JP7716334B2 - Aluminum clad material manufacturing method - Google Patents

Description

The present invention relates to a method for manufacturing aluminum clad materials.

Brazing sheets, which are used as a structural material for heat exchangers and are an example of aluminum clad materials, are constructed by laminating skin materials to one or both sides of a core material made of aluminum or an aluminum alloy. The core material is obtained by cutting a slab ingot made of aluminum or an aluminum alloy to a specified thickness and shape. The skin material is obtained by hot-rolling a slab ingot to a specified thickness and cutting it to an appropriate length. Examples of skin materials include brazing filler metals and sacrificial anode materials.
When producing this type of aluminum clad material, an aluminum alloy skin material having a different composition from the core material is integrated with the core material by rolling.
When producing aluminum clad materials by rolling, if the skin material is more elongated than the core material, the skin material will protrude from the end of the core material during rolling. If this protruding portion is long, it may fall off into the conveying path of the rolling mill, hindering production.

In the hot rolling of clad materials, the initial stage is a joining process to join the interface between the overlapping core and skin materials, and then the process moves to hot rolling to reduce the plate thickness. Until the interface is joined, the core and skin materials have little effect in restricting each other's deformation, so differences in strength are likely to result in differences in elongation.
As a result, the low-strength skin material, which has elongated more than the high-strength core material, protrudes from the end face of the core material in the rolling direction, which can cause the protruding skin material to fall off during the subsequent hot rolling process to reduce the plate thickness, potentially resulting in production problems.

For example, if the strength of the skin is lower than that of the core, the skin will stretch more than the core. At this time, the materials slide (shift) at the unbonded interface, causing the skin to protrude from the core. This is particularly likely at the front and rear ends in the rolling direction, and the protruding material may fall off from the point where it hits the edge of the core, hindering production.
Furthermore, if the low-strength cladding material elongates more than the core material, a large difference in the amount of elongation in the rolling direction between the cladding material and the core material will occur. The difference in the amount of elongation between the cladding material and the core material will directly result in a difference in the amount of change in the thickness of each layer. In other words, if the difference in elongation between the cladding material and the core material is large, a change in the cladding ratio will occur, and the desired cladding ratio will not be obtained.
As mentioned above, in low-strength skin materials, the front and rear ends in the rolling direction tend to protrude, which makes the change in cladding ratio at the front and rear ends particularly large. As a result, it becomes necessary to remove the front and rear ends that do not achieve the desired cladding ratio, which significantly impedes productivity.

The technologies described in the following Patent Documents 1 and 2 disclose a method of preventing misalignment between the skin and core materials by placing a mesh-like object or foil at the interface between the two.

Japanese Patent Application Laid-Open No. 2015-193012 Japanese Patent Application Laid-Open No. 2015-217403

However, the techniques described in Patent Documents 1 and 2 have the problem that the mesh or foil placed at the interface becomes an inclusion, causing a quality defect in the clad material, making them inapplicable to the production of high-quality clad materials that have been developed in recent years. Furthermore, there are concerns about the labor and cost involved in separately preparing the mesh or sheet.
Another known method is to weld the outer periphery of the interface between the core material and the skin material to suppress misalignment and differences in elongation at the interface, but this incurs costs for the welding process.

The present invention aims to provide a method for manufacturing aluminum clad material that can prevent skin material from extending and protruding from the core end during clad material manufacturing, without the need for welding or other methods, and can prevent skin material from falling off the core end. The present invention also aims to provide technology that can efficiently manufacture aluminum clad material with an appropriate clad ratio by reducing the amount of skin material that protrudes from the core end, thereby reducing the amount of skin material that needs to be discarded.

(1) The method for manufacturing an aluminum clad material of the present invention is a method for manufacturing an aluminum clad material in which a laminate of two or more overlapping aluminum or aluminum alloy materials is joined in a joining process using hot rolling rolls, and the joined body obtained by this joining is subsequently rolled using the hot rolling rolls to manufacture an aluminum clad material in which at least a core material and a skin material are joined, and the difference in the elongation in the rolling direction of the core material and the skin material that occurs in the joining process is predicted by numerical analysis that predicts the unknown actual hot rolling elongation, and assuming that the length of the core material does not change in the rolling direction, and when joining, the length of the skin material is determined in advance based on the length of the core material in the rolling direction so that the length of the skin material in the rolling direction after the joining process is approximately equal to the length of the core material in the rolling direction, a correction coefficient that can be approximated for the known actual hot rolling elongation is obtained using the predicted elongation value obtained by the same method as the numerical analysis, and when manufacturing the aluminum clad material, the rolling direction length of the skin material before being subjected to the joining process is determined based on the predicted elongation value of the skin material obtained by the same method as the numerical analysis based on the correction coefficient.

(2) In the manufacturing method of the aluminum clad material according to the present invention, when determining the correction coefficient α so that the predicted elongation value determined by the numerical analysis can approximate the measured elongation value determined by the actual hot rolling, it is preferable to determine the correction coefficient α by optimization calculation so that the sum of squared residuals expressed by the following equation (1) is minimized so that the relationship between the actual elongation values of the core material and the skin material determined by the actual hot rolling approximates the relationship between the predicted elongation value and the actual elongation values of the core material and the skin material determined by the actual hot rolling.

In equation (1), i represents the sample number, xi represents the predicted value for the corresponding sample number, and yi represents the actual value for the corresponding sample number.

(3) In the manufacturing method of the aluminum clad material according to the present invention, when the correction coefficients a, b, and c are determined so that the predicted elongation value determined by the numerical analysis can approximate the measured elongation value determined by the actual hot rolling, it is preferable to determine the correction coefficients a, b, and c by optimization calculation so that the sum of squared residuals expressed by the following equation (2) is minimized, so that the relationship between the actual elongation values of the core material and the skin material determined by the actual hot rolling approximates the relationship between the predicted elongation value and the actual elongation values of the core material and the skin material determined by the actual hot rolling.

In equation (2), i represents the sample number, xi represents the predicted value for the corresponding sample number, yi represents the actual value for the corresponding sample number, and a, b, and c represent correction coefficients.

