hours, and the demolding time for 20t steel ingot is ≥12 hours.

[0089] Furthermore, in step (5), the first stage temperature is 670℃ and maintained for 1.5h, then the temperature is raised to 850℃ after 2h and maintained for 1.5h, then the temperature is raised to 1250℃ after 5.5h and maintained for 25h. The starting temperature is ≥1100℃, the stopping temperature is ≥850℃, the forging ratio is ≥5, and the electric furnace ingot adopts a 2-upsetting and 2-drawing process. The first upsetting should be upsetting to less than 40% of the original ingot height to ensure internal forging. After the first forging is completed, the heating temperature of the remaining forgings is reduced to 1200±10℃ and held for 2h. After forging and upsetting, the ingot is widened, and the deformation of the last forging is ≥40%. The defects at the head and tail are cleaned by hot cutting after forging.

[0090] Furthermore, in step (6), the solution treatment process involves maintaining a temperature of 400°C for 4.5 hours, raising the temperature to 850°C at a rate of ≤80°C / h, maintaining the temperature for 3 hours, raising the temperature to 1050°C at a rate of ≤80°C / h, maintaining the temperature for 10 hours, finally cooling the surface to 350°C with water mist, and then continuing to air cool the surface to 230°C.

[0091] Furthermore, in step (6), the spheroidizing operation is to maintain the temperature at 625℃ for 2 hours, and then raise the temperature to 880℃ at a rate of ≤30℃ / h. At this time, the amount of spheroidizing treatment M is M=3+0.2Q+2D, where D is the diameter of the round steel or the thickness of the flat steel in decimeters, and Q is the actual amount of steel loaded into the furnace in tons. Then, the temperature is lowered to 730℃ at a rate of ≤30℃ / h. The spheroidizing is stopped when the amount reaches 10D. Finally, the temperature is lowered to 350℃ at a rate of ≤40℃ / h before the steel is removed from the furnace.

[0092] The large-diameter 4Cr5MoSiV1 (H13) forged bars in the above examples underwent multiple sampling inspections, and all indicators met the standard requirements. The microstructure was relatively uniform, and according to the North American Die Casting Association (NADCA) 207-2003 standard chart, it was grade AS5-AS9, with an unnotched impact value of 240-280J. The material passed flaw detection. The material can be directly used for dies in aluminum alloy extrusion, eliminating the need for reforging, which has been well-received by users. Compared to electroslag 4Cr5MoSiV1 (H13) steel, the cost is reduced by approximately 3000 RMB / ton, giving this product strong market competitiveness.

[0093] Example 4: The manufacturing method of Example 1 further includes, after the hot work die steel is manufactured, performing defect detection on the hot work die steel, and obtaining qualified hot work die steel when the detection results meet the requirements.

[0094] The defect detection of hot work die steel includes:

[0095] Select any hot work die steel as the hot work die steel to be tested; acquire images of the hot work die steel under different perspectives, and take the image under each perspective as the target image. The target image is marked with the target position and target flow direction corresponding to each target gate in the target gate set of the hot work die steel.

[0096] Based on Hough line detection, straight line edges in the target image are obtained, and straight line edges with a length greater than a preset length threshold are determined as the first connected edges.

[0097] The edges in the target image other than the first connecting edge are regarded as non-linear edges. Any non-linear edge is encoded based on the chain code method to obtain a chain code encoding sequence. When there is a local chain code group composed of at least two continuously distributed chain codes in the chain code encoding sequence and the number of consecutive occurrences of the local chain code group is greater than a preset regularity threshold, the non-linear edge is determined as the second connecting edge.

[0098] The target location and flow direction of the target gate are analyzed to identify molten metal confluence points, thus determining a set of target confluence points. Abnormal connection points are extracted from the first and second connecting edges to identify these points. The target image is then identified as an initial defect region. Based on the target confluence point set, abnormal connection points, and the initial defect region, initial seed points are determined. The target image is divided into several image blocks, and the integrity coefficient of each block is calculated. Based on the abnormal connection points and the initial defect region, the defect confidence score of each pixel within the initial defect region is calculated. Based on the initial seed points, the grayscale values, gradient direction angles, and defect confidence scores of the pixels within the initial defect region, the initial defect region is segmented using region growth to determine the target defect region. An image of standard hot work die steel from the same perspective is acquired, and the local features of the target defect region are used to verify whether it is a true defect region. If the verification result is negative, qualified hot work die steel is obtained.