The present invention provides a method for manufacturing aluminum clad material that can suppress the phenomenon of the skin material elongating and protruding from the end of the core material, and prevent the skin material from falling off at the end of the core material, even when the clad material is manufactured by hot rolling using skin materials that have different elongation and change length, while the core material has a constant length in the rolling direction during the joining process before hot rolling.
In addition, the present invention provides a method for manufacturing aluminum clad material that can efficiently produce aluminum clad material with an appropriate clad ratio by suppressing the amount of skin material protruding from the end of the core material, thereby reducing the amount of skin material that is cut off.

FIG. 2 is an explanatory diagram showing an example of a rolling start state employed when carrying out the method for manufacturing a clad material according to the present invention. FIG. 10 is an explanatory diagram showing another example of the rolling start state employed when carrying out the method for manufacturing a clad material according to the present invention. 1 is a side view showing an example of a clad material manufactured by a clad material manufacturing method according to the present invention. This is a graph in which the vertical axis shows the actual values of elongation of the skin material when the clad material is hot rolled, and the horizontal axis shows the predicted values of elongation calculated by numerical analysis to correspond to the actual values shown on the vertical axis. FIG. 10 is an explanatory diagram showing that when a correction coefficient α is calculated, there are multiple samples, and the values of the predicted value x, actual value y, predicted value x×α, and {actual value y−(predicted value x×α)} ² are calculated, respectively. When calculating the correction coefficient α for samples A to F shown in Table 1, multiple samples were used as shown in Table 4, and the predicted value x, actual value y, predicted value x × α, and {actual value y - (predicted value x × α)} ² values were calculated. This is an explanatory diagram showing the results.

Hereinafter, an example of an embodiment of the present invention will be described in detail with reference to the accompanying drawings. Note that the drawings used in the following description may show characteristic portions enlarged for convenience in order to make the characteristics easier to understand.
FIG. 1 shows an example of a material arrangement employed in the first rolling pass when implementing the clad material manufacturing method of the first embodiment of the present invention. A clad material 6 including a core material 3 and skin materials 4, 5 is arranged between hot rolling rolls (work rolls) 1, 2 spaced apart from one another. FIG. 1 shows a side view of the hot rolling rolls 1, 2, with the conveyance direction of the clad material 6 being rolled by the hot rolling rolls 1, 2 extending in the left-right direction. The core material 3 is obtained by cutting a slab ingot made of aluminum or an aluminum alloy to a specified thickness and shape. The skin materials 4, 5 are obtained by hot rolling a slab ingot to a specified thickness and cutting it to an appropriate length. Examples of materials include brazing filler metal and sacrificial anode material.

In this embodiment, it is assumed that the high-temperature deformation resistance (MPa) of the core material 3 is greater than the high-temperature deformation resistance (MPa) of the cladding materials 4 and 5, and that the high-temperature deformation resistance (MPa) of the cladding material 4 is less than the high-temperature deformation resistance (MPa) of the cladding material 5. Based on these assumptions, a case will be described in which a cladding material is formed by hot rolling.
In the state shown in Fig. 1, the core material 3 and the skin materials 4, 5 are installed so that they can be transported along transport paths (not shown) that are arranged before and after (left and right direction in Fig. 1) the hot rolling rolls 1, 2. Fig. 1 shows the state before the start of rolling, in which the core material 3 and the skin materials 4, 5 are stacked and unbonded before being bonded.

The arrangement shown in FIG. 1 (starting pattern 1) shows the case where the hot rolling rolls 1 and 2 are arranged near the center in the length direction of the clad material 6 (near the center in the length direction of the core material 3).
In the case of starting pattern 1 shown in Figure 1, the distance from the center of the core material 3 in the longitudinal direction to the left end of the core material 3 in the longitudinal direction is assumed to be a, and the distance from the center of the core material 3 in the longitudinal direction to the right end of the core material 3 in the longitudinal direction is assumed to be b. In the following explanation, it is assumed that the core material 3 is thicker than the skin materials 4 and 5, and that the skin materials 4 and 5 have the same thickness. It is assumed that the skin materials 4 and 5 all have smaller high-temperature deformation resistance at hot rolling temperatures than the core material 3, and that the high-temperature deformation resistance of the skin material 4 is smaller than that of the skin material 5.

In this embodiment, the high-temperature deformation resistance of the core material 3 is greater than that of the skin materials 4 and 5. Therefore, during the joining process, which is the initial stage of hot rolling, the elongation of the skin materials 4 and 5 is greater than that of the core material 3. Furthermore, if the core material 3 is stronger than the skin materials 4 and 5, the core material 3 undergoes rolling in an unjoined state during the joining process, resulting in little deformation. Therefore, to prevent the skin materials 4 and 5 from falling off at their ends due to elongation during subsequent hot rolling, the skin materials 4 and 5 are pre-set to be shorter than the core material 3 by a predetermined length at the clad material 6 stage, as shown in Figure 1. In Figure 1, hot rolling rolls 1 and 2 are positioned near the center of the clad material 6 in the longitudinal direction. The distance between the hot rolling rolls 1 and 2 is narrowed from the state shown in Figure 1, and the clad material 6 is hot rolled while being fed leftward or rightward along its longitudinal direction. For this reason, in Figure 1, the core material 3 has a section at the left end with length a' where the skin materials 4 and 5 are not present, and a section at the right end with length b' where the skin materials 4 and 5 are not present.

The clad material 6 is fed leftward along the lengthwise direction relative to the hot rolling rolls 1 and 2, and when the right half of the clad material 6 in the lengthwise direction is hot rolled, the clad material 6 is then fed rightward along the lengthwise direction, and the left half of the clad material 6 in the lengthwise direction is hot rolled.
The clad material 6 is fed to the right along the lengthwise direction relative to the hot rolling rolls 1 and 2, and when the left half of the clad material 6 in the lengthwise direction is hot rolled, the clad material 6 is then fed to the left along the lengthwise direction, and the right half of the clad material 6 in the lengthwise direction is hot rolled.

The first pass of hot rolling described above performs the joining process over the entire length of the clad material 6, and the skin materials 4 and 5 are tightly attached to the core material 3. After this, hot rolling is performed the required number of times to obtain an aluminum clad material 10 in which the skin materials 8 and 9 are integrated with the core material 7, as shown in Figure 3.
In an aluminum clad material 10 in which skin materials 8 and 9 are integrated with a core material 7, the target clad ratio can be, for example, 20%, 70%, 10%, etc., where A is the thickness of skin material 8, B is the thickness of core material 7, and C is the thickness of skin material 9. The target clad ratio can also be, for example, 30%, 60%, 10%, etc.
When the first pass of hot rolling is started from the start pattern 1 shown in FIG. 1, it is preferable that the relationship a:b≈a':b' is satisfied.

When hot rolling is performed according to the starting pattern shown in FIG. 1, numerical analysis is applied to predict the extent to which the skin materials 4 and 5 will stretch in the longitudinal direction, as will be described below, and the predicted value is calculated.
First, as an initial prerequisite, the lengths in the rolling direction of the core material and skin material before being subjected to the joining process are determined so that the length in the rolling direction of the skin material 8 or skin material 9 after the joining process and the length in the rolling direction of the core material 7 after the joining process are approximately equal.
Here, "almost equal" refers to a difference in the range in which the skin material does not fall off during hot rolling following the joining process, the amount of skin material discarded after hot rolling is reduced, and an appropriate cladding ratio is obtained. It is preferable that the difference between the rolling direction length of the skin material after the joining process and the rolling direction length of the core material after the joining process be 100 mm or less in order to prevent the skin material from falling off and obtain an appropriate cladding ratio. The difference is more preferably 50 mm or less, which further reduces the amount of skin material discarded after hot rolling, but it is clear that the closer the difference is to 0 mm, the better.
The reason why the difference is set to 100 mm or less is that when an aluminum clad material is manufactured in actual hot rolling equipment, if the skin materials 8, 9 expand relative to the core material 7 and protrude from the end face of the core material 7, a protruding length of 100 mm or less is a length that will not cause the skin materials 8, 9 to fall off in the conveyance path. In addition, since it is possible that the skin materials 8, 9 will be shorter than the core material 7 after hot rolling, in that case, it is assumed that the skin materials 8, 9 will be shorter than the length of the core material 7, and the difference between them will be 100 mm or less.
Considering both the case where the skin materials 8, 9 extend relative to the core material 7 and protrude from the end face of the core material 7, and the case where the skin materials 8, 9 are shorter than the length of the core material 7 and the difference is 100 mm or less, "almost equal" means that the difference between the length of the skin material in the rolling direction after the joining process and the length of the core material in the rolling direction after the joining process is within ±100 mm.

In this embodiment, the elongation of the skin material obtained by actual hot rolling using the core material and the skin material is calculated for the predicted elongation of the skin material obtained by numerical analysis, and a correction coefficient α is calculated so that the predicted elongation value obtained by numerical analysis can approximate the measured elongation value obtained by actual hot rolling.
Next, when manufacturing the aluminum clad material 10, the corrected predicted length of the skin materials 8, 9 is calculated by multiplying the predicted elongation amount by the correction coefficient α. By calculating the corrected predicted lengths of the skin materials 8, 9, it is possible to grasp the amount of elongation that each of them will experience due to hot rolling. By calculating the predicted length of the skin material 8 and subtracting it from the initial length of the core material 7, the predicted elongation amount of the skin material 8 relative to the core material 7 can be determined. Details of the method for calculating the correction coefficient α will be described later.

Therefore, as shown in Figure 1, it is advisable to use a skin material 4 that is shorter than the core material 3 in the stage before hot rolling to compensate for the amount of elongation that occurs in the subsequent joining process. In the case of Figure 1, a' + b' corresponds to the predicted amount of elongation.
Furthermore, if the skin material 4 and the skin material 5 are made of the same aluminum alloy and have the same thickness, the skin material 5 may be set to the same length as the skin material 4. However, if different aluminum alloys are used or if the thicknesses are different, the predicted length of the skin material 5 relative to the core material 3 can be obtained using the same method as described above, and the length of the skin material 5 before hot rolling can be determined.

In this embodiment, as an example, when calculating the predicted elongation amount in the joining process using numerical analysis (in the case of analytical hot rolling), the initial total plate thickness can be set to 450 to 650 mm, the cladding ratio of the skin materials 8 and 9 to the core material 7 can be set to a range of 5 to 30%, and the length of the initial core material can be set to a range of 3000 to 5000 mm. It is preferable to make the initial lengths of the skin materials 8 and 9 and the core material 7 the same. Furthermore, as an example of an analytical condition, the reduction amount is set to 20 mm, so that the analytical conditions match those of the actual rolling.

The method of numerical analysis is not particularly limited, but general-purpose nonlinear structural analysis software using the finite element method is preferred because it allows simple calculations for various rolled materials.
In this embodiment, as an example of numerical analysis, elastic-plastic finite element method software for structural analysis (LS-DYNA, Ver. R10.2.0, manufactured by Livermore Software Technology Corporation (LSTC)) can be used. The numerical analysis method used here may be elementary analysis such as the slab method.
Furthermore, when using the elastic-plastic finite element method for structural analysis, it is necessary to consider plastic deformation of the core material 3 and the skin materials 4 and 5, so that a rigid-plastic or elastic-plastic body can be used. Note that, since the amount of elastic deformation in the amount of elongation of the skin material is small and can be ignored, it is more preferable to use a rigid-plastic body from the viewpoint of calculation time. In this embodiment and in the examples described later, an elastic-plastic body is used.
The hot rolling rolls 1 and 2 are assumed to be rigid bodies that do not deform at all, and the high-temperature deformation resistance curves of the core material 3 and the skin materials 4 and 5 are determined by hot compression tests at 480°C and a strain rate of 1/s for each of the core material 3 and the skin materials 4 and 5.
Furthermore, the element type in the elastic-plastic finite element method for structural analysis software is not particularly limited as long as it is a so-called solid element that can consider triaxial stress. In plate rolling, the strain in the plate width direction is negligibly small compared to the strain in the longitudinal and thickness directions, so it is also possible to use a so-called plane strain element that does not consider strain in the plate width direction. In this embodiment, among the solid elements implemented in the above-mentioned LS-DYNA, a fully integrated S/R solid element was used, and the displacement of the core material 3 and the skin materials 4 and 5 in the plate width direction was constrained, resulting in a plane strain state that ignored deformation in the plate width direction.