[0099] In this embodiment, the target position and target flow direction of the target gate are analyzed to determine the molten metal confluence point and the target confluence point set is determined. This includes: determining the straight line that passes through the target position of each target gate and whose direction is consistent with the target flow direction of the target gate as the target straight line, and the intersection of all the target straight lines forms the target confluence point set.

[0100] In this embodiment, abnormal connection points are extracted from the first and second connecting edges to determine abnormal connection points. This includes: for each connecting pixel on the first and second connecting edges, a preset neighborhood window is constructed centered on the connecting pixel; the preset neighborhood window is divided into two regions by the tangent of the connecting edge where the connecting pixel is located; the gray-level mean and gray-level dispersion of each region are calculated, and the gray-level distribution feature value of each region is obtained based on the gray-level mean and gray-level dispersion; the difference between the gray-level distribution feature value of the current connecting pixel and the previous connecting pixel is normalized to obtain the neighborhood abnormal value of the current connecting pixel; connecting pixels with neighborhood abnormal values ​​greater than a preset first abnormal threshold are selected as abnormal connection points; wherein, the gray-level distribution feature value is directly proportional to the gray-level mean and inversely proportional to the gray-level dispersion.

[0101] In this embodiment, the initial defect region identification of the target image is determined as the initial defect region, including: calculating the grayscale histogram of the target image and using the grayscale histogram as the target histogram; obtaining the number of pixels in the interval corresponding to each peak in the target histogram, and selecting the peak with the most pixels in the interval as the large peak; determining the pixels in the intervals corresponding to other peaks in the target histogram (excluding the large peak) as candidate pixels; and determining the area formed by the candidate pixels in the target image as the initial defect region.

[0102] In this embodiment, the target image is divided into several image blocks, and the integrity coefficient of each image block is calculated, including: obtaining the edge lines within each image block; for each edge line within each image block, taking one endpoint pixel of the edge line as the analysis pixel and as the growth point for region growth, recording the number of growths as the feature expansion value, calculating the standard deviation of the expansion angle during the growth process as the angle discrete value, and calculating the edge continuity based on the feature expansion value and the angle discrete value.

[0103] For each edge line within each image block, the number of corner points is obtained through corner detection as a bending feature value, and the gradient direction and gradient value of each pixel on the edge line are obtained; the edge jaggedness is calculated based on the bending feature value, the difference in gradient direction and gradient value between adjacent pixels;

[0104] Calculate the difference between the maximum and minimum gray values ​​within each image block, excluding edge pixels, and combine this with the variance of gray values ​​of all pixels within the image block to calculate the surface uniformity.

[0105] The mean edge continuity of all edge lines within each image patch is used as the edge comprehensive continuity value, and the mean edge asymmetry of all edge lines is used as the edge comprehensive asymmetry value. The integrity coefficient of each image patch is calculated by combining the edge comprehensive continuity value, the edge comprehensive asymmetry value, and the surface asymmetry. Among them, the feature extension value is positively correlated with the edge continuity, and the angular discrepancy value is negatively correlated with the edge continuity.

[0106] In this embodiment, based on the abnormal connection points and the initial defect region, the defect confidence score of each pixel within the initial defect region is calculated, including:

[0107] The region formed by the preset neighborhood window of the abnormal connection point and the preset neighborhood window of the previous connected pixel of the abnormal connection point is used as the reference neighborhood. The gray average value of all pixels in the reference neighborhood is calculated as the target gray standard value. The gray average value of all pixels in the initial defect region is calculated as the abnormal gray standard value.