Next, it is assumed that the core material 3 and the skin materials 4, 5 do not slip relative to each other in the initial biting region where the hot rolling rolls 1, 2 first bite the core material 3 and the skin materials 4, 5, and a contact analysis of the core material 3 and the skin materials 4, 5 in other regions except for the initial biting region is calculated using a penalty method assuming Coulomb friction, and the friction coefficient used in the penalty method can be obtained by a friction behavior evaluation test at the hot rolling temperature.
1 and 2, the actual rolling method assumed in the present invention is to perform the first pass of rolling from a midpoint of the clad material (slab) 6, and to roll the skin materials 4 and 5 while making them shorter than the core material 3. In contrast, in the skin material elongation prediction analysis method of the embodiment, in order to simplify the prediction analysis, the joining conditions are set to roll in one pass over the entire length of the clad material (slab) 6 from one side to the other side of the clad material 6, and the end on the side that is first bitten is set as the initial biting region, and the initial lengths of the skin materials 4 and 5 and the core material 3 are made the same, thus simplifying the numerical analysis.

In this example, Coulomb friction was assumed as the contact condition at the interface between the core material 3 and the skin materials 4, 5, but this is not limited to Coulomb friction, and definitions such as shear friction can also be used. For Coulomb friction, a value of approximately 0.2 can be used as the Coulomb friction coefficient, and for shear friction, a value of approximately 0.9 can be used as the shear friction coefficient. The predicted elongation value of the skin material changes depending on the value of the friction coefficient, but since a correction coefficient α is ultimately used to make corrections so that the skin material elongation in actual rolling closely approximates the predicted elongation result, it is not necessary here to strictly match the friction coefficient with the actual phenomenon.
In this embodiment, CONTACT_AUTOMATIC_SURFACE_TO_SURFACE implemented in the above-mentioned LS-DYNA was selected and used as the contact condition for which the Coulomb friction coefficient can be set.

Next, a correction coefficient α is calculated so that the predicted elongation value calculated by the nonlinear structural analysis software can approximate the actually measured elongation value calculated by the hot rolling test.
The correction coefficient α that minimizes the sum of squared residuals expressed by the following equation (1) can be obtained by optimization calculation so that the actual elongation amounts of the core material 3 and the skin materials 4 and 5 in the hot rolling test and the predicted elongation amounts approximate each other.

In the formula (1), i represents the sample number, xi represents the predicted value of the corresponding sample number, and yi represents the actual value of the corresponding sample number.
In addition, in formula (1), the term (xi × α) means that correction is performed by first-order integration, but the form of the correction formula is not limited to the correction by first-order integration described above. For example, second-order integration or shifting using an intercept may also be performed.
For example, in the above formula for the sum of squared residuals, the following quadratic polynomial (2) can be used instead of formula (1): Other polynomials, exponential formulas, logarithmic formulas, etc. may also be used as appropriate for the above formula for the sum of squared residuals.

In equation (2), i represents the sample number, xi represents the predicted value for the corresponding sample number, yi represents the actual value for the corresponding sample number, and a, b, and c represent correction coefficients.

4 is a graph showing the actual values of elongation of the skin material when hot rolling a clad material using an actual hot rolling mill on the vertical axis, and the predicted values of elongation calculated by the numerical analysis described above so as to correspond to the actual values shown on the vertical axis on the horizontal axis. The diagonal solid line in Fig. 4 is a straight line where the values on the vertical axis and the values on the horizontal axis are 1:1 (i.e., the predicted values and the actual values are the same), and the purpose is to correct values that deviate from the solid line in Fig. 4.
When the relationship shown in Figure 4 is graphed, the actual elongation value when a clad material is manufactured by hot rolling using an actual hot rolling mill differs from the predicted elongation value obtained by the above-mentioned nonlinear structural analysis software. In the example of Figure 4, the predicted elongation value by the numerical analysis before correction is larger than the actual actual elongation value, and the predicted value before correction overestimates the elongation of the clad material compared to the actual phenomenon. Therefore, a correction coefficient α is calculated to minimize the above-mentioned residual sum of squares so that the two values approximate each other.

The method for determining the correction coefficient α will be described more specifically with reference to FIG.
For multiple clad materials A to G, the actual elongation value y of the skin material during rolling is measured. For these clad materials, elongation prediction is performed by numerical analysis to obtain a predicted elongation value x. Here, the predicted elongation value x is not corrected to approximate the actual measured elongation value, and therefore the elongation of the skin material cannot be accurately predicted, as shown in Figure 4, so a correction coefficient α must be determined.
Therefore, assuming that the correction coefficient α is 1, the value x×α obtained by multiplying the predicted value x by the correction coefficient α is calculated for each of the clad materials A to G. Here, since α is assumed to be 1, the value x×α coincides with the predicted elongation value x. Next, the square of the difference between the actual elongation value y and the predicted value x×α obtained by multiplying the correction coefficient α is calculated for each of the clad materials A to G, and the sum of the values of {y-(x×α)} ² calculated for each of the clad materials A to G is calculated.
Since the sum of squared residuals calculated in this way increases or decreases depending on the value of α, by varying α and calculating the sum of squared residuals, it is possible to determine the value of α that minimizes the sum of squared residuals. The predicted value x × α of the skin elongation amount after correction, multiplied by the correction coefficient α obtained by this method, closely approximates the actual skin elongation amount y.

Based on the above-mentioned numerical analysis, the predicted elongation (mm) of each of the core material 3 and the skin materials 4 and 5 is calculated, and then the lengths of the skin materials 4 and 5 are set to be short in advance so as to cancel the calculated results of the predicted elongation.
For example, assuming that core material 3 hardly stretches and the length of the core material is 3500 mm, if skin material 4 is predicted to stretch about 350 mm and skin material 5 is predicted to stretch 380 mm, the length of the core material is set to 3500 mm, skin material 4 is set to 3150 mm, and skin material 5 is set to 3120 mm.
Note that, based on the calculation results of the predicted elongation amount, the lengths of the skin materials 4 and 5 can be set short in advance so that the core material 7, skin material 8, and skin material 9 all have the same length after hot rolling. However, in this embodiment, it is also acceptable if the skin materials 8 and 9 are slightly longer or shorter than the core material 7 after hot rolling. In the hot rolling subsequent to the joining process, the difference between the rolling direction length of the skin material after the joining process and the rolling direction length of the core material after the joining process is 100 mm or less, which is a range within which no skin material falls off and within which the amount of skin material truncated after hot rolling can be reduced and an appropriate cladding ratio can be obtained. Therefore, as a rough guideline in this embodiment, if the difference in the rolling direction length of the skin materials 8 and 9 from the rolling direction length of the core material 7 after the joining process is 100 mm or less, it is determined that the joining result is good.
More preferably, the difference is 50 mm or less, which can further reduce the amount of skin material that is cut off after hot rolling, but it is clear that the closer the difference is to 0 mm, the more preferable it is.
If the skin materials 8, 9 are shorter than the core material 7 by more than 100 mm after hot rolling, a large amount of unclad waste will be generated, resulting in a decrease in yield. If the skin materials 8, 9 are longer than the core material 7 by more than 100 mm after hot rolling, the protruding parts of the skin materials 8, 9 may fall off into the transport path, hindering productivity.

As described above, the amount of elongation of the skin materials 4, 5 is calculated and predicted by numerical analysis, and the lengths of the skin materials 4, 5 are adjusted based on the elongation of the skin materials 4, 5 so that the lengths of the core material 3 and the skin materials 4, 5 are approximately the same after the first pass of hot rolling, or so that the skin materials 8, 9 after rolling are slightly shorter or slightly longer than the core material 7. This first pass of hot rolling allows the skin materials 4, 5 to be tightly attached to the core material 3.
Even if there is some error in the first pass of hot rolling, the skin material is shortened in advance according to the prediction made by the above-mentioned numerical analysis, so that the difference in length between the core material 3 and the skin materials 4, 5 after hot rolling can be suppressed to be shorter than before.
Therefore, even if the skin materials 4, 5 protrude from the longitudinal ends of the core material 3, the amount of protrusion can be significantly reduced compared to conventional methods, preventing the ends of the skin materials from falling off. Furthermore, even if the skin materials 4, 5 become shorter than the core material 3 after hot rolling, the amount of shortening can be reduced, thereby reducing the amount of core material 3 that is cut off and wasted after hot rolling.

After the first pass of hot rolling, the rolled material is subjected to the required number of hot rolling passes from the second pass onwards to obtain the clad material 10 having the structure shown in FIG. 3 and the desired thickness.
Furthermore, if the skin materials 4, 5 are brought into close contact with the core material 3 by the first pass of hot rolling, the skin materials 4, 5 can be hot rolled in the second pass and subsequent passes of hot rolling with almost no sliding (shifting) at the interface with the core material in their longitudinal direction, and a clad material with the desired clad ratio can be obtained.

If the high-temperature deformation resistance of the core material 3 is greater than that of the skin materials 4 and 5, it can be assumed that the core material 3 will hardly elongate during the joining process. When comparing the skin materials 4 and 5, the high-temperature deformation resistance of the skin material 5 may be greater or less than that of the skin material 4.
When comparing cladding materials 4 and 5, if the high-temperature deformation resistance of cladding material 5 is greater than that of cladding material 4, the amount of elongation of cladding materials 4 and 5 will change depending on the magnitude of this value. In the above-mentioned numerical analysis, the value of the predicted amount of elongation changes depending on the degree to which the high-temperature deformation resistance of cladding material 5 is greater than that of cladding material 4.

Since the clad ratios of the clad materials to be manufactured vary, the amount of elongation of each of the clad materials 4 and 5 will differ depending on the clad ratios of the clad materials 4 and 5, respectively.
In the above-mentioned numerical analysis, this relationship is taken into consideration and the amount of elongation can be predicted with high accuracy.

By starting the hot rolling process from the start pattern 1 shown in FIG. 1, it is possible to perform rolling while avoiding large misalignment that occurs when the work rolls 1 and 2 bite the skin materials 4 and 5 and start hot rolling.
Furthermore, when hot rolling is controlled by the load applied to the clad material 6 from the work rolls 1 and 2, rolling can be controlled by accurate load.
As a result, an aluminum clad material 10 can be obtained as a final product in which the skin materials 8, 9 are securely adhered to the core material 7 as shown in Figure 3. In the hot rolling following the joining process, the amount of protrusion of the skin materials 8, 9 from the longitudinal ends of the core material 7 is reduced, and the clad material 10 can be manufactured with the desired clad ratio while preventing the skin materials 8, 9 from falling off when they protrude from the core material 7.

FIG. 2 shows another example of a material arrangement (starting pattern 2) employed in the first rolling pass when carrying out the clad material manufacturing method of the first embodiment of the present invention, in which a clad material 6 comprising a core material 3 and skin materials 4, 5 is arranged between work rolls 1, 2 arranged spaced apart from each other above and below.
In this example of material arrangement, it is assumed that the high-temperature deformation resistance (MPa) of the core material 3 is greater than the high-temperature deformation resistance (MPa) of the skin materials 4 and 5, and that the high-temperature deformation resistance (MPa) of the skin material 4 is less than the high-temperature deformation resistance (MPa) of the skin material 5. Based on these assumptions, the conditions for forming the clad material by rolling are the same as those in the previous example.