[0108] For each pixel within the initial defect area, the absolute value of the difference between the pixel's grayscale value and the target grayscale standard value is calculated as the reference grayscale difference, and the absolute value of the difference between the abnormal grayscale standard value and the target grayscale standard value is calculated as the overall grayscale difference. The relative defect weight of the pixel is calculated based on the reference grayscale difference and the overall grayscale difference. Among them, the reference grayscale difference is positively correlated with the relative defect weight, and the overall grayscale difference is negatively correlated with the relative defect weight.

[0109] The ratio of the area of ​​the initial defect region to the area of ​​the minimum bounding rectangle is used as the rectangular regularity, and the circularity of the initial defect region is used as the circularity regularity. The edge regularity is calculated based on the rectangular regularity and the circularity regularity. Both the rectangular regularity and the circularity regularity are positively correlated with the edge regularity.

[0110] The defect confidence of a pixel is calculated based on the relative defect weight and edge regularity; where the relative defect weight is positively correlated with the defect confidence and the edge regularity is negatively correlated with the defect confidence.

[0111] In this embodiment, based on the initial seed point, the grayscale values ​​of pixels within the initial defect region, the gradient direction angle, and the defect confidence level, region growing and segmentation are performed on the initial defect region to determine the target defect region, including:

[0112] A preset target neighborhood is constructed with the initial seed point as the center. The absolute value of the gray level difference between each pixel in the target neighborhood and the initial seed point is calculated as the first gray level difference. The fifth gray level difference is calculated by combining the gradient direction angle difference between the pixel and the initial seed point. The product of the first gray level difference and the fifth gray level difference is used as the target gray level difference of the pixel.

[0113] Pixels with a target grayscale difference less than a preset grayscale difference threshold and a defect confidence level greater than a preset confidence threshold are selected as new seed points; the calculation is repeated for the new seed points until no new seed points are generated; the region composed of all seed points is determined as the target defect region.

[0114] In this embodiment, an image of standard hot work die steel from the same viewing angle is acquired, and the target defect area is verified as a real defect area based on local features of the target defect area. If the verification result is negative, qualified hot work die steel is obtained, including:

[0115] Obtain images of standard hot work die steel from the same viewpoint to obtain a standard image;

[0116] The target defect region is mapped to the standard image using a feature matching algorithm to obtain the corresponding target defect region in the standard image. Specifically, the BRISK algorithm is used to extract feature points from the target image and the standard image. Based on the feature point matching relationship, the projection position of the target defect region in the standard image is determined to obtain the corresponding region in the standard image.

[0117] Calculate the local feature difference values ​​between the target defect area and the corresponding area in the standard image. The local feature difference values ​​include the weighted sum of gray-level distribution differences, edge continuity differences, and surface unevenness differences.

[0118] If the local feature difference value is less than the preset difference threshold, the target defect area is confirmed to be a hot work die steel defect area; otherwise, it is confirmed as the target defect area.

[0119] If the target defect area is confirmed to be not a real defect area from all perspectives, then the hot work die steel to be tested is qualified hot work die steel.

[0120] The working principle and beneficial effects of the above technical solution are as follows: By acquiring images of the hot work die steel to be inspected from different perspectives and performing detailed analysis on each perspective, various features and potential defects on the surface of the hot work die steel can be comprehensively captured; images from different perspectives can provide richer information, avoiding the omission of certain defects due to the limitations of a single perspective, thereby improving the accuracy of defect detection; using Hough line detection to determine the first connecting edge and using chain code to encode non-linear edges to determine the second connecting edge, this combination of multiple edge detection methods can more accurately identify the edge features of the hot work die steel surface; different types of edges may correspond to different defect situations, and comprehensively considering straight edges and non-linear edges can more comprehensively discover potential defects; comprehensively considering multiple factors such as the target confluence point set, abnormal connection points, and initial defect areas to determine the initial seed point, and then based on... The target defect region is determined by region growing and segmentation based on the initial seed point, the grayscale value of pixels within the initial defect region, the gradient direction angle, and the defect confidence score. This multi-factor comprehensive analysis method can more accurately locate and identify defect regions, reducing false positives and false negatives. Images of standard hot work die steel from the same viewpoint are acquired, and the target defect region is verified as a true defect region based on its local features. This verification mechanism can further eliminate false detections and ensure the reliability of the detection results. Only verified defect regions are identified as true defect regions, thereby improving the credibility of the detection results. The entire defect detection process employs a series of automated algorithms and technologies, such as Hough line detection, chain code encoding, and region growing and segmentation. These automated detection methods can quickly inspect hot work die steel, reducing the time and workload of manual inspection. In large-scale production, this can significantly improve production efficiency and reduce production costs.