The starting pattern 2 shown in FIG. 2 indicates that the work rolls 1 and 2 are arranged on the longitudinal end side of the clad material 6 (the longitudinal end side of the core material 3).
2, the distance from the left end of the core material 3 in the longitudinal direction to the position where the work rolls 1 and 2 are arranged is assumed to be a, and the distance from the right end of the core material 3 in the longitudinal direction to the position where the work rolls 1 and 2 are arranged is assumed to be b. In the following explanation, the relationship between the thickness of the core material 3 and the skin materials 4 and 5, and the relationship between the high-temperature deformation resistance of the core material 3 and the skin materials 4 and 5 are also assumed to be the same as in the previous example.

To prevent the skin materials 4, 5 from falling off at the ends due to elongation during hot rolling, the skin materials 4, 5 are made shorter than the core material 3 in advance at the clad material 6 stage, as shown in Figure 2. In Figure 2, work rolls 1, 2 are positioned on the left end side of the length of the clad material 6, and hot rolling is performed while narrowing the gap between the work rolls 1, 2 from the state shown in Figure 2 and feeding the clad material 6 leftward along the length. Therefore, in the core material 3 in Figure 2, there is a section at the left end where the skin materials 4, 5 are not present, indicated by length a', and a section at the right end where the skin materials 4, 5 are not present, indicated by length b'.

The clad material 6 is fed leftward along its length relative to the work rolls 1 and 2, and when most of the right side of the clad material 6 in the length direction has been hot rolled, the clad material 6 is then fed rightward along its length direction, and the left end side of the clad material 6 in the length direction is hot rolled.
The above-described first pass of hot rolling is performed over the entire length of the core material 3 to adhere the skin materials 4 and 5 to the core material 3, and then hot rolling is performed the required number of times to obtain a clad material 10 of the desired thickness in which the skin materials 8 and 9 are integrated with the core material 7, as shown in Figure 3.

Even with start pattern 2 shown in Figure 2, when work rolls 1 and 2 bite into the clad material 4 and 5 and begin hot rolling, rolling can be performed while avoiding the large misalignment that occurs when the clad material 4 and 5 bite into each other. Therefore, when hot rolling is controlled by applying a load to the clad material 6 from the work rolls 1 and 2, accurate rolling control can be achieved while avoiding misalignment.

Even with starting pattern 2 shown in Figure 2, just as when hot rolling is started from starting pattern 1 shown in Figure 1, hot rolling can be performed with reduced elongation of the skin materials 4 and 5, preventing the end parts of the skin materials 4 and 5 from falling off and obtaining a clad material 10 with the desired clad ratio.

In the manufacturing method of the aluminum clad material according to the embodiment described above, under the prerequisites that the initial total plate thickness is set to 450 to 650 mm, the skin material clad ratio is set to the range of 5 to 30%, and the length of the initial core material is set to 3000 to 5000 mm, elastic-plastic finite element method software for structural analysis can be used as the nonlinear structural analysis software.
In addition, in elastic-plastic finite element method software for structural analysis, the core and skin materials are assumed to be elastic-plastic bodies, and the rolling rolls are assumed to be rigid bodies that do not deform at all. The high-temperature deformation resistance curves of the core and skin materials can be calculated by hot compression tests at 480°C for each material. Fully integrated S/R solid elements can be used as the element type, and six or more integration points can be set in the thickness direction of the core and skin materials. A plane strain state with displacement constraints in the width direction can be assumed as a constraint condition. However, when skin materials are clad on both sides of the core material, the cladding ratio of one skin material = thickness of one skin material / (thickness of one skin material + thickness of core material + thickness of the other skin material).
The element type is not limited to the fully integrated S/R solid element described above, but shell elements may also be used, and the numerical analysis is not limited to the elastic-plastic finite element method for structural analysis described above, but may be performed using elementary methods or slab methods.

Furthermore, when calculating the predicted elongation amount using the numerical analysis described above, it is assumed that the core and skin materials do not shift relative to each other in the rolling direction in the initial bite region where the rolling rolls first bite the core and skin materials, and the contact force between the core and skin materials in regions other than the initial bite region is calculated using the penalty method, and it is preferable to determine the friction coefficient used in the penalty method through a friction behavior evaluation test at the hot rolling temperature.

The following analysis was conducted on each of the clad materials Nos. A to E shown in Table 1, which have a three-layer structure consisting of a skin material (brazing filler metal) A made of an Al-Si alloy, a core material B made of an Al-Zn alloy, and a skin material (sacrificial material layer) C made of an Al-Mn alloy.

It was assumed that hot rolling would start from starting pattern 1 shown in Figure 1, and that skin material A, core material B, and skin material C had the high-temperature deformation resistance (MPa) and initial longitudinal dimensions (mm) shown in Table 1 below. When the target cladding ratios (%) of skin material A, core material B, and skin material C were as shown in Table 1, the predicted elongation amounts (mm) based on the finite element method described in the embodiment are shown in Table 1 below. The total thickness of the clad material including skin material A, core material B, and skin material C was set to 600 mm, and the reduction by the work rolls was set to 20 mm.
The prediction based on the finite element method is based on the calculation method of the finite element method based on the parameter settings described in detail above in the embodiment.

Table 1 also shows the calculation results for specimen No. B, which uses core material D made of an Al-Zn alloy, and the predicted elongation results for specimen No. C, which has a different initial length from specimen No. A.
Table 1 shows the predicted elongation amount for sample No. D, which has a three-layer structure that uses skin material (brazing material) E made of an Al-Si alloy, skin material (sacrificial material) F made of an Al-Mn alloy, and core material B made of an Al-Zn alloy.
Table 1 shows the predicted elongation amount for sample No. E, which has a three-layer structure using skin material A, core material B, and skin material C, with the target cladding ratio set to 30%:60%:10%.

As shown in Table 1, for sample No. A, the initial longitudinal dimensions are set to predetermined values of skin material A: core material B: skin material C = 3500 mm: 3500 mm: 3500 mm, and hot rolling is performed as follows.
As shown in Table 1, when the predicted elongation amount was calculated for sample No. A using the finite element method described in the embodiment, it was possible to calculate that skin material A: core material B: skin material C = 225 mm: 0 mm: 240 mm. It can be estimated from this calculation result that skin materials A and C will elongate more than core material B by approximately 225 to 240 mm, and that skin materials A and C will protrude from the end of core material B by approximately 125 to 140 mm.