[0121] Example 5: The manufacturing method of Example 2 further includes determining an initial seed point based on the target confluence point set, abnormal connection points, and initial defect region, including:

[0122] If there is a target convergence point in the target convergence point set that is located within the initial defect region, then the target convergence point located within the initial defect region is determined as the initial seed point;

[0123] If there are no target convergence points located within the initial defect region in the target convergence point set, then for each target convergence point in the target convergence point set, the angle formed by the two target lines intersecting at that target convergence point is determined as the target angle, and the angle bisector of the target angle is determined as the target feature line. All target feature lines form the target feature line set. The pixels where the target feature line set intersects with the initial defect region, as well as the abnormal connection points, are jointly determined as candidate seed points, forming the candidate seed point set.

[0124] For each candidate seed point in the candidate seed point set, select two adjacent candidate seed points that are on the same target feature line as the candidate seed point, and use them as the first reference point and the second reference point, respectively. Calculate the distribution feature value of the candidate seed point based on the gray values ​​of the candidate seed point, the first reference point, and the second reference point. Count the number of pixels in the initial defect area whose distribution feature value is the same as that of the candidate seed point, and use this as the feature quantity. Calculate the Euclidean distance between the candidate seed point and each target merging point in the target merging point set, select the smallest Euclidean distance as the first distance, and select the target merging point closest to the candidate seed point as the reference merging point. Calculate the Euclidean distance between each target merging point in the target merging point set and each pixel in the initial defect area, and select the largest Euclidean distance as the reference distance. Calculate the target optimization value of the candidate seed point based on the reference distance, the gray value of the reference merging point, the feature quantity, the first distance, and the gray value of the candidate seed point. Select the candidate seed point with the largest target optimization value from the candidate seed point set, and use it as the initial seed point.

[0125] The working principle and beneficial effects of the above technical solution are as follows: Firstly, target convergence points located within the initial defect region are identified as initial seed points, because these convergence points are key locations for molten metal convergence and are likely related to defect generation. When these convergence points are located within the initial defect region, their role as initial seed points more accurately reflects the core location of the defect, providing a precise starting point for subsequent region growth and segmentation, and helping to more accurately segment the target defect region. When no target convergence point is located within the initial defect region in the target convergence point set, the pixels intersecting the target feature line set with the initial defect region, along with abnormal connection points, are jointly identified as candidate seed points by calculating the target angle and target feature lines. This multi-factor comprehensive approach fully utilizes the physical properties of hot work die steel (such as the flow and convergence of molten metal) and the geometric features of the image (such as edge connection points), expanding the possible range of seed points and increasing the probability of finding seed points that truly reflect the starting location of the defect. For each candidate seed point in the candidate seed point set, its distribution characteristic value is calculated, and the number of pixels with the same distribution characteristic value within the initial defect region is counted. The distribution feature value reflects the gray-level distribution of pixels surrounding the candidate seed point, while the feature quantity reflects the similarity between the candidate seed point and its surrounding pixels. This method can filter out candidate seed points that are representative in terms of gray-level distribution, making the final determined initial seed points more representative of the defect region's features and improving the reliability of region growth segmentation. The Euclidean distance and reference distance between the candidate seed point and the target merging point are calculated, and the target optimization value is calculated by combining the gray-level values ​​of the reference merging point and the candidate seed point. The Euclidean distance reflects the spatial relationship between the candidate seed point and the target merging point, while the gray-level value reflects the brightness characteristics of the pixel. By comprehensively considering these information to calculate the target optimization value, multiple factors such as spatial location and gray-level distribution can be taken into account to select the most representative and advantageous initial seed point from the candidate seed point set, further improving the quality of the initial seed point and the accuracy of defect detection.