Therefore, in this invention, the objective is to absorb the predicted elongation amounts of 225 mm and 240 mm shown in Table 1, calculated based on the above-mentioned finite element method, and the initial longitudinal dimensions were set to skin A: core B: skin C = 3275 mm: 3500 mm: 3260 mm, as shown in sample No. A in Table 2. The initial cladding ratios were set to skin A: core B: skin C = 21.2%: 68.0%: 10.8%, and the total thickness was set to 600 mm.
The initial cladding ratio is adjusted so that the cladding ratio does not exceed the target value of ±1.5% when the predicted elongation occurs. Furthermore, since the volume does not change due to the law of conservation of mass, the reduction in the cladding ratio is calculated from the elongation. It is assumed that the cladding ratio is uniform and does not vary.
The sample No. A was subjected to hot rolling under the condition of a reduction of 20 mm in the first pass of hot rolling. The results are shown in Table 2 below.

As shown in Table 2, the actual elongation was cladding material A: core material B: cladding material C = 215 mm: 0 mm: 255 mm. As a result, the lengths after the first pass of hot rolling were cladding material A: core material B: cladding material C = 3,490 mm, 3,500 mm, and 3,515 mm, respectively, with cladding material A being 10 mm shorter than core material B and cladding material C being 15 mm longer than core material B.
This means that even though the end of skin material C protruded from the end of core material B, the protrusion was only 15 mm, and skin material A was 10 mm shorter, so no skin material fell off at the end of the core material.
Furthermore, the final cladding ratios of the product were skin A: core B: skin C = 19.3%, 71.6%, and 9.2%, respectively, which were within the range of ±1.5% of the target cladding ratio, and a good cladding ratio was obtained.

In contrast, when the first pass of hot rolling was performed under the conditions of skin material A: core material B: skin material C = 3400 mm, 3500 mm, and 3400 mm as shown in sample A' in Table 3 below, the actual elongation was 230 and 240 mm, the protrusion amount was 130 to 140 mm, skin material falling occurred, and the cladding ratio exceeded ±1.5% of the target cladding ratio.
The final clad ratio (%) of the product shown in each table means the clad ratio of the product after the first pass of hot rolling is followed by conventional hot rolling so that the final total thickness after hot rolling is in the range of 5 to 30 mm.

Table 4 and Fig. 6 show the correction coefficient α calculated based on the predicted value x, actual value y, predicted value x × α, and {actual value y - (predicted value x × α)} ² values from multiple samples when analyzing samples A to E, as shown in Fig. 5. In this example, the correction coefficient α is applied to all of samples A to E.
The relationship between the predicted values and the actual values shown in Table 4 can be displayed in a graph in Fig. 6, with the vertical axis (actual values) and the horizontal axis (predicted values, predicted values x α). The actual values indicate values when the samples were rolled using an actual hot rolling mill with a reduction of 20 mm.
The correction coefficient α obtained from the relationship shown in Table 4 and FIG. 6 is 0.6299995, but is determined to be approximately 0.63. As shown as an example in Table 5 below, if the analysis result of sample No. A is 355, multiplying this by 0.63 results in a predicted elongation amount of 225 mm, and if the analysis result of sample No. A is 380, multiplying this by 0.63 results in a predicted elongation amount of 240 mm.

Next, predictions were made for the case where a sample with a high-temperature deformation resistance of 30 MPa was used as core material D, as shown in sample No. B in Table 1. Skin materials A and C are equivalent to sample No. A.
As shown in sample No. B, if the initial longitudinal dimensions are skin A: core D: skin C = 3500 mm, 3500 mm, and 3500 mm, the predicted elongation amounts of skin A and skin C are 225 mm and 240 mm, as shown in Table 1, and therefore, as in the case of sample No. A, skin A and skin C are expected to protrude a considerable amount from the end of core D.
Therefore, as shown in sample No. B in Table 2, when the initial longitudinal dimensions were set to skin A: core D: skin C = 3275 mm: 3500 mm: 3260 mm, the actual elongation was skin A: core D: skin C = 225 mm: 0 mm: 235 mm. In the case of sample No. B in Table 2, as shown in Table 2, no skin shedding occurred and the final cladding rate of the product was also good.

In contrast, when the first pass of hot rolling was performed under the conditions of skin material A: core material D: skin material C = 3400 mm, 3500 mm, and 3400 mm, as shown in sample No. B' in Table 3, the actual elongation of skin material A and skin material C was 210 mm and 275 mm, respectively, and the protrusion amounts were large at 110 mm and 175 mm, resulting in skin shedding, and the cladding ratio also varied by more than ±1.5% from the target cladding ratio.

Next, as shown in sample No. C in Table 1, the high-temperature deformation resistance is equivalent to that of sample No. A in Table 1, but the initial lengths of the sample were set to skin material A:core material B:skin material C = 5000 mm:5000 mm:5000 mm, and predictions were made using the finite element method described in the embodiments.
As shown in sample No. C in Table 1, the predicted elongation values for skin materials A and C are large at 330 mm and 355 mm, respectively, so similar to the case of sample No. A, skin materials A and C will protrude a considerable amount from the end of core material B.
Therefore, as shown in sample No. C in Table 2, when the initial longitudinal dimensions were set to skin A: core B: skin C = 4670 mm: 5000 mm: 4645 mm, the actual elongation was skin A: core B: skin C = 335 mm: 0 mm: 355 mm. In the case of sample No. C in Table 2, as shown in Table 2, no skin shedding occurred and the final cladding rate of the product was also good.

In contrast, when the first pass of hot rolling was performed under the conditions of skin material A: core material B: skin material C = 4800 mm, 5000 mm, and 4800 mm, as shown in sample C' in Table 3, the actual elongation of skin material A and skin material C was 385 mm and 430 mm, and the protrusion amounts were large at 185 mm and 230 mm, resulting in skin shedding and a cladding ratio that exceeded ±1.5% of the target cladding ratio, resulting in variation.