[0126] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for manufacturing high-quality electric furnace hot work die steel, characterized in that, The method includes the following steps: (1) using an EAF furnace to perform primary refining of the mold steel raw material; (2) The metal raw materials obtained from the primary refining are then refined in an LF furnace. Lime is added during the refining process to form slag, and then deoxidation is carried out. (3) After refining in step (2), the steel material is then degassed in a VD furnace under vacuum, and magnesium cored wire is used to improve the carbide content. (4) The molten steel obtained in step (3) is poured into 13t or 20t steel ingots. If nodules are found at the sprue, they are removed immediately. Protective slag is used and the ingots are hung up. After pouring, a heating agent is used. After the mold cools, the ingots are demolded. After demolding, the steel ingots are hot-sent to the forging area for heating or hot-charging annealing. (5) Forging: Forging is carried out by programmed heating. The temperature of the first stage is 650-700℃, the temperature of the second stage is maintained at 650-700℃, the temperature of the third stage is raised to 840-860℃, the temperature of the fourth stage is maintained at 840-860℃, and the temperature of the final stage is 1235-1265℃. (6) Heat treatment: After forging, air cool to 200-350℃ on the surface, and then solution treatment and spheroidizing annealing are performed; In step (5), the first stage temperature is 650-700℃ and maintained for 1.5h. After 2-3h, the temperature is raised to 840-860℃ and maintained for 1-2h. After 5-6h, the temperature is raised to 1235-1265℃ and maintained for 15-28h. The starting temperature of forging is ≥1100℃, the stopping temperature of forging is ≥850℃, the forging ratio is ≥5, and the electric furnace ingot adopts a 2-upsetting and 2-drawing process. The first upsetting should be upsetting to less than 40% of the original ingot height to ensure internal forging. After the first forging is completed, the heating temperature of the remaining forgings is reduced to 1200±10℃ and held for ≥2h. After forging and upsetting, the ingot is widened, and the deformation of the last forging is ≥40%. The defects at the head and tail are cleaned by hot cutting after forging. The solution treatment process in step (6) involves maintaining a temperature of 350-450℃ for 4-5 hours, raising the temperature to 830-870℃ at a rate of ≤80℃ / h, maintaining it for 3 hours, raising the temperature to 1030-1060℃ at a rate of ≤80℃ / h, maintaining it for 10 hours, and finally cooling the surface with water mist to 300-400℃, followed by air cooling to 200-300℃. The mold steel is H13 steel.

2. The method according to claim 1, characterized in that, In step (1), the raw materials for the mold are selected from scrap steel with a phosphorus mass percentage of less than 0.025% and a carbon mass percentage of less than 0.37%, H13 or CrMoV mold steel cut-offs and high-quality pig iron, with a tapping temperature ≥1610℃.

3. The method according to claim 1, characterized in that, In step (2), the LF heating temperature is 1670-1700℃, lime is added 2-3 times, and the amount of lime added each time is 100-300kg. Deoxidation is carried out using SiFe powder and C powder, with an addition amount of 20-40kg of SiFe powder and C powder. The Ar gas pressure is controlled so that the molten steel does not overflow the slag surface. The mass percentage of aluminum in the mold raw materials is controlled at 0.02-0.05%. After refining, the slag with a mass percentage of 30-50% of the steel ladle is poured off. The temperature of the VD ladle is 1610-1640℃.

4. The method according to claim 1, characterized in that, In step (3), the ladle pumping time is ≤8min, the vacuum degree reaches 66.7Pa and the time is ≥20min, the Ar pressure is maintained at 0.20-1.00Mpa under vacuum conditions, the bottom blowing argon weak stirring time must be ≥10 minutes, the liquidus of the steel is 1483℃, and the ladle temperature is 1520~1540℃.

5. The method according to claim 1, characterized in that, In step (4), argon gas is used for protection during the casting of the steel ingot, and the flow rate of argon gas is controlled at 2 Nm. 3 / h-5Nm 3 / h, the exothermic agent used is SiO2-CaO-Al2O3-Na2O-K2O, the demolding time for 13t is ≥9 hours, and the demolding time for 20t steel ingot is ≥12 hours.

Images