Next, as shown in sample No. D in Table 1, skin material E and skin material F were added to sample No. A in Table 1, and a sample with a large difference in high-temperature deformation resistance was used, and elongation was predicted using the finite element method described in the embodiment under the conditions of the same initial length and the same target cladding ratio as sample No. A. As shown in sample No. D, the predicted elongation value for skin material E was 340 mm, which means that skin material E protrudes a considerable amount from the end of core material B.
Therefore, as shown in sample D in Table 2, when the initial longitudinal dimensions were set to skin E: core B: skin F = 3160 mm: 3500 mm: 3495 mm, the actual elongation was skin E: core B: skin F = 360 mm: 0 mm: 5 mm. In the case of sample No. D, as shown in Table 2, no skin shedding occurred, and the final cladding rate of the product was also good.

In contrast, as shown in sample D' in Table 3, when skin materials E and F, which have a large difference in high-temperature deformation resistance compared to skin materials A and C, were applied to sample No. A in Table 1, and the first pass of hot rolling was performed under the same initial length and target cladding ratio as sample No. A, the actual elongation of skin material D' was large at 410 mm, skin shedding occurred, and the cladding ratio also varied by more than ±1.5% from the target cladding ratio.

Next, as shown in sample No. E in Table 1, a sample in which the target cladding ratio for skin material A and skin material C was changed to 30:60:10 in comparison with sample No. A in Table 1 was applied, and elongation was predicted based on the above-mentioned finite element method on the basis of the same initial length and the same high-temperature deformation resistance as sample No. A.
As shown in sample No. E, the predicted elongation values are 145 mm for skin material A and 260 mm for skin material C, and it is predicted that skin material A will protrude 45 mm and skin material C will protrude 160 mm from the core material B.
Therefore, as shown in sample No. E in Table 2, when the initial longitudinal dimensions were set to skin A: core B: skin C = 3355 mm: 3500 mm: 3240 mm, the actual elongation was skin A: core B: skin C = 125 mm: 0 mm: 275 mm. In the case of sample No. E, as shown in Table 2, no skin shedding occurred and the final cladding rate of the product was also good.

In contrast, as shown in sample E' in Table 3, when a sample with a target cladding ratio of 30:60:10 was applied to sample No. A in Table 1 for skin material A and skin material C, and the first pass of hot rolling was performed under the same initial length conditions as sample A in Table 1, the actual elongation of skin material A was 140 mm and the actual elongation of skin material C was 300 mm, resulting in skin shedding and a cladding ratio variation of more than ±1.5% relative to the target cladding ratio.

As explained above based on Tables 1 to 4, in accordance with the present invention, the initial lengths of the core and the skin materials on both sides thereof are adjusted based on the elongation prediction results obtained by numerical analysis such as the finite element method, taking into account the high-temperature deformation resistance of the material, the cladding ratio, the length of the cladding material, etc. As a result, it has become clear that good elongation prediction can be made and skin material shedding can be prevented in all cases, including when skin materials with different high-temperature deformation resistances are provided on both sides of the core, when the cladding ratio of the skin material to the core is changed, and when the length of the cladding material is changed.
Furthermore, according to the present invention, it was found that a good cladding ratio can be obtained in the final product in any of the following cases: adjusting the strength of the skin material on both sides of the core material using numerical analysis such as the finite element method; adjusting the initial length of the core material and the skin material on both sides of the core material; adjusting the difference in strength of the skin material on both sides of the skin material; and adjusting the cladding ratio of the skin material to the core material.

1, 2...work roll, 3...core material, 4, 5...skin material, 6...clad material, 7...core material, 8, 9...skin material, 10...clad material.

Claims (3)

A method for producing an aluminum clad material in which at least a core material and a skin material are bonded together, the method comprising the steps of: bonding a laminate of two or more aluminum or aluminum alloy materials together in a bonding step using a hot rolling roll; and subsequently rolling the bonded body obtained by this bonding using the hot rolling roll;
When the difference in the elongation amounts of the core material and the skin material in the rolling direction that occurs in the joining process is predicted by a numerical analysis that predicts the unknown actual hot rolling elongation, and it is assumed that the length of the core material in the rolling direction does not change, and the length of the skin material is determined in advance based on the length of the core material in the rolling direction so that the length of the skin material in the rolling direction after the joining process is approximately equal to the length of the core material in the rolling direction,
A correction coefficient that can be approximated by using a predicted value of elongation obtained by the same method as the numerical analysis for a known elongation of actual hot rolling is calculated in advance,
When manufacturing the aluminum clad material, a length in the rolling direction of the skin material before being subjected to the joining step is determined based on a predicted value of elongation of the skin material obtained using the same method as the numerical analysis based on the correction coefficient. 2. The method for manufacturing an aluminum clad material according to claim 1, wherein when a correction coefficient α is determined so that the predicted elongation value determined by the numerical analysis can approximate the measured elongation value determined by the actual hot rolling, the correction coefficient α is determined by optimization calculation so that the sum of squared residuals expressed by the following equation (1) is minimized so that the relationship between the actual elongation values of the core material and the skin material determined by the actual hot rolling and the predicted elongation value approximates each other.
In the formula (1), i represents the sample number, xi represents the predicted value of the corresponding sample number, and yi represents the actual value of the corresponding sample number. 2. The method for manufacturing an aluminum clad material according to claim 1, wherein when the correction coefficients a, b, and c are determined so that the predicted elongation value determined by the numerical analysis can approximate the actually measured elongation value determined by the actual hot rolling, the correction coefficients a, b, and c are determined by optimization calculation so that the sum of squared residuals expressed by the following equation (2) is minimized so that the relationship between the actual elongation values of the core material and the skin material determined by the actual hot rolling and the predicted elongation value approximates each other.
In equation (2), i represents the sample number, xi represents the predicted value of the corresponding sample number, yi represents the actual value of the corresponding sample number, and a, b, and c represent correction coefficients.